From fabiano at fidencio.org Tue Jan 1 16:00:53 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 1 Jan 2013 14:00:53 -0200
Subject: [Libosinfo] [PATCHv3 1/3] debian: Fix .iso URL for Squeeze
Message-ID: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
---
data/oses/debian.xml.in | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/data/oses/debian.xml.in b/data/oses/debian.xml.in
index 67f27a9..d48702f 100644
--- a/data/oses/debian.xml.in
+++ b/data/oses/debian.xml.in
@@ -168,7 +168,7 @@
-
+ http://cdimage.debian.org/debian-cd/6.0.6/i386/iso-dvd/debian-6.0.6-i386-DVD-1.isoDebian 6.0.\d i386 1
@@ -176,21 +176,21 @@
/initrd.img
-
+ http://cdimage.debian.org/debian-cd/6.0.6/amd64/iso-dvd/debian-6.0.6-amd64-DVD-1.isoDebian 6.0.\d amd64 1
-
+ http://live.debian.net/cdimage/release/current/i386/iso-hybrid/debian-live-6.0.6-i386-gnome-desktop.isoDebian squeeze 20111014-08:13LINUX
-
+ http://live.debian.net/cdimage/release/current/amd64/iso-hybrid/debian-live-6.0.6-amd64-gnome-desktop.isoDebian squeeze 20111014-08:11LINUX
--
1.8.0.1
From fabiano at fidencio.org Tue Jan 1 16:00:55 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 1 Jan 2013 14:00:55 -0200
Subject: [Libosinfo] [PATCHv3 3/3] debian: Update volume-id for Squeeze
In-Reply-To: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
References: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1357056055-31236-3-git-send-email-fabiano@fidencio.org>
---
data/oses/debian.xml.in | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/data/oses/debian.xml.in b/data/oses/debian.xml.in
index fc0f6cc..349ba24 100644
--- a/data/oses/debian.xml.in
+++ b/data/oses/debian.xml.in
@@ -187,7 +187,7 @@
http://live.debian.net/cdimage/release/current/i386/iso-hybrid/debian-live-6.0.6-i386-gnome-desktop.iso
- Debian squeeze 20111014-08:13
+ Debian squeeze \d{8}-\d{2}:\d{2}LINUXinstall/vmlinuz
@@ -196,7 +196,7 @@
http://live.debian.net/cdimage/release/current/amd64/iso-hybrid/debian-live-6.0.6-amd64-gnome-desktop.iso
- Debian squeeze 20111014-08:11
+ Debian squeeze \d{8}-\d{2}:\d{2}LINUXinstall/vmlinuz
--
1.8.0.1
From fabiano at fidencio.org Tue Jan 1 16:00:54 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Tue, 1 Jan 2013 14:00:54 -0200
Subject: [Libosinfo] [PATCHv3 2/3] debian: Fix kernel/initrd path
In-Reply-To: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
References: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1357056055-31236-2-git-send-email-fabiano@fidencio.org>
---
data/oses/debian.xml.in | 10 ++++++++--
1 file changed, 8 insertions(+), 2 deletions(-)
diff --git a/data/oses/debian.xml.in b/data/oses/debian.xml.in
index d48702f..fc0f6cc 100644
--- a/data/oses/debian.xml.in
+++ b/data/oses/debian.xml.in
@@ -172,14 +172,16 @@
Debian 6.0.\d i386 1
- /vmlinuz
- /initrd.img
+ install.386/vmlinuz
+ install.386/initrd.gzhttp://cdimage.debian.org/debian-cd/6.0.6/amd64/iso-dvd/debian-6.0.6-amd64-DVD-1.isoDebian 6.0.\d amd64 1
+ install.amd/vmlinuz
+ install.amd/initrd.gz
@@ -188,6 +190,8 @@
Debian squeeze 20111014-08:13LINUX
+ install/vmlinuz
+ install/initrd.gzhttp://live.debian.net/cdimage/release/current/amd64/iso-hybrid/debian-live-6.0.6-amd64-gnome-desktop.iso
@@ -195,6 +199,8 @@
Debian squeeze 20111014-08:11LINUX
+ install/vmlinuz
+ install/initrd.gz
--
1.8.0.1
From fabiano at fidencio.org Wed Jan 2 02:07:20 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Wed, 2 Jan 2013 00:07:20 -0200
Subject: [Libosinfo] [PATCH 1/5] opensuse: Fix indentation
Message-ID: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
---
data/oses/opensuse.xml.in | 12 ++++++------
1 file changed, 6 insertions(+), 6 deletions(-)
diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
index 96ef638..a2451dc 100644
--- a/data/oses/opensuse.xml.in
+++ b/data/oses/opensuse.xml.in
@@ -130,12 +130,12 @@
- http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-i586.iso?mirrorlist
-
- openSUSE-DVD-i586-Build0024
- LINUX
- SUSE LINUX Products GmbH
-
+ http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-i586.iso?mirrorlist
+
+ openSUSE-DVD-i586-Build0024
+ LINUX
+ SUSE LINUX Products GmbH
+ http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-x86_64.iso?mirrorlist
--
1.8.0.1
From fabiano at fidencio.org Wed Jan 2 02:07:21 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Wed, 2 Jan 2013 00:07:21 -0200
Subject: [Libosinfo] [PATCH 2/5] opensuse: Fix kernel/initrd paths
In-Reply-To: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1357092444-4155-2-git-send-email-fabiano@fidencio.org>
---
data/oses/opensuse.xml.in | 16 ++++++++--------
1 file changed, 8 insertions(+), 8 deletions(-)
diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
index a2451dc..b1624e1 100644
--- a/data/oses/opensuse.xml.in
+++ b/data/oses/opensuse.xml.in
@@ -70,8 +70,8 @@
LINUXSUSE LINUX Products GmbH
- vmlinuz
- initrd.img
+ boot/i386/loader/linux
+ boot/i386/loader/initrdhttp://download.opensuse.org/distribution/11.3/iso/openSUSE-11.3-DVD-x86_64.iso?mirrorlist
@@ -80,8 +80,8 @@
LINUXSUSE LINUX Products GmbH
- vmlinuz
- initrd.img
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
@@ -91,8 +91,8 @@
LINUXSUSE LINUX Products GmbH
- vmlinuz
- initrd.img
+ boot/i386/loader/linux
+ boot/i386/loader/initrd
@@ -101,8 +101,8 @@
LINUXSUSE LINUX Products GmbH
- vmlinuz
- initrd.img
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
--
1.8.0.1
From fabiano at fidencio.org Wed Jan 2 02:07:22 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Wed, 2 Jan 2013 00:07:22 -0200
Subject: [Libosinfo] [PATCH 3/5] opensuse: Add missing kernel/initrd
information
In-Reply-To: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1357092444-4155-3-git-send-email-fabiano@fidencio.org>
---
data/oses/opensuse.xml.in | 16 ++++++++++++++++
1 file changed, 16 insertions(+)
diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
index b1624e1..63122b5 100644
--- a/data/oses/opensuse.xml.in
+++ b/data/oses/opensuse.xml.in
@@ -136,6 +136,8 @@
LINUXSUSE LINUX Products GmbH
+ boot/i386/loader/linux
+ boot/i386/loader/initrdhttp://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-x86_64.iso?mirrorlist
@@ -144,6 +146,8 @@
LINUXSUSE LINUX Products GmbH
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
@@ -153,6 +157,8 @@
LINUXSUSE LINUX Products GmbH
+ boot/i386/loader/linux
+ boot/i386/loader/initrd
@@ -161,6 +167,8 @@
LINUXSUSE LINUX Products GmbH
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
@@ -194,6 +202,8 @@
LINUXSUSE LINUX Products GmbH
+ boot/i386/loader/linux
+ boot/i386/loader/initrdhttp://download.opensuse.org/distribution/12.1/iso/openSUSE-12.1-DVD-x86_64.iso?mirrorlist
@@ -202,6 +212,8 @@
LINUXSUSE LINUX Products GmbH
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
@@ -211,6 +223,8 @@
LINUXSUSE LINUX Products GmbH
+ boot/i386/loader/linux
+ boot/i386/loader/initrd
@@ -219,6 +233,8 @@
LINUXSUSE LINUX Products GmbH
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
--
1.8.0.1
From fabiano at fidencio.org Wed Jan 2 02:07:23 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Wed, 2 Jan 2013 00:07:23 -0200
Subject: [Libosinfo] [PATCH 4/5] opensuse: Fix 12.1 resources
In-Reply-To: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1357092444-4155-4-git-send-email-fabiano@fidencio.org>
According with http://en.opensuse.org/Hardware_requirements
---
data/oses/opensuse.xml.in | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
index 63122b5..798d1a5 100644
--- a/data/oses/opensuse.xml.in
+++ b/data/oses/opensuse.xml.in
@@ -241,12 +241,13 @@
5000000001
- 536870912
+ 107374182432212254722400000000
- 1073741824
+ 2147483648
+ 5368709120
--
1.8.0.1
From fabiano at fidencio.org Wed Jan 2 02:07:24 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Wed, 2 Jan 2013 00:07:24 -0200
Subject: [Libosinfo] [PATCH 5/5] opensuse: Add 12.2 information
In-Reply-To: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1357092444-4155-5-git-send-email-fabiano@fidencio.org>
---
data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 67 insertions(+)
diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
index 798d1a5..c8e69b4 100644
--- a/data/oses/opensuse.xml.in
+++ b/data/oses/opensuse.xml.in
@@ -252,4 +252,71 @@
+
+ opensuse12.2
+ <_name>openSUSE 12.2
+ 12.2
+ <_vendor>openSUSE
+ linux
+ opensuse
+
+
+
+
+ http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-i586.iso?mirrorlist
+
+ openSUSE-DVD-i586-Build0039
+ LINUX
+ SUSE LINUX Products GmbH
+
+ boot/i386/loader/linux
+ boot/i386/loader/initrd
+
+
+ http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-x86_64.iso?mirrorlist
+
+ openSUSE-DVD-x86_640039
+ LINUX
+ SUSE LINUX Products GmbH
+
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
+
+
+
+
+
+ openSUSE Live CD.*
+ LINUX
+ SUSE LINUX Products GmbH
+
+ boot/i386/loader/linux
+ boot/i386/loader/initrd
+
+
+
+
+ openSUSE Live CD.*
+ LINUX
+ SUSE LINUX Products GmbH
+
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
+
+
+
+
+ 500000000
+ 1
+ 1073741824
+ 3221225472
+
+
+ 2400000000
+ 2147483648
+ 5368709120
+
+
+
+
--
1.8.0.1
From berrange at redhat.com Wed Jan 2 09:57:41 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 2 Jan 2013 09:57:41 +0000
Subject: [Libosinfo] [PATCH 1/4] rhel: No need for full version in name
In-Reply-To: <1356150057-24165-2-git-send-email-zeeshanak@gnome.org>
References: <1356150057-24165-1-git-send-email-zeeshanak@gnome.org>
<1356150057-24165-2-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130102095741.GD26146@redhat.com>
On Sat, Dec 22, 2012 at 06:20:54AM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> The proper full name 'Red Hat Enterprise Linux X' is pretty long enough
> already, there is really no need to make it longer by specifying full
> versions (e.g 6.3 instead of just 6) in it.
>
> The full version string is available separately for interested apps.
> ---
> data/oses/rhel.xml.in | 80 +++++++++++++++++++++++++--------------------------
> 1 file changed, 40 insertions(+), 40 deletions(-)
>
> diff --git a/data/oses/rhel.xml.in b/data/oses/rhel.xml.in
> index 12d4a02..92ccbbc 100644
> --- a/data/oses/rhel.xml.in
> +++ b/data/oses/rhel.xml.in
> @@ -4,7 +4,7 @@
>
>
> rhel2.1
> - <_name>Red Hat Enterprise Linux 2.1
> + <_name>Red Hat Enterprise Linux 2
This is absolutely not correct. There was never any '2', it was always
called '2.1'.
IMHO the current names are preferrable, since they are unique per
we record.
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From berrange at redhat.com Wed Jan 2 10:01:43 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 2 Jan 2013 10:01:43 +0000
Subject: [Libosinfo] translations
In-Reply-To:
References:
Message-ID: <20130102100143.GE26146@redhat.com>
On Sun, Dec 30, 2012 at 09:55:35PM +0200, Zeeshan Ali (Khattak) wrote:
> Hi everyone,
> Translations are just starting to pour in through transifex. I was
> wondering whats the policy for pushing them to our repo. Shall I just
> push them directly or send here for review first? If answer is latter,
> do I do that for each update or just the first time a new translation
> file is added?
I don't think there's any pointing in submitting for review. The only
people who can sensibly review them are the translators themselves.
I say just periodically pull from transifex to git.
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From zeeshanak at gnome.org Wed Jan 2 14:51:13 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 16:51:13 +0200
Subject: [Libosinfo] translations
In-Reply-To: <20130102100143.GE26146@redhat.com>
References:
<20130102100143.GE26146@redhat.com>
Message-ID:
On Wed, Jan 2, 2013 at 12:01 PM, Daniel P. Berrange wrote:
> On Sun, Dec 30, 2012 at 09:55:35PM +0200, Zeeshan Ali (Khattak) wrote:
>> Hi everyone,
>> Translations are just starting to pour in through transifex. I was
>> wondering whats the policy for pushing them to our repo. Shall I just
>> push them directly or send here for review first? If answer is latter,
>> do I do that for each update or just the first time a new translation
>> file is added?
>
> I don't think there's any pointing in submitting for review. The only
> people who can sensibly review them are the translators themselves.
> I say just periodically pull from transifex to git.
I agree.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 16:45:31 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 18:45:31 +0200
Subject: [Libosinfo] [PATCH] configure: Use LINGUAS file
Message-ID: <1357145131-25603-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Setting ALL_LINGUAS in configure.in is obsolete according to po.m4 in
favor of LINGUAS file. This is the reason why all (as far as I can see)
GNOME modules use LINGUAS file.
Also the rule in configure.ac creates broken Makefile if there is more
than one translation files:
[zeenix at z-laptop libosinfo]$ make
Makefile:166: *** missing separator. Stop.
[zeenix at z-laptop libosinfo]$ head -n 166 Makefile|tail -n 2
ALL_LINGUAS = pl
uk
---
configure.ac | 2 --
po/LINGUAS | 0
2 files changed, 2 deletions(-)
create mode 100644 po/LINGUAS
diff --git a/configure.ac b/configure.ac
index ade0ee7..f4b1044 100644
--- a/configure.ac
+++ b/configure.ac
@@ -21,8 +21,6 @@ AC_DEFINE_UNQUOTED([GETTEXT_PACKAGE],["$GETTEXT_PACKAGE"],[The name of the gette
AM_GNU_GETTEXT([external])
AM_GNU_GETTEXT_VERSION([0.17])
-ALL_LINGUAS=`cd "$srcdir/po" > /dev/null && ls *.po | sed 's+\.po$++'`
-
# Autoconf 2.61a.99 and earlier don't support linking a file only
# in VPATH builds. But since GNUmakefile is for maintainer use
# only, it does not matter if we skip the link with older autoconf.
diff --git a/po/LINGUAS b/po/LINGUAS
new file mode 100644
index 0000000..e69de29
--
1.8.0.2
From berrange at redhat.com Wed Jan 2 16:53:26 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 2 Jan 2013 16:53:26 +0000
Subject: [Libosinfo] [PATCH] configure: Use LINGUAS file
In-Reply-To: <1357145131-25603-1-git-send-email-zeeshanak@gnome.org>
References: <1357145131-25603-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130102165325.GT26146@redhat.com>
On Wed, Jan 02, 2013 at 06:45:31PM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> Setting ALL_LINGUAS in configure.in is obsolete according to po.m4 in
> favor of LINGUAS file. This is the reason why all (as far as I can see)
> GNOME modules use LINGUAS file.
>
> Also the rule in configure.ac creates broken Makefile if there is more
> than one translation files:
>
> [zeenix at z-laptop libosinfo]$ make
> Makefile:166: *** missing separator. Stop.
> [zeenix at z-laptop libosinfo]$ head -n 166 Makefile|tail -n 2
> ALL_LINGUAS = pl
> uk
> ---
> configure.ac | 2 --
> po/LINGUAS | 0
> 2 files changed, 2 deletions(-)
> create mode 100644 po/LINGUAS
>
> diff --git a/configure.ac b/configure.ac
> index ade0ee7..f4b1044 100644
> --- a/configure.ac
> +++ b/configure.ac
> @@ -21,8 +21,6 @@ AC_DEFINE_UNQUOTED([GETTEXT_PACKAGE],["$GETTEXT_PACKAGE"],[The name of the gette
> AM_GNU_GETTEXT([external])
> AM_GNU_GETTEXT_VERSION([0.17])
>
> -ALL_LINGUAS=`cd "$srcdir/po" > /dev/null && ls *.po | sed 's+\.po$++'`
> -
> # Autoconf 2.61a.99 and earlier don't support linking a file only
> # in VPATH builds. But since GNUmakefile is for maintainer use
> # only, it does not matter if we skip the link with older autoconf.
> diff --git a/po/LINGUAS b/po/LINGUAS
> new file mode 100644
> index 0000000..e69de29
ACK
Don't forget to actually populate LINGUAS when you pull in new xlations.
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From zeeshanak at gnome.org Wed Jan 2 16:58:56 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 18:58:56 +0200
Subject: [Libosinfo] [PATCHv3 3/3] debian: Update volume-id for Squeeze
In-Reply-To: <1357056055-31236-3-git-send-email-fabiano@fidencio.org>
References: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
<1357056055-31236-3-git-send-email-fabiano@fidencio.org>
Message-ID:
On Tue, Jan 1, 2013 at 6:00 PM, Fabiano Fid?ncio wrote:
> ---
> data/oses/debian.xml.in | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/data/oses/debian.xml.in b/data/oses/debian.xml.in
> index fc0f6cc..349ba24 100644
> --- a/data/oses/debian.xml.in
> +++ b/data/oses/debian.xml.in
> @@ -187,7 +187,7 @@
>
> http://live.debian.net/cdimage/release/current/i386/iso-hybrid/debian-live-6.0.6-i386-gnome-desktop.iso
>
> - Debian squeeze 20111014-08:13
> + Debian squeeze \d{8}-\d{2}:\d{2}
> LINUX
>
> install/vmlinuz
> @@ -196,7 +196,7 @@
>
> http://live.debian.net/cdimage/release/current/amd64/iso-hybrid/debian-live-6.0.6-amd64-gnome-desktop.iso
>
> - Debian squeeze 20111014-08:11
> + Debian squeeze \d{8}-\d{2}:\d{2}
I thought we agreed that there is no need for date recognition here.
i-e "Debian squeeze" should be both specific and generic enough?
I think thats also what teuf meant on IRC:
[12:17:02] fidencio: even a truncated regex (ie stop before the
date) would be good I think
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 16:59:42 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 18:59:42 +0200
Subject: [Libosinfo] [PATCH] configure: Use LINGUAS file
In-Reply-To: <20130102165325.GT26146@redhat.com>
References: <1357145131-25603-1-git-send-email-zeeshanak@gnome.org>
<20130102165325.GT26146@redhat.com>
Message-ID:
On Wed, Jan 2, 2013 at 6:53 PM, Daniel P. Berrange wrote:
> On Wed, Jan 02, 2013 at 06:45:31PM +0200, Zeeshan Ali (Khattak) wrote:
>> From: "Zeeshan Ali (Khattak)"
>>
>> Setting ALL_LINGUAS in configure.in is obsolete according to po.m4 in
>> favor of LINGUAS file. This is the reason why all (as far as I can see)
>> GNOME modules use LINGUAS file.
>>
>> Also the rule in configure.ac creates broken Makefile if there is more
>> than one translation files:
>>
>> [zeenix at z-laptop libosinfo]$ make
>> Makefile:166: *** missing separator. Stop.
>> [zeenix at z-laptop libosinfo]$ head -n 166 Makefile|tail -n 2
>> ALL_LINGUAS = pl
>> uk
>> ---
>> configure.ac | 2 --
>> po/LINGUAS | 0
>> 2 files changed, 2 deletions(-)
>> create mode 100644 po/LINGUAS
>>
>> diff --git a/configure.ac b/configure.ac
>> index ade0ee7..f4b1044 100644
>> --- a/configure.ac
>> +++ b/configure.ac
>> @@ -21,8 +21,6 @@ AC_DEFINE_UNQUOTED([GETTEXT_PACKAGE],["$GETTEXT_PACKAGE"],[The name of the gette
>> AM_GNU_GETTEXT([external])
>> AM_GNU_GETTEXT_VERSION([0.17])
>>
>> -ALL_LINGUAS=`cd "$srcdir/po" > /dev/null && ls *.po | sed 's+\.po$++'`
>> -
>> # Autoconf 2.61a.99 and earlier don't support linking a file only
>> # in VPATH builds. But since GNUmakefile is for maintainer use
>> # only, it does not matter if we skip the link with older autoconf.
>> diff --git a/po/LINGUAS b/po/LINGUAS
>> new file mode 100644
>> index 0000000..e69de29
>
> ACK
>
> Don't forget to actually populate LINGUAS when you pull in new xlations.
Yeah, already done for the 2 i have merged in local branch. :)
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 17:04:43 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 19:04:43 +0200
Subject: [Libosinfo] [PATCHv3 2/3] debian: Fix kernel/initrd path
In-Reply-To: <1357056055-31236-2-git-send-email-fabiano@fidencio.org>
References: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
<1357056055-31236-2-git-send-email-fabiano@fidencio.org>
Message-ID:
On Tue, Jan 1, 2013 at 6:00 PM, Fabiano Fid?ncio wrote:
> ---
2nd ACK and pushed.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 17:00:25 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 19:00:25 +0200
Subject: [Libosinfo] [PATCHv3 1/3] debian: Fix .iso URL for Squeeze
In-Reply-To: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
References: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
Message-ID:
On Tue, Jan 1, 2013 at 6:00 PM, Fabiano Fid?ncio wrote:
> ---
2nd ACK and pushed.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 17:21:08 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 19:21:08 +0200
Subject: [Libosinfo] [PATCH 1/5] opensuse: Fix indentation
In-Reply-To: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
Message-ID:
On Wed, Jan 2, 2013 at 4:07 AM, Fabiano Fid?ncio wrote:
> ---
ACK and pushed.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 17:21:42 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 19:21:42 +0200
Subject: [Libosinfo] [PATCH 3/5] opensuse: Add missing kernel/initrd
information
In-Reply-To: <1357092444-4155-3-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
<1357092444-4155-3-git-send-email-fabiano@fidencio.org>
Message-ID:
On Wed, Jan 2, 2013 at 4:07 AM, Fabiano Fid?ncio wrote:
> ---
ACK and pushed.
> data/oses/opensuse.xml.in | 16 ++++++++++++++++
> 1 file changed, 16 insertions(+)
>
> diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
> index b1624e1..63122b5 100644
> --- a/data/oses/opensuse.xml.in
> +++ b/data/oses/opensuse.xml.in
> @@ -136,6 +136,8 @@
> LINUX
> SUSE LINUX Products GmbH
>
> + boot/i386/loader/linux
> + boot/i386/loader/initrd
>
>
> http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-x86_64.iso?mirrorlist
While you are at this, would be nice to correct those URLs as well.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 17:32:21 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 19:32:21 +0200
Subject: [Libosinfo] [PATCH 5/5] opensuse: Add 12.2 information
In-Reply-To: <1357092444-4155-5-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
<1357092444-4155-5-git-send-email-fabiano@fidencio.org>
Message-ID:
On Wed, Jan 2, 2013 at 4:07 AM, Fabiano Fid?ncio wrote:
> ---
> data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 67 insertions(+)
>
> diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
> index 798d1a5..c8e69b4 100644
> --- a/data/oses/opensuse.xml.in
> +++ b/data/oses/opensuse.xml.in
> @@ -252,4 +252,71 @@
>
>
>
> +
> + opensuse12.2
> + <_name>openSUSE 12.2
> + 12.2
> + <_vendor>openSUSE
> + linux
> + opensuse
> +
> +
> +
> +
> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-i586.iso?mirrorlist
As stated in another mail, these URLs are wrong. I think you just need
to remove the '?mirrorlist'. I'm hoping you checked/tested the other
information you added?
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 17:30:16 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 19:30:16 +0200
Subject: [Libosinfo] [PATCH 4/5] opensuse: Fix 12.1 resources
In-Reply-To: <1357092444-4155-4-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
<1357092444-4155-4-git-send-email-fabiano@fidencio.org>
Message-ID:
On Wed, Jan 2, 2013 at 4:07 AM, Fabiano Fid?ncio wrote:
> According with http://en.opensuse.org/Hardware_requirements
> ---
ACK and pushed with minor corrections to log message.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 2 17:56:06 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 19:56:06 +0200
Subject: [Libosinfo] [PATCH] l10n: Add forgotten file to POTFILES.in
Message-ID: <1357149366-7234-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Add forgotten osinfo_install_config.c file to po/POTFILES.in.
Fixes `make check`.
Pushed under trivial rule.
---
po/POTFILES.in | 1 +
1 file changed, 1 insertion(+)
diff --git a/po/POTFILES.in b/po/POTFILES.in
index 1a209bb..216751a 100644
--- a/po/POTFILES.in
+++ b/po/POTFILES.in
@@ -28,6 +28,7 @@ osinfo/osinfo_deployment.c
osinfo/osinfo_devicelink.c
osinfo/osinfo_devicelinkfilter.c
osinfo/osinfo_entity.c
+osinfo/osinfo_install_config.c
osinfo/osinfo_install_config_param.c
osinfo/osinfo_install_script.c
osinfo/osinfo_list.c
--
1.8.0.2
From fabiano at fidencio.org Wed Jan 2 19:27:30 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Wed, 2 Jan 2013 17:27:30 -0200
Subject: [Libosinfo] [PATCH 5/5] opensuse: Add 12.2 information
In-Reply-To:
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
<1357092444-4155-5-git-send-email-fabiano@fidencio.org>
Message-ID:
On Wed, Jan 2, 2013 at 3:32 PM, Zeeshan Ali (Khattak)
wrote:
> On Wed, Jan 2, 2013 at 4:07 AM, Fabiano Fid?ncio wrote:
>> ---
>> data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
>> 1 file changed, 67 insertions(+)
>>
>> diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
>> index 798d1a5..c8e69b4 100644
>> --- a/data/oses/opensuse.xml.in
>> +++ b/data/oses/opensuse.xml.in
>> @@ -252,4 +252,71 @@
>>
>>
>>
>> +
>> + opensuse12.2
>> + <_name>openSUSE 12.2
>> + 12.2
>> + <_vendor>openSUSE
>> + linux
>> + opensuse
>> +
>> +
>> +
>> +
>> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-i586.iso?mirrorlist
>
> As stated in another mail, these URLs are wrong. I think you just need
> to remove the '?mirrorlist'. I'm hoping you checked/tested the other
> information you added?
This is a c&p from 12.1, and the URL was just sed'ed without a check.
I'll fix this and another URLs and resend the patches.
>
>
> --
> Regards,
>
> Zeeshan Ali (Khattak)
> FSF member#5124
Best Regards,
--
Fabiano Fid?ncio
From zeeshanak at gnome.org Wed Jan 2 21:11:56 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 23:11:56 +0200
Subject: [Libosinfo] [PATCH 1/2] Adapt windows-unattend.xml for Windows 8
In-Reply-To: <1356639790-1437-2-git-send-email-zeeshanak@gnome.org>
References: <1356639790-1437-2-git-send-email-zeeshanak@gnome.org>
Message-ID: <1357161117-22514-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
We still can't use this script directly for Windows 8 though as product
key is once again mandatory in Windows 8.
---
data/install-scripts/windows-unattend.xml | 74 +++++++++++++++++++------------
1 file changed, 46 insertions(+), 28 deletions(-)
diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
index d8c7563..baf4d59 100644
--- a/data/install-scripts/windows-unattend.xml
+++ b/data/install-scripts/windows-unattend.xml
@@ -129,12 +129,14 @@
-
-
- /IMAGE/INDEX
- 1
-
-
+
+
+
+ /IMAGE/INDEX
+ 1
+
+
+ 01
@@ -144,6 +146,11 @@
true
+
+
+
+
+
@@ -215,15 +222,17 @@
-
-
-
-
-
+
+
+
+
+
+
-
-
-
+
+
+
+
@@ -365,12 +374,14 @@
-
-
- /IMAGE/INDEX
- 1
-
-
+
+
+
+ /IMAGE/INDEX
+ 1
+
+
+ 01
@@ -380,6 +391,11 @@
true
+
+
+
+
+
@@ -460,15 +476,17 @@
-
-
-
-
-
+
+
+
+
+
+
-
-
-
+
+
+
+
--
1.8.0.2
From zeeshanak at gnome.org Wed Jan 2 21:11:57 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 23:11:57 +0200
Subject: [Libosinfo] [PATCH 2/2] win8: Add install script
In-Reply-To: <1357161117-22514-1-git-send-email-zeeshanak@gnome.org>
References: <1356639790-1437-2-git-send-email-zeeshanak@gnome.org>
<1357161117-22514-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1357161117-22514-2-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Alternative patch to add Windows 8 installer that makes use of
'xi:include' nodes to save space. However I failed to make it work.
Either the xpointer is wrong and/or (more likely) this node is not
automatically substituted by libxml2.
There is special xinclude API in libxml2 that probably solves our
problem but it works on DOM API and we are using XPath to load/parse our
XML files.
---
data/install-scripts/Makefile.am | 1 +
data/install-scripts/windows-8-unattend.xml | 45 +++++++++++++++++++++++++++++
data/oses/windows.xml.in | 3 +-
3 files changed, 48 insertions(+), 1 deletion(-)
create mode 100644 data/install-scripts/windows-8-unattend.xml
diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
index cb24e09..29715d2 100644
--- a/data/install-scripts/Makefile.am
+++ b/data/install-scripts/Makefile.am
@@ -6,6 +6,7 @@ database_DATA = \
windows-cmd.xml \
windows-reg.xml \
windows-unattend.xml \
+ windows-8-unattend.xml \
$(NULL)
EXTRA_DIST = $(database_DATA)
diff --git a/data/install-scripts/windows-8-unattend.xml b/data/install-scripts/windows-8-unattend.xml
new file mode 100644
index 0000000..fe7ea7b
--- /dev/null
+++ b/data/install-scripts/windows-8-unattend.xml
@@ -0,0 +1,45 @@
+
+
+ jeos
+ dos
+ $$$$$-$$$$$-$$$$$-$$$$$-$$$$$
+ autounattend.xml
+ true
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ desktop
+ dos
+ $$$$$-$$$$$-$$$$$-$$$$$-$$$$$
+ autounattend.xml
+ true
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
index 310de62..364fcaa 100644
--- a/data/oses/windows.xml.in
+++ b/data/oses/windows.xml.in
@@ -822,7 +822,8 @@
-
+
+
--
1.8.0.2
From zeeshanak at gnome.org Wed Jan 2 21:32:30 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 2 Jan 2013 23:32:30 +0200
Subject: [Libosinfo] [PATCH 2/5] opensuse: Fix kernel/initrd paths
In-Reply-To: <1357092444-4155-2-git-send-email-fabiano@fidencio.org>
References: <1357092444-4155-1-git-send-email-fabiano@fidencio.org>
<1357092444-4155-2-git-send-email-fabiano@fidencio.org>
Message-ID:
On Wed, Jan 2, 2013 at 4:07 AM, Fabiano Fid?ncio wrote:
> ---
ACK and pushed.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From fabiano at fidencio.org Fri Jan 4 05:45:46 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Fri, 4 Jan 2013 03:45:46 -0200
Subject: [Libosinfo] [PATCH 1/2] opensuse: Fix .iso URLs
Message-ID: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
---
data/oses/opensuse.xml.in | 24 ++++++++++++------------
1 file changed, 12 insertions(+), 12 deletions(-)
diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
index 798d1a5..0877b40 100644
--- a/data/oses/opensuse.xml.in
+++ b/data/oses/opensuse.xml.in
@@ -64,7 +64,7 @@
- http://download.opensuse.org/distribution/11.3/iso/openSUSE-11.3-DVD-i586.iso?mirrorlist
+ http://download.opensuse.org/distribution/11.3/iso/openSUSE-11.3-DVD-i586.isoopenSUSE-DVD-i586-Build0702..001LINUX
@@ -74,7 +74,7 @@
boot/i386/loader/initrd
- http://download.opensuse.org/distribution/11.3/iso/openSUSE-11.3-DVD-x86_64.iso?mirrorlist
+ http://download.opensuse.org/distribution/11.3/iso/openSUSE-11.3-DVD-x86_64.isoopenSUSE-DVD-x86_64.0702..001LINUX
@@ -85,7 +85,7 @@
-
+ http://download.opensuse.org/distribution/11.3/iso/openSUSE-11.3-GNOME-LiveCD-i686.isoopenSUSE Live CD.*LINUX
@@ -95,7 +95,7 @@
boot/i386/loader/initrd
-
+ http://download.opensuse.org/distribution/11.3/iso/openSUSE-11.3-GNOME-LiveCD-x86_64.isoopenSUSE Live CD.*LINUX
@@ -130,7 +130,7 @@
- http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-i586.iso?mirrorlist
+ http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-i586.isoopenSUSE-DVD-i586-Build0024LINUX
@@ -140,7 +140,7 @@
boot/i386/loader/initrd
- http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-x86_64.iso?mirrorlist
+ http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-DVD-x86_64.isoopenSUSE-DVD-x86_640024LINUX
@@ -151,7 +151,7 @@
-
+ http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-GNOME-LiveCD-i686.isoopenSUSE Live CD.*LINUX
@@ -161,7 +161,7 @@
boot/i386/loader/initrd
-
+ http://download.opensuse.org/distribution/11.4/iso/openSUSE-11.4-GNOME-LiveCD-x86_64.isoopenSUSE Live CD.*LINUX
@@ -196,7 +196,7 @@
- http://download.opensuse.org/distribution/12.1/iso/openSUSE-12.1-DVD-i586.iso?mirrorlist
+ http://download.opensuse.org/distribution/12.1/iso/openSUSE-12.1-DVD-i586.isoopenSUSE-DVD-i586-Build0039LINUX
@@ -206,7 +206,7 @@
boot/i386/loader/initrd
- http://download.opensuse.org/distribution/12.1/iso/openSUSE-12.1-DVD-x86_64.iso?mirrorlist
+ http://download.opensuse.org/distribution/12.1/iso/openSUSE-12.1-DVD-x86_64.isoopenSUSE-DVD-x86_640039LINUX
@@ -217,7 +217,7 @@
-
+ http://download.opensuse.org/distribution/12.1/iso/openSUSE-12.1-GNOME-LiveCD-i686.isoopenSUSE Live CD.*LINUX
@@ -227,7 +227,7 @@
boot/i386/loader/initrd
-
+ http://download.opensuse.org/distribution/12.1/iso/openSUSE-12.1-GNOME-LiveCD-x86_64.isoopenSUSE Live CD.*LINUX
--
1.8.0.1
From fabiano at fidencio.org Fri Jan 4 05:45:47 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Fri, 4 Jan 2013 03:45:47 -0200
Subject: [Libosinfo] [PATCHv2 2/2] opensuse: Add 12.2 information
In-Reply-To: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
References: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
Message-ID: <1357278347-10181-2-git-send-email-fabiano@fidencio.org>
---
data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 67 insertions(+)
diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
index 0877b40..abc5c2a 100644
--- a/data/oses/opensuse.xml.in
+++ b/data/oses/opensuse.xml.in
@@ -252,4 +252,71 @@
+
+ opensuse12.2
+ <_name>openSUSE 12.2
+ 12.2
+ <_vendor>openSUSE
+ linux
+ opensuse
+
+
+
+
+ http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-i586.iso
+
+ openSUSE-DVD-i586-Build0039
+ LINUX
+ SUSE LINUX Products GmbH
+
+ boot/i386/loader/linux
+ boot/i386/loader/initrd
+
+
+ http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-x86_64.iso
+
+ openSUSE-DVD-x86_640039
+ LINUX
+ SUSE LINUX Products GmbH
+
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
+
+
+
+ http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-GNOME-LiveCD-i686.iso
+
+ openSUSE Live CD.*
+ LINUX
+ SUSE LINUX Products GmbH
+
+ boot/i386/loader/linux
+ boot/i386/loader/initrd
+
+
+ http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-GNOME-LiveCD-x86_64.iso
+
+ openSUSE Live CD.*
+ LINUX
+ SUSE LINUX Products GmbH
+
+ boot/x86_64/loader/linux
+ boot/x86_64/loader/initrd
+
+
+
+
+ 500000000
+ 1
+ 1073741824
+ 3221225472
+
+
+ 2400000000
+ 2147483648
+ 5368709120
+
+
+
+
--
1.8.0.1
From fabiano at fidencio.org Fri Jan 4 06:03:16 2013
From: fabiano at fidencio.org (=?UTF-8?q?Fabiano=20Fid=C3=AAncio?=)
Date: Fri, 4 Jan 2013 04:03:16 -0200
Subject: [Libosinfo] [PATCHv4] debian: Update volume-id for Squeeze
Message-ID: <1357279396-12116-1-git-send-email-fabiano@fidencio.org>
---
data/oses/debian.xml.in | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/data/oses/debian.xml.in b/data/oses/debian.xml.in
index fc0f6cc..b2969aa 100644
--- a/data/oses/debian.xml.in
+++ b/data/oses/debian.xml.in
@@ -187,7 +187,7 @@
http://live.debian.net/cdimage/release/current/i386/iso-hybrid/debian-live-6.0.6-i386-gnome-desktop.iso
- Debian squeeze 20111014-08:13
+ Debian squeeze.*LINUXinstall/vmlinuz
@@ -196,7 +196,7 @@
http://live.debian.net/cdimage/release/current/amd64/iso-hybrid/debian-live-6.0.6-amd64-gnome-desktop.iso
- Debian squeeze 20111014-08:11
+ Debian squeeze.*LINUXinstall/vmlinuz
--
1.8.0.1
From mprivozn at redhat.com Fri Jan 4 10:01:31 2013
From: mprivozn at redhat.com (Michal Privoznik)
Date: Fri, 04 Jan 2013 11:01:31 +0100
Subject: [Libosinfo] [PATCHv4 06/11] Add
OsinfoInstallConfig:config-params property
In-Reply-To:
References: <1356021914-16369-1-git-send-email-cfergeau@redhat.com>
<1356021914-16369-6-git-send-email-cfergeau@redhat.com>
Message-ID: <50E6A87B.6090408@redhat.com>
On 31.12.2012 17:20, Zeeshan Ali (Khattak) wrote:
> On Thu, Dec 20, 2012 at 6:45 PM, Christophe Fergeau wrote:
>> This property lists the parameters that can be set for a given
>> OsinfoInstallConfig. This is not enforced, it's only there for
>> informative purpose. This will also be used in later commits
>> in order to automatically apply transformations on values
>> for parameters which have an associated OsinfoDatamap.
>
> Sorry to revive this thread again but I had an idea that I wanted to
> discuss before this API goes into a release: How about we have a
> OsinfoInstallConfig:install-script (of type OsinfoInstallScript)
> rather than having OsinfoInstallConfig:config-params (and
> 'path-format', 'avatar-format' etc) duplicated on both objects? IMO
> that makes a lot more sense since that not only avoids duplication of
> API (and some strings) but also makes things more clear: Whether a
> config is associated/specific to a script or not? If it is, which
> install script is exactly its tied to?
>
>
>
Right. I prefer OsinfoInstallConfig:install-script as well. The more
inheritance the better IMO in this case.
Michal
From zeeshanak at gnome.org Fri Jan 4 15:12:04 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Fri, 4 Jan 2013 17:12:04 +0200
Subject: [Libosinfo] [PATCH 1/2] opensuse: Fix .iso URLs
In-Reply-To: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
References: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
Message-ID:
On Fri, Jan 4, 2013 at 7:45 AM, Fabiano Fid?ncio wrote:
> ---
ACK.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Fri Jan 4 15:15:32 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Fri, 4 Jan 2013 17:15:32 +0200
Subject: [Libosinfo] [PATCHv2 2/2] opensuse: Add 12.2 information
In-Reply-To: <1357278347-10181-2-git-send-email-fabiano@fidencio.org>
References: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
<1357278347-10181-2-git-send-email-fabiano@fidencio.org>
Message-ID:
On Fri, Jan 4, 2013 at 7:45 AM, Fabiano Fid?ncio wrote:
> ---
> data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
> 1 file changed, 67 insertions(+)
>
> diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
> index 0877b40..abc5c2a 100644
> --- a/data/oses/opensuse.xml.in
> +++ b/data/oses/opensuse.xml.in
> @@ -252,4 +252,71 @@
>
>
>
> +
> + opensuse12.2
> + <_name>openSUSE 12.2
> + 12.2
> + <_vendor>openSUSE
> + linux
> + opensuse
> +
> +
> +
> +
> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-i586.iso
> +
> + openSUSE-DVD-i586-Build0039
> + LINUX
> + SUSE LINUX Products GmbH
> +
> + boot/i386/loader/linux
> + boot/i386/loader/initrd
> +
> +
> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-x86_64.iso
> +
> + openSUSE-DVD-x86_640039
> + LINUX
> + SUSE LINUX Products GmbH
> +
> + boot/x86_64/loader/linux
> + boot/x86_64/loader/initrd
> +
> +
> +
> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-GNOME-LiveCD-i686.iso
> +
> + openSUSE Live CD.*
> + LINUX
> + SUSE LINUX Products GmbH
> +
> + boot/i386/loader/linux
> + boot/i386/loader/initrd
> +
> +
> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-GNOME-LiveCD-x86_64.iso
> +
> + openSUSE Live CD.*
> + LINUX
> + SUSE LINUX Products GmbH
> +
> + boot/x86_64/loader/linux
> + boot/x86_64/loader/initrd
> +
> +
> +
> +
> + 500000000
> + 1
> + 1073741824
> + 3221225472
> +
> +
> + 2400000000
> + 2147483648
> + 5368709120
> +
> +
> +
Assuming you have checked that these resources remain same for 12.2, ACK!
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Fri Jan 4 15:18:15 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Fri, 4 Jan 2013 17:18:15 +0200
Subject: [Libosinfo] [PATCHv4] debian: Update volume-id for Squeeze
In-Reply-To: <1357279396-12116-1-git-send-email-fabiano@fidencio.org>
References: <1357279396-12116-1-git-send-email-fabiano@fidencio.org>
Message-ID:
On Fri, Jan 4, 2013 at 8:03 AM, Fabiano Fid?ncio wrote:
> ---
ACK
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From fabiano at fidencio.org Fri Jan 4 16:13:59 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Fri, 4 Jan 2013 14:13:59 -0200
Subject: [Libosinfo] [PATCHv2 2/2] opensuse: Add 12.2 information
In-Reply-To:
References: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
<1357278347-10181-2-git-send-email-fabiano@fidencio.org>
Message-ID:
On Fri, Jan 4, 2013 at 1:15 PM, Zeeshan Ali (Khattak)
wrote:
> On Fri, Jan 4, 2013 at 7:45 AM, Fabiano Fid?ncio wrote:
>> ---
>> data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
>> 1 file changed, 67 insertions(+)
>>
>> diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
>> index 0877b40..abc5c2a 100644
>> --- a/data/oses/opensuse.xml.in
>> +++ b/data/oses/opensuse.xml.in
>> @@ -252,4 +252,71 @@
>>
>>
>>
>> +
>> + opensuse12.2
>> + <_name>openSUSE 12.2
>> + 12.2
>> + <_vendor>openSUSE
>> + linux
>> + opensuse
>> +
>> +
>> +
>> +
>> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-i586.iso
>> +
>> + openSUSE-DVD-i586-Build0039
>> + LINUX
>> + SUSE LINUX Products GmbH
>> +
>> + boot/i386/loader/linux
>> + boot/i386/loader/initrd
>> +
>> +
>> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-x86_64.iso
>> +
>> + openSUSE-DVD-x86_640039
>> + LINUX
>> + SUSE LINUX Products GmbH
>> +
>> + boot/x86_64/loader/linux
>> + boot/x86_64/loader/initrd
>> +
>> +
>> +
>> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-GNOME-LiveCD-i686.iso
>> +
>> + openSUSE Live CD.*
>> + LINUX
>> + SUSE LINUX Products GmbH
>> +
>> + boot/i386/loader/linux
>> + boot/i386/loader/initrd
>> +
>> +
>> + http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-GNOME-LiveCD-x86_64.iso
>> +
>> + openSUSE Live CD.*
>> + LINUX
>> + SUSE LINUX Products GmbH
>> +
>> + boot/x86_64/loader/linux
>> + boot/x86_64/loader/initrd
>> +
>> +
>> +
>> +
>> + 500000000
>> + 1
>> + 1073741824
>> + 3221225472
>> +
>> +
>> + 2400000000
>> + 2147483648
>> + 5368709120
>> +
>> +
>> +
>
> Assuming you have checked that these resources remain same for 12.2, ACK!
I did! If you want to double check, please, take a look into:
http://en.opensuse.org/Hardware_requirements
>
> --
> Regards,
>
> Zeeshan Ali (Khattak)
> FSF member#5124
Best Regards,
--
Fabiano Fid?ncio
From berrange at redhat.com Fri Jan 4 16:17:40 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Fri, 4 Jan 2013 16:17:40 +0000
Subject: [Libosinfo] [PATCHv2 2/2] opensuse: Add 12.2 information
In-Reply-To:
References: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
<1357278347-10181-2-git-send-email-fabiano@fidencio.org>
Message-ID: <20130104161740.GC2105@redhat.com>
On Fri, Jan 04, 2013 at 02:13:59PM -0200, Fabiano Fid?ncio wrote:
> On Fri, Jan 4, 2013 at 1:15 PM, Zeeshan Ali (Khattak)
> wrote:
> > On Fri, Jan 4, 2013 at 7:45 AM, Fabiano Fid?ncio wrote:
> >> ---
> >> data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
> >> 1 file changed, 67 insertions(+)
> >>
> >> diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
> >> index 0877b40..abc5c2a 100644
> >> --- a/data/oses/opensuse.xml.in
> >> +++ b/data/oses/opensuse.xml.in
> >
> > Assuming you have checked that these resources remain same for 12.2, ACK!
>
> I did! If you want to double check, please, take a look into:
> http://en.opensuse.org/Hardware_requirements
It would be worth adding a comment to the top of opensuse.xml.in
so that we remember where to look in future. Likewise for any other
relevant URLs we're using for populating other pieces of data.
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From fabiano at fidencio.org Fri Jan 4 16:42:33 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Fri, 4 Jan 2013 14:42:33 -0200
Subject: [Libosinfo] [PATCHv2 2/2] opensuse: Add 12.2 information
In-Reply-To: <20130104161740.GC2105@redhat.com>
References: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
<1357278347-10181-2-git-send-email-fabiano@fidencio.org>
<20130104161740.GC2105@redhat.com>
Message-ID:
On Fri, Jan 4, 2013 at 2:17 PM, Daniel P. Berrange wrote:
> On Fri, Jan 04, 2013 at 02:13:59PM -0200, Fabiano Fid?ncio wrote:
>> On Fri, Jan 4, 2013 at 1:15 PM, Zeeshan Ali (Khattak)
>> wrote:
>> > On Fri, Jan 4, 2013 at 7:45 AM, Fabiano Fid?ncio wrote:
>> >> ---
>> >> data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
>> >> 1 file changed, 67 insertions(+)
>> >>
>> >> diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
>> >> index 0877b40..abc5c2a 100644
>> >> --- a/data/oses/opensuse.xml.in
>> >> +++ b/data/oses/opensuse.xml.in
>> >
>> > Assuming you have checked that these resources remain same for 12.2, ACK!
>>
>> I did! If you want to double check, please, take a look into:
>> http://en.opensuse.org/Hardware_requirements
>
> It would be worth adding a comment to the top of opensuse.xml.in
>
>
>
>
> so that we remember where to look in future. Likewise for any other
> relevant URLs we're using for populating other pieces of data.
Okay. I'll add the comment and push the patch ASAP.
>
> Daniel
> --
> |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
> |: http://libvirt.org -o- http://virt-manager.org :|
> |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
> |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
Best Regards,
--
Fabiano Fid?ncio
From fabiano at fidencio.org Fri Jan 4 17:04:48 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Fri, 4 Jan 2013 15:04:48 -0200
Subject: [Libosinfo] [PATCHv2 2/2] opensuse: Add 12.2 information
In-Reply-To:
References: <1357278347-10181-1-git-send-email-fabiano@fidencio.org>
<1357278347-10181-2-git-send-email-fabiano@fidencio.org>
<20130104161740.GC2105@redhat.com>
Message-ID:
On Fri, Jan 4, 2013 at 2:42 PM, Fabiano Fid?ncio wrote:
> On Fri, Jan 4, 2013 at 2:17 PM, Daniel P. Berrange wrote:
>> On Fri, Jan 04, 2013 at 02:13:59PM -0200, Fabiano Fid?ncio wrote:
>>> On Fri, Jan 4, 2013 at 1:15 PM, Zeeshan Ali (Khattak)
>>> wrote:
>>> > On Fri, Jan 4, 2013 at 7:45 AM, Fabiano Fid?ncio wrote:
>>> >> ---
>>> >> data/oses/opensuse.xml.in | 67 +++++++++++++++++++++++++++++++++++++++++++++++
>>> >> 1 file changed, 67 insertions(+)
>>> >>
>>> >> diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
>>> >> index 0877b40..abc5c2a 100644
>>> >> --- a/data/oses/opensuse.xml.in
>>> >> +++ b/data/oses/opensuse.xml.in
>>> >
>>> > Assuming you have checked that these resources remain same for 12.2, ACK!
>>>
>>> I did! If you want to double check, please, take a look into:
>>> http://en.opensuse.org/Hardware_requirements
>>
>> It would be worth adding a comment to the top of opensuse.xml.in
>>
>>
>>
>>
>> so that we remember where to look in future. Likewise for any other
>> relevant URLs we're using for populating other pieces of data.
>
> Okay. I'll add the comment and push the patch ASAP.
>
>>
>> Daniel
>> --
>> |: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
>> |: http://libvirt.org -o- http://virt-manager.org :|
>> |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
>> |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
>
>
> Best Regards,
> --
> Fabiano Fid?ncio
Pushed with a small modification in Daniel's comment, avoiding to
break make check.
Best Regards,
--
Fabiano Fid?ncio
From zeeshanak at gnome.org Mon Jan 7 00:12:57 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 7 Jan 2013 02:12:57 +0200
Subject: [Libosinfo] [PATCHv4 06/11] Add
OsinfoInstallConfig:config-params property
In-Reply-To: <50E6A87B.6090408@redhat.com>
References: <1356021914-16369-1-git-send-email-cfergeau@redhat.com>
<1356021914-16369-6-git-send-email-cfergeau@redhat.com>
<50E6A87B.6090408@redhat.com>
Message-ID:
On Fri, Jan 4, 2013 at 12:01 PM, Michal Privoznik wrote:
> On 31.12.2012 17:20, Zeeshan Ali (Khattak) wrote:
>> On Thu, Dec 20, 2012 at 6:45 PM, Christophe Fergeau wrote:
>>> This property lists the parameters that can be set for a given
>>> OsinfoInstallConfig. This is not enforced, it's only there for
>>> informative purpose. This will also be used in later commits
>>> in order to automatically apply transformations on values
>>> for parameters which have an associated OsinfoDatamap.
>>
>> Sorry to revive this thread again but I had an idea that I wanted to
>> discuss before this API goes into a release: How about we have a
>> OsinfoInstallConfig:install-script (of type OsinfoInstallScript)
>> rather than having OsinfoInstallConfig:config-params (and
>> 'path-format', 'avatar-format' etc) duplicated on both objects? IMO
>> that makes a lot more sense since that not only avoids duplication of
>> API (and some strings) but also makes things more clear: Whether a
>> config is associated/specific to a script or not? If it is, which
>> install script is exactly its tied to?
>>
>>
>>
>
> Right. I prefer OsinfoInstallConfig:install-script as well. The more
> inheritance the better IMO in this case.
As much as I'd like to see other developers agree with my suggestion,
I feel that there is a miscommunication here: The ':' in the GObject
world (especially when talking of C code/API) denotes property rather
than inheritance.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From fabiano at fidencio.org Mon Jan 7 02:46:45 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Mon, 7 Jan 2013 00:46:45 -0200
Subject: [Libosinfo] [PATCHv4 06/11] Add
OsinfoInstallConfig:config-params property
In-Reply-To:
References: <1356021914-16369-1-git-send-email-cfergeau@redhat.com>
<1356021914-16369-6-git-send-email-cfergeau@redhat.com>
Message-ID:
On Mon, Dec 31, 2012 at 2:20 PM, Zeeshan Ali (Khattak)
wrote:
> On Thu, Dec 20, 2012 at 6:45 PM, Christophe Fergeau wrote:
>> This property lists the parameters that can be set for a given
>> OsinfoInstallConfig. This is not enforced, it's only there for
>> informative purpose. This will also be used in later commits
>> in order to automatically apply transformations on values
>> for parameters which have an associated OsinfoDatamap.
>
> Sorry to revive this thread again but I had an idea that I wanted to
> discuss before this API goes into a release: How about we have a
> OsinfoInstallConfig:install-script (of type OsinfoInstallScript)
> rather than having OsinfoInstallConfig:config-params (and
> 'path-format', 'avatar-format' etc) duplicated on both objects? IMO
> that makes a lot more sense since that not only avoids duplication of
> API (and some strings) but also makes things more clear: Whether a
> config is associated/specific to a script or not? If it is, which
> install script is exactly its tied to?
After a long read of the topic, personally, I agree with Zeeshan's
suggestion, basically because it contributes for clearness.
However, I think we should wait Christophe's opinion (he is on vacations, IIRC).
>
>
>
> --
> Regards,
>
> Zeeshan Ali (Khattak)
> FSF member#5124
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
Best Regards,
--
Fabiano Fid?ncio
From zeeshanak at gnome.org Mon Jan 7 14:25:50 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 7 Jan 2013 16:25:50 +0200
Subject: [Libosinfo] [PATCH] ubuntu: Correct required/recommended resources
for 12.10
Message-ID: <1357568750-17053-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
This is as based on "Getting Started with Ubuntu 12.10 - Ubuntu Manual"
(page 11 in en_US version).
---
data/oses/ubuntu.xml.in | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/data/oses/ubuntu.xml.in b/data/oses/ubuntu.xml.in
index 8c3c8d2..300d0f8 100644
--- a/data/oses/ubuntu.xml.in
+++ b/data/oses/ubuntu.xml.in
@@ -1140,13 +1140,13 @@
10000000001
- 67108864
+ 107374182453687091201000000000
- 536870912
- 5368709120
+ 1073741824
+ 16106127360
--
1.8.0.2
From fabiano at fidencio.org Mon Jan 7 14:41:53 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Mon, 7 Jan 2013 12:41:53 -0200
Subject: [Libosinfo] [PATCH] ubuntu: Correct required/recommended
resources for 12.10
In-Reply-To: <1357568750-17053-1-git-send-email-zeeshanak@gnome.org>
References: <1357568750-17053-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Mon, Jan 7, 2013 at 12:25 PM, Zeeshan Ali (Khattak)
wrote:
> From: "Zeeshan Ali (Khattak)"
>
> This is as based on "Getting Started with Ubuntu 12.10 - Ubuntu Manual"
> (page 11 in en_US version).
ACK!
> ---
> data/oses/ubuntu.xml.in | 6 +++---
> 1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/data/oses/ubuntu.xml.in b/data/oses/ubuntu.xml.in
> index 8c3c8d2..300d0f8 100644
> --- a/data/oses/ubuntu.xml.in
> +++ b/data/oses/ubuntu.xml.in
> @@ -1140,13 +1140,13 @@
>
> 1000000000
> 1
> - 67108864
> + 1073741824
> 5368709120
>
>
> 1000000000
> - 536870912
> - 5368709120
> + 1073741824
> + 16106127360
>
>
>
> --
> 1.8.0.2
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
Best Regards,
--
Fabiano Fid?ncio
From fabiano at fidencio.org Mon Jan 7 14:44:27 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Mon, 7 Jan 2013 12:44:27 -0200
Subject: [Libosinfo] [PATCHv4 06/11] Add
OsinfoInstallConfig:config-params property
In-Reply-To:
References: <1356021914-16369-1-git-send-email-cfergeau@redhat.com>
<1356021914-16369-6-git-send-email-cfergeau@redhat.com>
Message-ID:
On Mon, Jan 7, 2013 at 12:40 PM, Zeeshan Ali (Khattak)
wrote:
> On Mon, Jan 7, 2013 at 4:46 AM, Fabiano Fid?ncio wrote:
>> On Mon, Dec 31, 2012 at 2:20 PM, Zeeshan Ali (Khattak)
>> wrote:
>>> On Thu, Dec 20, 2012 at 6:45 PM, Christophe Fergeau wrote:
>>>> This property lists the parameters that can be set for a given
>>>> OsinfoInstallConfig. This is not enforced, it's only there for
>>>> informative purpose. This will also be used in later commits
>>>> in order to automatically apply transformations on values
>>>> for parameters which have an associated OsinfoDatamap.
>>>
>>> Sorry to revive this thread again but I had an idea that I wanted to
>>> discuss before this API goes into a release: How about we have a
>>> OsinfoInstallConfig:install-script (of type OsinfoInstallScript)
>>> rather than having OsinfoInstallConfig:config-params (and
>>> 'path-format', 'avatar-format' etc) duplicated on both objects? IMO
>>> that makes a lot more sense since that not only avoids duplication of
>>> API (and some strings) but also makes things more clear: Whether a
>>> config is associated/specific to a script or not? If it is, which
>>> install script is exactly its tied to?
>>
>>
>> After a long read of the topic, personally, I agree with Zeeshan's
>> suggestion, basically because it contributes for clearness.
>> However, I think we should wait Christophe's opinion (he is on vacations, IIRC).
>
> Sure thing, I have been waiting for that. However, we'll be needing a
> release at the end of this week (or next monday) for Boxes and being
> API stable, we wont be able to change the API after wards. I-e we only
> got this week to get this done.
So I suggest go-ahead with your approach if Christophe does not answer
until the end of the day.
>
> --
> Regards,
>
> Zeeshan Ali (Khattak)
> FSF member#5124
Best Regards,
--
Fabiano Fid?ncio
From zeeshanak at gnome.org Mon Jan 7 14:40:08 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 7 Jan 2013 16:40:08 +0200
Subject: [Libosinfo] [PATCHv4 06/11] Add
OsinfoInstallConfig:config-params property
In-Reply-To:
References: <1356021914-16369-1-git-send-email-cfergeau@redhat.com>
<1356021914-16369-6-git-send-email-cfergeau@redhat.com>
Message-ID:
On Mon, Jan 7, 2013 at 4:46 AM, Fabiano Fid?ncio wrote:
> On Mon, Dec 31, 2012 at 2:20 PM, Zeeshan Ali (Khattak)
> wrote:
>> On Thu, Dec 20, 2012 at 6:45 PM, Christophe Fergeau wrote:
>>> This property lists the parameters that can be set for a given
>>> OsinfoInstallConfig. This is not enforced, it's only there for
>>> informative purpose. This will also be used in later commits
>>> in order to automatically apply transformations on values
>>> for parameters which have an associated OsinfoDatamap.
>>
>> Sorry to revive this thread again but I had an idea that I wanted to
>> discuss before this API goes into a release: How about we have a
>> OsinfoInstallConfig:install-script (of type OsinfoInstallScript)
>> rather than having OsinfoInstallConfig:config-params (and
>> 'path-format', 'avatar-format' etc) duplicated on both objects? IMO
>> that makes a lot more sense since that not only avoids duplication of
>> API (and some strings) but also makes things more clear: Whether a
>> config is associated/specific to a script or not? If it is, which
>> install script is exactly its tied to?
>
>
> After a long read of the topic, personally, I agree with Zeeshan's
> suggestion, basically because it contributes for clearness.
> However, I think we should wait Christophe's opinion (he is on vacations, IIRC).
Sure thing, I have been waiting for that. However, we'll be needing a
release at the end of this week (or next monday) for Boxes and being
API stable, we wont be able to change the API after wards. I-e we only
got this week to get this done.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 7 16:25:24 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 7 Jan 2013 18:25:24 +0200
Subject: [Libosinfo] [PATCH] debian,macos: Fix typos in OS names
Message-ID: <1357575924-20033-1-git-send-email-zeeshanak@gnome.org>
From: Yuri Chornoivan
Pushed under trivial rule.
---
data/oses/debian.xml.in | 4 ++--
data/oses/macos.xml.in | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/data/oses/debian.xml.in b/data/oses/debian.xml.in
index b2969aa..4e445ad 100644
--- a/data/oses/debian.xml.in
+++ b/data/oses/debian.xml.in
@@ -48,8 +48,8 @@
debian2.1
- debiansilnk
- <_name>Debian Silnk
+ debianslink
+ <_name>Debian Slink
2.1
<_vendor>Debian Project
linux
diff --git a/data/oses/macos.xml.in b/data/oses/macos.xml.in
index 9ec868c..4b4a931 100644
--- a/data/oses/macos.xml.in
+++ b/data/oses/macos.xml.in
@@ -21,7 +21,7 @@
macosx10.2
- <_name>MacOS X Jaghuar
+ <_name>MacOS X Jaguar
10.2
<_vendor>Apple Inc.
darwin
--
1.8.0.2
From fabiano at fidencio.org Tue Jan 8 21:45:54 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Tue, 8 Jan 2013 19:45:54 -0200
Subject: [Libosinfo] [PATCH 2/2] win8: Add install script
In-Reply-To: <1357161117-22514-2-git-send-email-zeeshanak@gnome.org>
References: <1356639790-1437-2-git-send-email-zeeshanak@gnome.org>
<1357161117-22514-1-git-send-email-zeeshanak@gnome.org>
<1357161117-22514-2-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Wed, Jan 2, 2013 at 7:11 PM, Zeeshan Ali (Khattak)
wrote:
> From: "Zeeshan Ali (Khattak)"
>
> Alternative patch to add Windows 8 installer that makes use of
> 'xi:include' nodes to save space. However I failed to make it work.
> Either the xpointer is wrong and/or (more likely) this node is not
> automatically substituted by libxml2.
>
> There is special xinclude API in libxml2 that probably solves our
> problem but it works on DOM API and we are using XPath to load/parse our
> XML files.
At least for now, I'm inclined to say to commit your first patch.
There are many parts in libosinfo that we will need to rework to avoid
duplications and I don't think the rework is a dependency of this
patch.
So, go ahead with your first approach.
> ---
> data/install-scripts/Makefile.am | 1 +
> data/install-scripts/windows-8-unattend.xml | 45 +++++++++++++++++++++++++++++
> data/oses/windows.xml.in | 3 +-
> 3 files changed, 48 insertions(+), 1 deletion(-)
> create mode 100644 data/install-scripts/windows-8-unattend.xml
>
> diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
> index cb24e09..29715d2 100644
> --- a/data/install-scripts/Makefile.am
> +++ b/data/install-scripts/Makefile.am
> @@ -6,6 +6,7 @@ database_DATA = \
> windows-cmd.xml \
> windows-reg.xml \
> windows-unattend.xml \
> + windows-8-unattend.xml \
> $(NULL)
>
> EXTRA_DIST = $(database_DATA)
> diff --git a/data/install-scripts/windows-8-unattend.xml b/data/install-scripts/windows-8-unattend.xml
> new file mode 100644
> index 0000000..fe7ea7b
> --- /dev/null
> +++ b/data/install-scripts/windows-8-unattend.xml
> @@ -0,0 +1,45 @@
> +
> +
> + jeos
> + dos
> + $$$$$-$$$$$-$$$$$-$$$$$-$$$$$
> + autounattend.xml
> + true
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + desktop
> + dos
> + $$$$$-$$$$$-$$$$$-$$$$$-$$$$$
> + autounattend.xml
> + true
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
> index 310de62..364fcaa 100644
> --- a/data/oses/windows.xml.in
> +++ b/data/oses/windows.xml.in
> @@ -822,7 +822,8 @@
>
>
>
> -
> +
> +
>
>
>
> --
> 1.8.0.2
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
Best Regards,
--
Fabiano Fid?ncio
From zeeshanak at gnome.org Tue Jan 8 23:12:38 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 01:12:38 +0200
Subject: [Libosinfo] [PATCH 4/4] Add install script for RHEL 6.3
In-Reply-To:
References: <1356150057-24165-1-git-send-email-zeeshanak@gnome.org>
<1356150057-24165-5-git-send-email-zeeshanak@gnome.org>
<20121223105029.GB6273@teriyaki.redhat.com>
<20121223171535.GG6273@teriyaki.redhat.com>
<20121227130904.GC6522@teriyaki.redhat.com>
<20121229111114.GJ27991@teriyaki.redhat.com>
Message-ID:
On Sat, Dec 29, 2012 at 6:59 PM, Zeeshan Ali (Khattak)
wrote:
> On Sat, Dec 29, 2012 at 1:11 PM, Christophe Fergeau wrote:
>> On Sat, Dec 29, 2012 at 02:01:08AM +0200, Zeeshan Ali (Khattak) wrote:
>>> On Thu, Dec 27, 2012 at 3:09 PM, Christophe Fergeau wrote:
>>> > On Thu, Dec 27, 2012 at 02:55:04PM +0200, Zeeshan Ali (Khattak) wrote:
>>> >> On Sun, Dec 23, 2012 at 7:15 PM, Christophe Fergeau wrote:
>>> >> > clearparts is the only command still trying to use target-disk,
>>> >> > my guess is that this should be hardcoded to be done on the first disk some
>>> >> > how, and 'target-disk' should never appear in this script.
>>> >>
>>> >> I don't think thats wise. RHEL 7.x will most likely support disk
>>> >> arguments to 'part' commands (as Fedora17-18 supports it) and we
>>> >> can't have conditions on configuration parameters but only on their
>>> >> usage in the templates.
>>> >
>>> > Well, solve this the way you want, but having target-disk used only with
>>> > clearparts in the script is wrong.
>>>
>>> Its not exactly wrong.
>>
>> It is wrong, passing 'target-disk' is really uninteresting and won't do
>> what the user expects.
>
> Thanks for completely ignoring my arguments.
>
>> Anyway, after reading the RHEL kickstart documentation, I'm not sure why
>> you got rid of the --ondisk/--ondrive arguments to the 'part' command as
>> they are present in the documentation.
>
> I'll check again..
I checked now and seems I remembered incorrectly. I'll update the patch..
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Tue Jan 8 23:23:20 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 01:23:20 +0200
Subject: [Libosinfo] [PATCH v2] Add install script for RHEL 6.x
In-Reply-To:
References:
Message-ID: <1357687400-853-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
The same script might work for other RHEL 6.x too but I have only tested
it against RHEL 6.3 for now. Still some issues though:
1. For some reason user's avatar is ignored although according to the docs
I could find*, copying the avatar to /home/${USER}/.face should be
enough.
2. We don't set the keyboard layout properly yet and just hardcode it to
'us'. This will require the same kind of mapping as Fedora 17 and older.
* http://projects.gnome.org/gdm/docs/2.14/configuration.html
---
data/install-scripts/Makefile.am | 1 +
data/install-scripts/rhel.xml | 204 +++++++++++++++++++++++++++++++++++++++
data/oses/rhel.xml.in | 20 ++++
3 files changed, 225 insertions(+)
create mode 100644 data/install-scripts/rhel.xml
diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
index cb24e09..c310468 100644
--- a/data/install-scripts/Makefile.am
+++ b/data/install-scripts/Makefile.am
@@ -2,6 +2,7 @@
databasedir = $(pkgdatadir)/db/install-scripts/
database_DATA = \
fedora.xml \
+ rhel.xml \
windows-sif.xml \
windows-cmd.xml \
windows-reg.xml \
diff --git a/data/install-scripts/rhel.xml b/data/install-scripts/rhel.xml
new file mode 100644
index 0000000..9c17f11
--- /dev/null
+++ b/data/install-scripts/rhel.xml
@@ -0,0 +1,204 @@
+
+
+
+ jeos
+ fedora.ks
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ /dev/vda
+
+
+
+ /dev/sda
+
+
+
+
+
+# Install script for profile
+install
+text
+
+keyboard us
+lang
+skipx
+network --device eth0 --bootproto dhcp
+rootpw
+firewall --disabled
+authconfig --enableshadow --enablemd5
+selinux --enforcing
+timezone --utc
+bootloader --location=mbr
+zerombr
+
+clearpart --all --drives=
+
+part /boot --fstype ext4 --size=200 --ondisk=
+part pv.2 --size=1 --grow --ondisk=
+volgroup VolGroup00 --pesize=32768 pv.2
+logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 --size=768 --grow --maxsize=1536
+logvol / --fstype ext4 --name=LogVol00 --vgname=VolGroup00 --size=1024 --grow
+reboot
+
+%packages
+
+
+ at base
+
+
+ at standard
+
+
+ at core
+
+%end
+
+
+
+
+
+
+
+ desktop
+ fedora.ks
+
+
+
+
+
+
+
+
+
+
+
+
+
+ image/png
+
+
+
+
+
+
+
+
+
+
+
+
+
+ /dev/vda
+
+
+
+ /dev/sda
+
+
+
+
+
+# Install script for profile
+install
+keyboard us
+lang
+network --onboot yes --device eth0 --bootproto dhcp --noipv6 --hostname= --activate
+rootpw dummyPa55w0rd # Actual password set (or unset) in %post below
+firewall --disabled
+authconfig --enableshadow --enablemd5
+timezone --utc
+bootloader --location=mbr
+zerombr
+
+clearpart --all --drives=
+
+firstboot --disable
+
+part /boot --fstype ext4 --size=200 --ondisk=
+part pv.2 --size=1 --grow --ondisk=
+volgroup VolGroup00 --pesize=32768 pv.2
+logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 --size=768 --grow --maxsize=1536
+logvol / --fstype ext4 --name=LogVol00 --vgname=VolGroup00 --size=1024 --grow
+reboot
+
+%packages
+ at core
+ at x11
+ at basic-desktop
+ at general-desktop
+ at desktop-platform
+ at internet-applications
+ at internet-browser
+ at input-methods
+ at office-suite
+
+%end
+
+%post --erroronfail
+
+useradd -G wheel # Add user
+if test -z ; then
+ passwd -d # Make user account passwordless
+else
+ echo |passwd --stdin
+fi
+
+if test -z ; then
+ passwd -d root # Make root account passwordless
+else
+ echo |passwd --stdin root
+fi
+
+# Set user avatar
+if test -n ; then
+mkdir /mnt/unattended-media
+mount /mnt/unattended-media
+cp /mnt/unattended-media /home//.face
+umount /mnt/unattended-media
+fi
+
+# Enable autologin
+echo "[daemon]
+AutomaticLoginEnable=true
+AutomaticLogin=
+
+[security]
+
+[xdmcp]
+
+[greeter]
+
+[chooser]
+
+[debug]
+" > /etc/gdm/custom.conf
+
+%end
+
+
+
+
+
+
diff --git a/data/oses/rhel.xml.in b/data/oses/rhel.xml.in
index 1b34704..7e0db5b 100644
--- a/data/oses/rhel.xml.in
+++ b/data/oses/rhel.xml.in
@@ -708,6 +708,11 @@
9663676416
+
+
+
+
+
@@ -753,6 +758,11 @@
9663676416
+
+
+
+
+
@@ -798,6 +808,11 @@
9663676416
+
+
+
+
+
@@ -843,5 +858,10 @@
9663676416
+
+
+
+
+
--
1.8.0.2
From fabiano at fidencio.org Tue Jan 8 23:43:19 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Tue, 8 Jan 2013 21:43:19 -0200
Subject: [Libosinfo] [PATCH v2] Add install script for RHEL 6.x
In-Reply-To: <1357687400-853-1-git-send-email-zeeshanak@gnome.org>
References:
<1357687400-853-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Tue, Jan 8, 2013 at 9:23 PM, Zeeshan Ali (Khattak)
wrote:
> From: "Zeeshan Ali (Khattak)"
>
> The same script might work for other RHEL 6.x too but I have only tested
> it against RHEL 6.3 for now. Still some issues though:
>
> 1. For some reason user's avatar is ignored although according to the docs
> I could find*, copying the avatar to /home/${USER}/.face should be
> enough.
>
> 2. We don't set the keyboard layout properly yet and just hardcode it to
> 'us'. This will require the same kind of mapping as Fedora 17 and older.
>
> * http://projects.gnome.org/gdm/docs/2.14/configuration.html
> ---
> data/install-scripts/Makefile.am | 1 +
> data/install-scripts/rhel.xml | 204 +++++++++++++++++++++++++++++++++++++++
> data/oses/rhel.xml.in | 20 ++++
> 3 files changed, 225 insertions(+)
> create mode 100644 data/install-scripts/rhel.xml
>
> diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
> index cb24e09..c310468 100644
> --- a/data/install-scripts/Makefile.am
> +++ b/data/install-scripts/Makefile.am
> @@ -2,6 +2,7 @@
> databasedir = $(pkgdatadir)/db/install-scripts/
> database_DATA = \
> fedora.xml \
> + rhel.xml \
> windows-sif.xml \
> windows-cmd.xml \
> windows-reg.xml \
> diff --git a/data/install-scripts/rhel.xml b/data/install-scripts/rhel.xml
> new file mode 100644
> index 0000000..9c17f11
> --- /dev/null
> +++ b/data/install-scripts/rhel.xml
> @@ -0,0 +1,204 @@
> +
> +
> +
> + jeos
> + fedora.ks
> +
> +
> +
> +
> +
> +
> +
> +
> + + xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
> + version="1.0">
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + /dev/vda
> +
> +
> +
> + /dev/sda
> +
> +
> +
> +
> +
> +# Install script for profile
> +install
> +text
> +
> +keyboard us
> +lang
> +skipx
> +network --device eth0 --bootproto dhcp
> +rootpw
> +firewall --disabled
> +authconfig --enableshadow --enablemd5
> +selinux --enforcing
> +timezone --utc
> +bootloader --location=mbr
> +zerombr
> +
> +clearpart --all --drives=
According with: https://bugzilla.gnome.org/show_bug.cgi?id=676537#c12
JeOS scripts should be identical to scripts provided by:
https://github.com/clalancette/oz/tree/master/oz/auto
So, if the script works properly without "--drives=...", remove it
from JeOS profile.
ACK, otherwise.
> +
> +part /boot --fstype ext4 --size=200 --ondisk=
> +part pv.2 --size=1 --grow --ondisk=
> +volgroup VolGroup00 --pesize=32768 pv.2
> +logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 --size=768 --grow --maxsize=1536
> +logvol / --fstype ext4 --name=LogVol00 --vgname=VolGroup00 --size=1024 --grow
> +reboot
> +
> +%packages
> +
> +
> + at base
> +
> +
> + at standard
> +
> +
> + at core
> +
> +%end
> +
> +
> +
> +
> +
> +
> +
> + desktop
> + fedora.ks
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + image/png
> +
> +
> + + xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
> + version="1.0">
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + /dev/vda
> +
> +
> +
> + /dev/sda
> +
> +
> +
> +
> +
> +# Install script for profile
> +install
> +keyboard us
> +lang
> +network --onboot yes --device eth0 --bootproto dhcp --noipv6 --hostname= --activate
> +rootpw dummyPa55w0rd # Actual password set (or unset) in %post below
> +firewall --disabled
> +authconfig --enableshadow --enablemd5
> +timezone --utc
> +bootloader --location=mbr
> +zerombr
> +
> +clearpart --all --drives=
> +
> +firstboot --disable
> +
> +part /boot --fstype ext4 --size=200 --ondisk=
> +part pv.2 --size=1 --grow --ondisk=
> +volgroup VolGroup00 --pesize=32768 pv.2
> +logvol swap --fstype swap --name=LogVol01 --vgname=VolGroup00 --size=768 --grow --maxsize=1536
> +logvol / --fstype ext4 --name=LogVol00 --vgname=VolGroup00 --size=1024 --grow
> +reboot
> +
> +%packages
> + at core
> + at x11
> + at basic-desktop
> + at general-desktop
> + at desktop-platform
> + at internet-applications
> + at internet-browser
> + at input-methods
> + at office-suite
> +
> +%end
> +
> +%post --erroronfail
> +
> +useradd -G wheel # Add user
> +if test -z ; then
> + passwd -d # Make user account passwordless
> +else
> + echo |passwd --stdin
> +fi
> +
> +if test -z ; then
> + passwd -d root # Make root account passwordless
> +else
> + echo |passwd --stdin root
> +fi
> +
> +# Set user avatar
> +if test -n ; then
> +mkdir /mnt/unattended-media
> +mount /mnt/unattended-media
> +cp /mnt/unattended-media /home//.face
> +umount /mnt/unattended-media
> +fi
> +
> +# Enable autologin
> +echo "[daemon]
> +AutomaticLoginEnable=true
> +AutomaticLogin=
> +
> +[security]
> +
> +[xdmcp]
> +
> +[greeter]
> +
> +[chooser]
> +
> +[debug]
> +" > /etc/gdm/custom.conf
> +
> +%end
> +
> +
> +
> +
> +
> +
> diff --git a/data/oses/rhel.xml.in b/data/oses/rhel.xml.in
> index 1b34704..7e0db5b 100644
> --- a/data/oses/rhel.xml.in
> +++ b/data/oses/rhel.xml.in
> @@ -708,6 +708,11 @@
> 9663676416
>
>
> +
> +
> +
> +
> +
>
>
>
> @@ -753,6 +758,11 @@
> 9663676416
>
>
> +
> +
> +
> +
> +
>
>
>
> @@ -798,6 +808,11 @@
> 9663676416
>
>
> +
> +
> +
> +
> +
>
>
>
> @@ -843,5 +858,10 @@
> 9663676416
>
>
> +
> +
> +
> +
> +
>
>
> --
> 1.8.0.2
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
Best Regards,
--
Fabiano Fid?ncio
From zeeshanak at gnome.org Wed Jan 9 00:27:40 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 02:27:40 +0200
Subject: [Libosinfo] [PATCH v2] Add install script for RHEL 6.x
In-Reply-To:
References:
<1357687400-853-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Wed, Jan 9, 2013 at 1:43 AM, Fabiano Fid?ncio wrote:
> On Tue, Jan 8, 2013 at 9:23 PM, Zeeshan Ali (Khattak)
> wrote:
>> From: "Zeeshan Ali (Khattak)"
>>
>> The same script might work for other RHEL 6.x too but I have only tested
>> it against RHEL 6.3 for now. Still some issues though:
>>
>> 1. For some reason user's avatar is ignored although according to the docs
>> I could find*, copying the avatar to /home/${USER}/.face should be
>> enough.
>>
>> 2. We don't set the keyboard layout properly yet and just hardcode it to
>> 'us'. This will require the same kind of mapping as Fedora 17 and older.
>>
>> * http://projects.gnome.org/gdm/docs/2.14/configuration.html
>> ---
>> data/install-scripts/Makefile.am | 1 +
>> data/install-scripts/rhel.xml | 204 +++++++++++++++++++++++++++++++++++++++
>> data/oses/rhel.xml.in | 20 ++++
>> 3 files changed, 225 insertions(+)
>> create mode 100644 data/install-scripts/rhel.xml
>>
>> diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
>> index cb24e09..c310468 100644
>> --- a/data/install-scripts/Makefile.am
>> +++ b/data/install-scripts/Makefile.am
>> @@ -2,6 +2,7 @@
>> databasedir = $(pkgdatadir)/db/install-scripts/
>> database_DATA = \
>> fedora.xml \
>> + rhel.xml \
>> windows-sif.xml \
>> windows-cmd.xml \
>> windows-reg.xml \
>> diff --git a/data/install-scripts/rhel.xml b/data/install-scripts/rhel.xml
>> new file mode 100644
>> index 0000000..9c17f11
>> --- /dev/null
>> +++ b/data/install-scripts/rhel.xml
>> @@ -0,0 +1,204 @@
>> +
>> +
>> +
>> + jeos
>> + fedora.ks
>> +
>> +
>> +
>> +
>> +
>> +
>> +
>> +
>> + > + xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>> + version="1.0">
>> +
>> +
>> +
>> +
>> +
>> +
>> +
>> +
>> +
>> +
>> + /dev/vda
>> +
>> +
>> +
>> + /dev/sda
>> +
>> +
>> +
>> +
>> +
>> +# Install script for profile
>> +install
>> +text
>> +
>> +keyboard us
>> +lang
>> +skipx
>> +network --device eth0 --bootproto dhcp
>> +rootpw
>> +firewall --disabled
>> +authconfig --enableshadow --enablemd5
>> +selinux --enforcing
>> +timezone --utc
>> +bootloader --location=mbr
>> +zerombr
>> +
>> +clearpart --all --drives=
>
> According with: https://bugzilla.gnome.org/show_bug.cgi?id=676537#c12
> JeOS scripts should be identical to scripts provided by:
> https://github.com/clalancette/oz/tree/master/oz/auto
>
> So, if the script works properly without "--drives=...", remove it
> from JeOS profile.
Rationale?
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From fabiano at fidencio.org Wed Jan 9 00:44:20 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Tue, 8 Jan 2013 22:44:20 -0200
Subject: [Libosinfo] [PATCH v2] Add install script for RHEL 6.x
In-Reply-To:
References:
<1357687400-853-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Tue, Jan 8, 2013 at 10:27 PM, Zeeshan Ali (Khattak)
wrote:
> On Wed, Jan 9, 2013 at 1:43 AM, Fabiano Fid?ncio wrote:
>> On Tue, Jan 8, 2013 at 9:23 PM, Zeeshan Ali (Khattak)
>> wrote:
>>> From: "Zeeshan Ali (Khattak)"
>>>
>>> The same script might work for other RHEL 6.x too but I have only tested
>>> it against RHEL 6.3 for now. Still some issues though:
>>>
>>> 1. For some reason user's avatar is ignored although according to the docs
>>> I could find*, copying the avatar to /home/${USER}/.face should be
>>> enough.
>>>
>>> 2. We don't set the keyboard layout properly yet and just hardcode it to
>>> 'us'. This will require the same kind of mapping as Fedora 17 and older.
>>>
>>> * http://projects.gnome.org/gdm/docs/2.14/configuration.html
>>> ---
>>> data/install-scripts/Makefile.am | 1 +
>>> data/install-scripts/rhel.xml | 204 +++++++++++++++++++++++++++++++++++++++
>>> data/oses/rhel.xml.in | 20 ++++
>>> 3 files changed, 225 insertions(+)
>>> create mode 100644 data/install-scripts/rhel.xml
>>>
>>> diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
>>> index cb24e09..c310468 100644
>>> --- a/data/install-scripts/Makefile.am
>>> +++ b/data/install-scripts/Makefile.am
>>> @@ -2,6 +2,7 @@
>>> databasedir = $(pkgdatadir)/db/install-scripts/
>>> database_DATA = \
>>> fedora.xml \
>>> + rhel.xml \
>>> windows-sif.xml \
>>> windows-cmd.xml \
>>> windows-reg.xml \
>>> diff --git a/data/install-scripts/rhel.xml b/data/install-scripts/rhel.xml
>>> new file mode 100644
>>> index 0000000..9c17f11
>>> --- /dev/null
>>> +++ b/data/install-scripts/rhel.xml
>>> @@ -0,0 +1,204 @@
>>> +
>>> +
>>> +
>>> + jeos
>>> + fedora.ks
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +
>>> + >> + xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>>> + version="1.0">
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +
>>> + /dev/vda
>>> +
>>> +
>>> +
>>> + /dev/sda
>>> +
>>> +
>>> +
>>> +
>>> +
>>> +# Install script for profile
>>> +install
>>> +text
>>> +
>>> +keyboard us
>>> +lang
>>> +skipx
>>> +network --device eth0 --bootproto dhcp
>>> +rootpw
>>> +firewall --disabled
>>> +authconfig --enableshadow --enablemd5
>>> +selinux --enforcing
>>> +timezone --utc
>>> +bootloader --location=mbr
>>> +zerombr
>>> +
>>> +clearpart --all --drives=
>>
>> According with: https://bugzilla.gnome.org/show_bug.cgi?id=676537#c12
>> JeOS scripts should be identical to scripts provided by:
>> https://github.com/clalancette/oz/tree/master/oz/auto
>>
>> So, if the script works properly without "--drives=...", remove it
>> from JeOS profile.
>
> Rationale?
Again, "According with: https://bugzilla.gnome.org/show_bug.cgi?id=676537#c12"
IMHO, go ahead with your patch. *If* needed, we can change this in the
future, once the script was tested and is working (consider this an
ACK).
>
> --
> Regards,
>
> Zeeshan Ali (Khattak)
> FSF member#5124
Best Regards,
--
Fabiano Fid?ncio
From fabiano at fidencio.org Wed Jan 9 01:07:15 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Tue, 8 Jan 2013 23:07:15 -0200
Subject: [Libosinfo] [PATCH] rhel: Specify resources for all RHEL 6
releases
In-Reply-To: <1356805897-23400-1-git-send-email-zeeshanak@gnome.org>
References: <1356615528-9641-1-git-send-email-zeeshanak@gnome.org>
<1356805897-23400-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Sat, Dec 29, 2012 at 4:31 PM, Zeeshan Ali (Khattak)
wrote:
> From: "Zeeshan Ali (Khattak)"
>
> ---
> data/oses/rhel.xml.in | 39 +++++++++++++++++++++++++++++++++++++++
> 1 file changed, 39 insertions(+)
>
> diff --git a/data/oses/rhel.xml.in b/data/oses/rhel.xml.in
> index f2d3736..1b34704 100644
> --- a/data/oses/rhel.xml.in
> +++ b/data/oses/rhel.xml.in
> @@ -695,6 +695,19 @@
> isolinux/vmlinuz
> isolinux/initrd.img
>
> +
> +
> +
> + 1
> + 536870912
> +
> +
> +
> + 400000000
> + 1073741824
> + 9663676416
> +
> +
>
>
>
> @@ -727,6 +740,19 @@
> isolinux/vmlinuz
> isolinux/initrd.img
>
> +
> +
> +
> + 1
> + 536870912
> +
> +
> +
> + 400000000
> + 1073741824
> + 9663676416
> +
> +
>
>
>
> @@ -759,6 +785,19 @@
> isolinux/vmlinuz
> isolinux/initrd.img
>
> +
> +
> +
> + 1
> + 536870912
> +
> +
> +
> + 400000000
> + 1073741824
> + 9663676416
> +
> +
>
>
>
> --
> 1.8.0.2
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
As suggested by Daniel in OpenSUSE's patch, if you can put the
reference on the top of this file, just to remember where to look in
the future.
ACK!
--
Fabiano Fid?ncio
From zeeshanak at gnome.org Wed Jan 9 00:52:45 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 02:52:45 +0200
Subject: [Libosinfo] [PATCH v2] Add install script for RHEL 6.x
In-Reply-To:
References:
<1357687400-853-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Wed, Jan 9, 2013 at 2:44 AM, Fabiano Fid?ncio wrote:
> On Tue, Jan 8, 2013 at 10:27 PM, Zeeshan Ali (Khattak)
> wrote:
>> On Wed, Jan 9, 2013 at 1:43 AM, Fabiano Fid?ncio wrote:
>>> On Tue, Jan 8, 2013 at 9:23 PM, Zeeshan Ali (Khattak)
>>> wrote:
>>>> From: "Zeeshan Ali (Khattak)"
>>>>
>>>> The same script might work for other RHEL 6.x too but I have only tested
>>>> it against RHEL 6.3 for now. Still some issues though:
>>>>
>>>> 1. For some reason user's avatar is ignored although according to the docs
>>>> I could find*, copying the avatar to /home/${USER}/.face should be
>>>> enough.
>>>>
>>>> 2. We don't set the keyboard layout properly yet and just hardcode it to
>>>> 'us'. This will require the same kind of mapping as Fedora 17 and older.
>>>>
>>>> * http://projects.gnome.org/gdm/docs/2.14/configuration.html
>>>> ---
>>>> data/install-scripts/Makefile.am | 1 +
>>>> data/install-scripts/rhel.xml | 204 +++++++++++++++++++++++++++++++++++++++
>>>> data/oses/rhel.xml.in | 20 ++++
>>>> 3 files changed, 225 insertions(+)
>>>> create mode 100644 data/install-scripts/rhel.xml
>>>>
>>>> diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
>>>> index cb24e09..c310468 100644
>>>> --- a/data/install-scripts/Makefile.am
>>>> +++ b/data/install-scripts/Makefile.am
>>>> @@ -2,6 +2,7 @@
>>>> databasedir = $(pkgdatadir)/db/install-scripts/
>>>> database_DATA = \
>>>> fedora.xml \
>>>> + rhel.xml \
>>>> windows-sif.xml \
>>>> windows-cmd.xml \
>>>> windows-reg.xml \
>>>> diff --git a/data/install-scripts/rhel.xml b/data/install-scripts/rhel.xml
>>>> new file mode 100644
>>>> index 0000000..9c17f11
>>>> --- /dev/null
>>>> +++ b/data/install-scripts/rhel.xml
>>>> @@ -0,0 +1,204 @@
>>>> +
>>>> +
>>>> +
>>>> + jeos
>>>> + fedora.ks
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> + >>> + xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
>>>> + version="1.0">
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> + /dev/vda
>>>> +
>>>> +
>>>> +
>>>> + /dev/sda
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +
>>>> +# Install script for profile
>>>> +install
>>>> +text
>>>> +
>>>> +keyboard us
>>>> +lang
>>>> +skipx
>>>> +network --device eth0 --bootproto dhcp
>>>> +rootpw
>>>> +firewall --disabled
>>>> +authconfig --enableshadow --enablemd5
>>>> +selinux --enforcing
>>>> +timezone --utc
>>>> +bootloader --location=mbr
>>>> +zerombr
>>>> +
>>>> +clearpart --all --drives=
>>>
>>> According with: https://bugzilla.gnome.org/show_bug.cgi?id=676537#c12
>>> JeOS scripts should be identical to scripts provided by:
>>> https://github.com/clalancette/oz/tree/master/oz/auto
>>>
>>> So, if the script works properly without "--drives=...", remove it
>>> from JeOS profile.
>>
>> Rationale?
>
> Again, "According with: https://bugzilla.gnome.org/show_bug.cgi?id=676537#c12"
Ouch, I missed the link. In that context, I think Daniel was trying to
explain the difference and more appropriate word would have been "very
similar". Daniel can clarify if I got it wrong.
> IMHO, go ahead with your patch. *If* needed, we can change this in the
> future, once the script was tested and is working (consider this an
> ACK).
Yeah, even if I'm wrong about what Daniel meant, its not like our
other jeos scripts are identical to that of Oz's either.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 9 01:19:08 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 03:19:08 +0200
Subject: [Libosinfo] [PATCH v2] win8: Add install script
In-Reply-To: <1356639790-1437-2-git-send-email-zeeshanak@gnome.org>
References: <1356639790-1437-2-git-send-email-zeeshanak@gnome.org>
Message-ID: <1357694349-5833-1-git-send-email-zeeshanak@gnome.org>
Rebased on top of current master to make use of datamaps for language setting, just like win7.
From zeeshanak at gnome.org Wed Jan 9 01:19:09 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 03:19:09 +0200
Subject: [Libosinfo] [PATCH v2] win8: Add install script
In-Reply-To: <1357694349-5833-1-git-send-email-zeeshanak@gnome.org>
References: <1356639790-1437-2-git-send-email-zeeshanak@gnome.org>
<1357694349-5833-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1357694349-5833-2-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Using the desktop profile, I was able to do an express installation
against GNOME Boxes.
I'm adding a seperate script for windows 8 rather than re-using the
existing unattended.xml we have for windows 7 because windows 8 requires
product key to be specified while it remains optional for windows 7.
---
data/install-scripts/Makefile.am | 1 +
data/install-scripts/windows-8-unattend.xml | 419 ++++++++++++++++++++++++++++
data/oses/windows.xml.in | 3 +-
3 files changed, 422 insertions(+), 1 deletion(-)
create mode 100644 data/install-scripts/windows-8-unattend.xml
diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
index cb24e09..29715d2 100644
--- a/data/install-scripts/Makefile.am
+++ b/data/install-scripts/Makefile.am
@@ -6,6 +6,7 @@ database_DATA = \
windows-cmd.xml \
windows-reg.xml \
windows-unattend.xml \
+ windows-8-unattend.xml \
$(NULL)
EXTRA_DIST = $(database_DATA)
diff --git a/data/install-scripts/windows-8-unattend.xml b/data/install-scripts/windows-8-unattend.xml
new file mode 100644
index 0000000..cc6472b
--- /dev/null
+++ b/data/install-scripts/windows-8-unattend.xml
@@ -0,0 +1,419 @@
+
+
+ jeos
+ dos
+ $$$$$-$$$$$-$$$$$-$$$$$-$$$$$
+ autounattend.xml
+ true
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ C
+
+
+
+
+
+
+
+
+
+
+ x86
+
+
+
+
+
+
+
+
+
+
+ A
+
+
+
+
+
+
+
+
+
+
+ \
+
+
+
+
+
+
+
+
+
+
+
+
+
+ :
+
+
+
+
+
+
+
+
+ OnError
+
+
+
+ 1
+ 1
+ Primary
+
+
+ 0
+ true
+
+
+ true
+ true
+ NTFS
+
+
+ 1
+ 1
+
+
+
+
+
+
+
+ 0
+ 1
+
+ OnError
+
+
+
+ true
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ true
+
+
+
+
+
+
+
+ false
+
+
+
+ users
+
+
+
+
+
+
+
+
+ true
+
+ true
+ 5
+
+
+
+
+ Work
+ true
+ 3
+ true
+ true
+
+
+
+ 1
+ Turn Off Network Selection pop-up
+ cmd /c reg add "HKLM\SYSTEM\CurrentControlSet\Control\Network\NewNetworkWindowOff"
+
+
+ 2
+ Shutting down Windows
+ cmd /C shutdown /s /t 0
+
+
+
+
+
+
+
+
+
+
+
+
+ desktop
+ dos
+ $$$$$-$$$$$-$$$$$-$$$$$-$$$$$
+ autounattend.xml
+ true
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ C
+
+
+
+
+
+
+
+ amd64
+
+
+ x86
+
+
+ x86
+
+
+
+
+
+
+
+
+
+
+ A
+
+
+
+
+
+
+
+
+
+
+ \
+
+
+
+
+
+
+
+
+
+
+
+
+
+ :
+
+
+
+
+
+
+
+
+ OnError
+
+
+
+ 1
+ 1
+ Primary
+
+
+ 0
+ true
+
+
+ true
+ true
+ NTFS
+
+
+ 1
+ 1
+
+
+
+
+
+
+
+ 0
+ 1
+
+ OnError
+
+
+
+ true
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ true
+
+
+
+
+
+
+ true
+
+
+
+ Administrators;Power Users
+
+
+
+
+
+
+
+
+
+ true
+
+ true
+ 5
+
+
+
+
+
+ Work
+ true
+ true
+
+ true
+ 3
+
+
+
+
+ 1
+ Turn Off Network Selection pop-up
+ cmd /c reg add "HKLM\SYSTEM\CurrentControlSet\Control\Network\NewNetworkWindowOff"
+
+
+
+
+
+
+
+
+
+
+
diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
index 310de62..364fcaa 100644
--- a/data/oses/windows.xml.in
+++ b/data/oses/windows.xml.in
@@ -822,7 +822,8 @@
-
+
+
--
1.8.0.2
From zeeshanak at gnome.org Wed Jan 9 01:22:54 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 03:22:54 +0200
Subject: [Libosinfo] [PATCH] rhel: Specify resources for all RHEL 6
releases
In-Reply-To:
References: <1356615528-9641-1-git-send-email-zeeshanak@gnome.org>
<1356805897-23400-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Wed, Jan 9, 2013 at 3:07 AM, Fabiano Fid?ncio wrote:
> On Sat, Dec 29, 2012 at 4:31 PM, Zeeshan Ali (Khattak)
> wrote:
>> From: "Zeeshan Ali (Khattak)"
>>
>> ---
>> data/oses/rhel.xml.in | 39 +++++++++++++++++++++++++++++++++++++++
>> 1 file changed, 39 insertions(+)
>>
>> diff --git a/data/oses/rhel.xml.in b/data/oses/rhel.xml.in
>> index f2d3736..1b34704 100644
>> --- a/data/oses/rhel.xml.in
>> +++ b/data/oses/rhel.xml.in
>> @@ -695,6 +695,19 @@
>> isolinux/vmlinuz
>> isolinux/initrd.img
>>
>> +
>> +
>> +
>> + 1
>> + 536870912
>> +
>> +
>> +
>> + 400000000
>> + 1073741824
>> + 9663676416
>> +
>> +
>>
>>
>>
>> @@ -727,6 +740,19 @@
>> isolinux/vmlinuz
>> isolinux/initrd.img
>>
>> +
>> +
>> +
>> + 1
>> + 536870912
>> +
>> +
>> +
>> + 400000000
>> + 1073741824
>> + 9663676416
>> +
>> +
>>
>>
>>
>> @@ -759,6 +785,19 @@
>> isolinux/vmlinuz
>> isolinux/initrd.img
>>
>> +
>> +
>> +
>> + 1
>> + 536870912
>> +
>> +
>> +
>> + 400000000
>> + 1073741824
>> + 9663676416
>> +
>> +
>>
>>
>>
>> --
>> 1.8.0.2
>>
>> _______________________________________________
>> Libosinfo mailing list
>> Libosinfo at redhat.com
>> https://www.redhat.com/mailman/listinfo/libosinfo
>
> As suggested by Daniel in OpenSUSE's patch, if you can put the
> reference on the top of this file, just to remember where to look in
> the future.
Yikes, forgot about this before pushing the two RHEL patches. :(
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From fabiano at fidencio.org Wed Jan 9 01:33:10 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Tue, 8 Jan 2013 23:33:10 -0200
Subject: [Libosinfo] [PATCH v2] win8: Add install script
In-Reply-To: <1357694349-5833-2-git-send-email-zeeshanak@gnome.org>
References: <1356639790-1437-2-git-send-email-zeeshanak@gnome.org>
<1357694349-5833-1-git-send-email-zeeshanak@gnome.org>
<1357694349-5833-2-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Tue, Jan 8, 2013 at 11:19 PM, Zeeshan Ali (Khattak)
wrote:
> From: "Zeeshan Ali (Khattak)"
>
> Using the desktop profile, I was able to do an express installation
> against GNOME Boxes.
>
> I'm adding a seperate script for windows 8 rather than re-using the
> existing unattended.xml we have for windows 7 because windows 8 requires
> product key to be specified while it remains optional for windows 7.
ACK!
> ---
> data/install-scripts/Makefile.am | 1 +
> data/install-scripts/windows-8-unattend.xml | 419 ++++++++++++++++++++++++++++
> data/oses/windows.xml.in | 3 +-
> 3 files changed, 422 insertions(+), 1 deletion(-)
> create mode 100644 data/install-scripts/windows-8-unattend.xml
>
> diff --git a/data/install-scripts/Makefile.am b/data/install-scripts/Makefile.am
> index cb24e09..29715d2 100644
> --- a/data/install-scripts/Makefile.am
> +++ b/data/install-scripts/Makefile.am
> @@ -6,6 +6,7 @@ database_DATA = \
> windows-cmd.xml \
> windows-reg.xml \
> windows-unattend.xml \
> + windows-8-unattend.xml \
> $(NULL)
>
> EXTRA_DIST = $(database_DATA)
> diff --git a/data/install-scripts/windows-8-unattend.xml b/data/install-scripts/windows-8-unattend.xml
> new file mode 100644
> index 0000000..cc6472b
> --- /dev/null
> +++ b/data/install-scripts/windows-8-unattend.xml
> @@ -0,0 +1,419 @@
> +
> +
> + jeos
> + dos
> + $$$$$-$$$$$-$$$$$-$$$$$-$$$$$
> + autounattend.xml
> + true
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + + xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
> + version="1.0">
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + C
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + x86
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + A
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + \
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + :
> +
> +
> +
> +
> +
> +
> +
> +
> + OnError
> +
> +
> +
> + 1
> + 1
> + Primary
> +
> +
> + 0
> + true
> +
> +
> + true
> + true
> + NTFS
> +
> +
> + 1
> + 1
> +
> +
> +
> +
> +
> +
> +
> + 0
> + 1
> +
> + OnError
> +
> +
> +
> + true
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + true
> +
> +
> +
> +
> +
> +
> +
> + false
> +
> +
> +
> + users
> +
> +
> +
> +
> +
> +
> +
> +
> + true
> +
> + true
> + 5
> +
> +
> +
> +
> + Work
> + true
> + 3
> + true
> + true
> +
> +
> +
> + 1
> + Turn Off Network Selection pop-up
> + cmd /c reg add "HKLM\SYSTEM\CurrentControlSet\Control\Network\NewNetworkWindowOff"
> +
> +
> + 2
> + Shutting down Windows
> + cmd /C shutdown /s /t 0
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + desktop
> + dos
> + $$$$$-$$$$$-$$$$$-$$$$$-$$$$$
> + autounattend.xml
> + true
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + + xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
> + version="1.0">
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + C
> +
> +
> +
> +
> +
> +
> +
> + amd64
> +
> +
> + x86
> +
> +
> + x86
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + A
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + \
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + :
> +
> +
> +
> +
> +
> +
> +
> +
> + OnError
> +
> +
> +
> + 1
> + 1
> + Primary
> +
> +
> + 0
> + true
> +
> +
> + true
> + true
> + NTFS
> +
> +
> + 1
> + 1
> +
> +
> +
> +
> +
> +
> +
> + 0
> + 1
> +
> + OnError
> +
> +
> +
> + true
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + true
> +
> +
> +
> +
> +
> +
> + true
> +
> +
> +
> + Administrators;Power Users
> +
> +
> +
> +
> +
> +
> +
> +
> +
> + true
> +
> + true
> + 5
> +
> +
> +
> +
> +
> + Work
> + true
> + true
> +
> + true
> + 3
> +
> +
> +
> +
> + 1
> + Turn Off Network Selection pop-up
> + cmd /c reg add "HKLM\SYSTEM\CurrentControlSet\Control\Network\NewNetworkWindowOff"
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
> index 310de62..364fcaa 100644
> --- a/data/oses/windows.xml.in
> +++ b/data/oses/windows.xml.in
> @@ -822,7 +822,8 @@
>
>
>
> -
> +
> +
>
>
>
> --
> 1.8.0.2
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
--
Fabiano Fid?ncio
From cfergeau at redhat.com Wed Jan 9 10:49:26 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 9 Jan 2013 05:49:26 -0500 (EST)
Subject: [Libosinfo] [PATCHv4 06/11] Add
OsinfoInstallConfig:config-params property
In-Reply-To:
Message-ID: <1168163406.1831361.1357728566356.JavaMail.root@redhat.com>
Hey,
----- ???????? ????????? -----
> On Thu, Dec 20, 2012 at 6:45 PM, Christophe Fergeau
> wrote:
> > This property lists the parameters that can be set for a given
> > OsinfoInstallConfig. This is not enforced, it's only there for
> > informative purpose. This will also be used in later commits
> > in order to automatically apply transformations on values
> > for parameters which have an associated OsinfoDatamap.
>
> Sorry to revive this thread again but I had an idea that I wanted to
> discuss before this API goes into a release: How about we have a
> OsinfoInstallConfig:install-script (of type OsinfoInstallScript)
> rather than having OsinfoInstallConfig:config-params (and
> 'path-format', 'avatar-format' etc) duplicated on both objects?
I don't understand your 'path-format', 'avatar-format', ... comment. The OsinfoInstallConfig::config-params property appears on both OsinfoInstallConfig and OsinfoInstallScript, but the actual OsinfoInstallConfigParams instance is shared between these 2 classes, so not much duplication either.
>IMO
> that makes a lot more sense since that not only avoids duplication of
> API (and some strings) but also makes things more clear: Whether a
> config is associated/specific to a script or not? If it is, which
> install script is exactly its tied to?
In my opinion, this is completely backward. When you have an OsinfoInstallScript, it's interesting to know the OsinfoInstallConfig
that is set on it. When you have an OsinfoInstallConfig instance, it's interesting to know the schema that is valid for this config, ie to have access to the OsinfoInstallConfigParams associated with it. If anything, I'd add an OsinfoInstallScript::install-config property of type OsinfoInstallConfig, which becomes a bit harder after your suggested change if you want to avoid cyclic references.
As I see it, OsinfoInstallScript ties an OsinfoInstallConfig with an install script template, and when filling OsinfoInstallConfig, you should not need to care at all about the other stuff that OsinfoInstallScript can do, all the info you need should be in OsinfoInstallConfigParams.
So -1 on this change from me.
Christophe
From berrange at redhat.com Wed Jan 9 11:45:05 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 11:45:05 +0000
Subject: [Libosinfo] [PATCH] Add ability to list install script config
params & profiles
Message-ID: <1357731905-7118-1-git-send-email-berrange@redhat.com>
From: "Daniel P. Berrange"
When using 'osinfo-install-script' it is useful to know what
profiles and config parameters are available. Add two new
flags --list-configs and --list-profiles to output this
information
# osinfo-install-script --list-profiles win7
desktop: autounattend.xml
jeos: autounattend.xml
# osinfo-install-script --list-config winxp
admin-password: optional
user-realname: required
reg-product-key: required
Signed-off-by: Daniel P. Berrange
---
tools/osinfo-install-script.c | 112 ++++++++++++++++++++++++++++++++++++++++--
1 file changed, 108 insertions(+), 4 deletions(-)
diff --git a/tools/osinfo-install-script.c b/tools/osinfo-install-script.c
index 2311e68..7cdb842 100644
--- a/tools/osinfo-install-script.c
+++ b/tools/osinfo-install-script.c
@@ -28,10 +28,13 @@
#include
#include
-static const gchar *profile;
+static const gchar *profile = "jeos";
static const gchar *output_dir;
static const gchar *prefix;
+static gboolean list_config = FALSE;
+static gboolean list_profile = FALSE;
+
static OsinfoInstallConfig *config;
static gboolean handle_config(const gchar *option_name G_GNUC_UNUSED,
@@ -71,6 +74,10 @@ static GOptionEntry entries[] =
{ "config", 'c', 0, G_OPTION_ARG_CALLBACK,
handle_config,
N_("Set configuration parameter"), "key=value" },
+ { "list-config", '\0', 0, G_OPTION_ARG_NONE, (void*)&list_config,
+ N_("List configuration parameters"), NULL },
+ { "list-profiles", '\0', 0, G_OPTION_ARG_NONE, (void*)&list_profile,
+ N_("List install script profiles"), NULL },
{ NULL }
};
@@ -108,6 +115,76 @@ static OsinfoOs *find_os(OsinfoDb *db,
}
+static gboolean list_script_config(OsinfoOs *os)
+{
+ OsinfoInstallScriptList *scripts = osinfo_os_get_install_script_list(os);
+ OsinfoInstallScriptList *profile_scripts;
+ OsinfoFilter *filter;
+ GList *l, *tmp;
+ gboolean ret = FALSE;
+
+ filter = osinfo_filter_new();
+ osinfo_filter_add_constraint(filter,
+ OSINFO_INSTALL_SCRIPT_PROP_PROFILE,
+ profile ? profile :
+ OSINFO_INSTALL_SCRIPT_PROFILE_JEOS);
+ profile_scripts = OSINFO_INSTALL_SCRIPTLIST(osinfo_list_new_filtered(OSINFO_LIST(scripts),
+ filter));
+ l = osinfo_list_get_elements(OSINFO_LIST(profile_scripts));
+ if (!l) {
+ g_printerr(_("No install script for profile '%s' and OS '%s'"),
+ profile, osinfo_product_get_name(OSINFO_PRODUCT(os)));
+ goto cleanup;
+ }
+
+ for (tmp = l; tmp != NULL; tmp = tmp->next) {
+ OsinfoInstallScript *script = tmp->data;
+ GList *params = osinfo_install_script_get_config_param_list(script);
+ GList *tmp2;
+
+ for (tmp2 = params ; tmp2 != NULL ; tmp2 = tmp2->next) {
+ OsinfoInstallConfigParam *param = OSINFO_INSTALL_CONFIG_PARAM(tmp2->data);
+
+ g_print("%s: %s\n",
+ osinfo_install_config_param_get_name(param),
+ osinfo_install_config_param_is_required(param) ?
+ _("required") : _("optional"));
+ }
+ }
+ ret = TRUE;
+
+ cleanup:
+ g_list_free(l);
+ g_object_unref(scripts);
+ g_object_unref(filter);
+ g_object_unref(profile_scripts);
+ return ret;
+}
+
+
+static gboolean list_script_profile(OsinfoOs *os)
+{
+ OsinfoInstallScriptList *scripts = osinfo_os_get_install_script_list(os);
+ GList *l, *tmp;
+ gboolean ret = FALSE;
+
+ l = osinfo_list_get_elements(OSINFO_LIST(scripts));
+
+ for (tmp = l; tmp != NULL; tmp = tmp->next) {
+ OsinfoInstallScript *script = tmp->data;
+
+ g_print("%s: %s\n",
+ osinfo_install_script_get_profile(script),
+ osinfo_install_script_get_expected_filename(script));
+ }
+ ret = TRUE;
+
+ g_list_free(l);
+ g_object_unref(scripts);
+ return ret;
+}
+
+
static gboolean generate_script(OsinfoOs *os)
{
OsinfoInstallScriptList *scripts = osinfo_os_get_install_script_list(os);
@@ -125,6 +202,13 @@ static gboolean generate_script(OsinfoOs *os)
profile_scripts = OSINFO_INSTALL_SCRIPTLIST(osinfo_list_new_filtered(OSINFO_LIST(scripts),
filter));
l = osinfo_list_get_elements(OSINFO_LIST(profile_scripts));
+
+ if (!l) {
+ g_printerr(_("No install script for profile '%s' and OS '%s'"),
+ profile, osinfo_product_get_name(OSINFO_PRODUCT(os)));
+ goto cleanup;
+ }
+
for (tmp = l; tmp != NULL; tmp = tmp->next) {
OsinfoInstallScript *script = tmp->data;
GFile *dir = g_file_new_for_commandline_arg(output_dir ?
@@ -191,6 +275,14 @@ gint main(gint argc, gchar **argv)
goto EXIT;
}
+ if (list_profile && list_config) {
+ g_printerr("%s",
+ _("Only one of --list-propfile and --list-config can be requested"));
+ ret = -2;
+ goto EXIT;
+ }
+
+
loader = osinfo_loader_new();
osinfo_loader_process_default_path(loader, &error);
if (error != NULL) {
@@ -208,9 +300,21 @@ gint main(gint argc, gchar **argv)
goto EXIT;
}
- if (!generate_script(os)) {
- ret = -5;
- goto EXIT;
+ if (list_config) {
+ if (!list_script_config(os)) {
+ ret = -5;
+ goto EXIT;
+ }
+ } else if (list_profile) {
+ if (!list_script_profile(os)) {
+ ret = -5;
+ goto EXIT;
+ }
+ } else {
+ if (!generate_script(os)) {
+ ret = -5;
+ goto EXIT;
+ }
}
EXIT:
--
1.8.1
From berrange at redhat.com Wed Jan 9 12:37:43 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 12:37:43 +0000
Subject: [Libosinfo] [PATCH 0/6] Simplify application of datamaps
Message-ID: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Zeeshan pointed me at this thread
https://www.redhat.com/archives/libosinfo/2012-December/msg00174.html
asking whether I thought it is better for the OsinfoInstallConfig
class to have a 'config-params' property (as per that patch), or
a more general 'install-script' property.
Of course I thought it should have neither! IMHO the whole design is
overly convoluted. The InstallScript class is creating a copy of the
InstallConfig object instance, passing in the InstallScript so that
the new InstallConfig object can have a 'config-params' set associated
with it. This is so that the InstalLScript can then use that to apply
datamap translations.
This is all complete overkill - the InstallScript class already has
info about the datamaps and can apply them to the original InstallConfig
object instance it has, without needing to create a throwaway copy.
This series reverts the last 5 patches in that quoted series and
applies one simpler patch at the end which does the same job without
needing to directly associate the InstallConfig & InstallScript
classes at all.
From berrange at redhat.com Wed Jan 9 12:37:44 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 12:37:44 +0000
Subject: [Libosinfo] [PATCH 1/6] Revert "Use OS-specific config in
OsinfoInstallScript"
In-Reply-To: <1357735069-29235-1-git-send-email-berrange@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Message-ID: <1357735069-29235-2-git-send-email-berrange@redhat.com>
From: "Daniel P. Berrange"
This reverts commit 99beb54c7b9a0cfa6285fb2988ed93c406c452a3.
---
osinfo/osinfo_install_script.c | 10 ++--------
1 file changed, 2 insertions(+), 8 deletions(-)
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index 70d6b1d..0591708 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -637,15 +637,9 @@ static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallConf
tmp1 = keys = osinfo_entity_get_param_keys(entity);
while (tmp1) {
- GList *values;
- GList *tmp2;
-
- if (OSINFO_IS_INSTALL_CONFIG(entity))
- values = osinfo_install_config_get_param_value_list(OSINFO_INSTALL_CONFIG(entity), tmp1->data);
- else
- values = osinfo_entity_get_param_value_list(entity, tmp1->data);
+ GList *values = osinfo_entity_get_param_value_list(entity, tmp1->data);
+ GList *tmp2 = values;
- tmp2 = values;
while (tmp2) {
if (!(data = xmlNewDocNode(NULL, NULL, (const xmlChar*)tmp1->data,
(const xmlChar*)tmp2->data))) {
--
1.8.1
From berrange at redhat.com Wed Jan 9 12:37:45 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 12:37:45 +0000
Subject: [Libosinfo] [PATCH 2/6] Revert "Set
OsinfoInstallConfig:config-params in
osinfo_install_script_apply_template"
In-Reply-To: <1357735069-29235-1-git-send-email-berrange@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Message-ID: <1357735069-29235-3-git-send-email-berrange@redhat.com>
From: "Daniel P. Berrange"
This reverts commit 04f95d473865ed257a4d98a9f0ca17cdee3900f3.
---
osinfo/osinfo_install_script.c | 32 +-------------------------------
1 file changed, 1 insertion(+), 31 deletions(-)
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index 0591708..b82c0f9 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -30,7 +30,6 @@
#include
#include
#include
-#include "osinfo_install_config_private.h"
#include "osinfo_install_script_private.h"
G_DEFINE_TYPE (OsinfoInstallScript, osinfo_install_script, OSINFO_TYPE_ENTITY);
@@ -759,33 +758,6 @@ static gchar *osinfo_install_script_apply_xslt(xsltStylesheetPtr ss,
}
-static OsinfoInstallConfig *create_config_for_script(OsinfoInstallScript *script, OsinfoInstallConfig *config)
-{
- OsinfoInstallConfig *os_config;
- GList *params;
- GList *param;
- const gchar *id;
-
- id = osinfo_entity_get_id(OSINFO_ENTITY(config));
- os_config = osinfo_install_config_new_for_script(id, script);
- params = osinfo_entity_get_param_keys(OSINFO_ENTITY(config));
- for (param = params; param != NULL; param = param->next) {
- GList *values;
- GList *value;
-
- osinfo_entity_clear_param(OSINFO_ENTITY(os_config), param->data);
- values = osinfo_entity_get_param_value_list(OSINFO_ENTITY(config), param->data);
- for (value = values; value != NULL; value = value->next) {
- osinfo_entity_add_param(OSINFO_ENTITY(os_config), param->data, value->data);
- }
- g_list_free(values);
- }
- g_list_free(params);
-
- return os_config;
-}
-
-
static gboolean osinfo_install_script_apply_template(OsinfoInstallScript *script,
OsinfoOs *os,
const gchar *templateUri,
@@ -795,10 +767,8 @@ static gboolean osinfo_install_script_apply_template(OsinfoInstallScript *script
GError **error)
{
gboolean ret = FALSE;
- OsinfoInstallConfig *os_config = create_config_for_script(script, config);
xsltStylesheetPtr templateXsl = osinfo_install_script_load_template(templateUri, template, error);
- xmlDocPtr configXml = osinfo_install_script_generate_config_xml(script, os, os_config, error);
- g_object_unref(G_OBJECT(os_config));
+ xmlDocPtr configXml = osinfo_install_script_generate_config_xml(script, os, config, error);
if (!templateXsl || !configXml)
goto cleanup;
--
1.8.1
From berrange at redhat.com Wed Jan 9 12:37:46 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 12:37:46 +0000
Subject: [Libosinfo] [PATCH 3/6] Revert "Add
osinfo_install_config_new_for_script"
In-Reply-To: <1357735069-29235-1-git-send-email-berrange@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Message-ID: <1357735069-29235-4-git-send-email-berrange@redhat.com>
From: "Daniel P. Berrange"
This reverts commit 82784b5f446e2b0cc47704df6ebddff804b4b667.
---
osinfo/libosinfo.syms | 1 -
osinfo/osinfo_install_config.c | 31 -------------------------------
osinfo/osinfo_install_config.h | 4 ----
3 files changed, 36 deletions(-)
diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index efea371..a32a8b7 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -388,7 +388,6 @@ LIBOSINFO_0.2.3 {
osinfo_db_get_datamap_list;
osinfo_db_identify_media;
- osinfo_install_config_new_for_script;
osinfo_install_config_get_config_params;
osinfo_install_config_paramlist_get_type;
diff --git a/osinfo/osinfo_install_config.c b/osinfo/osinfo_install_config.c
index 8572e1a..a650014 100644
--- a/osinfo/osinfo_install_config.c
+++ b/osinfo/osinfo_install_config.c
@@ -188,39 +188,8 @@ osinfo_install_config_init (OsinfoInstallConfig *config)
*/
OsinfoInstallConfig *osinfo_install_config_new(const gchar *id)
{
- g_return_val_if_fail(id != NULL, NULL);
-
- return g_object_new(OSINFO_TYPE_INSTALL_CONFIG, "id", id, NULL);
-}
-
-
-/**
- * osinfo_install_config_new_for_script:
- * @id: the unique identifier
- * @script: the #OsinfoInstallScript we are creating the configuration for
- *
- * Construct a new install configuration associated with @script.
- * OsinfoInstallConfig:config-params will contain the
- * #OsinfoInstallConfigParamList describing the parameters that can be set
- * on the config object when creating a configuration for @script. See
- * osinfo_install_config_new() for a description of the default values that
- * will be set on the newly created #OsinfoInstallConfig.
- *
- * Returns: (transfer full): an install configuration
- */
-OsinfoInstallConfig *osinfo_install_config_new_for_script(const gchar *id,
- OsinfoInstallScript *script)
-{
- OsinfoInstallConfigParamList *params;
-
- g_return_val_if_fail(id != NULL, NULL);
- g_return_val_if_fail(script != NULL, NULL);
-
- params = osinfo_install_script_get_config_params(script);
-
return g_object_new(OSINFO_TYPE_INSTALL_CONFIG,
"id", id,
- "config-params", params,
NULL);
}
diff --git a/osinfo/osinfo_install_config.h b/osinfo/osinfo_install_config.h
index 112860b..caf5518 100644
--- a/osinfo/osinfo_install_config.h
+++ b/osinfo/osinfo_install_config.h
@@ -72,8 +72,6 @@ typedef struct _OsinfoInstallConfig OsinfoInstallConfig;
typedef struct _OsinfoInstallConfigClass OsinfoInstallConfigClass;
typedef struct _OsinfoInstallConfigPrivate OsinfoInstallConfigPrivate;
-#include
-
/* object */
struct _OsinfoInstallConfig
{
@@ -97,8 +95,6 @@ struct _OsinfoInstallConfigClass
GType osinfo_install_config_get_type(void);
OsinfoInstallConfig *osinfo_install_config_new(const gchar *id);
-OsinfoInstallConfig *osinfo_install_config_new_for_script(const gchar *id,
- OsinfoInstallScript *script);
void osinfo_install_config_set_hardware_arch(OsinfoInstallConfig *config,
const gchar *arch);
--
1.8.1
From berrange at redhat.com Wed Jan 9 12:37:47 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 12:37:47 +0000
Subject: [Libosinfo] [PATCH 4/6] Revert "OsinfoInstallConfig: Use
config-params if set"
In-Reply-To: <1357735069-29235-1-git-send-email-berrange@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Message-ID: <1357735069-29235-5-git-send-email-berrange@redhat.com>
From: "Daniel P. Berrange"
This reverts commit 72f00ce4673d1f3cc18f8dddd73f75a2a05a306a.
---
osinfo/osinfo_install_config.c | 50 ----------------------------------
osinfo/osinfo_install_config_private.h | 1 -
2 files changed, 51 deletions(-)
diff --git a/osinfo/osinfo_install_config.c b/osinfo/osinfo_install_config.c
index a650014..d395ee9 100644
--- a/osinfo/osinfo_install_config.c
+++ b/osinfo/osinfo_install_config.c
@@ -736,56 +736,6 @@ OsinfoInstallConfigParamList *osinfo_install_config_get_config_params(OsinfoInst
return config->priv->config_params;
}
-
-static OsinfoDatamap *
-osinfo_install_config_get_param_datamap(OsinfoInstallConfig *config,
- const gchar *param_name)
-{
- OsinfoEntity *entity;
- OsinfoInstallConfigParam *param;
-
- if (!config->priv->config_params)
- return NULL;
-
- entity = osinfo_list_find_by_id(OSINFO_LIST(config->priv->config_params),
- param_name);
- if (entity == NULL) {
- g_debug("%s is not a known parameter for this config", param_name);
- return NULL;
- }
-
- param = OSINFO_INSTALL_CONFIG_PARAM(entity);;
- return osinfo_install_config_param_get_value_map(param);
-}
-
-
-GList *
-osinfo_install_config_get_param_value_list(OsinfoInstallConfig *config,
- const gchar *key)
-{
- GList *values;
- GList *it;
- OsinfoDatamap *map;
-
- values = osinfo_entity_get_param_value_list(OSINFO_ENTITY(config), key);
- if (values == NULL)
- return NULL;
-
- map = osinfo_install_config_get_param_datamap(config, key);
- if (map != NULL) {
- for (it = values; it != NULL; it = it->next) {
- const char *transformed_value;
- transformed_value = osinfo_datamap_lookup(map, it->data);
- if (transformed_value == NULL) {
- continue;
- }
- it->data = (gpointer)transformed_value;
- }
- }
-
- return values;
-}
-
/*
* Local variables:
* indent-tabs-mode: nil
diff --git a/osinfo/osinfo_install_config_private.h b/osinfo/osinfo_install_config_private.h
index 5ad2162..5a1edd3 100644
--- a/osinfo/osinfo_install_config_private.h
+++ b/osinfo/osinfo_install_config_private.h
@@ -28,7 +28,6 @@
void osinfo_install_config_set_config_params(OsinfoInstallConfig *config,
OsinfoInstallConfigParamList *config_params);
-GList *osinfo_install_config_get_param_value_list(OsinfoInstallConfig *config, const gchar *key);
#endif /* __OSINFO_INSTALL_CONFIG_PRIVATE_H__ */
/*
--
1.8.1
From berrange at redhat.com Wed Jan 9 12:37:48 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 12:37:48 +0000
Subject: [Libosinfo] [PATCH 5/6] Revert "Add
OsinfoInstallConfig:config-params property"
In-Reply-To: <1357735069-29235-1-git-send-email-berrange@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Message-ID: <1357735069-29235-6-git-send-email-berrange@redhat.com>
From: "Daniel P. Berrange"
This reverts commit 36e5364c1a3e67f8656bfd2db570717b5d2e6e15.
---
osinfo/Makefile.am | 1 -
osinfo/libosinfo.syms | 2 -
osinfo/osinfo_install_config.c | 96 +---------------------------------
osinfo/osinfo_install_config.h | 5 --
osinfo/osinfo_install_config_private.h | 39 --------------
5 files changed, 1 insertion(+), 142 deletions(-)
delete mode 100644 osinfo/osinfo_install_config_private.h
diff --git a/osinfo/Makefile.am b/osinfo/Makefile.am
index 9f3c20d..9d03a34 100644
--- a/osinfo/Makefile.am
+++ b/osinfo/Makefile.am
@@ -116,7 +116,6 @@ libosinfo_1_0_la_SOURCES = \
osinfo_install_config.c \
osinfo_install_config_param.c \
osinfo_install_config_paramlist.c \
- osinfo_install_config_private.h \
osinfo_install_script.c \
osinfo_install_script_private.h \
osinfo_install_scriptlist.c \
diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index a32a8b7..4615829 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -388,8 +388,6 @@ LIBOSINFO_0.2.3 {
osinfo_db_get_datamap_list;
osinfo_db_identify_media;
- osinfo_install_config_get_config_params;
-
osinfo_install_config_paramlist_get_type;
osinfo_install_config_paramlist_new;
diff --git a/osinfo/osinfo_install_config.c b/osinfo/osinfo_install_config.c
index d395ee9..93dc039 100644
--- a/osinfo/osinfo_install_config.c
+++ b/osinfo/osinfo_install_config.c
@@ -24,7 +24,6 @@
#include
#include
-#include "osinfo/osinfo_install_config_private.h"
#include
G_DEFINE_TYPE (OsinfoInstallConfig, osinfo_install_config, OSINFO_TYPE_ENTITY);
@@ -43,91 +42,14 @@ G_DEFINE_TYPE (OsinfoInstallConfig, osinfo_install_config, OSINFO_TYPE_ENTITY);
struct _OsinfoInstallConfigPrivate
{
- OsinfoInstallConfigParamList *config_params;
+ gboolean unused;
};
-enum {
- PROP_0,
-
- PROP_CONFIG_PARAMS,
-};
-
-static void
-osinfo_install_config_set_property(GObject *object,
- guint property_id,
- const GValue *value,
- GParamSpec *pspec)
-{
- OsinfoInstallConfig *config = OSINFO_INSTALL_CONFIG(object);
-
- switch (property_id) {
- case PROP_CONFIG_PARAMS:
- osinfo_install_config_set_config_params(config, g_value_get_object(value));
- break;
-
- default:
- /* We don't have any other property... */
- G_OBJECT_WARN_INVALID_PROPERTY_ID (object, property_id, pspec);
- break;
- }
-}
-
-static void
-osinfo_install_config_get_property(GObject *object,
- guint property_id,
- GValue *value,
- GParamSpec *pspec)
-{
- OsinfoInstallConfig *config = OSINFO_INSTALL_CONFIG(object);
-
- switch (property_id) {
- case PROP_CONFIG_PARAMS:
- g_value_set_object(value, osinfo_install_config_get_config_params(config));
- break;
-
- default:
- /* We don't have any other property... */
- G_OBJECT_WARN_INVALID_PROPERTY_ID (object, property_id, pspec);
- break;
- }
-}
-
-
-static void
-osinfo_install_config_finalize (GObject *object)
-{
- OsinfoInstallConfig *config = OSINFO_INSTALL_CONFIG (object);
-
- if (config->priv->config_params)
- g_object_unref(config->priv->config_params);
-
- /* Chain up to the parent class */
- G_OBJECT_CLASS (osinfo_install_config_parent_class)->finalize (object);
-}
-
/* Init functions */
static void
osinfo_install_config_class_init (OsinfoInstallConfigClass *klass)
{
- GObjectClass *g_klass = G_OBJECT_CLASS (klass);
- GParamSpec *pspec;
-
- g_klass->get_property = osinfo_install_config_get_property;
- g_klass->set_property = osinfo_install_config_set_property;
- g_klass->finalize = osinfo_install_config_finalize;
-
- pspec = g_param_spec_object("config-params",
- "Config Parameters",
- _("Valid configuration parameters"),
- OSINFO_TYPE_INSTALL_CONFIG_PARAMLIST,
- G_PARAM_READWRITE |
- G_PARAM_CONSTRUCT_ONLY |
- G_PARAM_STATIC_STRINGS);
- g_object_class_install_property(g_klass,
- PROP_CONFIG_PARAMS,
- pspec);
-
g_type_class_add_private (klass, sizeof (OsinfoInstallConfigPrivate));
}
@@ -720,22 +642,6 @@ const gchar *osinfo_install_config_get_post_install_drivers_location(OsinfoInsta
OSINFO_INSTALL_CONFIG_PROP_POST_INSTALL_DRIVERS_LOCATION);
}
-void osinfo_install_config_set_config_params(OsinfoInstallConfig *config,
- OsinfoInstallConfigParamList *config_params)
-{
- if (config->priv->config_params != NULL)
- g_object_unref(config->priv->config_params);
- if (config_params != NULL)
- config->priv->config_params = g_object_ref(G_OBJECT(config_params));
- else
- config->priv->config_params = NULL;
-}
-
-OsinfoInstallConfigParamList *osinfo_install_config_get_config_params(OsinfoInstallConfig *config)
-{
- return config->priv->config_params;
-}
-
/*
* Local variables:
* indent-tabs-mode: nil
diff --git a/osinfo/osinfo_install_config.h b/osinfo/osinfo_install_config.h
index caf5518..d650a0a 100644
--- a/osinfo/osinfo_install_config.h
+++ b/osinfo/osinfo_install_config.h
@@ -22,7 +22,6 @@
*/
#include
-#include
#ifndef __OSINFO_INSTALL_CONFIG_H__
#define __OSINFO_INSTALL_CONFIG_H__
@@ -183,7 +182,6 @@ const gchar *osinfo_install_config_get_avatar_disk(OsinfoInstallConfig *config);
void osinfo_install_config_set_pre_install_drivers_disk(OsinfoInstallConfig *config,
const gchar *disk);
const gchar *osinfo_install_config_get_pre_install_drivers_disk(OsinfoInstallConfig *config);
-
void osinfo_install_config_set_pre_install_drivers_location(OsinfoInstallConfig *config,
const gchar *location);
const gchar *osinfo_install_config_get_pre_install_drivers_location(OsinfoInstallConfig *config);
@@ -191,13 +189,10 @@ const gchar *osinfo_install_config_get_pre_install_drivers_location(OsinfoInstal
void osinfo_install_config_set_post_install_drivers_disk(OsinfoInstallConfig *config,
const gchar *disk);
const gchar *osinfo_install_config_get_post_install_drivers_disk(OsinfoInstallConfig *config);
-
void osinfo_install_config_set_post_install_drivers_location(OsinfoInstallConfig *config,
const gchar *location);
const gchar *osinfo_install_config_get_post_install_drivers_location(OsinfoInstallConfig *config);
-OsinfoInstallConfigParamList *osinfo_install_config_get_config_params(OsinfoInstallConfig *config);
-
#endif /* __OSINFO_INSTALL_CONFIG_H__ */
/*
* Local variables:
diff --git a/osinfo/osinfo_install_config_private.h b/osinfo/osinfo_install_config_private.h
deleted file mode 100644
index 5a1edd3..0000000
--- a/osinfo/osinfo_install_config_private.h
+++ /dev/null
@@ -1,39 +0,0 @@
-/*
- * libosinfo: OS installation config
- *
- * Copyright (C) 2012 Red Hat, Inc.
- *
- * This library is free software; you can redistribute it and/or
- * modify it under the terms of the GNU Lesser General Public
- * License as published by the Free Software Foundation; either
- * version 2.1 of the License, or (at your option) any later version.
- *
- * This library is distributed in the hope that it will be useful,
- * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * Lesser General Public License for more details.
- *
- * You should have received a copy of the GNU Lesser General Public
- * License along with this library; if not, write to the Free Software
- * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
- *
- * Authors:
- * Christophe Fergeau
- */
-
-#include
-
-#ifndef __OSINFO_INSTALL_CONFIG_PRIVATE_H__
-#define __OSINFO_INSTALL_CONFIG_PRIVATE_H__
-
-void osinfo_install_config_set_config_params(OsinfoInstallConfig *config,
- OsinfoInstallConfigParamList *config_params);
-
-#endif /* __OSINFO_INSTALL_CONFIG_PRIVATE_H__ */
-/*
- * Local variables:
- * indent-tabs-mode: nil
- * c-indent-level: 4
- * c-basic-offset: 4
- * End:
- */
--
1.8.1
From berrange at redhat.com Wed Jan 9 12:37:49 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 12:37:49 +0000
Subject: [Libosinfo] [PATCH 6/6] Apply datamap to config parameters when
generating install script
In-Reply-To: <1357735069-29235-1-git-send-email-berrange@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Message-ID: <1357735069-29235-7-git-send-email-berrange@redhat.com>
From: "Daniel P. Berrange"
When creating the XML to use in the install script XSL transform,
apply any datamap associated with the config parameters.
---
osinfo/osinfo_install_script.c | 72 ++++++++++++++++++++++++++++++++++++++----
1 file changed, 66 insertions(+), 6 deletions(-)
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index b82c0f9..47be667 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -603,7 +603,59 @@ static xsltStylesheetPtr osinfo_install_script_load_template(const gchar *uri,
return xslt;
}
-static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallConfig *config,
+
+static OsinfoDatamap *
+osinfo_install_script_get_param_datamap(OsinfoInstallScript *script,
+ const gchar *param_name)
+{
+ OsinfoEntity *entity;
+ OsinfoInstallConfigParam *param;
+
+ if (!script->priv->config_params)
+ return NULL;
+
+ entity = osinfo_list_find_by_id(OSINFO_LIST(script->priv->config_params),
+ param_name);
+ if (entity == NULL) {
+ g_debug("%s is not a known parameter for this config", param_name);
+ return NULL;
+ }
+
+ param = OSINFO_INSTALL_CONFIG_PARAM(entity);
+ return osinfo_install_config_param_get_value_map(param);
+}
+
+
+static GList *
+osinfo_install_script_get_param_value_list(OsinfoInstallScript *script,
+ OsinfoInstallConfig *config,
+ const gchar *key)
+{
+ GList *values;
+ GList *it;
+ OsinfoDatamap *map;
+
+ values = osinfo_entity_get_param_value_list(OSINFO_ENTITY(config), key);
+ if (values == NULL)
+ return NULL;
+
+ map = osinfo_install_script_get_param_datamap(script, key);
+ if (map != NULL) {
+ for (it = values; it != NULL; it = it->next) {
+ const char *transformed_value;
+ transformed_value = osinfo_datamap_lookup(map, it->data);
+ if (transformed_value == NULL) {
+ continue;
+ }
+ it->data = (gpointer)transformed_value;
+ }
+ }
+
+ return values;
+}
+
+
+static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallScript *script,
OsinfoEntity *entity,
const gchar *name,
GError **error)
@@ -636,9 +688,17 @@ static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallConf
tmp1 = keys = osinfo_entity_get_param_keys(entity);
while (tmp1) {
- GList *values = osinfo_entity_get_param_value_list(entity, tmp1->data);
- GList *tmp2 = values;
+ GList *values;
+ GList *tmp2;
+
+ if (OSINFO_IS_INSTALL_CONFIG(entity))
+ values = osinfo_install_script_get_param_value_list(script,
+ OSINFO_INSTALL_CONFIG(entity),
+ tmp1->data);
+ else
+ values = osinfo_entity_get_param_value_list(entity, tmp1->data);
+ tmp2 = values;
while (tmp2) {
if (!(data = xmlNewDocNode(NULL, NULL, (const xmlChar*)tmp1->data,
(const xmlChar*)tmp2->data))) {
@@ -686,7 +746,7 @@ static xmlDocPtr osinfo_install_script_generate_config_xml(OsinfoInstallScript *
NULL);
xmlDocSetRootElement(doc, root);
- if (!(node = osinfo_install_script_generate_entity_config(config,
+ if (!(node = osinfo_install_script_generate_entity_config(script,
OSINFO_ENTITY(script),
"script",
error)))
@@ -697,7 +757,7 @@ static xmlDocPtr osinfo_install_script_generate_config_xml(OsinfoInstallScript *
goto error;
}
- if (!(node = osinfo_install_script_generate_entity_config(config,
+ if (!(node = osinfo_install_script_generate_entity_config(script,
OSINFO_ENTITY(os),
"os",
error)))
@@ -708,7 +768,7 @@ static xmlDocPtr osinfo_install_script_generate_config_xml(OsinfoInstallScript *
goto error;
}
- if (!(node = osinfo_install_script_generate_entity_config(config,
+ if (!(node = osinfo_install_script_generate_entity_config(script,
OSINFO_ENTITY(config),
"config",
error)))
--
1.8.1
From zeeshanak at gnome.org Wed Jan 9 14:21:23 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 16:21:23 +0200
Subject: [Libosinfo] [PATCHv4 06/11] Add
OsinfoInstallConfig:config-params property
In-Reply-To: <1168163406.1831361.1357728566356.JavaMail.root@redhat.com>
References:
<1168163406.1831361.1357728566356.JavaMail.root@redhat.com>
Message-ID:
On Wed, Jan 9, 2013 at 12:49 PM, Christophe Fergeau wrote:
> Hey,
>
> ----- ???????? ????????? -----
>> On Thu, Dec 20, 2012 at 6:45 PM, Christophe Fergeau
>> wrote:
>> > This property lists the parameters that can be set for a given
>> > OsinfoInstallConfig. This is not enforced, it's only there for
>> > informative purpose. This will also be used in later commits
>> > in order to automatically apply transformations on values
>> > for parameters which have an associated OsinfoDatamap.
>>
>> Sorry to revive this thread again but I had an idea that I wanted to
>> discuss before this API goes into a release: How about we have a
>> OsinfoInstallConfig:install-script (of type OsinfoInstallScript)
>> rather than having OsinfoInstallConfig:config-params (and
>> 'path-format', 'avatar-format' etc) duplicated on both objects?
>
> I don't understand your 'path-format', 'avatar-format', ... comment. The OsinfoInstallConfig::config-params property appears on both OsinfoInstallConfig and OsinfoInstallScript, but the actual OsinfoInstallConfigParams instance is shared between these 2 classes, so not much duplication either.
Actually it was you who mentioned 'path-format' on
OsinfoInstallConfig being a good idea and gave an example of how it
will be useful to Boxes in this thread and you presented that as an
argument in favor of your patch.
>>IMO
>> that makes a lot more sense since that not only avoids duplication of
>> API (and some strings) but also makes things more clear: Whether a
>> config is associated/specific to a script or not? If it is, which
>> install script is exactly its tied to?
>
> In my opinion, this is completely backward. When you have an OsinfoInstallScript, it's interesting to know the OsinfoInstallConfig
> that is set on it. When you have an OsinfoInstallConfig instance, it's interesting to know the schema that is valid for this config, ie to have access to the OsinfoInstallConfigParams associated with it. If anything, I'd add an OsinfoInstallScript::install-config property of type OsinfoInstallConfig, which becomes a bit harder after your suggested change if you want to avoid cyclic references.
>
> As I see it, OsinfoInstallScript ties an OsinfoInstallConfig with an install script template, and when filling OsinfoInstallConfig, you should not need to care at all about the other stuff that OsinfoInstallScript can do, all the info you need should be in OsinfoInstallConfigParams.
Well, I think Daniel got it sorted out now so we can stop arguing on
these obsolete patches. :)
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 9 14:24:06 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 16:24:06 +0200
Subject: [Libosinfo] [PATCH 0/6] Simplify application of datamaps
In-Reply-To: <1357735069-29235-1-git-send-email-berrange@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Message-ID:
On Wed, Jan 9, 2013 at 2:37 PM, Daniel P. Berrange wrote:
> Zeeshan pointed me at this thread
>
> https://www.redhat.com/archives/libosinfo/2012-December/msg00174.html
>
> asking whether I thought it is better for the OsinfoInstallConfig
> class to have a 'config-params' property (as per that patch), or
> a more general 'install-script' property.
>
> Of course I thought it should have neither! IMHO the whole design is
> overly convoluted. The InstallScript class is creating a copy of the
> InstallConfig object instance, passing in the InstallScript so that
> the new InstallConfig object can have a 'config-params' set associated
> with it. This is so that the InstalLScript can then use that to apply
> datamap translations.
>
> This is all complete overkill - the InstallScript class already has
> info about the datamaps and can apply them to the original InstallConfig
> object instance it has, without needing to create a throwaway copy.
Agreed.
> This series reverts the last 5 patches in that quoted series and
> applies one simpler patch at the end which does the same job without
> needing to directly associate the InstallConfig & InstallScript
> classes at all.
I'll look into your last patch but you can already push the reverting
patches now.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From cfergeau at redhat.com Wed Jan 9 15:47:16 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 9 Jan 2013 10:47:16 -0500 (EST)
Subject: [Libosinfo] [PATCH 0/6] Simplify application of datamaps
In-Reply-To:
Message-ID: <1611455453.2098149.1357746436797.JavaMail.root@redhat.com>
----- ???????? ????????? -----
> On Wed, Jan 9, 2013 at 2:37 PM, Daniel P. Berrange
> wrote:
> > Zeeshan pointed me at this thread
> >
> > https://www.redhat.com/archives/libosinfo/2012-December/msg00174.html
> >
> > asking whether I thought it is better for the OsinfoInstallConfig
> > class to have a 'config-params' property (as per that patch), or
> > a more general 'install-script' property.
> >
> > Of course I thought it should have neither! IMHO the whole design
> > is
> > overly convoluted. The InstallScript class is creating a copy of
> > the
> > InstallConfig object instance, passing in the InstallScript so that
> > the new InstallConfig object can have a 'config-params' set
> > associated
> > with it. This is so that the InstalLScript can then use that to
> > apply
> > datamap translations.
> >
> > This is all complete overkill - the InstallScript class already has
> > info about the datamaps and can apply them to the original
> > InstallConfig
> > object instance it has, without needing to create a throwaway copy.
>
> Agreed.
Well, you insisted on the creation of the throw away copy which was not there initially...
Christophe
From cfergeau at redhat.com Wed Jan 9 16:02:14 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 9 Jan 2013 11:02:14 -0500 (EST)
Subject: [Libosinfo] [PATCH 0/6] Simplify application of datamaps
In-Reply-To: <1357735069-29235-1-git-send-email-berrange@redhat.com>
Message-ID: <626761469.2108012.1357747334242.JavaMail.root@redhat.com>
----- ???????? ????????? -----
> This is all complete overkill - the InstallScript class already has
> info about the datamaps and can apply them to the original
> InstallConfig
> object instance it has, without needing to create a throwaway copy.
>
> This series reverts the last 5 patches in that quoted series and
> applies one simpler patch at the end which does the same job without
> needing to directly associate the InstallConfig & InstallScript
> classes at all.
While having an OsinfoInstallConfigParams property in the OsinfoInstallConfig class, it imo makes sense to have it as the disconnect between OsinfoInstallConfig and OsinfoInstallConfigParams is weird. It's more convenient to have the configuration schema (OsinfoInstallConfigParams) directly available while filling an OsinfoInstallConfig object, and would potentially allow in the future to check which parameters are allowed/disallowed/... when they are set.
Regarding the osinfo_install_script_get_param_value_list code, imo it fits better in OsinfoInstallConfig as a special getter similar to osinfo_entity_get_param_value_list, but with a specialized behaviour.
As for the throw-away copy, I'm also all for getting rid of it, it was added after I got an objection during review. The code was initially doing a straight g_object_set(config, "config-params", params, NULL); during the generation of the install script instead of a copy.
Christophe
From zeeshanak at gnome.org Wed Jan 9 16:03:17 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 18:03:17 +0200
Subject: [Libosinfo] [PATCH 0/6] Simplify application of datamaps
In-Reply-To: <1611455453.2098149.1357746436797.JavaMail.root@redhat.com>
References:
<1611455453.2098149.1357746436797.JavaMail.root@redhat.com>
Message-ID:
On Wed, Jan 9, 2013 at 5:47 PM, Christophe Fergeau wrote:
>
>
> ----- ???????? ????????? -----
>> On Wed, Jan 9, 2013 at 2:37 PM, Daniel P. Berrange
>> wrote:
>> > Zeeshan pointed me at this thread
>> >
>> > https://www.redhat.com/archives/libosinfo/2012-December/msg00174.html
>> >
>> > asking whether I thought it is better for the OsinfoInstallConfig
>> > class to have a 'config-params' property (as per that patch), or
>> > a more general 'install-script' property.
>> >
>> > Of course I thought it should have neither! IMHO the whole design
>> > is
>> > overly convoluted. The InstallScript class is creating a copy of
>> > the
>> > InstallConfig object instance, passing in the InstallScript so that
>> > the new InstallConfig object can have a 'config-params' set
>> > associated
>> > with it. This is so that the InstalLScript can then use that to
>> > apply
>> > datamap translations.
>> >
>> > This is all complete overkill - the InstallScript class already has
>> > info about the datamaps and can apply them to the original
>> > InstallConfig
>> > object instance it has, without needing to create a throwaway copy.
>>
>> Agreed.
>
> Well, you insisted on the creation of the throw away copy which was not there initially...
As alternative to your approach of having config-params on InstallConfig.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From berrange at redhat.com Wed Jan 9 16:10:41 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 9 Jan 2013 16:10:41 +0000
Subject: [Libosinfo] [PATCH 0/6] Simplify application of datamaps
In-Reply-To: <626761469.2108012.1357747334242.JavaMail.root@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
<626761469.2108012.1357747334242.JavaMail.root@redhat.com>
Message-ID: <20130109161041.GP20716@redhat.com>
On Wed, Jan 09, 2013 at 11:02:14AM -0500, Christophe Fergeau wrote:
>
>
> ----- ???????? ????????? -----
> > This is all complete overkill - the InstallScript class already has
> > info about the datamaps and can apply them to the original
> > InstallConfig
> > object instance it has, without needing to create a throwaway copy.
> >
> > This series reverts the last 5 patches in that quoted series and
> > applies one simpler patch at the end which does the same job without
> > needing to directly associate the InstallConfig & InstallScript
> > classes at all.
>
> While having an OsinfoInstallConfigParams property in the
> OsinfoInstallConfig class, it imo makes sense to have it as
> the disconnect between OsinfoInstallConfig and OsinfoInstallConfigParams
> is weird. It's more convenient to have the configuration schema
> (OsinfoInstallConfigParams) directly available while filling an
> OsinfoInstallConfig object, and would potentially allow in the
> future to check which parameters are allowed/disallowed/... when
> they are set.
If you're filling in a InstallConfig object you'll have to have
an InstallScript object around already, so you have the schema
available via that object if you really need it, without needing
to duplicate it in the InstallConfig object. I think it makes
the InstallConfig object design more futureproof if we don't
uneccessarily tie it to a InstallScript instance.
> As for the throw-away copy, I'm also all for getting rid of it, it
> was added after I got an objection during review. The code was initially
> doing a straight g_object_set(config, "config-params", params, NULL);
> during the generation of the install script instead of a copy.
I think that would be a bad - as a user of this API, I'd really not be
expecting it to modify the InstallConfig instance I pass in - it should
be considered 'const' IMHO.
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From zeeshanak at gnome.org Wed Jan 9 16:28:23 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 18:28:23 +0200
Subject: [Libosinfo] [PATCH 6/6] Apply datamap to config parameters when
generating install script
In-Reply-To: <1357735069-29235-7-git-send-email-berrange@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
<1357735069-29235-7-git-send-email-berrange@redhat.com>
Message-ID:
On Wed, Jan 9, 2013 at 2:37 PM, Daniel P. Berrange wrote:
> From: "Daniel P. Berrange"
>
> When creating the XML to use in the install script XSL transform,
> apply any datamap associated with the config parameters.
Looks good. Some minor concerns below. I assume you tested it?
> ---
> osinfo/osinfo_install_script.c | 72 ++++++++++++++++++++++++++++++++++++++----
> 1 file changed, 66 insertions(+), 6 deletions(-)
>
> diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
> index b82c0f9..47be667 100644
> --- a/osinfo/osinfo_install_script.c
> +++ b/osinfo/osinfo_install_script.c
> @@ -603,7 +603,59 @@ static xsltStylesheetPtr osinfo_install_script_load_template(const gchar *uri,
> return xslt;
> }
>
> -static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallConfig *config,
> +
> +static OsinfoDatamap *
> +osinfo_install_script_get_param_datamap(OsinfoInstallScript *script,
> + const gchar *param_name)
> +{
> + OsinfoEntity *entity;
> + OsinfoInstallConfigParam *param;
> +
> + if (!script->priv->config_params)
> + return NULL;
> +
> + entity = osinfo_list_find_by_id(OSINFO_LIST(script->priv->config_params),
> + param_name);
> + if (entity == NULL) {
> + g_debug("%s is not a known parameter for this config", param_name);
> + return NULL;
> + }
> +
> + param = OSINFO_INSTALL_CONFIG_PARAM(entity);
> + return osinfo_install_config_param_get_value_map(param);
> +}
> +
> +
> +static GList *
> +osinfo_install_script_get_param_value_list(OsinfoInstallScript *script,
> + OsinfoInstallConfig *config,
> + const gchar *key)
> +{
> + GList *values;
> + GList *it;
> + OsinfoDatamap *map;
> +
> + values = osinfo_entity_get_param_value_list(OSINFO_ENTITY(config), key);
> + if (values == NULL)
> + return NULL;
> +
> + map = osinfo_install_script_get_param_datamap(script, key);
> + if (map != NULL) {
> + for (it = values; it != NULL; it = it->next) {
> + const char *transformed_value;
> + transformed_value = osinfo_datamap_lookup(map, it->data);
> + if (transformed_value == NULL) {
> + continue;
> + }
> + it->data = (gpointer)transformed_value;
Is the cast really needed?
> + }
> + }
> +
> + return values;
> +}
> +
> +
> +static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallScript *script,
The name was already a bit misleading IMHO but now that it doesn't
take a 'config', its seems even more misleading. Could we name it to
osinfo_install_script_generate_entity_xml instead?
> OsinfoEntity *entity,
> const gchar *name,
> GError **error)
> @@ -636,9 +688,17 @@ static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallConf
>
> tmp1 = keys = osinfo_entity_get_param_keys(entity);
> while (tmp1) {
> - GList *values = osinfo_entity_get_param_value_list(entity, tmp1->data);
> - GList *tmp2 = values;
> + GList *values;
> + GList *tmp2;
> +
> + if (OSINFO_IS_INSTALL_CONFIG(entity))
> + values = osinfo_install_script_get_param_value_list(script,
> + OSINFO_INSTALL_CONFIG(entity),
> + tmp1->data);
> + else
> + values = osinfo_entity_get_param_value_list(entity, tmp1->data);
>
> + tmp2 = values;
> while (tmp2) {
> if (!(data = xmlNewDocNode(NULL, NULL, (const xmlChar*)tmp1->data,
> (const xmlChar*)tmp2->data))) {
> @@ -686,7 +746,7 @@ static xmlDocPtr osinfo_install_script_generate_config_xml(OsinfoInstallScript *
> NULL);
> xmlDocSetRootElement(doc, root);
>
> - if (!(node = osinfo_install_script_generate_entity_config(config,
> + if (!(node = osinfo_install_script_generate_entity_config(script,
> OSINFO_ENTITY(script),
> "script",
> error)))
> @@ -697,7 +757,7 @@ static xmlDocPtr osinfo_install_script_generate_config_xml(OsinfoInstallScript *
> goto error;
> }
>
> - if (!(node = osinfo_install_script_generate_entity_config(config,
> + if (!(node = osinfo_install_script_generate_entity_config(script,
> OSINFO_ENTITY(os),
> "os",
> error)))
> @@ -708,7 +768,7 @@ static xmlDocPtr osinfo_install_script_generate_config_xml(OsinfoInstallScript *
> goto error;
> }
>
> - if (!(node = osinfo_install_script_generate_entity_config(config,
> + if (!(node = osinfo_install_script_generate_entity_config(script,
> OSINFO_ENTITY(config),
> "config",
> error)))
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Wed Jan 9 16:31:46 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Wed, 9 Jan 2013 18:31:46 +0200
Subject: [Libosinfo] [PATCH 0/6] Simplify application of datamaps
In-Reply-To: <20130109161041.GP20716@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
<626761469.2108012.1357747334242.JavaMail.root@redhat.com>
<20130109161041.GP20716@redhat.com>
Message-ID:
On Wed, Jan 9, 2013 at 6:10 PM, Daniel P. Berrange wrote:
>> As for the throw-away copy, I'm also all for getting rid of it, it
>> was added after I got an objection during review. The code was initially
>> doing a straight g_object_set(config, "config-params", params, NULL);
>> during the generation of the install script instead of a copy.
>
> I think that would be a bad - as a user of this API, I'd really not be
> expecting it to modify the InstallConfig instance I pass in - it should
> be considered 'const' IMHO.
I fully agree. Hence the reason I suggested on the throw-away copy.
Probably didn't think of the fact that we can transform without doing
so.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From cfergeau at redhat.com Wed Jan 9 16:32:42 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 9 Jan 2013 11:32:42 -0500 (EST)
Subject: [Libosinfo] [PATCH 0/6] Simplify application of datamaps
In-Reply-To: <20130109161041.GP20716@redhat.com>
Message-ID: <1209478219.2153420.1357749162191.JavaMail.root@redhat.com>
----- ???????? ????????? -----
> On Wed, Jan 09, 2013 at 11:02:14AM -0500, Christophe Fergeau wrote:
> > While having an OsinfoInstallConfigParams property in the
> > OsinfoInstallConfig class, it imo makes sense to have it as
> > the disconnect between OsinfoInstallConfig and
> > OsinfoInstallConfigParams
> > is weird. It's more convenient to have the configuration schema
> > (OsinfoInstallConfigParams) directly available while filling an
> > OsinfoInstallConfig object, and would potentially allow in the
> > future to check which parameters are allowed/disallowed/... when
> > they are set.
>
> If you're filling in a InstallConfig object you'll have to have
> an InstallScript object around already, so you have the schema
> available via that object if you really need it, without needing
> to duplicate it in the InstallConfig object.
Yup, you can access it this way, but this means passing around 2 different objects (InstallConfig and InstallScript) to be able to access the InstallConfigParams data while you fill the InstallConfig object. When I first looked at that code, I was actually very confused to find no link at all in the API between OsinfoInstallConfig and OsinfoInstallConfigParams, contrary to what their very similar names could imply.
> I think it makes the InstallConfig object design more futureproof if we don't
> unnecessarily tie it to a InstallScript instance.
It's currently only tied to an OsinfoInstallConfigParams instance (which indeed indirectly links it to an InstallScript instance in the current code). But I'm fine with not adding this property until a clear need for it appears, if ever.
Christophe
From berrange at redhat.com Thu Jan 10 12:04:32 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Thu, 10 Jan 2013 12:04:32 +0000
Subject: [Libosinfo] [PATCH 6/6] Apply datamap to config parameters when
generating install script
In-Reply-To:
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
<1357735069-29235-7-git-send-email-berrange@redhat.com>
Message-ID: <20130110120432.GL6021@redhat.com>
On Wed, Jan 09, 2013 at 06:28:23PM +0200, Zeeshan Ali (Khattak) wrote:
> On Wed, Jan 9, 2013 at 2:37 PM, Daniel P. Berrange wrote:
> > From: "Daniel P. Berrange"
> >
> > When creating the XML to use in the install script XSL transform,
> > apply any datamap associated with the config parameters.
>
> Looks good. Some minor concerns below. I assume you tested it?
>
> > ---
> > osinfo/osinfo_install_script.c | 72 ++++++++++++++++++++++++++++++++++++++----
> > 1 file changed, 66 insertions(+), 6 deletions(-)
> >
> > diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
> > index b82c0f9..47be667 100644
> > --- a/osinfo/osinfo_install_script.c
> > +++ b/osinfo/osinfo_install_script.c
> > @@ -603,7 +603,59 @@ static xsltStylesheetPtr osinfo_install_script_load_template(const gchar *uri,
> > return xslt;
> > }
> >
> > -static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallConfig *config,
> > +
> > +static OsinfoDatamap *
> > +osinfo_install_script_get_param_datamap(OsinfoInstallScript *script,
> > + const gchar *param_name)
> > +{
> > + OsinfoEntity *entity;
> > + OsinfoInstallConfigParam *param;
> > +
> > + if (!script->priv->config_params)
> > + return NULL;
> > +
> > + entity = osinfo_list_find_by_id(OSINFO_LIST(script->priv->config_params),
> > + param_name);
> > + if (entity == NULL) {
> > + g_debug("%s is not a known parameter for this config", param_name);
> > + return NULL;
> > + }
> > +
> > + param = OSINFO_INSTALL_CONFIG_PARAM(entity);
> > + return osinfo_install_config_param_get_value_map(param);
> > +}
> > +
> > +
> > +static GList *
> > +osinfo_install_script_get_param_value_list(OsinfoInstallScript *script,
> > + OsinfoInstallConfig *config,
> > + const gchar *key)
> > +{
> > + GList *values;
> > + GList *it;
> > + OsinfoDatamap *map;
> > +
> > + values = osinfo_entity_get_param_value_list(OSINFO_ENTITY(config), key);
> > + if (values == NULL)
> > + return NULL;
> > +
> > + map = osinfo_install_script_get_param_datamap(script, key);
> > + if (map != NULL) {
> > + for (it = values; it != NULL; it = it->next) {
> > + const char *transformed_value;
> > + transformed_value = osinfo_datamap_lookup(map, it->data);
> > + if (transformed_value == NULL) {
> > + continue;
> > + }
> > + it->data = (gpointer)transformed_value;
>
> Is the cast really needed?
Shouldn't be - anything casts to/from void * in C.
> > + }
> > + }
> > +
> > + return values;
> > +}
> > +
> > +
> > +static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallScript *script,
>
> The name was already a bit misleading IMHO but now that it doesn't
> take a 'config', its seems even more misleading. Could we name it to
> osinfo_install_script_generate_entity_xml instead?
Ok
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From berrange at redhat.com Thu Jan 10 12:07:13 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Thu, 10 Jan 2013 12:07:13 +0000
Subject: [Libosinfo] [PATCH 6/6] Apply datamap to config parameters when
generating install script
In-Reply-To: <20130110120432.GL6021@redhat.com>
References: <1357735069-29235-1-git-send-email-berrange@redhat.com>
<1357735069-29235-7-git-send-email-berrange@redhat.com>
<20130110120432.GL6021@redhat.com>
Message-ID: <20130110120713.GM6021@redhat.com>
On Thu, Jan 10, 2013 at 12:04:32PM +0000, Daniel P. Berrange wrote:
> On Wed, Jan 09, 2013 at 06:28:23PM +0200, Zeeshan Ali (Khattak) wrote:
> > On Wed, Jan 9, 2013 at 2:37 PM, Daniel P. Berrange wrote:
> > > From: "Daniel P. Berrange"
> > >
> > > When creating the XML to use in the install script XSL transform,
> > > apply any datamap associated with the config parameters.
> >
> > Looks good. Some minor concerns below. I assume you tested it?
> >
> > > ---
> > > osinfo/osinfo_install_script.c | 72 ++++++++++++++++++++++++++++++++++++++----
> > > 1 file changed, 66 insertions(+), 6 deletions(-)
> > >
> > > diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
> > > index b82c0f9..47be667 100644
> > > --- a/osinfo/osinfo_install_script.c
> > > +++ b/osinfo/osinfo_install_script.c
> > > @@ -603,7 +603,59 @@ static xsltStylesheetPtr osinfo_install_script_load_template(const gchar *uri,
> > > return xslt;
> > > }
> > >
> > > -static xmlNodePtr osinfo_install_script_generate_entity_config(OsinfoInstallConfig *config,
> > > +
> > > +static OsinfoDatamap *
> > > +osinfo_install_script_get_param_datamap(OsinfoInstallScript *script,
> > > + const gchar *param_name)
> > > +{
> > > + OsinfoEntity *entity;
> > > + OsinfoInstallConfigParam *param;
> > > +
> > > + if (!script->priv->config_params)
> > > + return NULL;
> > > +
> > > + entity = osinfo_list_find_by_id(OSINFO_LIST(script->priv->config_params),
> > > + param_name);
> > > + if (entity == NULL) {
> > > + g_debug("%s is not a known parameter for this config", param_name);
> > > + return NULL;
> > > + }
> > > +
> > > + param = OSINFO_INSTALL_CONFIG_PARAM(entity);
> > > + return osinfo_install_config_param_get_value_map(param);
> > > +}
> > > +
> > > +
> > > +static GList *
> > > +osinfo_install_script_get_param_value_list(OsinfoInstallScript *script,
> > > + OsinfoInstallConfig *config,
> > > + const gchar *key)
> > > +{
> > > + GList *values;
> > > + GList *it;
> > > + OsinfoDatamap *map;
> > > +
> > > + values = osinfo_entity_get_param_value_list(OSINFO_ENTITY(config), key);
> > > + if (values == NULL)
> > > + return NULL;
> > > +
> > > + map = osinfo_install_script_get_param_datamap(script, key);
> > > + if (map != NULL) {
> > > + for (it = values; it != NULL; it = it->next) {
> > > + const char *transformed_value;
> > > + transformed_value = osinfo_datamap_lookup(map, it->data);
> > > + if (transformed_value == NULL) {
> > > + continue;
> > > + }
> > > + it->data = (gpointer)transformed_value;
> >
> > Is the cast really needed?
>
> Shouldn't be - anything casts to/from void * in C.
Opps I'm wrong. We're using it to cast-away const-ness.
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From cfergeau at redhat.com Thu Jan 10 13:44:12 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Thu, 10 Jan 2013 08:44:12 -0500 (EST)
Subject: [Libosinfo] [PATCH 2/2] win8: Add install script
In-Reply-To: <1357161117-22514-2-git-send-email-zeeshanak@gnome.org>
Message-ID: <615015871.2858188.1357825452203.JavaMail.root@redhat.com>
----- ???????? ????????? -----
> From: "Zeeshan Ali (Khattak)"
>
> Alternative patch to add Windows 8 installer that makes use of
> 'xi:include' nodes to save space. However I failed to make it work.
> Either the xpointer is wrong and/or (more likely) this node is not
> automatically substituted by libxml2.
>
> There is special xinclude API in libxml2 that probably solves our
> problem but it works on DOM API and we are using XPath to load/parse
> our
> XML files.
libxml2 XPath API works on top of a DOM tree, see osinfo_loader_process_xml. Did you try to call xmlXIncludeProcessFlags on the xmlDocPtr xml variable available in this function?
Christophe
From zeeshanak at gnome.org Thu Jan 10 13:59:54 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Thu, 10 Jan 2013 15:59:54 +0200
Subject: [Libosinfo] [PATCH 2/2] win8: Add install script
In-Reply-To: <615015871.2858188.1357825452203.JavaMail.root@redhat.com>
References: <1357161117-22514-2-git-send-email-zeeshanak@gnome.org>
<615015871.2858188.1357825452203.JavaMail.root@redhat.com>
Message-ID:
On Thu, Jan 10, 2013 at 3:44 PM, Christophe Fergeau wrote:
>
>
> ----- ???????? ????????? -----
>> From: "Zeeshan Ali (Khattak)"
>>
>> Alternative patch to add Windows 8 installer that makes use of
>> 'xi:include' nodes to save space. However I failed to make it work.
>> Either the xpointer is wrong and/or (more likely) this node is not
>> automatically substituted by libxml2.
>>
>> There is special xinclude API in libxml2 that probably solves our
>> problem but it works on DOM API and we are using XPath to load/parse
>> our
>> XML files.
>
> libxml2 XPath API works on top of a DOM tree, see osinfo_loader_process_xml. Did you try to call xmlXIncludeProcessFlags on the xmlDocPtr xml variable available in this function?
Oh, I had missed the fact that we have the xmlDocPtr is available.
I'll give it another try..
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Thu Jan 10 22:21:05 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Fri, 11 Jan 2013 00:21:05 +0200
Subject: [Libosinfo] [PATCH 1/2] fedora,
installer: Don't drop 'base' package group
Message-ID: <1357856466-25840-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
In commit e625641, we dropped 'base' package group from destkop profile
script thinking that this essential package is pulled as dep by other
groups. This turned out to be not true and Fedora 18 documentation states
that this package must be explicitly requested.
So instead, we now do the same as we do in jeos profile: Use 'standard'
group if dealing with >= F18, 'base' otherwise.
---
data/install-scripts/fedora.xml | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
index a0a3f35..d36f4b5 100644
--- a/data/install-scripts/fedora.xml
+++ b/data/install-scripts/fedora.xml
@@ -237,6 +237,14 @@ logvol / --fstype --name=LogVol00 --vgname=Vo
reboot
%packages
+
+
+ at base
+
+
+ at standard
+
+
@core
@hardware-support
@base-x
--
1.8.0.2
From zeeshanak at gnome.org Thu Jan 10 22:21:06 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Fri, 11 Jan 2013 00:21:06 +0200
Subject: [Libosinfo] [PATCH 2/2] fedora,installer: Adapt to Fedora 18
In-Reply-To: <1357856466-25840-1-git-send-email-zeeshanak@gnome.org>
References: <1357856466-25840-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1357856466-25840-2-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Package groups have changed in Fedora 18 yet again. This patch makes the
script work against Fedora 18 without breaking it against older releases.
---
data/install-scripts/fedora.xml | 7 +++++--
1 file changed, 5 insertions(+), 2 deletions(-)
diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
index d36f4b5..6e8a92d 100644
--- a/data/install-scripts/fedora.xml
+++ b/data/install-scripts/fedora.xml
@@ -240,17 +240,20 @@ reboot
@base
+ at graphical-internet
+ at sound-and-video
@standard
+ at epiphany
+ at firefox
+ at multimedia
@core
@hardware-support
@base-x
@gnome-desktop
- at graphical-internet
- at sound-and-video
# QXL driver and SPICE vdagent
xorg-x11-drv-qxl
--
1.8.0.2
From fabiano at fidencio.org Thu Jan 10 23:03:22 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Thu, 10 Jan 2013 21:03:22 -0200
Subject: [Libosinfo] [PATCH 1/2] fedora,
installer: Don't drop 'base' package group
In-Reply-To: <1357856466-25840-1-git-send-email-zeeshanak@gnome.org>
References: <1357856466-25840-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Thu, Jan 10, 2013 at 8:21 PM, Zeeshan Ali (Khattak)
wrote:
> From: "Zeeshan Ali (Khattak)"
>
> In commit e625641, we dropped 'base' package group from destkop profile
> script thinking that this essential package is pulled as dep by other
> groups. This turned out to be not true and Fedora 18 documentation states
> that this package must be explicitly requested.
>
> So instead, we now do the same as we do in jeos profile: Use 'standard'
> group if dealing with >= F18, 'base' otherwise.
ACK both
> ---
> data/install-scripts/fedora.xml | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/data/install-scripts/fedora.xml b/data/install-scripts/fedora.xml
> index a0a3f35..d36f4b5 100644
> --- a/data/install-scripts/fedora.xml
> +++ b/data/install-scripts/fedora.xml
> @@ -237,6 +237,14 @@ logvol / --fstype --name=LogVol00 --vgname=Vo
> reboot
>
> %packages
> +
> +
> + at base
> +
> +
> + at standard
> +
> +
> @core
> @hardware-support
> @base-x
> --
> 1.8.0.2
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
--
Fabiano Fid?ncio
From fabiano at fidencio.org Thu Jan 10 23:05:10 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Thu, 10 Jan 2013 21:05:10 -0200
Subject: [Libosinfo] [PATCH 2/2] win8: Add install script
In-Reply-To:
References: <1357161117-22514-2-git-send-email-zeeshanak@gnome.org>
<615015871.2858188.1357825452203.JavaMail.root@redhat.com>
Message-ID:
On Thu, Jan 10, 2013 at 11:59 AM, Zeeshan Ali (Khattak)
wrote:
> On Thu, Jan 10, 2013 at 3:44 PM, Christophe Fergeau wrote:
>>
>>
>> ----- ???????? ????????? -----
>>> From: "Zeeshan Ali (Khattak)"
>>>
>>> Alternative patch to add Windows 8 installer that makes use of
>>> 'xi:include' nodes to save space. However I failed to make it work.
>>> Either the xpointer is wrong and/or (more likely) this node is not
>>> automatically substituted by libxml2.
>>>
>>> There is special xinclude API in libxml2 that probably solves our
>>> problem but it works on DOM API and we are using XPath to load/parse
>>> our
>>> XML files.
>>
>> libxml2 XPath API works on top of a DOM tree, see osinfo_loader_process_xml. Did you try to call xmlXIncludeProcessFlags on the xmlDocPtr xml variable available in this function?
>
> Oh, I had missed the fact that we have the xmlDocPtr is available.
> I'll give it another try..
Don't want to start a discussion, but, once you're on it ... could you
check another scripts/xml files as well? :)
>
> --
> Regards,
>
> Zeeshan Ali (Khattak)
> FSF member#5124
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
Best Regards,
--
Fabiano Fid?ncio
From zeeshanak at gnome.org Fri Jan 11 17:17:18 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Fri, 11 Jan 2013 19:17:18 +0200
Subject: [Libosinfo] [PATCH] Add ability to list install script config
params & profiles
In-Reply-To: <1357731905-7118-1-git-send-email-berrange@redhat.com>
References: <1357731905-7118-1-git-send-email-berrange@redhat.com>
Message-ID:
On Wed, Jan 9, 2013 at 1:45 PM, Daniel P. Berrange wrote:
> From: "Daniel P. Berrange"
>
> When using 'osinfo-install-script' it is useful to know what
> profiles and config parameters are available. Add two new
> flags --list-configs and --list-profiles to output this
> information
>
> # osinfo-install-script --list-profiles win7
> desktop: autounattend.xml
> jeos: autounattend.xml
>
> # osinfo-install-script --list-config winxp
> admin-password: optional
> user-realname: required
> reg-product-key: required
Nice! ACK.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 14 15:04:22 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 14 Jan 2013 17:04:22 +0200
Subject: [Libosinfo] [PATCH] test: Distribute test DB data
Message-ID: <1358175862-6959-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Pushed under trivial rule.
---
test/Makefile.am | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/test/Makefile.am b/test/Makefile.am
index 2ceb7b1..5c05162 100644
--- a/test/Makefile.am
+++ b/test/Makefile.am
@@ -1,5 +1,5 @@
-EXTRA_DIST = isodata
+EXTRA_DIST = isodata dbdata
check_PROGRAMS = \
test-entity \
--
1.8.1
From zeeshanak at gnome.org Mon Jan 14 15:00:00 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 14 Jan 2013 17:00:00 +0200
Subject: [Libosinfo] [PATCH] build: Add forgotten header file
Message-ID: <1358175600-27818-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Pushed under trivial rule.
---
osinfo/Makefile.am | 1 +
1 file changed, 1 insertion(+)
diff --git a/osinfo/Makefile.am b/osinfo/Makefile.am
index 9d03a34..0419030 100644
--- a/osinfo/Makefile.am
+++ b/osinfo/Makefile.am
@@ -112,6 +112,7 @@ libosinfo_1_0_la_SOURCES = \
osinfo_devicelinklist.c \
osinfo_devicelinkfilter.c \
osinfo_device_driver.c \
+ osinfo_device_driver_private.h \
osinfo_device_driverlist.c \
osinfo_install_config.c \
osinfo_install_config_param.c \
--
1.8.1
From zeeshanak at gnome.org Mon Jan 14 16:02:18 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 14 Jan 2013 18:02:18 +0200
Subject: [Libosinfo] [PATCH] Fix `make syntax-check`
Message-ID: <1358179338-13282-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
* Add Yuri Chornoivan to AUTHORS file.
* Remove osinfo/osinfo_install_config.c from translated sources.
Pushed under trivial rule.
---
AUTHORS | 1 +
po/POTFILES.in | 1 -
2 files changed, 1 insertion(+), 1 deletion(-)
diff --git a/AUTHORS b/AUTHORS
index 320b5f2..65fc04a 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -20,6 +20,7 @@ Patches contributed by:
Michal Privoznik
David Zeuthen
Eric Blake
+ Yuri Chornoivan
...send patches to get your name here...
-- End
diff --git a/po/POTFILES.in b/po/POTFILES.in
index 216751a..1a209bb 100644
--- a/po/POTFILES.in
+++ b/po/POTFILES.in
@@ -28,7 +28,6 @@ osinfo/osinfo_deployment.c
osinfo/osinfo_devicelink.c
osinfo/osinfo_devicelinkfilter.c
osinfo/osinfo_entity.c
-osinfo/osinfo_install_config.c
osinfo/osinfo_install_config_param.c
osinfo/osinfo_install_script.c
osinfo/osinfo_list.c
--
1.8.1
From zeeshanak at gnome.org Mon Jan 14 16:51:42 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 14 Jan 2013 18:51:42 +0200
Subject: [Libosinfo] Libosinfo 0.2.3
Message-ID:
Libosinfo 0.2.3 is out!
Changes since 0.2.2:
- Add datamaps for translating OS-neutral values to OS-specific ones, e.g some
installer configuration parameters like keyboard, language and timezone etc.
- New API to detect media that makes it possible to also query languages
supported by the media.
- Add install scripts for:
- RHEL 6.x.
- Microsoft Windows 8
- Fix install script for Fedora 18.
- Drop support for encoding in l10n install script configuration parameters.
- Fix test build issues.
- Fixes and improvements to documentation.
- Fix potential issues spotted by Coverity.
- Fix build for translations.
- osinfo-install-script tool now has options to list available configuration
parameters and profiles.
- Add/improve data on:
- RHEL
- Debian
- openSUSE
- Microsoft Windows 7
- Ubuntu
- MacOS X
- Added translations:
- Ukrainian
- Polish
- Many other fixes and improvements.
What is libosinfo?
=============
libosinfo is a GObject based library API for managing information about
operating systems, hypervisors and the (virtual) hardware devices they can
support. It includes a database containing device metadata and provides APIs
to match/identify optimal devices for deploying an operating system on a
hypervisor. Via the magic of GObject Introspection, the API is available in all
common programming languages with demos for javascript (GJS/Seed) and python
(PyGObject). Also provided are Vala bindings.
libosinfo is Free Software and licenced under LGPLv2+.
The latest official releases can be found at:
https://fedorahosted.org/released/libosinfo/
Dependencies
============
- Required:
- gobject-2.0
- gio-2.0
- libxml-2.0
- Optional:
- gobject-introspection
- Vala (build-time only)
For further information about libosinfo please consult the project homepage
https://fedorahosted.org/libosinfo/
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Tue Jan 15 15:50:15 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Tue, 15 Jan 2013 17:50:15 +0200
Subject: [Libosinfo] [PATCH] gnome: Add release & eol dates
Message-ID: <1358265015-15301-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
eol date is the due release date of GNOME 3.8.
---
data/oses/gnome.xml.in | 3 +++
1 file changed, 3 insertions(+)
diff --git a/data/oses/gnome.xml.in b/data/oses/gnome.xml.in
index 2200aa7..eb25ec0 100644
--- a/data/oses/gnome.xml.in
+++ b/data/oses/gnome.xml.in
@@ -9,6 +9,9 @@
gnome
+ 2012-09-26
+ 2013-03-27
+
http://ftp.gnome.org/pub/gnome/misc/promo-usb/GNOME-3.6.0.iso
--
1.8.1
From fabiano at fidencio.org Tue Jan 15 16:41:57 2013
From: fabiano at fidencio.org (=?ISO-8859-1?Q?Fabiano_Fid=EAncio?=)
Date: Tue, 15 Jan 2013 14:41:57 -0200
Subject: [Libosinfo] [PATCH] gnome: Add release & eol dates
In-Reply-To: <1358265015-15301-1-git-send-email-zeeshanak@gnome.org>
References: <1358265015-15301-1-git-send-email-zeeshanak@gnome.org>
Message-ID:
On Tue, Jan 15, 2013 at 1:50 PM, Zeeshan Ali (Khattak)
wrote:
> From: "Zeeshan Ali (Khattak)"
>
> eol date is the due release date of GNOME 3.8.
ACK!
> ---
> data/oses/gnome.xml.in | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/data/oses/gnome.xml.in b/data/oses/gnome.xml.in
> index 2200aa7..eb25ec0 100644
> --- a/data/oses/gnome.xml.in
> +++ b/data/oses/gnome.xml.in
> @@ -9,6 +9,9 @@
> gnome
>
>
> + 2012-09-26
> + 2013-03-27
> +
>
> http://ftp.gnome.org/pub/gnome/misc/promo-usb/GNOME-3.6.0.iso
>
> --
> 1.8.1
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
--
Fabiano Fid?ncio
From berrange at redhat.com Tue Jan 15 16:57:01 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Tue, 15 Jan 2013 16:57:01 +0000
Subject: [Libosinfo] [PATCH] gnome: Add release & eol dates
In-Reply-To: <1358265015-15301-1-git-send-email-zeeshanak@gnome.org>
References: <1358265015-15301-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130115165701.GE9079@redhat.com>
On Tue, Jan 15, 2013 at 05:50:15PM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> eol date is the due release date of GNOME 3.8.
> ---
> data/oses/gnome.xml.in | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/data/oses/gnome.xml.in b/data/oses/gnome.xml.in
> index 2200aa7..eb25ec0 100644
> --- a/data/oses/gnome.xml.in
> +++ b/data/oses/gnome.xml.in
> @@ -9,6 +9,9 @@
> gnome
>
>
> + 2012-09-26
> + 2013-03-27
I wouldn't try to second-guess an EOL date. Just leave it out
until it actually happens.
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From marcandre.lureau at gmail.com Mon Jan 21 13:16:12 2013
From: marcandre.lureau at gmail.com (=?UTF-8?q?Marc-Andr=C3=A9=20Lureau?=)
Date: Mon, 21 Jan 2013 14:16:12 +0100
Subject: [Libosinfo] [PATCH libosinfo] build-sys: g_type_init() is
deprecated in 2.36
Message-ID: <1358774172-15863-1-git-send-email-marcandre.lureau@redhat.com>
---
test/test-db.c | 2 ++
test/test-device.c | 2 ++
test/test-devicelist.c | 2 ++
test/test-entity.c | 2 ++
test/test-filter.c | 2 ++
test/test-install-script.c | 2 ++
test/test-isodetect.c | 2 ++
test/test-list.c | 2 ++
test/test-loader.c | 2 ++
test/test-mediauris.c | 2 ++
test/test-os.c | 2 ++
test/test-oslist.c | 2 ++
test/test-platform.c | 2 ++
test/test-platformlist.c | 2 ++
test/test-product.c | 2 ++
test/test-productfilter.c | 2 ++
test/test-treeuris.c | 2 ++
tools/osinfo-db-validate.c | 2 ++
tools/osinfo-detect.c | 2 ++
tools/osinfo-install-script.c | 2 ++
tools/osinfo-query.c | 2 ++
21 files changed, 42 insertions(+)
diff --git a/test/test-db.c b/test/test-db.c
index 50e29fe..576f40f 100644
--- a/test/test-db.c
+++ b/test/test-db.c
@@ -422,7 +422,9 @@ int main(void)
Suite *s = list_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_entity_get_type();
diff --git a/test/test-device.c b/test/test-device.c
index 4612d27..4b9d3c6 100644
--- a/test/test-device.c
+++ b/test/test-device.c
@@ -55,7 +55,9 @@ int main(void)
Suite *s = device_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_device_get_type();
diff --git a/test/test-devicelist.c b/test/test-devicelist.c
index 1fd001a..c3aa65f 100644
--- a/test/test-devicelist.c
+++ b/test/test-devicelist.c
@@ -224,7 +224,9 @@ int main(void)
Suite *s = list_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_device_get_type();
diff --git a/test/test-entity.c b/test/test-entity.c
index 199307b..aeee513 100644
--- a/test/test-entity.c
+++ b/test/test-entity.c
@@ -335,7 +335,9 @@ int main(void)
Suite *s = entity_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_dummy_get_type();
diff --git a/test/test-filter.c b/test/test-filter.c
index c6fbf4b..1a51370 100644
--- a/test/test-filter.c
+++ b/test/test-filter.c
@@ -172,7 +172,9 @@ int main(void)
Suite *s = filter_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_device_get_type();
diff --git a/test/test-install-script.c b/test/test-install-script.c
index 4a3f16a..1068da9 100644
--- a/test/test-install-script.c
+++ b/test/test-install-script.c
@@ -277,7 +277,9 @@ int main(void)
Suite *s = list_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_entity_get_type();
diff --git a/test/test-isodetect.c b/test/test-isodetect.c
index a3460ff..efbc56a 100644
--- a/test/test-isodetect.c
+++ b/test/test-isodetect.c
@@ -403,7 +403,9 @@ int main(void)
Suite *s = list_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_entity_get_type();
diff --git a/test/test-list.c b/test/test-list.c
index 0656799..bd82183 100644
--- a/test/test-list.c
+++ b/test/test-list.c
@@ -384,7 +384,9 @@ int main(void)
Suite *s = list_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_dummy_get_type();
diff --git a/test/test-loader.c b/test/test-loader.c
index 7755600..3fe5c26 100644
--- a/test/test-loader.c
+++ b/test/test-loader.c
@@ -55,7 +55,9 @@ int main(void)
Suite *s = loader_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_entity_get_type();
diff --git a/test/test-mediauris.c b/test/test-mediauris.c
index 85a2239..191e8c6 100644
--- a/test/test-mediauris.c
+++ b/test/test-mediauris.c
@@ -134,7 +134,9 @@ int main(void)
if (!getenv("LIBOSINFO_NETWORK_TESTS"))
return 77; /* Skip */
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_entity_get_type();
diff --git a/test/test-os.c b/test/test-os.c
index ecec8f5..8ea86df 100644
--- a/test/test-os.c
+++ b/test/test-os.c
@@ -143,7 +143,9 @@ int main(void)
Suite *s = os_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_platform_get_type();
diff --git a/test/test-oslist.c b/test/test-oslist.c
index c3d114c..2fa97c6 100644
--- a/test/test-oslist.c
+++ b/test/test-oslist.c
@@ -224,7 +224,9 @@ int main(void)
Suite *s = list_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_os_get_type();
diff --git a/test/test-platform.c b/test/test-platform.c
index 97c4c8b..92772a7 100644
--- a/test/test-platform.c
+++ b/test/test-platform.c
@@ -131,7 +131,9 @@ int main(void)
Suite *s = platform_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_platform_get_type();
diff --git a/test/test-platformlist.c b/test/test-platformlist.c
index d191480..4baf4ff 100644
--- a/test/test-platformlist.c
+++ b/test/test-platformlist.c
@@ -224,7 +224,9 @@ int main(void)
Suite *s = list_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_platform_get_type();
diff --git a/test/test-product.c b/test/test-product.c
index 412b6e1..f8f654c 100644
--- a/test/test-product.c
+++ b/test/test-product.c
@@ -205,7 +205,9 @@ int main(void)
Suite *s = product_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_platform_get_type();
diff --git a/test/test-productfilter.c b/test/test-productfilter.c
index 7c9ce47..f05ce67 100644
--- a/test/test-productfilter.c
+++ b/test/test-productfilter.c
@@ -243,7 +243,9 @@ int main(void)
Suite *s = productfilter_suite ();
SRunner *sr = srunner_create (s);
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_entity_get_type();
diff --git a/test/test-treeuris.c b/test/test-treeuris.c
index 584cda8..07f14bf 100644
--- a/test/test-treeuris.c
+++ b/test/test-treeuris.c
@@ -134,7 +134,9 @@ int main(void)
if (!getenv("LIBOSINFO_NETWORK_TESTS"))
return 77; /* Skip */
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
/* Upfront so we don't confuse valgrind */
osinfo_entity_get_type();
diff --git a/tools/osinfo-db-validate.c b/tools/osinfo-db-validate.c
index a4e025b..20a77c2 100644
--- a/tools/osinfo-db-validate.c
+++ b/tools/osinfo-db-validate.c
@@ -256,7 +256,9 @@ gint main(gint argc, gchar **argv)
bindtextdomain(GETTEXT_PACKAGE, LOCALEDIR);
bind_textdomain_codeset (GETTEXT_PACKAGE, "UTF-8");
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
context = g_option_context_new(_("- Validate XML documents "));
diff --git a/tools/osinfo-detect.c b/tools/osinfo-detect.c
index 0caa4ca..1262783 100644
--- a/tools/osinfo-detect.c
+++ b/tools/osinfo-detect.c
@@ -212,7 +212,9 @@ gint main(gint argc, gchar **argv)
goto EXIT;
}
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
loader = osinfo_loader_new();
osinfo_loader_process_default_path(loader, &error);
diff --git a/tools/osinfo-install-script.c b/tools/osinfo-install-script.c
index 7cdb842..fabce96 100644
--- a/tools/osinfo-install-script.c
+++ b/tools/osinfo-install-script.c
@@ -254,7 +254,9 @@ gint main(gint argc, gchar **argv)
bindtextdomain(GETTEXT_PACKAGE, LOCALEDIR);
bind_textdomain_codeset (GETTEXT_PACKAGE, "UTF-8");
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
config = osinfo_install_config_new("http://libosinfo.fedorahosted.org/config");
diff --git a/tools/osinfo-query.c b/tools/osinfo-query.c
index 7a679b0..8ceba3f 100644
--- a/tools/osinfo-query.c
+++ b/tools/osinfo-query.c
@@ -366,7 +366,9 @@ gint main(gint argc, gchar **argv)
bindtextdomain(GETTEXT_PACKAGE, LOCALEDIR);
bind_textdomain_codeset (GETTEXT_PACKAGE, "UTF-8");
+#if !GLIB_CHECK_VERSION(2,35,1)
g_type_init();
+#endif
struct OsinfoType types[] = {
{ "os",
--
1.8.1.rc1.17.g75ed918
From berrange at redhat.com Mon Jan 21 13:19:35 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Mon, 21 Jan 2013 13:19:35 +0000
Subject: [Libosinfo] [PATCH libosinfo] build-sys: g_type_init() is
deprecated in 2.36
In-Reply-To: <1358774172-15863-1-git-send-email-marcandre.lureau@redhat.com>
References: <1358774172-15863-1-git-send-email-marcandre.lureau@redhat.com>
Message-ID: <20130121131935.GK16437@redhat.com>
On Mon, Jan 21, 2013 at 02:16:12PM +0100, Marc-Andr? Lureau wrote:
> ---
> test/test-db.c | 2 ++
> test/test-device.c | 2 ++
> test/test-devicelist.c | 2 ++
> test/test-entity.c | 2 ++
> test/test-filter.c | 2 ++
> test/test-install-script.c | 2 ++
> test/test-isodetect.c | 2 ++
> test/test-list.c | 2 ++
> test/test-loader.c | 2 ++
> test/test-mediauris.c | 2 ++
> test/test-os.c | 2 ++
> test/test-oslist.c | 2 ++
> test/test-platform.c | 2 ++
> test/test-platformlist.c | 2 ++
> test/test-product.c | 2 ++
> test/test-productfilter.c | 2 ++
> test/test-treeuris.c | 2 ++
> tools/osinfo-db-validate.c | 2 ++
> tools/osinfo-detect.c | 2 ++
> tools/osinfo-install-script.c | 2 ++
> tools/osinfo-query.c | 2 ++
> 21 files changed, 42 insertions(+)
Urgh, I'm getting really fed up of the recent glib deprecations.
All they're doing is making life harder for applications developers
by making them litter code with conditionals :-( Thankfully this
isn't as horrific as the mutex ones they did.
ACK
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From zeeshanak at gnome.org Mon Jan 21 20:39:05 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 21 Jan 2013 22:39:05 +0200
Subject: [Libosinfo] [PATCH] openSUSE: Correct C&P mistakes in volume IDs
Message-ID: <1358800745-27893-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Pushed under trivial rule.
---
data/oses/opensuse.xml.in | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/data/oses/opensuse.xml.in b/data/oses/opensuse.xml.in
index 1fd6966..9a11b34 100644
--- a/data/oses/opensuse.xml.in
+++ b/data/oses/opensuse.xml.in
@@ -270,7 +270,7 @@
http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-i586.iso
- openSUSE-DVD-i586-Build0039
+ openSUSE-DVD-i586-Build0167LINUXSUSE LINUX Products GmbH
@@ -280,7 +280,7 @@
http://download.opensuse.org/distribution/12.2/iso/openSUSE-12.2-DVD-x86_64.iso
- openSUSE-DVD-x86_640039
+ openSUSE-DVD-x86_640167LINUXSUSE LINUX Products GmbH
--
1.8.1
From cfergeau at redhat.com Tue Jan 22 11:26:03 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 22 Jan 2013 12:26:03 +0100
Subject: [Libosinfo] [PATCHv3 3/3] debian: Update volume-id for Squeeze
In-Reply-To:
References: <1357056055-31236-1-git-send-email-fabiano@fidencio.org>
<1357056055-31236-3-git-send-email-fabiano@fidencio.org>
Message-ID: <20130122112602.GE31793@teriyaki.redhat.com>
On Wed, Jan 02, 2013 at 06:58:56PM +0200, Zeeshan Ali (Khattak) wrote:
> On Tue, Jan 1, 2013 at 6:00 PM, Fabiano Fid?ncio wrote:
> > ---
> > data/oses/debian.xml.in | 4 ++--
> > 1 file changed, 2 insertions(+), 2 deletions(-)
> >
> > diff --git a/data/oses/debian.xml.in b/data/oses/debian.xml.in
> > index fc0f6cc..349ba24 100644
> > --- a/data/oses/debian.xml.in
> > +++ b/data/oses/debian.xml.in
> > @@ -187,7 +187,7 @@
> >
> > http://live.debian.net/cdimage/release/current/i386/iso-hybrid/debian-live-6.0.6-i386-gnome-desktop.iso
> >
> > - Debian squeeze 20111014-08:13
> > + Debian squeeze \d{8}-\d{2}:\d{2}
> > LINUX
> >
> > install/vmlinuz
> > @@ -196,7 +196,7 @@
> >
> > http://live.debian.net/cdimage/release/current/amd64/iso-hybrid/debian-live-6.0.6-amd64-gnome-desktop.iso
> >
> > - Debian squeeze 20111014-08:11
> > + Debian squeeze \d{8}-\d{2}:\d{2}
>
> I thought we agreed that there is no need for date recognition here.
> i-e "Debian squeeze" should be both specific and generic enough?
>
> I think thats also what teuf meant on IRC:
>
> [12:17:02] fidencio: even a truncated regex (ie stop before the
> date) would be good I think
For what it's worth, I mentioned this because it makes the regex simpler,
but the more specific matching from this patch would have been fine with me
too.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Tue Jan 22 15:47:42 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 22 Jan 2013 16:47:42 +0100
Subject: [Libosinfo] RFC: Add platform info for qemu 1.2.2
Message-ID: <1358869663-24025-1-git-send-email-cfergeau@redhat.com>
Hey,
This patch adds hypervisor data about qemu-kvm 1.2.2, which would be useful
as a basis for adding data about other QEMU version. However, I'm not exactly sure
how we want to expose things with respect to qemu VS qemu-kvm, nor about what we want
in this file, so sending what I have so far for feedback ;)
Christophe
From cfergeau at redhat.com Tue Jan 22 15:47:43 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 22 Jan 2013 16:47:43 +0100
Subject: [Libosinfo] [libosinfo] Add QEMU-KVM 1.2.2 hypervisor data
In-Reply-To: <1358869663-24025-1-git-send-email-cfergeau@redhat.com>
References: <1358869663-24025-1-git-send-email-cfergeau@redhat.com>
Message-ID: <1358869663-24025-2-git-send-email-cfergeau@redhat.com>
---
data/hypervisors/Makefile.am | 2 +-
data/hypervisors/kvm.xml.in | 57 ++++++++++++++++++++++++++++++++++++++++++++
2 files changed, 58 insertions(+), 1 deletion(-)
create mode 100644 data/hypervisors/kvm.xml.in
diff --git a/data/hypervisors/Makefile.am b/data/hypervisors/Makefile.am
index 19ac267..2f7dd45 100644
--- a/data/hypervisors/Makefile.am
+++ b/data/hypervisors/Makefile.am
@@ -1,6 +1,6 @@
databasedir = $(pkgdatadir)/db/hypervisors/
-database_in_files = rhel-xen.xml.in xen.xml.in
+database_in_files = kvm.xml.in rhel-xen.xml.in xen.xml.in
database_DATA = $(database_in_files:.xml.in=.xml)
diff --git a/data/hypervisors/kvm.xml.in b/data/hypervisors/kvm.xml.in
new file mode 100644
index 0000000..3653c46
--- /dev/null
+++ b/data/hypervisors/kvm.xml.in
@@ -0,0 +1,57 @@
+
+
+
+
+ kvm-1.2.2
+ <_name>QEMU-KVM 1.2.2
+ <_vendor>qemu.org
+ 1.2.2
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
--
1.8.1
From mprivozn at redhat.com Tue Jan 22 18:53:13 2013
From: mprivozn at redhat.com (Michal Privoznik)
Date: Tue, 22 Jan 2013 19:53:13 +0100
Subject: [Libosinfo] [PATCH] Don't redefine _FORTIFY_SOURCE macro
Message-ID: <7d5d7ab7ddabc80fa430122350c5c841558c6b5a.1358880752.git.mprivozn@redhat.com>
If the _FORTIFY_SOURCE has been already defined, we unconditionally
redefine it, leaving us with warning/error thrown at compilation time.
---
m4/libosinfo-compile-warnings.m4 | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/m4/libosinfo-compile-warnings.m4 b/m4/libosinfo-compile-warnings.m4
index 508ff80..fb58607 100644
--- a/m4/libosinfo-compile-warnings.m4
+++ b/m4/libosinfo-compile-warnings.m4
@@ -89,7 +89,7 @@ AC_DEFUN([LIBOSINFO_COMPILE_WARNINGS],[
AH_VERBATIM([FORTIFY_SOURCE],
[/* Enable compile-time and run-time bounds-checking, and some warnings,
without upsetting newer glibc. */
- #if defined __OPTIMIZE__ && __OPTIMIZE__
+ #if !defined _FORTIFY_SOURCE && defined __OPTIMIZE__ && __OPTIMIZE__
# define _FORTIFY_SOURCE 2
#endif
])
--
1.8.0.2
From cfergeau at redhat.com Tue Jan 22 19:01:20 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 22 Jan 2013 20:01:20 +0100
Subject: [Libosinfo] [PATCH] Don't redefine _FORTIFY_SOURCE macro
In-Reply-To: <7d5d7ab7ddabc80fa430122350c5c841558c6b5a.1358880752.git.mprivozn@redhat.com>
References: <7d5d7ab7ddabc80fa430122350c5c841558c6b5a.1358880752.git.mprivozn@redhat.com>
Message-ID: <20130122190120.GN31793@teriyaki.redhat.com>
On Tue, Jan 22, 2013 at 07:53:13PM +0100, Michal Privoznik wrote:
> If the _FORTIFY_SOURCE has been already defined, we unconditionally
> redefine it, leaving us with warning/error thrown at compilation time.
Same change as what is done in libvirt, ACK.
Christophe
> ---
> m4/libosinfo-compile-warnings.m4 | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/m4/libosinfo-compile-warnings.m4 b/m4/libosinfo-compile-warnings.m4
> index 508ff80..fb58607 100644
> --- a/m4/libosinfo-compile-warnings.m4
> +++ b/m4/libosinfo-compile-warnings.m4
> @@ -89,7 +89,7 @@ AC_DEFUN([LIBOSINFO_COMPILE_WARNINGS],[
> AH_VERBATIM([FORTIFY_SOURCE],
> [/* Enable compile-time and run-time bounds-checking, and some warnings,
> without upsetting newer glibc. */
> - #if defined __OPTIMIZE__ && __OPTIMIZE__
> + #if !defined _FORTIFY_SOURCE && defined __OPTIMIZE__ && __OPTIMIZE__
> # define _FORTIFY_SOURCE 2
> #endif
> ])
> --
> 1.8.0.2
>
> _______________________________________________
> Libosinfo mailing list
> Libosinfo at redhat.com
> https://www.redhat.com/mailman/listinfo/libosinfo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From zeeshanak at gnome.org Mon Jan 28 03:18:39 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:39 +0200
Subject: [Libosinfo] More device drivers
Message-ID: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
These patches add information about virtio and QXL device drivers, make it possible for apps to automatically install them as part of Windows XP installation and improve drivers-related API.
From zeeshanak at gnome.org Mon Jan 28 03:18:40 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:40 +0200
Subject: [Libosinfo] [PATCH 1/8] winxp,
installer: Don't assume script disk is specified
In-Reply-To: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1359343127-27272-2-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Declare that script optionally uses script disk if specified and hardcode
a fallback for the case when its not specified.
---
data/install-scripts/windows-sif.xml | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
index 2947efa..c072f4b 100644
--- a/data/install-scripts/windows-sif.xml
+++ b/data/install-scripts/windows-sif.xml
@@ -77,6 +77,7 @@
+
+
+
+
+
+
+
+ A
+
+
+
+
@@ -182,7 +194,7 @@
[SetupParams]
- UserExecute = ":\windows.cmd"
+ UserExecute = ":\windows.cmd"
[WindowsFirewall]
Profiles = WindowsFirewall.TurnOffFirewall
--
1.8.1
From zeeshanak at gnome.org Mon Jan 28 03:18:41 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:41 +0200
Subject: [Libosinfo] [PATCH 2/8] winxp,installer: Ignore unsigned drivers
In-Reply-To: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
We are unlikely to find any signed free drivers for windows that are
signed. Better just ask windows to ignore.
---
data/install-scripts/windows-sif.xml | 2 ++
1 file changed, 2 insertions(+)
diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
index c072f4b..3f947d0 100644
--- a/data/install-scripts/windows-sif.xml
+++ b/data/install-scripts/windows-sif.xml
@@ -29,6 +29,7 @@
UnattendMode=FullUnattended
OemSkipEula=Yes
OemPreinstall=No
+ DriverSigningPolicy=Ignore
TargetPath=\WINDOWS
Repartition=Yes
WaitForReboot=No
@@ -134,6 +135,7 @@
UnattendMode=FullUnattended
OemSkipEula=Yes
OemPreinstall=No
+ DriverSigningPolicy=Ignore
TargetPath=\WINDOWS
--
1.8.1
From zeeshanak at gnome.org Mon Jan 28 03:18:42 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:42 +0200
Subject: [Libosinfo] [PATCH 3/8] API to query format of device driver
In-Reply-To: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1359343127-27272-4-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
---
data/oses/windows.xml.in | 8 ++++----
data/schemas/libosinfo.rng | 11 +++++++++++
osinfo/libosinfo.syms | 6 ++++++
osinfo/osinfo_device_driver.c | 9 +++++++++
osinfo/osinfo_device_driver.h | 16 ++++++++++++++++
osinfo/osinfo_loader.c | 8 ++++++++
6 files changed, 54 insertions(+), 4 deletions(-)
diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
index 310de62..1b762f8 100644
--- a/data/oses/windows.xml.in
+++ b/data/oses/windows.xml.in
@@ -383,7 +383,7 @@
-
+ viostor.catviostor.infviostor.sys
@@ -392,7 +392,7 @@
-
+ viostor.catviostor.infviostor.sys
@@ -754,14 +754,14 @@
-
+ viostor.catviostor.infviostor.sys
-
+ viostor.catviostor.infviostor.sys
diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
index 90b0dfb..05a34be 100644
--- a/data/schemas/libosinfo.rng
+++ b/data/schemas/libosinfo.rng
@@ -451,6 +451,11 @@
+
+
+
+
+
@@ -677,4 +682,10 @@
dos|unix
+
+
+
+ windows|executable|package
+
+
diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index 4615829..41d3756 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -399,6 +399,12 @@ LIBOSINFO_0.2.3 {
} LIBOSINFO_0.2.2;
+LIBOSINFO_0.2.4 {
+ global:
+ osinfo_device_driver_format_get_type;
+ osinfo_device_driver_get_format;
+} LIBOSINFO_0.2.3;
+
/* Symbols in next release...
LIBOSINFO_0.0.2 {
diff --git a/osinfo/osinfo_device_driver.c b/osinfo/osinfo_device_driver.c
index d402bc7..d88273a 100644
--- a/osinfo/osinfo_device_driver.c
+++ b/osinfo/osinfo_device_driver.c
@@ -162,6 +162,15 @@ OsinfoDeviceList *osinfo_device_driver_get_devices(OsinfoDeviceDriver *driver)
return driver->priv->devices;
}
+OsinfoDeviceDriverFormat osinfo_device_driver_get_format(OsinfoDeviceDriver *driver)
+{
+ return osinfo_entity_get_param_value_enum
+ (OSINFO_ENTITY(driver),
+ OSINFO_DEVICE_DRIVER_PROP_FORMAT,
+ OSINFO_TYPE_DEVICE_DRIVER_FORMAT,
+ OSINFO_DEVICE_DRIVER_FORMAT_EXECUTABLE);
+}
+
void osinfo_device_driver_add_device(OsinfoDeviceDriver *driver,
OsinfoDevice *device)
{
diff --git a/osinfo/osinfo_device_driver.h b/osinfo/osinfo_device_driver.h
index c894fe8..b45f0ab 100644
--- a/osinfo/osinfo_device_driver.h
+++ b/osinfo/osinfo_device_driver.h
@@ -55,6 +55,7 @@ typedef struct _OsinfoDeviceDriverPrivate OsinfoDeviceDriverPrivate;
#define OSINFO_DEVICE_DRIVER_PROP_PRE_INSTALLABLE "pre-installable"
#define OSINFO_DEVICE_DRIVER_PROP_FILE "file"
#define OSINFO_DEVICE_DRIVER_PROP_DEVICE "device"
+#define OSINFO_DEVICE_DRIVER_PROP_FORMAT "format"
/* object */
struct _OsinfoDeviceDriver
@@ -76,6 +77,20 @@ struct _OsinfoDeviceDriverClass
/* class members */
};
+/**
+ * OsinfoDeviceDriverFormat:
+ * @OSINFO_DEVICE_DRIVER_FORMAT_WINDOWS: Windows specific driver in the form of
+ * .cat, .inf, .sys and (optionally) .dll
+ * files.
+ * @OSINFO_DEVICE_DRIVER_FORMAT_EXECUTABLE: Executable installer script/binary.
+ * @OSINFO_DEVICE_DRIVER_FORMAT_PACKAGE: OS-specific installer package (e.g RPM or DEB etc).
+ */
+typedef enum {
+ OSINFO_DEVICE_DRIVER_FORMAT_WINDOWS,
+ OSINFO_DEVICE_DRIVER_FORMAT_EXECUTABLE,
+ OSINFO_DEVICE_DRIVER_FORMAT_PACKAGE
+} OsinfoDeviceDriverFormat;
+
GType osinfo_device_driver_get_type(void);
const gchar *osinfo_device_driver_get_architecture(OsinfoDeviceDriver *driver);
@@ -83,6 +98,7 @@ const gchar *osinfo_device_driver_get_location(OsinfoDeviceDriver *driver);
gboolean osinfo_device_driver_get_pre_installable(OsinfoDeviceDriver *driver);
GList *osinfo_device_driver_get_files(OsinfoDeviceDriver *driver);
OsinfoDeviceList *osinfo_device_driver_get_devices(OsinfoDeviceDriver *driver);
+OsinfoDeviceDriverFormat osinfo_device_driver_get_format(OsinfoDeviceDriver *driver);
#endif /* __OSINFO_DEVICE_DRIVER_H__ */
/*
diff --git a/osinfo/osinfo_loader.c b/osinfo/osinfo_loader.c
index efbbc97..d186073 100644
--- a/osinfo/osinfo_loader.c
+++ b/osinfo/osinfo_loader.c
@@ -1018,6 +1018,7 @@ static OsinfoDeviceDriver *osinfo_loader_driver(OsinfoLoader *loader,
xmlChar *arch = xmlGetProp(root, BAD_CAST OSINFO_DEVICE_DRIVER_PROP_ARCHITECTURE);
xmlChar *location = xmlGetProp(root, BAD_CAST OSINFO_DEVICE_DRIVER_PROP_LOCATION);
xmlChar *preinst = xmlGetProp(root, BAD_CAST OSINFO_DEVICE_DRIVER_PROP_PRE_INSTALLABLE);
+ xmlChar *format = xmlGetProp(root, BAD_CAST OSINFO_DEVICE_DRIVER_PROP_FORMAT);
OsinfoDeviceDriver *driver = osinfo_device_driver_new(id);
@@ -1042,6 +1043,13 @@ static OsinfoDeviceDriver *osinfo_loader_driver(OsinfoLoader *loader,
xmlFree(preinst);
}
+ if (format) {
+ osinfo_entity_set_param(OSINFO_ENTITY(driver),
+ OSINFO_DEVICE_DRIVER_PROP_FORMAT,
+ (gchar *)format);
+ xmlFree(format);
+ }
+
gint nnodes = osinfo_loader_nodeset("./*", ctxt, &nodes, err);
if (error_is_set(err)) {
g_object_unref(G_OBJECT(driver));
--
1.8.1
From zeeshanak at gnome.org Mon Jan 28 03:18:43 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:43 +0200
Subject: [Libosinfo] [PATCH 4/8] installer: API to query supported device
driver formats
In-Reply-To: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1359343127-27272-5-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
---
data/install-scripts/windows-sif.xml | 2 ++
data/install-scripts/windows-unattend.xml | 2 ++
data/schemas/libosinfo.rng | 10 ++++++++++
osinfo/libosinfo.syms | 2 ++
osinfo/osinfo_install_script.c | 18 ++++++++++++++++++
osinfo/osinfo_install_script.h | 5 +++++
osinfo/osinfo_loader.c | 2 ++
7 files changed, 41 insertions(+)
diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
index 3f947d0..8634e3a 100644
--- a/data/install-scripts/windows-sif.xml
+++ b/data/install-scripts/windows-sif.xml
@@ -6,6 +6,7 @@
$$$$$-$$$$$-$$$$$-$$$$$-$$$$$winnt.siftrue
+ windows
@@ -73,6 +74,7 @@
$$$$$-$$$$$-$$$$$-$$$$$-$$$$$winnt.siftrue
+ windows
diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
index e140617..e306ec4 100644
--- a/data/install-scripts/windows-unattend.xml
+++ b/data/install-scripts/windows-unattend.xml
@@ -5,6 +5,7 @@
$$$$$-$$$$$-$$$$$-$$$$$-$$$$$autounattend.xmltrue
+ windows
@@ -220,6 +221,7 @@
$$$$$-$$$$$-$$$$$-$$$$$-$$$$$autounattend.xmltrue
+ windows
diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
index 05a34be..c193bcc 100644
--- a/data/schemas/libosinfo.rng
+++ b/data/schemas/libosinfo.rng
@@ -580,10 +580,20 @@
+
+
+
+
+
+
+
+
+
+
diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index 41d3756..3e71d9e 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -403,6 +403,8 @@ LIBOSINFO_0.2.4 {
global:
osinfo_device_driver_format_get_type;
osinfo_device_driver_get_format;
+ osinfo_install_script_get_pre_install_driver_format;
+ osinfo_install_script_get_post_install_driver_format;
} LIBOSINFO_0.2.3;
/* Symbols in next release...
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index 7105430..ca45b60 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -1232,6 +1232,24 @@ gboolean osinfo_install_script_get_can_post_install_drivers(OsinfoInstallScript
OSINFO_INSTALL_SCRIPT_PROP_CAN_POST_INSTALL_DRIVERS);
}
+OsinfoDeviceDriverFormat osinfo_install_script_get_pre_install_driver_format(OsinfoInstallScript *script)
+{
+ return osinfo_entity_get_param_value_enum
+ (OSINFO_ENTITY(script),
+ OSINFO_INSTALL_SCRIPT_PROP_PRE_INSTALL_DEVICE_DRIVER_FORMAT,
+ OSINFO_TYPE_DEVICE_DRIVER_FORMAT,
+ OSINFO_DEVICE_DRIVER_FORMAT_EXECUTABLE);
+}
+
+OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(OsinfoInstallScript *script)
+{
+ return osinfo_entity_get_param_value_enum
+ (OSINFO_ENTITY(script),
+ OSINFO_INSTALL_SCRIPT_PROP_POST_INSTALL_DEVICE_DRIVER_FORMAT,
+ OSINFO_TYPE_DEVICE_DRIVER_FORMAT,
+ OSINFO_DEVICE_DRIVER_FORMAT_EXECUTABLE);
+}
+
/*
* Local variables:
* indent-tabs-mode: nil
diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
index d91751e..82486ef 100644
--- a/osinfo/osinfo_install_script.h
+++ b/osinfo/osinfo_install_script.h
@@ -55,6 +55,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
#define OSINFO_INSTALL_SCRIPT_PROP_PATH_FORMAT "path-format"
#define OSINFO_INSTALL_SCRIPT_PROP_CAN_PRE_INSTALL_DRIVERS "can-pre-install-drivers"
#define OSINFO_INSTALL_SCRIPT_PROP_CAN_POST_INSTALL_DRIVERS "can-post-install-drivers"
+#define OSINFO_INSTALL_SCRIPT_PROP_PRE_INSTALL_DEVICE_DRIVER_FORMAT "pre-install-driver-format"
+#define OSINFO_INSTALL_SCRIPT_PROP_POST_INSTALL_DEVICE_DRIVER_FORMAT "post-install-driver-format"
/* object */
struct _OsinfoInstallScript
@@ -163,6 +165,9 @@ OsinfoPathFormat osinfo_install_script_get_path_format(OsinfoInstallScript *scri
gboolean osinfo_install_script_get_can_pre_install_drivers(OsinfoInstallScript *script);
gboolean osinfo_install_script_get_can_post_install_drivers(OsinfoInstallScript *script);
+OsinfoDeviceDriverFormat osinfo_install_script_get_pre_install_driver_format(OsinfoInstallScript *script);
+OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(OsinfoInstallScript *script);
+
#endif /* __OSINFO_INSTALL_SCRIPT_H__ */
/*
* Local variables:
diff --git a/osinfo/osinfo_loader.c b/osinfo/osinfo_loader.c
index d186073..aad5a61 100644
--- a/osinfo/osinfo_loader.c
+++ b/osinfo/osinfo_loader.c
@@ -702,6 +702,8 @@ static void osinfo_loader_install_script(OsinfoLoader *loader,
OSINFO_INSTALL_SCRIPT_PROP_EXPECTED_FILENAME,
OSINFO_INSTALL_SCRIPT_PROP_CAN_PRE_INSTALL_DRIVERS,
OSINFO_INSTALL_SCRIPT_PROP_CAN_POST_INSTALL_DRIVERS,
+ OSINFO_INSTALL_SCRIPT_PROP_PRE_INSTALL_DEVICE_DRIVER_FORMAT,
+ OSINFO_INSTALL_SCRIPT_PROP_POST_INSTALL_DEVICE_DRIVER_FORMAT,
NULL
};
gchar *value = NULL;
--
1.8.1
From zeeshanak at gnome.org Mon Jan 28 03:18:44 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:44 +0200
Subject: [Libosinfo] [PATCH 5/8] API to query signed status of device drivers
In-Reply-To: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1359343127-27272-6-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Some vendors recommend or require device drivers to be signed by them
before these device drivers could be installed on their OS.
---
data/oses/windows.xml.in | 8 ++++----
data/schemas/libosinfo.rng | 5 +++++
osinfo/libosinfo.syms | 1 +
osinfo/osinfo_device_driver.c | 16 ++++++++++++++++
osinfo/osinfo_device_driver.h | 2 ++
osinfo/osinfo_loader.c | 8 ++++++++
6 files changed, 36 insertions(+), 4 deletions(-)
diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
index 1b762f8..f60786c 100644
--- a/data/oses/windows.xml.in
+++ b/data/oses/windows.xml.in
@@ -383,7 +383,7 @@
-
+ viostor.catviostor.infviostor.sys
@@ -392,7 +392,7 @@
-
+ viostor.catviostor.infviostor.sys
@@ -754,14 +754,14 @@
-
+ viostor.catviostor.infviostor.sys
-
+ viostor.catviostor.infviostor.sys
diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
index c193bcc..a3eb01c 100644
--- a/data/schemas/libosinfo.rng
+++ b/data/schemas/libosinfo.rng
@@ -452,6 +452,11 @@
+
+
+
+
+
diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index 3e71d9e..8998bf4 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -403,6 +403,7 @@ LIBOSINFO_0.2.4 {
global:
osinfo_device_driver_format_get_type;
osinfo_device_driver_get_format;
+ osinfo_device_driver_get_signed;
osinfo_install_script_get_pre_install_driver_format;
osinfo_install_script_get_post_install_driver_format;
} LIBOSINFO_0.2.3;
diff --git a/osinfo/osinfo_device_driver.c b/osinfo/osinfo_device_driver.c
index d88273a..ef2c204 100644
--- a/osinfo/osinfo_device_driver.c
+++ b/osinfo/osinfo_device_driver.c
@@ -181,6 +181,22 @@ void osinfo_device_driver_add_device(OsinfoDeviceDriver *driver,
OSINFO_ENTITY(device));
}
+/**
+ * osinfo_device_driver_get_signed:
+ * @driver: an #OsinfoDeviceDriver instance
+ *
+ * Some vendors recommend or require device drivers to be signed by them before
+ * these device drivers could be installed on their OS.
+ *
+ * Returns: TRUE if @driver is signed, FALSE otherwise.
+ */
+gboolean osinfo_device_driver_get_signed(OsinfoDeviceDriver *driver)
+{
+ return osinfo_entity_get_param_value_boolean
+ (OSINFO_ENTITY(driver),
+ OSINFO_DEVICE_DRIVER_PROP_SIGNED);
+}
+
/*
* Local variables:
* indent-tabs-mode: nil
diff --git a/osinfo/osinfo_device_driver.h b/osinfo/osinfo_device_driver.h
index b45f0ab..4ee2322 100644
--- a/osinfo/osinfo_device_driver.h
+++ b/osinfo/osinfo_device_driver.h
@@ -56,6 +56,7 @@ typedef struct _OsinfoDeviceDriverPrivate OsinfoDeviceDriverPrivate;
#define OSINFO_DEVICE_DRIVER_PROP_FILE "file"
#define OSINFO_DEVICE_DRIVER_PROP_DEVICE "device"
#define OSINFO_DEVICE_DRIVER_PROP_FORMAT "format"
+#define OSINFO_DEVICE_DRIVER_PROP_SIGNED "signed"
/* object */
struct _OsinfoDeviceDriver
@@ -99,6 +100,7 @@ gboolean osinfo_device_driver_get_pre_installable(OsinfoDeviceDriver *driver);
GList *osinfo_device_driver_get_files(OsinfoDeviceDriver *driver);
OsinfoDeviceList *osinfo_device_driver_get_devices(OsinfoDeviceDriver *driver);
OsinfoDeviceDriverFormat osinfo_device_driver_get_format(OsinfoDeviceDriver *driver);
+gboolean osinfo_device_driver_get_signed(OsinfoDeviceDriver *driver);
#endif /* __OSINFO_DEVICE_DRIVER_H__ */
/*
diff --git a/osinfo/osinfo_loader.c b/osinfo/osinfo_loader.c
index aad5a61..776a4b6 100644
--- a/osinfo/osinfo_loader.c
+++ b/osinfo/osinfo_loader.c
@@ -1021,6 +1021,7 @@ static OsinfoDeviceDriver *osinfo_loader_driver(OsinfoLoader *loader,
xmlChar *location = xmlGetProp(root, BAD_CAST OSINFO_DEVICE_DRIVER_PROP_LOCATION);
xmlChar *preinst = xmlGetProp(root, BAD_CAST OSINFO_DEVICE_DRIVER_PROP_PRE_INSTALLABLE);
xmlChar *format = xmlGetProp(root, BAD_CAST OSINFO_DEVICE_DRIVER_PROP_FORMAT);
+ xmlChar *is_signed = xmlGetProp(root, BAD_CAST OSINFO_DEVICE_DRIVER_PROP_SIGNED);
OsinfoDeviceDriver *driver = osinfo_device_driver_new(id);
@@ -1052,6 +1053,13 @@ static OsinfoDeviceDriver *osinfo_loader_driver(OsinfoLoader *loader,
xmlFree(format);
}
+ if (is_signed) {
+ osinfo_entity_set_param(OSINFO_ENTITY(driver),
+ OSINFO_DEVICE_DRIVER_PROP_SIGNED,
+ (gchar *)is_signed);
+ xmlFree(is_signed);
+ }
+
gint nnodes = osinfo_loader_nodeset("./*", ctxt, &nodes, err);
if (error_is_set(err)) {
g_object_unref(G_OBJECT(driver));
--
1.8.1
From zeeshanak at gnome.org Mon Jan 28 03:18:46 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:46 +0200
Subject: [Libosinfo] [PATCH 7/8] winxp,
win7: Add virtio & QXL device drivers info
In-Reply-To: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1359343127-27272-8-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Add an all-in-one virtio and QXL device driver setup binary to Windows
XP and 7. This needs to be listed separately from virtio block driver as
not all of these are pre-installable.
As bonus, the same binary also installs spice-vdagent for us.
Unfortunately we wont be able to actually use this driver in Windows 7
installer script as:
* 64-bit (the most commonly used version AFAIK) Windows 7 simply refuses
to install unsigned drivers.
* I failed to find any way to tell windows 7 to ignore unsigned status
of the drivers.
More (but rather confusing information) here:
http://technet.microsoft.com/en-us/library/dd744517(WS.10).aspx#SignatureReqs
---
data/oses/windows.xml.in | 39 +++++++++++++++++++++++++++++++++++++++
1 file changed, 39 insertions(+)
diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
index f60786c..bfc3cbc 100644
--- a/data/oses/windows.xml.in
+++ b/data/oses/windows.xml.in
@@ -400,6 +400,25 @@
txtsetup.oem
+
+
+
+ spice-guest-tools-0.3.exe
+
+
+
+
+
+
+
+
+ spice-guest-tools-0.3.exe
+
+
+
+
+
+
@@ -765,7 +784,27 @@
viostor.catviostor.infviostor.sys
+
+
+
+
+
+
+ spice-guest-tools-0.3.exe
+
+
+
+
+
+
+
+
+ spice-guest-tools-0.3.exe
+
+
+
+
--
1.8.1
From zeeshanak at gnome.org Mon Jan 28 03:18:45 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:45 +0200
Subject: [Libosinfo] [PATCH 6/8] installer: API to query signed device
driver requirement
In-Reply-To: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1359343127-27272-7-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
---
data/install-scripts/windows-sif.xml | 2 ++
data/install-scripts/windows-unattend.xml | 2 ++
data/schemas/libosinfo.rng | 10 +++++++++
osinfo/libosinfo.syms | 2 ++
osinfo/osinfo_install_script.c | 34 +++++++++++++++++++++++++++++++
osinfo/osinfo_install_script.h | 5 +++++
osinfo/osinfo_loader.c | 2 ++
7 files changed, 57 insertions(+)
diff --git a/data/install-scripts/windows-sif.xml b/data/install-scripts/windows-sif.xml
index 8634e3a..972d9b1 100644
--- a/data/install-scripts/windows-sif.xml
+++ b/data/install-scripts/windows-sif.xml
@@ -6,6 +6,7 @@
$$$$$-$$$$$-$$$$$-$$$$$-$$$$$winnt.siftrue
+ truewindows
@@ -74,6 +75,7 @@
$$$$$-$$$$$-$$$$$-$$$$$-$$$$$winnt.siftrue
+ truewindows
diff --git a/data/install-scripts/windows-unattend.xml b/data/install-scripts/windows-unattend.xml
index e306ec4..43020bc 100644
--- a/data/install-scripts/windows-unattend.xml
+++ b/data/install-scripts/windows-unattend.xml
@@ -5,6 +5,7 @@
$$$$$-$$$$$-$$$$$-$$$$$-$$$$$autounattend.xmltrue
+ truewindows
@@ -221,6 +222,7 @@
$$$$$-$$$$$-$$$$$-$$$$$-$$$$$autounattend.xmltrue
+ truewindows
diff --git a/data/schemas/libosinfo.rng b/data/schemas/libosinfo.rng
index a3eb01c..62a7243 100644
--- a/data/schemas/libosinfo.rng
+++ b/data/schemas/libosinfo.rng
@@ -590,6 +590,11 @@
+
+
+
+
+
@@ -599,6 +604,11 @@
+
+
+
+
+
diff --git a/osinfo/libosinfo.syms b/osinfo/libosinfo.syms
index 8998bf4..a70c7de 100644
--- a/osinfo/libosinfo.syms
+++ b/osinfo/libosinfo.syms
@@ -404,6 +404,8 @@ LIBOSINFO_0.2.4 {
osinfo_device_driver_format_get_type;
osinfo_device_driver_get_format;
osinfo_device_driver_get_signed;
+ osinfo_install_script_get_can_pre_install_unsigned_drivers;
+ osinfo_install_script_get_can_post_install_unsigned_drivers;
osinfo_install_script_get_pre_install_driver_format;
osinfo_install_script_get_post_install_driver_format;
} LIBOSINFO_0.2.3;
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index ca45b60..67d4391 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -1250,6 +1250,40 @@ OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(Os
OSINFO_DEVICE_DRIVER_FORMAT_EXECUTABLE);
}
+/**
+ * osinfo_install_script_get_can_pre_install_unsigned_drivers:
+ * @script: the install script
+ *
+ * If install script can install drivers at the very beginning of installation,
+ * this function can be used to query if it can handle unsigned drivers.
+ *
+ * Returns: TRUE if install script supports unsigned pre-installable drivers,
+ * FASLSE otherwise.
+ */
+gboolean osinfo_install_script_get_can_pre_install_unsigned_drivers(OsinfoInstallScript *script)
+{
+ return osinfo_entity_get_param_value_boolean
+ (OSINFO_ENTITY(script),
+ OSINFO_INSTALL_SCRIPT_PROP_CAN_PRE_INSTALL_UNSIGNED_DRIVERS);
+}
+
+/**
+ * osinfo_install_script_get_can_post_install_unsigned_drivers:
+ * @script: the install script
+ *
+ * If install script can install drivers at the end of installation, this
+ * function can be used to query if it can handle unsigned drivers.
+ *
+ * Returns: TRUE if install script supports unsigned post-installable drivers,
+ * FASLSE otherwise.
+ */
+gboolean osinfo_install_script_get_can_post_install_unsigned_drivers(OsinfoInstallScript *script)
+{
+ return osinfo_entity_get_param_value_boolean
+ (OSINFO_ENTITY(script),
+ OSINFO_INSTALL_SCRIPT_PROP_CAN_POST_INSTALL_UNSIGNED_DRIVERS);
+}
+
/*
* Local variables:
* indent-tabs-mode: nil
diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
index 82486ef..35a85e4 100644
--- a/osinfo/osinfo_install_script.h
+++ b/osinfo/osinfo_install_script.h
@@ -57,6 +57,8 @@ typedef struct _OsinfoInstallScriptPrivate OsinfoInstallScriptPrivate;
#define OSINFO_INSTALL_SCRIPT_PROP_CAN_POST_INSTALL_DRIVERS "can-post-install-drivers"
#define OSINFO_INSTALL_SCRIPT_PROP_PRE_INSTALL_DEVICE_DRIVER_FORMAT "pre-install-driver-format"
#define OSINFO_INSTALL_SCRIPT_PROP_POST_INSTALL_DEVICE_DRIVER_FORMAT "post-install-driver-format"
+#define OSINFO_INSTALL_SCRIPT_PROP_CAN_PRE_INSTALL_UNSIGNED_DRIVERS "can-pre-install-unsigned-drivers"
+#define OSINFO_INSTALL_SCRIPT_PROP_CAN_POST_INSTALL_UNSIGNED_DRIVERS "can-post-install-unsigned-drivers"
/* object */
struct _OsinfoInstallScript
@@ -168,6 +170,9 @@ gboolean osinfo_install_script_get_can_post_install_drivers(OsinfoInstallScript
OsinfoDeviceDriverFormat osinfo_install_script_get_pre_install_driver_format(OsinfoInstallScript *script);
OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(OsinfoInstallScript *script);
+gboolean osinfo_install_script_get_can_pre_install_unsigned_drivers(OsinfoInstallScript *script);
+gboolean osinfo_install_script_get_can_post_install_unsigned_drivers(OsinfoInstallScript *script);
+
#endif /* __OSINFO_INSTALL_SCRIPT_H__ */
/*
* Local variables:
diff --git a/osinfo/osinfo_loader.c b/osinfo/osinfo_loader.c
index 776a4b6..19780a1 100644
--- a/osinfo/osinfo_loader.c
+++ b/osinfo/osinfo_loader.c
@@ -704,6 +704,8 @@ static void osinfo_loader_install_script(OsinfoLoader *loader,
OSINFO_INSTALL_SCRIPT_PROP_CAN_POST_INSTALL_DRIVERS,
OSINFO_INSTALL_SCRIPT_PROP_PRE_INSTALL_DEVICE_DRIVER_FORMAT,
OSINFO_INSTALL_SCRIPT_PROP_POST_INSTALL_DEVICE_DRIVER_FORMAT,
+ OSINFO_INSTALL_SCRIPT_PROP_CAN_PRE_INSTALL_UNSIGNED_DRIVERS,
+ OSINFO_INSTALL_SCRIPT_PROP_CAN_POST_INSTALL_UNSIGNED_DRIVERS,
NULL
};
gchar *value = NULL;
--
1.8.1
From zeeshanak at gnome.org Mon Jan 28 03:18:47 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 05:18:47 +0200
Subject: [Libosinfo] [PATCH 8/8] winxp,
installer: Make use of post-installation drivers
In-Reply-To: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <1359343127-27272-9-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
---
data/install-scripts/windows-cmd.xml | 17 +++++++++++++++++
1 file changed, 17 insertions(+)
diff --git a/data/install-scripts/windows-cmd.xml b/data/install-scripts/windows-cmd.xml
index 85aae12..0288394 100644
--- a/data/install-scripts/windows-cmd.xml
+++ b/data/install-scripts/windows-cmd.xml
@@ -4,6 +4,9 @@
desktopdoswindows.cmd
+ true
+ true
+ executable
@@ -11,6 +14,8 @@
+
+
image/bmp
@@ -46,6 +51,17 @@
+
+
+
+
+
+
+ A
+
+
+
+
sc config TlntSvr start= auto
net user /add /passwordreq:no
@@ -53,6 +69,7 @@ net localgroup administrators /add
net accounts /maxpwage:unlimited
if not ""=="" copy ":" ":\Documents and Settings\All Users\Application Data\Microsoft\User Account Pictures\.bmp"
REGEDIT /S :\windows.reg
+for %%i in (":\*.exe") do %%i /S
EXIT
--
1.8.1
From cfergeau at redhat.com Mon Jan 28 10:14:41 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 11:14:41 +0100
Subject: [Libosinfo] [PATCH 1/8] winxp,
installer: Don't assume script disk is specified
In-Reply-To: <1359343127-27272-2-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-2-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130128101441.GB2332@teriyaki.redhat.com>
Hey,
On Mon, Jan 28, 2013 at 05:18:40AM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> Declare that script optionally uses script disk if specified and hardcode
> a fallback for the case when its not specified.
Looks good to me, ACK. Can WinXP pick its winnt.sif file from places that
are not A: ?
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 10:19:53 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 11:19:53 +0100
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To: <1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130128101953.GC2332@teriyaki.redhat.com>
Hey,
On Mon, Jan 28, 2013 at 05:18:41AM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> We are unlikely to find any signed free drivers for windows that are
> signed. Better just ask windows to ignore.
As this disables some kind of security measures, I'd rather that we don't
silently disable this, but let the user control this setting and have
signature checking enabled by default (the world is not just virtio and qxl
drivers).
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 10:27:35 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 11:27:35 +0100
Subject: [Libosinfo] [PATCH 4/8] installer: API to query supported
device driver formats
In-Reply-To: <1359343127-27272-5-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-5-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130128102735.GD2332@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 05:18:43AM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
> index d91751e..82486ef 100644
> --- a/osinfo/osinfo_install_script.h
> +++ b/osinfo/osinfo_install_script.h
> @@ -163,6 +165,9 @@ OsinfoPathFormat osinfo_install_script_get_path_format(OsinfoInstallScript *scri
> gboolean osinfo_install_script_get_can_pre_install_drivers(OsinfoInstallScript *script);
> gboolean osinfo_install_script_get_can_post_install_drivers(OsinfoInstallScript *script);
>
> +OsinfoDeviceDriverFormat osinfo_install_script_get_pre_install_driver_format(OsinfoInstallScript *script);
> +OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(OsinfoInstallScript *script);
I don't think assuming that a given installer will support only one driver
format is expressive enough. For Windows post-install drivers, supporting
unpacked Windows drivers in addition to running a .exe shouldn't be very
hard, and this API would not work there.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 10:40:35 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 11:40:35 +0100
Subject: [Libosinfo] [PATCH 5/8] API to query signed status of device
drivers
In-Reply-To: <1359343127-27272-6-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-6-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130128104035.GE2332@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 05:18:44AM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> Some vendors recommend or require device drivers to be signed by them
> before these device drivers could be installed on their OS.
It would be nice to explain somewhere what 'signed' means for Windows, as
there can be multiple signatures on a driver (vendor and MS). Also, this
could be interesting to know if this scheme would work for
https://fedoraproject.org/wiki/Features/PackageSignatureCheckingDuringOSInstall?rd=Features/PackageSignatureCheckingDuringInstall
(if relevant), but it's probably a bit early for that :(
> diff --git a/data/oses/windows.xml.in b/data/oses/windows.xml.in
> index 1b762f8..f60786c 100644
> --- a/data/oses/windows.xml.in
> +++ b/data/oses/windows.xml.in
> @@ -383,7 +383,7 @@
>
>
>
> -
> +
These changes are not strictly needed as FALSE is the default for 'signed'
Looks good otherwise,
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 10:48:49 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 11:48:49 +0100
Subject: [Libosinfo] [PATCH 6/8] installer: API to query signed device
driver requirement
In-Reply-To: <1359343127-27272-7-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-7-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130128104849.GF2332@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 05:18:45AM +0200, Zeeshan Ali (Khattak) wrote:
> +/**
> + * osinfo_install_script_get_can_post_install_unsigned_drivers:
> + * @script: the install script
> + *
> + * If install script can install drivers at the end of installation, this
> + * function can be used to query if it can handle unsigned drivers.
All of this is a bit vague to me, this seems more tri-state than just a
boolean, most Windows will warn when trying to install unsigned-by-MS
drivers but will let you go through after saying you are ok with installing
this driver. Starting with win7 64 bit, the OS will refuse to install the
drivers without letting you click your way around it.
This commit seems to assume that 'having to click' = 'cannot post install
the driver'. While inconvenient during post-install, I can see this being
better than nothing to some libosinfo users.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 11:01:22 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 12:01:22 +0100
Subject: [Libosinfo] [PATCH 7/8] winxp,
win7: Add virtio & QXL device drivers info
In-Reply-To: <1359343127-27272-8-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-8-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130128110122.GG2332@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 05:18:46AM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> Add an all-in-one virtio and QXL device driver setup binary to Windows
> XP and 7. This needs to be listed separately from virtio block driver as
> not all of these are pre-installable.
>
> As bonus, the same binary also installs spice-vdagent for us.
I see this as an unfortunate side-effect, not as a bonus. I don't think
we have a way to express that this agent is nice to have when using
spice, and that it needs vioserial.
>
> Unfortunately we wont be able to actually use this driver in Windows 7
> installer script as:
>
> * 64-bit (the most commonly used version AFAIK) Windows 7 simply refuses
> to install unsigned drivers.
> * I failed to find any way to tell windows 7 to ignore unsigned status
> of the drivers.
I think we should still have a way for libosinfo user to have this binary
run at the end of the install process even though this means having to
click through a few dialogs. For what it's worth, I would have preferred
this series to first add support for post-install drivers while ignoring
driver signature, and then adding the signature stuff on top of it.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 13:14:53 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 14:14:53 +0100
Subject: [Libosinfo] [PATCH 8/8] winxp,
installer: Make use of post-installation drivers
In-Reply-To: <1359343127-27272-9-git-send-email-zeeshanak@gnome.org>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-9-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130128131453.GN2332@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 05:18:47AM +0200, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> ---
> data/install-scripts/windows-cmd.xml | 17 +++++++++++++++++
> 1 file changed, 17 insertions(+)
>
> diff --git a/data/install-scripts/windows-cmd.xml b/data/install-scripts/windows-cmd.xml
> index 85aae12..0288394 100644
> --- a/data/install-scripts/windows-cmd.xml
> +++ b/data/install-scripts/windows-cmd.xml
> @@ -4,6 +4,9 @@
> desktop
> dos
> windows.cmd
> + true
> + true
> + executable
>
>
>
> @@ -11,6 +14,8 @@
>
>
>
> +
> +
>
>
> image/bmp
> @@ -46,6 +51,17 @@
>
>
>
> +
> +
> +
> +
> +
> +
> + A
> +
> +
> +
> +
>
> sc config TlntSvr start= auto
> net user /add /passwordreq:no
> @@ -53,6 +69,7 @@ net localgroup administrators /add
> net accounts /maxpwage:unlimited
> if not ""=="" copy ":" ":\Documents and Settings\All Users\Application Data\Microsoft\User Account Pictures\.bmp"
> REGEDIT /S :\windows.reg
> +for %%i in (":\*.exe") do %%i /S
You can't expect every possible Windows exe file to honour the /S switch.
I'd tend to go with a list of files + a .cmd file which contains the actual
commands to run (osinfo-*.cmd would be even better but I'm not 100% sure
Windows accept such wildcards).
This would let us handle .msi file, unpacked drivers should be possible too
with an appropriate .cmd file, ...
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From zeeshanak at gnome.org Mon Jan 28 13:35:51 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 15:35:51 +0200
Subject: [Libosinfo] [PATCH 1/8] winxp,
installer: Don't assume script disk is specified
In-Reply-To: <20130128101441.GB2332@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-2-git-send-email-zeeshanak@gnome.org>
<20130128101441.GB2332@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 12:14 PM, Christophe Fergeau
wrote:
> Hey,
Hi,
> On Mon, Jan 28, 2013 at 05:18:40AM +0200, Zeeshan Ali (Khattak) wrote:
>> From: "Zeeshan Ali (Khattak)"
>>
>> Declare that script optionally uses script disk if specified and hardcode
>> a fallback for the case when its not specified.
>
> Looks good to me, ACK. Can WinXP pick its winnt.sif file from places that
> are not A: ?
It might be able to pick it up from B: as well but I haven't tried.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 14:40:06 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 16:40:06 +0200
Subject: [Libosinfo] [PATCH 4/8] installer: API to query supported
device driver formats
In-Reply-To: <20130128102735.GD2332@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-5-git-send-email-zeeshanak@gnome.org>
<20130128102735.GD2332@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 12:27 PM, Christophe Fergeau
wrote:
> On Mon, Jan 28, 2013 at 05:18:43AM +0200, Zeeshan Ali (Khattak) wrote:
>> From: "Zeeshan Ali (Khattak)"
>> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
>> index d91751e..82486ef 100644
>> --- a/osinfo/osinfo_install_script.h
>> +++ b/osinfo/osinfo_install_script.h
>> @@ -163,6 +165,9 @@ OsinfoPathFormat osinfo_install_script_get_path_format(OsinfoInstallScript *scri
>> gboolean osinfo_install_script_get_can_pre_install_drivers(OsinfoInstallScript *script);
>> gboolean osinfo_install_script_get_can_post_install_drivers(OsinfoInstallScript *script);
>>
>> +OsinfoDeviceDriverFormat osinfo_install_script_get_pre_install_driver_format(OsinfoInstallScript *script);
>> +OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(OsinfoInstallScript *script);
>
> I don't think assuming that a given installer will support only one driver
> format is expressive enough. For Windows post-install drivers, supporting
> unpacked Windows drivers in addition to running a .exe shouldn't be very
> hard, and this API would not work there.
I see you point. I can make it a list. Would that be good?
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 14:51:04 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 16:51:04 +0200
Subject: [Libosinfo] [PATCH 6/8] installer: API to query signed device
driver requirement
In-Reply-To: <20130128104849.GF2332@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-7-git-send-email-zeeshanak@gnome.org>
<20130128104849.GF2332@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 12:48 PM, Christophe Fergeau
wrote:
> On Mon, Jan 28, 2013 at 05:18:45AM +0200, Zeeshan Ali (Khattak) wrote:
>> +/**
>> + * osinfo_install_script_get_can_post_install_unsigned_drivers:
>> + * @script: the install script
>> + *
>> + * If install script can install drivers at the end of installation, this
>> + * function can be used to query if it can handle unsigned drivers.
>
> All of this is a bit vague to me, this seems more tri-state than just a
> boolean, most Windows will warn when trying to install unsigned-by-MS
> drivers but will let you go through after saying you are ok with installing
> this driver. Starting with win7 64 bit, the OS will refuse to install the
> drivers without letting you click your way around it.
> This commit seems to assume that 'having to click' = 'cannot post install
> the driver'. While inconvenient during post-install, I can see this being
> better than nothing to some libosinfo users.
While in Boxes, we certainly wouldn't want a warning message as it
breaks the whole concept of express installation, I guess other apps
could be given a choice here.
How about an enum like this:
enum DeviceDriverSignReq {
DEVICE_DRIVER_SIGN_REQ_STRICT, /* Script must only be given signed
device drivers. */
DEVICE_DRIVER_SIGN_REQ_WARN, /* A warning will be issued by OS
installer if device drivers are not signed. */
DEVICE_DRIVER_SIGN_REQ_NONE, /* Script do not require device drivers
to be signed. */
}
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 15:05:28 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 17:05:28 +0200
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To: <20130128101953.GC2332@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
<20130128101953.GC2332@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 12:19 PM, Christophe Fergeau
wrote:
> Hey,
>
> On Mon, Jan 28, 2013 at 05:18:41AM +0200, Zeeshan Ali (Khattak) wrote:
>> From: "Zeeshan Ali (Khattak)"
>>
>> We are unlikely to find any signed free drivers for windows that are
>> signed. Better just ask windows to ignore.
>
> As this disables some kind of security measures, I'd rather that we don't
> silently disable this, but let the user control this setting and have
> signature checking enabled by default
The problem is that having it configurable would mean that we can't
have a static info on the script about the signature requirement (see
"installer: API to query signed device driver requirement" patch).
I don't really see this a real security feature of windows but more a
control feature of Microsoft as they have made it impossible for free
drivers to be signed. So not very much motivated to do a lot of rework
to accommodate configurability here.
> (the world is not just virtio and qxl
> drivers).
If it was possible to have free (as in Free Software) drivers that
were signed, we wouldn't have issues with virtio and qxl either.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 15:27:16 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 17:27:16 +0200
Subject: [Libosinfo] [PATCH 7/8] winxp,
win7: Add virtio & QXL device drivers info
In-Reply-To: <20130128110122.GG2332@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-8-git-send-email-zeeshanak@gnome.org>
<20130128110122.GG2332@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 1:01 PM, Christophe Fergeau wrote:
> On Mon, Jan 28, 2013 at 05:18:46AM +0200, Zeeshan Ali (Khattak) wrote:
>> From: "Zeeshan Ali (Khattak)"
>>
>> Add an all-in-one virtio and QXL device driver setup binary to Windows
>> XP and 7. This needs to be listed separately from virtio block driver as
>> not all of these are pre-installable.
>>
>> As bonus, the same binary also installs spice-vdagent for us.
>
> I see this as an unfortunate side-effect, not as a bonus. I don't think
> we have a way to express that this agent is nice to have when using
> spice, and that it needs vioserial.
>
>>
>> Unfortunately we wont be able to actually use this driver in Windows 7
>> installer script as:
>>
>> * 64-bit (the most commonly used version AFAIK) Windows 7 simply refuses
>> to install unsigned drivers.
>> * I failed to find any way to tell windows 7 to ignore unsigned status
>> of the drivers.
>
> I think we should still have a way for libosinfo user to have this binary
> run at the end of the install process even though this means having to
> click through a few dialogs.
Apart from the 3-state (rather than boolean) API I proposed in a
previous mail, we could also have a separate API for adding random
binaries that are put on user's desktop and apps can then use that for
things like these. In this specific case, app will see that there is a
driver available for the os in executable format but script warns on
it during install so they can decide to use this API to add the binary
to the desktop with filename "Installer for X".
Such an API could be useful in general and allow us to not complicate
drivers-related API and most importantly not break the concept of
automated/express installation.
>For what it's worth, I would have preferred
> this series to first add support for post-install drivers while ignoring
> driver signature, and then adding the signature stuff on top of it.
I must admit, these patches weren't very well planned.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From cfergeau at redhat.com Mon Jan 28 15:41:58 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 16:41:58 +0100
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
<20130128101953.GC2332@teriyaki.redhat.com>
Message-ID: <20130128154157.GB22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 05:05:28PM +0200, Zeeshan Ali (Khattak) wrote:
> I don't really see this a real security feature of windows but more a
> control feature of Microsoft as they have made it impossible for free
> drivers to be signed.
[snip]
> If it was possible to have free (as in Free Software) drivers that
> were signed, we wouldn't have issues with virtio and qxl either.
It is possible to have signed free drivers, it's copyleft licences that
they are rejecting.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From zeeshanak at gnome.org Mon Jan 28 15:45:25 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 17:45:25 +0200
Subject: [Libosinfo] [PATCH 8/8] winxp,
installer: Make use of post-installation drivers
In-Reply-To: <20130128131453.GN2332@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-9-git-send-email-zeeshanak@gnome.org>
<20130128131453.GN2332@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 3:14 PM, Christophe Fergeau wrote:
> On Mon, Jan 28, 2013 at 05:18:47AM +0200, Zeeshan Ali (Khattak) wrote:
>> From: "Zeeshan Ali (Khattak)"
>>
>> ---
>> data/install-scripts/windows-cmd.xml | 17 +++++++++++++++++
>> 1 file changed, 17 insertions(+)
>>
>> diff --git a/data/install-scripts/windows-cmd.xml b/data/install-scripts/windows-cmd.xml
>> index 85aae12..0288394 100644
>> --- a/data/install-scripts/windows-cmd.xml
>> +++ b/data/install-scripts/windows-cmd.xml
>> @@ -4,6 +4,9 @@
>> desktop
>> dos
>> windows.cmd
>> + true
>> + true
>> + executable
>>
>>
>>
>> @@ -11,6 +14,8 @@
>>
>>
>>
>> +
>> +
>>
>>
>> image/bmp
>> @@ -46,6 +51,17 @@
>>
>>
>>
>> +
>> +
>> +
>> +
>> +
>> +
>> + A
>> +
>> +
>> +
>> +
>>
>> sc config TlntSvr start= auto
>> net user /add /passwordreq:no
>> @@ -53,6 +69,7 @@ net localgroup administrators /add
>> net accounts /maxpwage:unlimited
>> if not ""=="" copy ":" ":\Documents and Settings\All Users\Application Data\Microsoft\User Account Pictures\.bmp"
>> REGEDIT /S :\windows.reg
>> +for %%i in (":\*.exe") do %%i /S
>
> You can't expect every possible Windows exe file to honour the /S switch.
If a binary doesn't support '/S' switch, it will ignore it?
> I'd tend to go with a list of files + a .cmd file which contains the actual
> commands to run (osinfo-*.cmd would be even better but I'm not 100% sure
> Windows accept such wildcards).
> This would let us handle .msi file, unpacked drivers should be possible too
> with an appropriate .cmd file, ...
This would mean that either us or the driver setup distributor will
have to provide a cmd file with it?
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 16:05:53 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 18:05:53 +0200
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To: <20130128154157.GB22845@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
<20130128101953.GC2332@teriyaki.redhat.com>
<20130128154157.GB22845@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 5:41 PM, Christophe Fergeau wrote:
> On Mon, Jan 28, 2013 at 05:05:28PM +0200, Zeeshan Ali (Khattak) wrote:
>> I don't really see this a real security feature of windows but more a
>> control feature of Microsoft as they have made it impossible for free
>> drivers to be signed.
>
> [snip]
>
>> If it was possible to have free (as in Free Software) drivers that
>> were signed, we wouldn't have issues with virtio and qxl either.
>
> It is possible to have signed free drivers, it's copyleft licences that
> they are rejecting.
whats the difference between copyleft and free, as in Free Software?
Reading this, I don't see any:
http://en.wikipedia.org/wiki/Copyleft
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From cfergeau at redhat.com Mon Jan 28 16:13:27 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 17:13:27 +0100
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
<20130128101953.GC2332@teriyaki.redhat.com>
<20130128154157.GB22845@teriyaki.redhat.com>
Message-ID: <20130128161327.GC22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 06:05:53PM +0200, Zeeshan Ali (Khattak) wrote:
> whats the difference between copyleft and free, as in Free Software?
> Reading this, I don't see any:
>
> http://en.wikipedia.org/wiki/Copyleft
http://en.wikipedia.org/wiki/Copyleft#Types_of_copyleft_and_relation_to_other_licenses
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 17:19:43 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 18:19:43 +0100
Subject: [Libosinfo] [PATCH 8/8] winxp,
installer: Make use of post-installation drivers
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-9-git-send-email-zeeshanak@gnome.org>
<20130128131453.GN2332@teriyaki.redhat.com>
Message-ID: <20130128171943.GD22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 05:45:25PM +0200, Zeeshan Ali (Khattak) wrote:
> On Mon, Jan 28, 2013 at 3:14 PM, Christophe Fergeau wrote:
> > You can't expect every possible Windows exe file to honour the /S switch.
>
> If a binary doesn't support '/S' switch, it will ignore it?
Or fail saying it's an unknown parameter?
>
> > I'd tend to go with a list of files + a .cmd file which contains the actual
> > commands to run (osinfo-*.cmd would be even better but I'm not 100% sure
> > Windows accept such wildcards).
> > This would let us handle .msi file, unpacked drivers should be possible too
> > with an appropriate .cmd file, ...
>
> This would mean that either us or the driver setup distributor will
> have to provide a cmd file with it?
Yup, but we currently only have a single binary to support, and there is
already a need for a flag, so we need a way to support this in a more
flexible way than hardcoding it in libosinfo imo.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 17:25:06 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 18:25:06 +0100
Subject: [Libosinfo] [PATCH 7/8] winxp,
win7: Add virtio & QXL device drivers info
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-8-git-send-email-zeeshanak@gnome.org>
<20130128110122.GG2332@teriyaki.redhat.com>
Message-ID: <20130128172506.GE22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 05:27:16PM +0200, Zeeshan Ali (Khattak) wrote:
> On Mon, Jan 28, 2013 at 1:01 PM, Christophe Fergeau wrote:
> > I think we should still have a way for libosinfo user to have this binary
> > run at the end of the install process even though this means having to
> > click through a few dialogs.
>
> Apart from the 3-state (rather than boolean) API I proposed in a
> previous mail, we could also have a separate API for adding random
> binaries that are put on user's desktop and apps can then use that for
> things like these. In this specific case, app will see that there is a
> driver available for the os in executable format but script warns on
> it during install so they can decide to use this API to add the binary
> to the desktop with filename "Installer for X".
I don't think this is really useful... And I'd prefer to have the binary
be run for me at the end of installation even if it means it's showing
dialog.
> Such an API could be useful in general and allow us to not complicate
> drivers-related API and most importantly not break the concept of
> automated/express installation.
Depends what you value most, fully unattended install, or some guarantee
that the drivers will be installed when the install completes even if this
means blocking. While we want the former to be the default, I think it
would be nice to have some support for the latter.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 17:28:30 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 18:28:30 +0100
Subject: [Libosinfo] [PATCH 4/8] installer: API to query supported
device driver formats
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-5-git-send-email-zeeshanak@gnome.org>
<20130128102735.GD2332@teriyaki.redhat.com>
Message-ID: <20130128172830.GF22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 04:40:06PM +0200, Zeeshan Ali (Khattak) wrote:
> On Mon, Jan 28, 2013 at 12:27 PM, Christophe Fergeau
> wrote:
> > On Mon, Jan 28, 2013 at 05:18:43AM +0200, Zeeshan Ali (Khattak) wrote:
> >> From: "Zeeshan Ali (Khattak)"
> >> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
> >> index d91751e..82486ef 100644
> >> --- a/osinfo/osinfo_install_script.h
> >> +++ b/osinfo/osinfo_install_script.h
> >> @@ -163,6 +165,9 @@ OsinfoPathFormat osinfo_install_script_get_path_format(OsinfoInstallScript *scri
> >> gboolean osinfo_install_script_get_can_pre_install_drivers(OsinfoInstallScript *script);
> >> gboolean osinfo_install_script_get_can_post_install_drivers(OsinfoInstallScript *script);
> >>
> >> +OsinfoDeviceDriverFormat osinfo_install_script_get_pre_install_driver_format(OsinfoInstallScript *script);
> >> +OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(OsinfoInstallScript *script);
> >
> > I don't think assuming that a given installer will support only one driver
> > format is expressive enough. For Windows post-install drivers, supporting
> > unpacked Windows drivers in addition to running a .exe shouldn't be very
> > hard, and this API would not work there.
>
> I see you point. I can make it a list. Would that be good?
I think so. Though concretely why do we need to expose this information? In
all cases user of this information will need to drop it to a disk image
which will be passed to the VM, no?
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Mon Jan 28 17:31:36 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Mon, 28 Jan 2013 18:31:36 +0100
Subject: [Libosinfo] [PATCH 6/8] installer: API to query signed device
driver requirement
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-7-git-send-email-zeeshanak@gnome.org>
<20130128104849.GF2332@teriyaki.redhat.com>
Message-ID: <20130128173135.GG22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 04:51:04PM +0200, Zeeshan Ali (Khattak) wrote:
> On Mon, Jan 28, 2013 at 12:48 PM, Christophe Fergeau
> wrote:
> > On Mon, Jan 28, 2013 at 05:18:45AM +0200, Zeeshan Ali (Khattak) wrote:
> >> +/**
> >> + * osinfo_install_script_get_can_post_install_unsigned_drivers:
> >> + * @script: the install script
> >> + *
> >> + * If install script can install drivers at the end of installation, this
> >> + * function can be used to query if it can handle unsigned drivers.
> >
> > All of this is a bit vague to me, this seems more tri-state than just a
> > boolean, most Windows will warn when trying to install unsigned-by-MS
> > drivers but will let you go through after saying you are ok with installing
> > this driver. Starting with win7 64 bit, the OS will refuse to install the
> > drivers without letting you click your way around it.
> > This commit seems to assume that 'having to click' = 'cannot post install
> > the driver'. While inconvenient during post-install, I can see this being
> > better than nothing to some libosinfo users.
>
> While in Boxes, we certainly wouldn't want a warning message as it
> breaks the whole concept of express installation, I guess other apps
> could be given a choice here.
>
> How about an enum like this:
>
> enum DeviceDriverSignReq {
> DEVICE_DRIVER_SIGN_REQ_STRICT, /* Script must only be given signed
> device drivers. */
> DEVICE_DRIVER_SIGN_REQ_WARN, /* A warning will be issued by OS
> installer if device drivers are not signed. */
> DEVICE_DRIVER_SIGN_REQ_NONE, /* Script do not require device drivers
> to be signed. */
> }
I'd change 'Sign' to 'Signing', and make it explicit that the 'REQ_WARN'
case can break unattended installation and require user input. Same for
'REQ_STRICT' actually, in addition to not installing unsigned drivers,
passing an unsigned driver may also break unattended install (?)
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From zeeshanak at gnome.org Mon Jan 28 20:54:52 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 22:54:52 +0200
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To: <20130128161327.GC22845@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
<20130128101953.GC2332@teriyaki.redhat.com>
<20130128154157.GB22845@teriyaki.redhat.com>
<20130128161327.GC22845@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 6:13 PM, Christophe Fergeau wrote:
> On Mon, Jan 28, 2013 at 06:05:53PM +0200, Zeeshan Ali (Khattak) wrote:
>> whats the difference between copyleft and free, as in Free Software?
>> Reading this, I don't see any:
>>
>> http://en.wikipedia.org/wiki/Copyleft
>
> http://en.wikipedia.org/wiki/Copyleft#Types_of_copyleft_and_relation_to_other_licenses
OK, I stand corrected about the exact definition of copyleft. :)
Since you are stressing on this distinction, I imagine you know some
free drivers out there that are signed?
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 21:06:53 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 23:06:53 +0200
Subject: [Libosinfo] [PATCH 8/8] winxp,
installer: Make use of post-installation drivers
In-Reply-To: <20130128171943.GD22845@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-9-git-send-email-zeeshanak@gnome.org>
<20130128131453.GN2332@teriyaki.redhat.com>
<20130128171943.GD22845@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 7:19 PM, Christophe Fergeau wrote:
> On Mon, Jan 28, 2013 at 05:45:25PM +0200, Zeeshan Ali (Khattak) wrote:
>> On Mon, Jan 28, 2013 at 3:14 PM, Christophe Fergeau wrote:
>> > You can't expect every possible Windows exe file to honour the /S switch.
>>
>> If a binary doesn't support '/S' switch, it will ignore it?
>
> Or fail saying it's an unknown parameter?
According to Vmware docs, this is a standard MSI option:
http://pubs.vmware.com/view-50/index.jsp?topic=/com.vmware.view.installation.doc/GUID-F19EBDF6-20A3-4A8B-95AE-786CC74F26AF.html
If that is the case, its supposed to be guaranteed on every windows
installer setup app?
>> > I'd tend to go with a list of files + a .cmd file which contains the actual
>> > commands to run (osinfo-*.cmd would be even better but I'm not 100% sure
>> > Windows accept such wildcards).
>> > This would let us handle .msi file, unpacked drivers should be possible too
>> > with an appropriate .cmd file, ...
>>
>> This would mean that either us or the driver setup distributor will
>> have to provide a cmd file with it?
>
> Yup, but we currently only have a single binary to support, and there is
> already a need for a flag, so we need a way to support this in a more
> flexible way than hardcoding it in libosinfo imo.
I'd want to avoid hardcoding as well but at the same time I would hate
to ask driver distributors (or anyone who wants to add drivers to db
in future) to have to write a .cmd file per driver so I'll first want
to know if we are worrying for a good reason.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 21:11:56 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 23:11:56 +0200
Subject: [Libosinfo] [PATCH 7/8] winxp,
win7: Add virtio & QXL device drivers info
In-Reply-To: <20130128172506.GE22845@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-8-git-send-email-zeeshanak@gnome.org>
<20130128110122.GG2332@teriyaki.redhat.com>
<20130128172506.GE22845@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 7:25 PM, Christophe Fergeau wrote:
> On Mon, Jan 28, 2013 at 05:27:16PM +0200, Zeeshan Ali (Khattak) wrote:
>> On Mon, Jan 28, 2013 at 1:01 PM, Christophe Fergeau wrote:
>> > I think we should still have a way for libosinfo user to have this binary
>> > run at the end of the install process even though this means having to
>> > click through a few dialogs.
>>
>> Apart from the 3-state (rather than boolean) API I proposed in a
>> previous mail, we could also have a separate API for adding random
>> binaries that are put on user's desktop and apps can then use that for
>> things like these. In this specific case, app will see that there is a
>> driver available for the os in executable format but script warns on
>> it during install so they can decide to use this API to add the binary
>> to the desktop with filename "Installer for X".
>
> I don't think this is really useful... And I'd prefer to have the binary
> be run for me at the end of installation even if it means it's showing
> dialog.
>
>> Such an API could be useful in general and allow us to not complicate
>> drivers-related API and most importantly not break the concept of
>> automated/express installation.
>
> Depends what you value most, fully unattended install, or some guarantee
> that the drivers will be installed when the install completes even if this
> means blocking. While we want the former to be the default, I think it
> would be nice to have some support for the latter.
I value not breaking unattended installation. It will look especially
bad if in the UI we ask users if they want unattended install and then
take steps that needs user input before the installation is really
complete. One approach I tried before this was to copy the binary to C
drive and setup registry keys to run it as first login but I failed
get it to work. If we can make that to work, I'm all for it but
breaking unattended install is not an option for me. Its not like
users' windows installation is broken without virtio/qxl drivers.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 21:24:55 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 23:24:55 +0200
Subject: [Libosinfo] [PATCH 4/8] installer: API to query supported
device driver formats
In-Reply-To: <20130128172830.GF22845@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-5-git-send-email-zeeshanak@gnome.org>
<20130128102735.GD2332@teriyaki.redhat.com>
<20130128172830.GF22845@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 7:28 PM, Christophe Fergeau wrote:
> On Mon, Jan 28, 2013 at 04:40:06PM +0200, Zeeshan Ali (Khattak) wrote:
>> On Mon, Jan 28, 2013 at 12:27 PM, Christophe Fergeau
>> wrote:
>> > On Mon, Jan 28, 2013 at 05:18:43AM +0200, Zeeshan Ali (Khattak) wrote:
>> >> From: "Zeeshan Ali (Khattak)"
>> >> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
>> >> index d91751e..82486ef 100644
>> >> --- a/osinfo/osinfo_install_script.h
>> >> +++ b/osinfo/osinfo_install_script.h
>> >> @@ -163,6 +165,9 @@ OsinfoPathFormat osinfo_install_script_get_path_format(OsinfoInstallScript *scri
>> >> gboolean osinfo_install_script_get_can_pre_install_drivers(OsinfoInstallScript *script);
>> >> gboolean osinfo_install_script_get_can_post_install_drivers(OsinfoInstallScript *script);
>> >>
>> >> +OsinfoDeviceDriverFormat osinfo_install_script_get_pre_install_driver_format(OsinfoInstallScript *script);
>> >> +OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(OsinfoInstallScript *script);
>> >
>> > I don't think assuming that a given installer will support only one driver
>> > format is expressive enough. For Windows post-install drivers, supporting
>> > unpacked Windows drivers in addition to running a .exe shouldn't be very
>> > hard, and this API would not work there.
>>
>> I see you point. I can make it a list. Would that be good?
>
> I think so. Though concretely why do we need to expose this information? In
> all cases user of this information will need to drop it to a disk image
> which will be passed to the VM, no?
Without this information, apps not only have to copy unnecessary
driver files but most probably (as is the case with spice-guest-tools
binary in Boxes) also need to create a redundant disk image to copy
the files to when drivers are incompatible with scripts. Also same
drivers can be available in multiple formats so Apps should be able to
pick one in compatible (with script) format.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From zeeshanak at gnome.org Mon Jan 28 21:29:10 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Mon, 28 Jan 2013 23:29:10 +0200
Subject: [Libosinfo] [PATCH 6/8] installer: API to query signed device
driver requirement
In-Reply-To: <20130128173135.GG22845@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-7-git-send-email-zeeshanak@gnome.org>
<20130128104849.GF2332@teriyaki.redhat.com>
<20130128173135.GG22845@teriyaki.redhat.com>
Message-ID:
On Mon, Jan 28, 2013 at 7:31 PM, Christophe Fergeau wrote:
> On Mon, Jan 28, 2013 at 04:51:04PM +0200, Zeeshan Ali (Khattak) wrote:
>> On Mon, Jan 28, 2013 at 12:48 PM, Christophe Fergeau
>> wrote:
>> > On Mon, Jan 28, 2013 at 05:18:45AM +0200, Zeeshan Ali (Khattak) wrote:
>> >> +/**
>> >> + * osinfo_install_script_get_can_post_install_unsigned_drivers:
>> >> + * @script: the install script
>> >> + *
>> >> + * If install script can install drivers at the end of installation, this
>> >> + * function can be used to query if it can handle unsigned drivers.
>> >
>> > All of this is a bit vague to me, this seems more tri-state than just a
>> > boolean, most Windows will warn when trying to install unsigned-by-MS
>> > drivers but will let you go through after saying you are ok with installing
>> > this driver. Starting with win7 64 bit, the OS will refuse to install the
>> > drivers without letting you click your way around it.
>> > This commit seems to assume that 'having to click' = 'cannot post install
>> > the driver'. While inconvenient during post-install, I can see this being
>> > better than nothing to some libosinfo users.
>>
>> While in Boxes, we certainly wouldn't want a warning message as it
>> breaks the whole concept of express installation, I guess other apps
>> could be given a choice here.
>>
>> How about an enum like this:
>>
>> enum DeviceDriverSignReq {
>> DEVICE_DRIVER_SIGN_REQ_STRICT, /* Script must only be given signed
>> device drivers. */
>> DEVICE_DRIVER_SIGN_REQ_WARN, /* A warning will be issued by OS
>> installer if device drivers are not signed. */
>> DEVICE_DRIVER_SIGN_REQ_NONE, /* Script do not require device drivers
>> to be signed. */
>> }
>
> I'd change 'Sign' to 'Signing', and make it explicit that the 'REQ_WARN'
> case can break unattended installation and require user input. Same for
> 'REQ_STRICT' actually, in addition to not installing unsigned drivers,
> passing an unsigned driver may also break unattended install (?)
So far I haven't seen that happening. Windows 7 simply gets installed
without drivers installed. With lack of any information in log files,
I believe that it just terminates the installer for trying to install
unsigned driver.
But yes, I can provide more info.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From cfergeau at redhat.com Tue Jan 29 09:13:31 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 29 Jan 2013 10:13:31 +0100
Subject: [Libosinfo] [PATCH 8/8] winxp,
installer: Make use of post-installation drivers
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-9-git-send-email-zeeshanak@gnome.org>
<20130128131453.GN2332@teriyaki.redhat.com>
<20130128171943.GD22845@teriyaki.redhat.com>
Message-ID: <20130129091331.GL22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 11:06:53PM +0200, Zeeshan Ali (Khattak) wrote:
> On Mon, Jan 28, 2013 at 7:19 PM, Christophe Fergeau wrote:
> > On Mon, Jan 28, 2013 at 05:45:25PM +0200, Zeeshan Ali (Khattak) wrote:
> >> On Mon, Jan 28, 2013 at 3:14 PM, Christophe Fergeau wrote:
> >> > You can't expect every possible Windows exe file to honour the /S switch.
> >>
> >> If a binary doesn't support '/S' switch, it will ignore it?
> >
> > Or fail saying it's an unknown parameter?
>
> According to Vmware docs, this is a standard MSI option:
>
> http://pubs.vmware.com/view-50/index.jsp?topic=/com.vmware.view.installation.doc/GUID-F19EBDF6-20A3-4A8B-95AE-786CC74F26AF.html
>
> If that is the case, its supposed to be guaranteed on every windows
> installer setup app?
I doubt it's guaranteed... And the way the code currently is, MSIs can't be
installed either.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Tue Jan 29 09:15:07 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 29 Jan 2013 10:15:07 +0100
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
<20130128101953.GC2332@teriyaki.redhat.com>
<20130128154157.GB22845@teriyaki.redhat.com>
<20130128161327.GC22845@teriyaki.redhat.com>
Message-ID: <20130129091507.GM22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 10:54:52PM +0200, Zeeshan Ali (Khattak) wrote:
> On Mon, Jan 28, 2013 at 6:13 PM, Christophe Fergeau wrote:
> > On Mon, Jan 28, 2013 at 06:05:53PM +0200, Zeeshan Ali (Khattak) wrote:
> >> whats the difference between copyleft and free, as in Free Software?
> >> Reading this, I don't see any:
> >>
> >> http://en.wikipedia.org/wiki/Copyleft
> >
> > http://en.wikipedia.org/wiki/Copyleft#Types_of_copyleft_and_relation_to_other_licenses
>
> OK, I stand corrected about the exact definition of copyleft. :)
>
> Since you are stressing on this distinction, I imagine you know some
> free drivers out there that are signed?
Nope, but that does not mean they do not exist.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Tue Jan 29 09:19:05 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 29 Jan 2013 10:19:05 +0100
Subject: [Libosinfo] [PATCH 7/8] winxp,
win7: Add virtio & QXL device drivers info
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-8-git-send-email-zeeshanak@gnome.org>
<20130128110122.GG2332@teriyaki.redhat.com>
<20130128172506.GE22845@teriyaki.redhat.com>
Message-ID: <20130129091905.GO22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 11:11:56PM +0200, Zeeshan Ali (Khattak) wrote:
> I value not breaking unattended installation.
Once again, I'm not pushing for any of this to be the default, I just don't
want we close that (API/ABI) door if people (even us!) need to do that in
the future.
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From cfergeau at redhat.com Tue Jan 29 09:31:14 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Tue, 29 Jan 2013 10:31:14 +0100
Subject: [Libosinfo] [PATCH 4/8] installer: API to query supported
device driver formats
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-5-git-send-email-zeeshanak@gnome.org>
<20130128102735.GD2332@teriyaki.redhat.com>
<20130128172830.GF22845@teriyaki.redhat.com>
Message-ID: <20130129093114.GP22845@teriyaki.redhat.com>
On Mon, Jan 28, 2013 at 11:24:55PM +0200, Zeeshan Ali (Khattak) wrote:
> On Mon, Jan 28, 2013 at 7:28 PM, Christophe Fergeau wrote:
> > On Mon, Jan 28, 2013 at 04:40:06PM +0200, Zeeshan Ali (Khattak) wrote:
> >> On Mon, Jan 28, 2013 at 12:27 PM, Christophe Fergeau
> >> wrote:
> >> > On Mon, Jan 28, 2013 at 05:18:43AM +0200, Zeeshan Ali (Khattak) wrote:
> >> >> From: "Zeeshan Ali (Khattak)"
> >> >> diff --git a/osinfo/osinfo_install_script.h b/osinfo/osinfo_install_script.h
> >> >> index d91751e..82486ef 100644
> >> >> --- a/osinfo/osinfo_install_script.h
> >> >> +++ b/osinfo/osinfo_install_script.h
> >> >> @@ -163,6 +165,9 @@ OsinfoPathFormat osinfo_install_script_get_path_format(OsinfoInstallScript *scri
> >> >> gboolean osinfo_install_script_get_can_pre_install_drivers(OsinfoInstallScript *script);
> >> >> gboolean osinfo_install_script_get_can_post_install_drivers(OsinfoInstallScript *script);
> >> >>
> >> >> +OsinfoDeviceDriverFormat osinfo_install_script_get_pre_install_driver_format(OsinfoInstallScript *script);
> >> >> +OsinfoDeviceDriverFormat osinfo_install_script_get_post_install_driver_format(OsinfoInstallScript *script);
> >> >
> >> > I don't think assuming that a given installer will support only one driver
> >> > format is expressive enough. For Windows post-install drivers, supporting
> >> > unpacked Windows drivers in addition to running a .exe shouldn't be very
> >> > hard, and this API would not work there.
> >>
> >> I see you point. I can make it a list. Would that be good?
> >
> > I think so. Though concretely why do we need to expose this information? In
> > all cases user of this information will need to drop it to a disk image
> > which will be passed to the VM, no?
>
> Without this information, apps not only have to copy unnecessary
> driver files but most probably (as is the case with spice-guest-tools
> binary in Boxes) also need to create a redundant disk image to copy
> the files to when drivers are incompatible with scripts. Also same
> drivers can be available in multiple formats so Apps should be able to
> pick one in compatible (with script) format.
At this point, this is all theoritical, isn't it? We support one
post-install format, which must be a .exe supporting the /S switch, and
which has to be copied to an ISO image. Wouldn't it be better to postpone
this API until there's a need for it?
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From zeeshanak at gnome.org Tue Jan 29 14:49:05 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Tue, 29 Jan 2013 16:49:05 +0200
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To: <20130129091507.GM22845@teriyaki.redhat.com>
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
<20130128101953.GC2332@teriyaki.redhat.com>
<20130128154157.GB22845@teriyaki.redhat.com>
<20130128161327.GC22845@teriyaki.redhat.com>
<20130129091507.GM22845@teriyaki.redhat.com>
Message-ID:
On Tue, Jan 29, 2013 at 11:15 AM, Christophe Fergeau
wrote:
> On Mon, Jan 28, 2013 at 10:54:52PM +0200, Zeeshan Ali (Khattak) wrote:
>> On Mon, Jan 28, 2013 at 6:13 PM, Christophe Fergeau wrote:
>> > On Mon, Jan 28, 2013 at 06:05:53PM +0200, Zeeshan Ali (Khattak) wrote:
>> >> whats the difference between copyleft and free, as in Free Software?
>> >> Reading this, I don't see any:
>> >>
>> >> http://en.wikipedia.org/wiki/Copyleft
>> >
>> > http://en.wikipedia.org/wiki/Copyleft#Types_of_copyleft_and_relation_to_other_licenses
>>
>> OK, I stand corrected about the exact definition of copyleft. :)
>>
>> Since you are stressing on this distinction, I imagine you know some
>> free drivers out there that are signed?
>
> Nope, but that does not mean they do not exist.
Well then I'm just not being very precise in my commit log. With the
API I'm proposing, Apps can easily decide to not use unsigned drivers
even if script supports it. So I don't see the need for Apps to have
to use a configuration parameter to disable this manually.
--
Regards,
Zeeshan Ali (Khattak)
FSF member#5124
From cfergeau at redhat.com Wed Jan 30 09:26:19 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Wed, 30 Jan 2013 10:26:19 +0100
Subject: [Libosinfo] [PATCH 2/8] winxp,
installer: Ignore unsigned drivers
In-Reply-To:
References: <1359343127-27272-1-git-send-email-zeeshanak@gnome.org>
<1359343127-27272-3-git-send-email-zeeshanak@gnome.org>
<20130128101953.GC2332@teriyaki.redhat.com>
<20130128154157.GB22845@teriyaki.redhat.com>
<20130128161327.GC22845@teriyaki.redhat.com>
<20130129091507.GM22845@teriyaki.redhat.com>
Message-ID: <20130130092619.GJ22845@teriyaki.redhat.com>
On Tue, Jan 29, 2013 at 04:49:05PM +0200, Zeeshan Ali (Khattak) wrote:
> On Tue, Jan 29, 2013 at 11:15 AM, Christophe Fergeau
> wrote:
> > On Mon, Jan 28, 2013 at 10:54:52PM +0200, Zeeshan Ali (Khattak) wrote:
> >> On Mon, Jan 28, 2013 at 6:13 PM, Christophe Fergeau wrote:
> >> > On Mon, Jan 28, 2013 at 06:05:53PM +0200, Zeeshan Ali (Khattak) wrote:
> >> >> whats the difference between copyleft and free, as in Free Software?
> >> >> Reading this, I don't see any:
> >> >>
> >> >> http://en.wikipedia.org/wiki/Copyleft
> >> >
> >> > http://en.wikipedia.org/wiki/Copyleft#Types_of_copyleft_and_relation_to_other_licenses
> >>
> >> OK, I stand corrected about the exact definition of copyleft. :)
> >>
> >> Since you are stressing on this distinction, I imagine you know some
> >> free drivers out there that are signed?
> >
> > Nope, but that does not mean they do not exist.
>
> Well then I'm just not being very precise in my commit log. With the
> API I'm proposing, Apps can easily decide to not use unsigned drivers
> even if script supports it. So I don't see the need for Apps to have
> to use a configuration parameter to disable this manually.
https://www.redhat.com/archives/libosinfo/2013-January/msg00108.html
Christophe
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL:
From berrange at redhat.com Wed Jan 30 13:48:40 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 30 Jan 2013 13:48:40 +0000
Subject: [Libosinfo] [libosinfo] Add QEMU-KVM 1.2.2 hypervisor data
In-Reply-To: <1358869663-24025-2-git-send-email-cfergeau@redhat.com>
References: <1358869663-24025-1-git-send-email-cfergeau@redhat.com>
<1358869663-24025-2-git-send-email-cfergeau@redhat.com>
Message-ID: <20130130134840.GI8449@redhat.com>
On Tue, Jan 22, 2013 at 04:47:43PM +0100, Christophe Fergeau wrote:
> ---
> data/hypervisors/Makefile.am | 2 +-
> data/hypervisors/kvm.xml.in | 57 ++++++++++++++++++++++++++++++++++++++++++++
> 2 files changed, 58 insertions(+), 1 deletion(-)
> create mode 100644 data/hypervisors/kvm.xml.in
>
> diff --git a/data/hypervisors/Makefile.am b/data/hypervisors/Makefile.am
> index 19ac267..2f7dd45 100644
> --- a/data/hypervisors/Makefile.am
> +++ b/data/hypervisors/Makefile.am
> @@ -1,6 +1,6 @@
>
> databasedir = $(pkgdatadir)/db/hypervisors/
> -database_in_files = rhel-xen.xml.in xen.xml.in
> +database_in_files = kvm.xml.in rhel-xen.xml.in xen.xml.in
>
> database_DATA = $(database_in_files:.xml.in=.xml)
>
> diff --git a/data/hypervisors/kvm.xml.in b/data/hypervisors/kvm.xml.in
> new file mode 100644
> index 0000000..3653c46
> --- /dev/null
> +++ b/data/hypervisors/kvm.xml.in
> @@ -0,0 +1,57 @@
> +
> +
> +
> +
> + kvm-1.2.2
> + <_name>QEMU-KVM 1.2.2
> + <_vendor>qemu.org
> + 1.2.2
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
> +
So nothing wrong with the data you've added here.
At some point we'll need to get a way to express different devices
per architecture, since not all qemu-system-XXXX emulators provide
the same device list. This isn't a new problem, so lets just pretend
only x86 exists for now.
In terms of QEMU vs QEMU-KVM, for all intents & purposes they support
the same devices. QEMU-KVM is the fork, so I'd say we should be adding
devices against entries for plain QEMU, and then the
entries for QEMU-KVM can be marked as "derives-from" or
"clones" QEMU.
FYI that device list is accurate back to at least QEMU 1.0.0, so I'd
just add the list against that relase, and then create
entries with "upgrades" for every QEMU major/minor release since
then.
Regards,
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From berrange at redhat.com Wed Jan 30 14:21:17 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Wed, 30 Jan 2013 14:21:17 +0000
Subject: [Libosinfo] New libosinfo website
Message-ID: <20130130142117.GK8449@redhat.com>
I've finally got my act together and finished creating the new
libosinfo website
http://libosinfo.org/
Please use this link in preference to the old Fedora hosted wiki.
NB, Fedora is still used for the GIT repos - we're not just not
bothering with the fugly Trac wiki.
Regards,
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From cfergeau at redhat.com Thu Jan 31 09:44:22 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Thu, 31 Jan 2013 10:44:22 +0100
Subject: [Libosinfo] [PATCHv2 0/1] Add QEMU/QEMU-KVM hypervisor data
Message-ID: <1359625463-10215-1-git-send-email-cfergeau@redhat.com>
Hey,
This is a follow-up patch to my previous RFC about QEMU hypervisor data.
This patch is much more extensive with the data added as it adds all
qemu/qemu-kvm versions newer than 1.0.
I've added the qemu 1.1.1-1 release, but as this is just the 1.1.1 tarball
repacked as .tar.bz2 I'm not 100% sure we want to have it.
I've checked that make check still passes with these new data files (this
validates the XML against the .rng schema)
Christophe
From cfergeau at redhat.com Thu Jan 31 09:44:23 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Thu, 31 Jan 2013 10:44:23 +0100
Subject: [Libosinfo] [PATCHv2] Add QEMU/QEMU-KVM hypervisor data
In-Reply-To: <1359625463-10215-1-git-send-email-cfergeau@redhat.com>
References: <1359625463-10215-1-git-send-email-cfergeau@redhat.com>
Message-ID: <1359625463-10215-2-git-send-email-cfergeau@redhat.com>
---
data/hypervisors/Makefile.am | 2 +-
data/hypervisors/kvm.xml.in | 63 ++++++++++++++++++
data/hypervisors/qemu.xml.in | 148 +++++++++++++++++++++++++++++++++++++++++++
po/POTFILES.in | 2 +
4 files changed, 214 insertions(+), 1 deletion(-)
create mode 100644 data/hypervisors/kvm.xml.in
create mode 100644 data/hypervisors/qemu.xml.in
diff --git a/data/hypervisors/Makefile.am b/data/hypervisors/Makefile.am
index 19ac267..4b84d3e 100644
--- a/data/hypervisors/Makefile.am
+++ b/data/hypervisors/Makefile.am
@@ -1,6 +1,6 @@
databasedir = $(pkgdatadir)/db/hypervisors/
-database_in_files = rhel-xen.xml.in xen.xml.in
+database_in_files = kvm.xml.in qemu.xml.in rhel-xen.xml.in xen.xml.in
database_DATA = $(database_in_files:.xml.in=.xml)
diff --git a/data/hypervisors/kvm.xml.in b/data/hypervisors/kvm.xml.in
new file mode 100644
index 0000000..3833943
--- /dev/null
+++ b/data/hypervisors/kvm.xml.in
@@ -0,0 +1,63 @@
+
+
+
+
+ qemu-kvm-1.0
+ <_name>QEMU-KVM 1.0
+ <_vendor>linux-kvm.org
+ 1.0
+ 2011-12-04
+
+
+
+
+ qemu-kvm-1.0.1
+ <_name>QEMU-KVM 1.0.1
+ <_vendor>linux-kvm.org
+ 1.0.1
+ 2012-04-17
+
+
+
+
+
+ qemu-kvm-1.1.0
+ <_name>QEMU-KVM 1.1.0
+ <_vendor>linux-kvm.org
+ 1.1.0
+ 2012-07-02
+
+
+
+
+
+ qemu-kvm-1.1.1
+ <_name>QEMU-KVM 1.1.1
+ <_vendor>linux-kvm.org
+ 1.1.1
+ 2012-08-27
+
+
+
+
+
+ qemu-kvm-1.1.2
+ <_name>QEMU-KVM 1.1.2
+ <_vendor>linux-kvm.org
+ 1.1.2
+ 2012-09-10
+
+
+
+
+
+ qemu-kvm-1.2.0
+ <_name>QEMU-KVM 1.2.0
+ <_vendor>linux-kvm.org
+ 1.2.0
+ 2012-09-10
+
+
+
+
+
diff --git a/data/hypervisors/qemu.xml.in b/data/hypervisors/qemu.xml.in
new file mode 100644
index 0000000..e57e867
--- /dev/null
+++ b/data/hypervisors/qemu.xml.in
@@ -0,0 +1,148 @@
+
+
+
+
+ qemu-1.0
+ <_name>QEMU 1.0
+ <_vendor>qemu.org
+ 1.0
+ 2011-12-01
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ qemu-1.0.1
+ <_name>QEMU 1.0.1
+ <_vendor>qemu.org
+ 1.0.1
+ 2012-02-17
+
+
+
+
+ qemu-1.1
+ <_name>QEMU 1.1
+ <_vendor>qemu.org
+ 1.1
+ 2012-06-01
+
+
+
+
+ qemu-1.1.1
+ <_name>QEMU 1.1.1
+ <_vendor>qemu.org
+ 1.1.1
+ 2012-07-12
+
+
+
+
+ qemu-1.1.1-1
+ <_name>QEMU 1.1.1-1
+ <_vendor>qemu.org
+ 1.1.1-1
+ 2012-07-17
+
+
+
+
+ qemu-1.1.2
+ <_name>QEMU 1.1.2
+ <_vendor>qemu.org
+ 1.1.2
+ 2012-09-05
+
+
+
+
+ qemu-1.2.0
+ <_name>QEMU 1.2.0
+ <_vendor>qemu.org
+ 1.2.0
+ 2012-09-05
+
+
+
+
+ qemu-1.2.1
+ <_name>QEMU 1.2.1
+ <_vendor>qemu.org
+ 1.2.1
+ 2012-11-20
+
+
+
+
+ qemu-1.2.2
+ <_name>QEMU 1.2.2
+ <_vendor>qemu.org
+ 1.2.2
+ 2012-12-11
+
+
+
+
+ qemu-1.3.0
+ <_name>QEMU 1.3.0
+ <_vendor>qemu.org
+ 1.3.0
+ 2012-12-03
+
+
+
+
+ qemu-1.3.1
+ <_name>QEMU 1.3.1
+ <_vendor>qemu.org
+ 1.3.1
+ 2013-01-28
+
+
+
+
diff --git a/po/POTFILES.in b/po/POTFILES.in
index 1a209bb..fc1c01b 100644
--- a/po/POTFILES.in
+++ b/po/POTFILES.in
@@ -21,6 +21,8 @@ data/devices/qemu-ps2.xml.in
data/devices/qemu-pci.xml.in
data/devices/virtio-pci.xml.in
data/devices/qemu-usb.xml.in
+data/hypervisors/kvm.xml.in
+data/hypervisors/qemu.xml.in
data/hypervisors/rhel-xen.xml.in
data/hypervisors/xen.xml.in
osinfo/osinfo_avatar_format.c
--
1.8.1
From berrange at redhat.com Thu Jan 31 09:45:43 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Thu, 31 Jan 2013 09:45:43 +0000
Subject: [Libosinfo] [PATCHv2] Add QEMU/QEMU-KVM hypervisor data
In-Reply-To: <1359625463-10215-2-git-send-email-cfergeau@redhat.com>
References: <1359625463-10215-1-git-send-email-cfergeau@redhat.com>
<1359625463-10215-2-git-send-email-cfergeau@redhat.com>
Message-ID: <20130131094543.GA23534@redhat.com>
On Thu, Jan 31, 2013 at 10:44:23AM +0100, Christophe Fergeau wrote:
> ---
> data/hypervisors/Makefile.am | 2 +-
> data/hypervisors/kvm.xml.in | 63 ++++++++++++++++++
> data/hypervisors/qemu.xml.in | 148 +++++++++++++++++++++++++++++++++++++++++++
> po/POTFILES.in | 2 +
> 4 files changed, 214 insertions(+), 1 deletion(-)
> create mode 100644 data/hypervisors/kvm.xml.in
> create mode 100644 data/hypervisors/qemu.xml.in
ACK, looks like a good starting point to me.
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From cfergeau at redhat.com Thu Jan 31 10:13:52 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Thu, 31 Jan 2013 11:13:52 +0100
Subject: [Libosinfo] [libosinfo] Remove unused 'priv' variable from _init
methods
Message-ID: <1359627232-13578-1-git-send-email-cfergeau@redhat.com>
In most instance _init() methods, the instance priv member
is initialized, but a local 'priv' variable was also declared and
assigned, but not actually used. Kill this intermediate variable
as this causes Coverity warnings.
---
osinfo/osinfo_datamap.c | 4 +---
osinfo/osinfo_datamaplist.c | 4 +---
osinfo/osinfo_db.c | 4 +---
osinfo/osinfo_deployment.c | 4 +---
osinfo/osinfo_deploymentlist.c | 4 +---
osinfo/osinfo_device.c | 3 +--
osinfo/osinfo_device_driver.c | 6 ++----
osinfo/osinfo_device_driverlist.c | 4 +---
osinfo/osinfo_devicelink.c | 3 +--
osinfo/osinfo_devicelinkfilter.c | 4 +---
osinfo/osinfo_devicelinklist.c | 4 +---
osinfo/osinfo_devicelist.c | 4 +---
osinfo/osinfo_entity.c | 4 +---
osinfo/osinfo_filter.c | 4 +---
osinfo/osinfo_install_config.c | 3 +--
osinfo/osinfo_install_config_paramlist.c | 4 +---
osinfo/osinfo_install_script.c | 4 +---
osinfo/osinfo_install_scriptlist.c | 4 +---
osinfo/osinfo_list.c | 4 +---
osinfo/osinfo_loader.c | 4 +---
osinfo/osinfo_media.c | 3 +--
osinfo/osinfo_medialist.c | 4 +---
osinfo/osinfo_os.c | 3 +--
osinfo/osinfo_oslist.c | 4 +---
osinfo/osinfo_platform.c | 4 +---
osinfo/osinfo_platformlist.c | 4 +---
osinfo/osinfo_product.c | 4 +---
osinfo/osinfo_productfilter.c | 5 +----
osinfo/osinfo_productlist.c | 4 +---
osinfo/osinfo_resources.c | 3 +--
osinfo/osinfo_resourceslist.c | 3 +--
osinfo/osinfo_tree.c | 3 +--
osinfo/osinfo_treelist.c | 4 +---
33 files changed, 34 insertions(+), 93 deletions(-)
diff --git a/osinfo/osinfo_datamap.c b/osinfo/osinfo_datamap.c
index 50c9c7c..98ce0a6 100644
--- a/osinfo/osinfo_datamap.c
+++ b/osinfo/osinfo_datamap.c
@@ -76,9 +76,7 @@ osinfo_datamap_class_init (OsinfoDatamapClass *klass)
static void
osinfo_datamap_init (OsinfoDatamap *list)
{
- OsinfoDatamapPrivate *priv;
- list->priv = priv = OSINFO_DATAMAP_GET_PRIVATE(list);
-
+ list->priv = OSINFO_DATAMAP_GET_PRIVATE(list);
list->priv->map = g_hash_table_new_full(g_str_hash,
g_str_equal,
g_free,
diff --git a/osinfo/osinfo_datamaplist.c b/osinfo/osinfo_datamaplist.c
index db25bbe..5d0d7b6 100644
--- a/osinfo/osinfo_datamaplist.c
+++ b/osinfo/osinfo_datamaplist.c
@@ -64,9 +64,7 @@ osinfo_datamaplist_class_init (OsinfoDatamapListClass *klass)
static void
osinfo_datamaplist_init (OsinfoDatamapList *list)
{
- OsinfoDatamapListPrivate *priv;
- list->priv = priv = OSINFO_DATAMAPLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_DATAMAPLIST_GET_PRIVATE(list);
}
/**
diff --git a/osinfo/osinfo_db.c b/osinfo/osinfo_db.c
index adf2737..413db8f 100644
--- a/osinfo/osinfo_db.c
+++ b/osinfo/osinfo_db.c
@@ -172,9 +172,7 @@ osinfo_db_class_init (OsinfoDbClass *klass)
static void
osinfo_db_init (OsinfoDb *db)
{
- OsinfoDbPrivate *priv;
- db->priv = priv = OSINFO_DB_GET_PRIVATE(db);
-
+ db->priv = OSINFO_DB_GET_PRIVATE(db);
db->priv->devices = osinfo_devicelist_new();
db->priv->platforms = osinfo_platformlist_new();
db->priv->oses = osinfo_oslist_new();
diff --git a/osinfo/osinfo_deployment.c b/osinfo/osinfo_deployment.c
index 3678fa2..3e511e2 100644
--- a/osinfo/osinfo_deployment.c
+++ b/osinfo/osinfo_deployment.c
@@ -182,9 +182,7 @@ osinfo_deployment_class_init (OsinfoDeploymentClass *klass)
static void
osinfo_deployment_init (OsinfoDeployment *deployment)
{
- OsinfoDeploymentPrivate *priv;
- deployment->priv = priv = OSINFO_DEPLOYMENT_GET_PRIVATE(deployment);
-
+ deployment->priv = OSINFO_DEPLOYMENT_GET_PRIVATE(deployment);
deployment->priv->deviceLinks = NULL;
}
diff --git a/osinfo/osinfo_deploymentlist.c b/osinfo/osinfo_deploymentlist.c
index d411cff..41a5575 100644
--- a/osinfo/osinfo_deploymentlist.c
+++ b/osinfo/osinfo_deploymentlist.c
@@ -65,9 +65,7 @@ osinfo_deploymentlist_class_init (OsinfoDeploymentListClass *klass)
static void
osinfo_deploymentlist_init (OsinfoDeploymentList *list)
{
- OsinfoDeploymentListPrivate *priv;
- list->priv = priv = OSINFO_DEPLOYMENTLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_DEPLOYMENTLIST_GET_PRIVATE(list);
}
diff --git a/osinfo/osinfo_device.c b/osinfo/osinfo_device.c
index d3fb3d7..183ae41 100644
--- a/osinfo/osinfo_device.c
+++ b/osinfo/osinfo_device.c
@@ -70,8 +70,7 @@ osinfo_device_class_init (OsinfoDeviceClass *klass)
static void
osinfo_device_init (OsinfoDevice *device)
{
- OsinfoDevicePrivate *priv;
- device->priv = priv = OSINFO_DEVICE_GET_PRIVATE(device);
+ device->priv = OSINFO_DEVICE_GET_PRIVATE(device);
}
OsinfoDevice *osinfo_device_new(const gchar *id)
diff --git a/osinfo/osinfo_device_driver.c b/osinfo/osinfo_device_driver.c
index d402bc7..23731bf 100644
--- a/osinfo/osinfo_device_driver.c
+++ b/osinfo/osinfo_device_driver.c
@@ -76,10 +76,8 @@ osinfo_device_driver_class_init (OsinfoDeviceDriverClass *klass)
static void
osinfo_device_driver_init (OsinfoDeviceDriver *driver)
{
- OsinfoDeviceDriverPrivate *priv;
- driver->priv = priv = OSINFO_DEVICE_DRIVER_GET_PRIVATE(driver);
-
- priv->devices = osinfo_devicelist_new ();
+ driver->priv = OSINFO_DEVICE_DRIVER_GET_PRIVATE(driver);
+ driver->priv->devices = osinfo_devicelist_new ();
}
OsinfoDeviceDriver *osinfo_device_driver_new(const gchar *id)
diff --git a/osinfo/osinfo_device_driverlist.c b/osinfo/osinfo_device_driverlist.c
index b4f6f0f..11f29db 100644
--- a/osinfo/osinfo_device_driverlist.c
+++ b/osinfo/osinfo_device_driverlist.c
@@ -66,9 +66,7 @@ osinfo_device_driverlist_class_init (OsinfoDeviceDriverListClass *klass)
static void
osinfo_device_driverlist_init (OsinfoDeviceDriverList *list)
{
- OsinfoDeviceDriverListPrivate *priv;
- list->priv = priv = OSINFO_DEVICE_DRIVERLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_DEVICE_DRIVERLIST_GET_PRIVATE(list);
}
/**
diff --git a/osinfo/osinfo_devicelink.c b/osinfo/osinfo_devicelink.c
index 52dea06..95b5770 100644
--- a/osinfo/osinfo_devicelink.c
+++ b/osinfo/osinfo_devicelink.c
@@ -143,8 +143,7 @@ osinfo_devicelink_class_init (OsinfoDeviceLinkClass *klass)
static void
osinfo_devicelink_init (OsinfoDeviceLink *devlink)
{
- OsinfoDeviceLinkPrivate *priv;
- devlink->priv = priv = OSINFO_DEVICELINK_GET_PRIVATE(devlink);
+ devlink->priv = OSINFO_DEVICELINK_GET_PRIVATE(devlink);
}
diff --git a/osinfo/osinfo_devicelinkfilter.c b/osinfo/osinfo_devicelinkfilter.c
index 004d38a..6e30823 100644
--- a/osinfo/osinfo_devicelinkfilter.c
+++ b/osinfo/osinfo_devicelinkfilter.c
@@ -166,9 +166,7 @@ OsinfoDeviceLinkFilter *osinfo_devicelinkfilter_new(OsinfoFilter *filter)
static void
osinfo_devicelinkfilter_init (OsinfoDeviceLinkFilter *devicelinkfilter)
{
- OsinfoDeviceLinkFilterPrivate *priv;
- priv = OSINFO_DEVICELINKFILTER_GET_PRIVATE(devicelinkfilter);
- devicelinkfilter->priv = priv;
+ devicelinkfilter->priv = OSINFO_DEVICELINKFILTER_GET_PRIVATE(devicelinkfilter);
}
diff --git a/osinfo/osinfo_devicelinklist.c b/osinfo/osinfo_devicelinklist.c
index 89b6ce6..672934f 100644
--- a/osinfo/osinfo_devicelinklist.c
+++ b/osinfo/osinfo_devicelinklist.c
@@ -65,9 +65,7 @@ osinfo_devicelinklist_class_init (OsinfoDeviceLinkListClass *klass)
static void
osinfo_devicelinklist_init (OsinfoDeviceLinkList *list)
{
- OsinfoDeviceLinkListPrivate *priv;
- list->priv = priv = OSINFO_DEVICELINKLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_DEVICELINKLIST_GET_PRIVATE(list);
}
diff --git a/osinfo/osinfo_devicelist.c b/osinfo/osinfo_devicelist.c
index 96ad2fc..ca2ceef 100644
--- a/osinfo/osinfo_devicelist.c
+++ b/osinfo/osinfo_devicelist.c
@@ -65,9 +65,7 @@ osinfo_devicelist_class_init (OsinfoDeviceListClass *klass)
static void
osinfo_devicelist_init (OsinfoDeviceList *list)
{
- OsinfoDeviceListPrivate *priv;
- list->priv = priv = OSINFO_DEVICELIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_DEVICELIST_GET_PRIVATE(list);
}
diff --git a/osinfo/osinfo_entity.c b/osinfo/osinfo_entity.c
index b291252..543c710 100644
--- a/osinfo/osinfo_entity.c
+++ b/osinfo/osinfo_entity.c
@@ -161,9 +161,7 @@ static void osinfo_entity_param_values_free(gpointer values)
static void
osinfo_entity_init (OsinfoEntity *entity)
{
- OsinfoEntityPrivate *priv;
- entity->priv = priv = OSINFO_ENTITY_GET_PRIVATE(entity);
-
+ entity->priv = OSINFO_ENTITY_GET_PRIVATE(entity);
entity->priv->params = g_hash_table_new_full(g_str_hash,
g_str_equal,
g_free,
diff --git a/osinfo/osinfo_filter.c b/osinfo/osinfo_filter.c
index 6f19ec1..959ab6e 100644
--- a/osinfo/osinfo_filter.c
+++ b/osinfo/osinfo_filter.c
@@ -105,9 +105,7 @@ osinfo_filter_prop_constraints_free(gpointer props)
static void
osinfo_filter_init (OsinfoFilter *filter)
{
- OsinfoFilterPrivate *priv;
- priv = OSINFO_FILTER_GET_PRIVATE(filter);
- filter->priv = priv;
+ filter->priv = OSINFO_FILTER_GET_PRIVATE(filter);
filter->priv->propertyConstraints =
g_hash_table_new_full(g_str_hash,
diff --git a/osinfo/osinfo_install_config.c b/osinfo/osinfo_install_config.c
index 93dc039..1712be5 100644
--- a/osinfo/osinfo_install_config.c
+++ b/osinfo/osinfo_install_config.c
@@ -67,11 +67,10 @@ static const gchar valid[] = {
static void
osinfo_install_config_init (OsinfoInstallConfig *config)
{
- OsinfoInstallConfigPrivate *priv;
gchar pass[9];
gsize i;
- config->priv = priv = OSINFO_INSTALL_CONFIG_GET_PRIVATE(config);
+ config->priv = OSINFO_INSTALL_CONFIG_GET_PRIVATE(config);
osinfo_entity_set_param(OSINFO_ENTITY(config),
OSINFO_INSTALL_CONFIG_PROP_L10N_KEYBOARD,
diff --git a/osinfo/osinfo_install_config_paramlist.c b/osinfo/osinfo_install_config_paramlist.c
index d3eeffc..740de70 100644
--- a/osinfo/osinfo_install_config_paramlist.c
+++ b/osinfo/osinfo_install_config_paramlist.c
@@ -66,9 +66,7 @@ osinfo_install_config_paramlist_class_init (OsinfoInstallConfigParamListClass *k
static void
osinfo_install_config_paramlist_init (OsinfoInstallConfigParamList *list)
{
- OsinfoInstallConfigParamListPrivate *priv;
- list->priv = priv = OSINFO_INSTALL_CONFIG_PARAMLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_INSTALL_CONFIG_PARAMLIST_GET_PRIVATE(list);
}
/**
diff --git a/osinfo/osinfo_install_script.c b/osinfo/osinfo_install_script.c
index 7105430..276c756 100644
--- a/osinfo/osinfo_install_script.c
+++ b/osinfo/osinfo_install_script.c
@@ -337,9 +337,7 @@ osinfo_install_script_get_config_param(const OsinfoInstallScript *script,
static void
osinfo_install_script_init (OsinfoInstallScript *list)
{
- OsinfoInstallScriptPrivate *priv;
- list->priv = priv = OSINFO_INSTALL_SCRIPT_GET_PRIVATE(list);
-
+ list->priv = OSINFO_INSTALL_SCRIPT_GET_PRIVATE(list);
list->priv->config_params = osinfo_install_config_paramlist_new();
}
diff --git a/osinfo/osinfo_install_scriptlist.c b/osinfo/osinfo_install_scriptlist.c
index 99f49dd..9edc4fa 100644
--- a/osinfo/osinfo_install_scriptlist.c
+++ b/osinfo/osinfo_install_scriptlist.c
@@ -65,9 +65,7 @@ osinfo_install_scriptlist_class_init (OsinfoInstallScriptListClass *klass)
static void
osinfo_install_scriptlist_init (OsinfoInstallScriptList *list)
{
- OsinfoInstallScriptListPrivate *priv;
- list->priv = priv = OSINFO_INSTALL_SCRIPTLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_INSTALL_SCRIPTLIST_GET_PRIVATE(list);
}
/**
diff --git a/osinfo/osinfo_list.c b/osinfo/osinfo_list.c
index c32a5ff..ddfd58d 100644
--- a/osinfo/osinfo_list.c
+++ b/osinfo/osinfo_list.c
@@ -145,9 +145,7 @@ osinfo_list_class_init (OsinfoListClass *klass)
static void
osinfo_list_init (OsinfoList *list)
{
- OsinfoListPrivate *priv;
- list->priv = priv = OSINFO_LIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_LIST_GET_PRIVATE(list);
list->priv->array = g_ptr_array_new_with_free_func(NULL);
list->priv->entities = g_hash_table_new_full(g_str_hash,
g_str_equal,
diff --git a/osinfo/osinfo_loader.c b/osinfo/osinfo_loader.c
index efbbc97..d713767 100644
--- a/osinfo/osinfo_loader.c
+++ b/osinfo/osinfo_loader.c
@@ -84,9 +84,7 @@ osinfo_loader_class_init (OsinfoLoaderClass *klass)
static void
osinfo_loader_init (OsinfoLoader *loader)
{
- OsinfoLoaderPrivate *priv;
- loader->priv = priv = OSINFO_LOADER_GET_PRIVATE(loader);
-
+ loader->priv = OSINFO_LOADER_GET_PRIVATE(loader);
loader->priv->db = osinfo_db_new();
}
diff --git a/osinfo/osinfo_media.c b/osinfo/osinfo_media.c
index 8f03046..67dd6af 100644
--- a/osinfo/osinfo_media.c
+++ b/osinfo/osinfo_media.c
@@ -552,8 +552,7 @@ osinfo_media_class_init (OsinfoMediaClass *klass)
static void
osinfo_media_init (OsinfoMedia *media)
{
- OsinfoMediaPrivate *priv;
- media->priv = priv = OSINFO_MEDIA_GET_PRIVATE(media);
+ media->priv = OSINFO_MEDIA_GET_PRIVATE(media);
g_weak_ref_init(&media->priv->os, NULL);
}
diff --git a/osinfo/osinfo_medialist.c b/osinfo/osinfo_medialist.c
index 3894177..9ec06bf 100644
--- a/osinfo/osinfo_medialist.c
+++ b/osinfo/osinfo_medialist.c
@@ -65,9 +65,7 @@ osinfo_medialist_class_init (OsinfoMediaListClass *klass)
static void
osinfo_medialist_init (OsinfoMediaList *list)
{
- OsinfoMediaListPrivate *priv;
- list->priv = priv = OSINFO_MEDIALIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_MEDIALIST_GET_PRIVATE(list);
}
/**
diff --git a/osinfo/osinfo_os.c b/osinfo/osinfo_os.c
index 97c33f7..25f7a4c 100644
--- a/osinfo/osinfo_os.c
+++ b/osinfo/osinfo_os.c
@@ -171,8 +171,7 @@ osinfo_os_class_init (OsinfoOsClass *klass)
static void
osinfo_os_init (OsinfoOs *os)
{
- OsinfoOsPrivate *priv;
- os->priv = priv = OSINFO_OS_GET_PRIVATE(os);
+ os->priv = OSINFO_OS_GET_PRIVATE(os);
os->priv->deviceLinks = NULL;
os->priv->medias = osinfo_medialist_new ();
diff --git a/osinfo/osinfo_oslist.c b/osinfo/osinfo_oslist.c
index fbb2f30..e2f5c42 100644
--- a/osinfo/osinfo_oslist.c
+++ b/osinfo/osinfo_oslist.c
@@ -65,9 +65,7 @@ osinfo_oslist_class_init (OsinfoOsListClass *klass)
static void
osinfo_oslist_init (OsinfoOsList *list)
{
- OsinfoOsListPrivate *priv;
- list->priv = priv = OSINFO_OSLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_OSLIST_GET_PRIVATE(list);
}
diff --git a/osinfo/osinfo_platform.c b/osinfo/osinfo_platform.c
index 73411aa..fdec4c6 100644
--- a/osinfo/osinfo_platform.c
+++ b/osinfo/osinfo_platform.c
@@ -81,9 +81,7 @@ osinfo_platform_class_init (OsinfoPlatformClass *klass)
static void
osinfo_platform_init (OsinfoPlatform *platform)
{
- OsinfoPlatformPrivate *priv;
- platform->priv = priv = OSINFO_PLATFORM_GET_PRIVATE(platform);
-
+ platform->priv = OSINFO_PLATFORM_GET_PRIVATE(platform);
platform->priv->deviceLinks = NULL;
}
diff --git a/osinfo/osinfo_platformlist.c b/osinfo/osinfo_platformlist.c
index ef197e3..cdd7052 100644
--- a/osinfo/osinfo_platformlist.c
+++ b/osinfo/osinfo_platformlist.c
@@ -65,9 +65,7 @@ osinfo_platformlist_class_init (OsinfoPlatformListClass *klass)
static void
osinfo_platformlist_init (OsinfoPlatformList *list)
{
- OsinfoPlatformListPrivate *priv;
- list->priv = priv = OSINFO_PLATFORMLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_PLATFORMLIST_GET_PRIVATE(list);
}
diff --git a/osinfo/osinfo_product.c b/osinfo/osinfo_product.c
index 8f0a799..47edb9c 100644
--- a/osinfo/osinfo_product.c
+++ b/osinfo/osinfo_product.c
@@ -239,9 +239,7 @@ osinfo_product_class_init (OsinfoProductClass *klass)
static void
osinfo_product_init (OsinfoProduct *product)
{
- OsinfoProductPrivate *priv;
- product->priv = priv = OSINFO_PRODUCT_GET_PRIVATE(product);
-
+ product->priv = OSINFO_PRODUCT_GET_PRIVATE(product);
product->priv->productLinks = NULL;
}
diff --git a/osinfo/osinfo_productfilter.c b/osinfo/osinfo_productfilter.c
index 5961707..075ccc6 100644
--- a/osinfo/osinfo_productfilter.c
+++ b/osinfo/osinfo_productfilter.c
@@ -110,10 +110,7 @@ osinfo_productfilter_product_constraints_free(gpointer relshps)
static void
osinfo_productfilter_init (OsinfoProductFilter *productfilter)
{
- OsinfoProductFilterPrivate *priv;
- priv = OSINFO_PRODUCTFILTER_GET_PRIVATE(productfilter);
- productfilter->priv = priv;
-
+ productfilter->priv = OSINFO_PRODUCTFILTER_GET_PRIVATE(productfilter);
productfilter->priv->productConstraints =
g_hash_table_new_full(g_direct_hash,
g_direct_equal,
diff --git a/osinfo/osinfo_productlist.c b/osinfo/osinfo_productlist.c
index adbcb3b..5ebcc90 100644
--- a/osinfo/osinfo_productlist.c
+++ b/osinfo/osinfo_productlist.c
@@ -65,9 +65,7 @@ osinfo_productlist_class_init (OsinfoProductListClass *klass)
static void
osinfo_productlist_init (OsinfoProductList *list)
{
- OsinfoProductListPrivate *priv;
- list->priv = priv = OSINFO_PRODUCTLIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_PRODUCTLIST_GET_PRIVATE(list);
}
diff --git a/osinfo/osinfo_resources.c b/osinfo/osinfo_resources.c
index 67d9952..7dfb3d1 100644
--- a/osinfo/osinfo_resources.c
+++ b/osinfo/osinfo_resources.c
@@ -248,8 +248,7 @@ osinfo_resources_class_init (OsinfoResourcesClass *klass)
static void
osinfo_resources_init (OsinfoResources *resources)
{
- OsinfoResourcesPrivate *priv;
- resources->priv = priv = OSINFO_RESOURCES_GET_PRIVATE(resources);
+ resources->priv = OSINFO_RESOURCES_GET_PRIVATE(resources);
}
OsinfoResources *osinfo_resources_new(const gchar *id,
diff --git a/osinfo/osinfo_resourceslist.c b/osinfo/osinfo_resourceslist.c
index 45c54f6..407ccf9 100644
--- a/osinfo/osinfo_resourceslist.c
+++ b/osinfo/osinfo_resourceslist.c
@@ -67,8 +67,7 @@ osinfo_resourceslist_class_init (OsinfoResourcesListClass *klass)
static void
osinfo_resourceslist_init (OsinfoResourcesList *list)
{
- OsinfoResourcesListPrivate *priv;
- list->priv = priv = OSINFO_RESOURCESLIST_GET_PRIVATE(list);
+ list->priv = OSINFO_RESOURCESLIST_GET_PRIVATE(list);
}
diff --git a/osinfo/osinfo_tree.c b/osinfo/osinfo_tree.c
index fe8e659..96db760 100644
--- a/osinfo/osinfo_tree.c
+++ b/osinfo/osinfo_tree.c
@@ -389,8 +389,7 @@ osinfo_tree_class_init(OsinfoTreeClass *klass)
static void
osinfo_tree_init(OsinfoTree *tree)
{
- OsinfoTreePrivate *priv;
- tree->priv = priv = OSINFO_TREE_GET_PRIVATE(tree);
+ tree->priv = OSINFO_TREE_GET_PRIVATE(tree);
}
OsinfoTree *osinfo_tree_new(const gchar *id,
diff --git a/osinfo/osinfo_treelist.c b/osinfo/osinfo_treelist.c
index 9b1e47e..d3508a2 100644
--- a/osinfo/osinfo_treelist.c
+++ b/osinfo/osinfo_treelist.c
@@ -65,9 +65,7 @@ osinfo_treelist_class_init (OsinfoTreeListClass *klass)
static void
osinfo_treelist_init (OsinfoTreeList *list)
{
- OsinfoTreeListPrivate *priv;
- list->priv = priv = OSINFO_TREELIST_GET_PRIVATE(list);
-
+ list->priv = OSINFO_TREELIST_GET_PRIVATE(list);
}
/**
--
1.8.1
From berrange at redhat.com Thu Jan 31 11:45:21 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Thu, 31 Jan 2013 11:45:21 +0000
Subject: [Libosinfo] [libosinfo] Remove unused 'priv' variable from
_init methods
In-Reply-To: <1359627232-13578-1-git-send-email-cfergeau@redhat.com>
References: <1359627232-13578-1-git-send-email-cfergeau@redhat.com>
Message-ID: <20130131114521.GG23534@redhat.com>
On Thu, Jan 31, 2013 at 11:13:52AM +0100, Christophe Fergeau wrote:
> In most instance _init() methods, the instance priv member
> is initialized, but a local 'priv' variable was also declared and
> assigned, but not actually used. Kill this intermediate variable
> as this causes Coverity warnings.
ACK
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From cfergeau at redhat.com Thu Jan 31 12:43:22 2013
From: cfergeau at redhat.com (Christophe Fergeau)
Date: Thu, 31 Jan 2013 13:43:22 +0100
Subject: [Libosinfo] [libosinfo] Use gnulib's ignore-value.h to silence gcc
Message-ID: <1359636202-25516-1-git-send-email-cfergeau@redhat.com>
osinfo_loader.c contains a few local hacks to avoid gcc warnings
about set-but-not-read variables. While they are good to silence
gcc, they also cause warnings from the Coverity checker.
Use the ignore_value() macro from gnulib to avoid these warnings
as they don't cause Coverity warnings.
---
osinfo/ignore-value.h | 47 +++++++++++++++++++++++++++++++++++++++++++++++
osinfo/osinfo_loader.c | 19 ++++++-------------
2 files changed, 53 insertions(+), 13 deletions(-)
create mode 100644 osinfo/ignore-value.h
diff --git a/osinfo/ignore-value.h b/osinfo/ignore-value.h
new file mode 100644
index 0000000..6bab971
--- /dev/null
+++ b/osinfo/ignore-value.h
@@ -0,0 +1,47 @@
+/* ignore a function return without a compiler warning
+
+ Copyright (C) 2008-2013 Free Software Foundation, Inc.
+
+ This program is free software: you can redistribute it and/or modify
+ it under the terms of the GNU Lesser General Public License as published by
+ the Free Software Foundation; either version 2.1 of the License, or
+ (at your option) any later version.
+
+ This program is distributed in the hope that it will be useful,
+ but WITHOUT ANY WARRANTY; without even the implied warranty of
+ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ GNU Lesser General Public License for more details.
+
+ You should have received a copy of the GNU Lesser General Public License
+ along with this program. If not, see . */
+
+/* Written by Jim Meyering, Eric Blake and P?draig Brady. */
+
+/* Use "ignore_value" to avoid a warning when using a function declared with
+ gcc's warn_unused_result attribute, but for which you really do want to
+ ignore the result. Traditionally, people have used a "(void)" cast to
+ indicate that a function's return value is deliberately unused. However,
+ if the function is declared with __attribute__((warn_unused_result)),
+ gcc issues a warning even with the cast.
+
+ Caution: most of the time, you really should heed gcc's warning, and
+ check the return value. However, in those exceptional cases in which
+ you're sure you know what you're doing, use this function.
+
+ For the record, here's one of the ignorable warnings:
+ "copy.c:233: warning: ignoring return value of 'fchown',
+ declared with attribute warn_unused_result". */
+
+#ifndef _GL_IGNORE_VALUE_H
+# define _GL_IGNORE_VALUE_H
+
+/* The __attribute__((__warn_unused_result__)) feature
+ is available in gcc versions 3.4 and newer,
+ while the typeof feature has been available since 2.7 at least. */
+# if __GNUC__ < 3 || (__GNUC__ == 3 && __GNUC_MINOR__ < 4)
+# define ignore_value(x) ((void) (x))
+# else
+# define ignore_value(x) (({ __typeof__ (x) __x = (x); (void) __x; }))
+# endif
+
+#endif
diff --git a/osinfo/osinfo_loader.c b/osinfo/osinfo_loader.c
index d713767..76e9bc2 100644
--- a/osinfo/osinfo_loader.c
+++ b/osinfo/osinfo_loader.c
@@ -34,6 +34,7 @@
#include
#include
#include
+#include "ignore-value.h"
#include "osinfo_install_script_private.h"
#include "osinfo_device_driver_private.h"
@@ -1486,20 +1487,15 @@ osinfo_loader_process_file_reg_ids(OsinfoLoader *loader,
WANT_ID(subvendor_id);
WANT_ID(subdevice_id);
WANT_REST(subsystem);
-
- /* Pretend we 'use' these variables to get around
- * 'gcc' warnings about set-but-not-read vars */
- if (subvendor_id || subdevice_id || subsystem)
- subsystem = subsystem;
+ ignore_value(subvendor_id);
+ ignore_value(subdevice_id);
+ ignore_value(subsystem);
} else {
FREE_BUF(device_buf);
WANT_ID(device_id);
WANT_REST(device);
SAVE_BUF(device_buf);
-
- /* Trick gcc set-by-not-read-vars warning */
- if (device)
- device = device;
+ ignore_value(device);
gchar *id = g_strdup_printf("%s/%s/%s",
baseURI, vendor_id, device_id);
@@ -1521,10 +1517,7 @@ osinfo_loader_process_file_reg_ids(OsinfoLoader *loader,
WANT_ID(vendor_id);
WANT_REST(vendor);
SAVE_BUF(vendor_buf);
-
- /* Trick gcc set-by-not-read-vars warning */
- if (vendor)
- vendor = vendor;
+ ignore_value(vendor);
}
done:
--
1.8.1
From zeeshanak at gnome.org Thu Jan 31 14:33:08 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Thu, 31 Jan 2013 15:33:08 +0100
Subject: [Libosinfo] [PATCH] install-script tool: Display names of generated
files
Message-ID: <1359642788-12597-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
---
tools/osinfo-install-script.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/tools/osinfo-install-script.c b/tools/osinfo-install-script.c
index fabce96..0f5656b 100644
--- a/tools/osinfo-install-script.c
+++ b/tools/osinfo-install-script.c
@@ -228,6 +228,7 @@ static gboolean generate_script(OsinfoOs *os)
error->message ? error->message : "unknown");
goto cleanup;
}
+ g_print ("%s\n", osinfo_install_script_get_output_filename(script));
}
ret = TRUE;
--
1.8.1
From zeeshanak at gnome.org Thu Jan 31 14:56:24 2013
From: zeeshanak at gnome.org (Zeeshan Ali (Khattak))
Date: Thu, 31 Jan 2013 15:56:24 +0100
Subject: [Libosinfo] [PATCH] install-script tool: Take GFile creation out of
the loop
Message-ID: <1359644184-13735-1-git-send-email-zeeshanak@gnome.org>
From: "Zeeshan Ali (Khattak)"
Also, unref the GFile when done.
---
tools/osinfo-install-script.c | 6 ++++--
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/tools/osinfo-install-script.c b/tools/osinfo-install-script.c
index 0f5656b..9eabea6 100644
--- a/tools/osinfo-install-script.c
+++ b/tools/osinfo-install-script.c
@@ -190,6 +190,7 @@ static gboolean generate_script(OsinfoOs *os)
OsinfoInstallScriptList *scripts = osinfo_os_get_install_script_list(os);
OsinfoInstallScriptList *profile_scripts;
OsinfoFilter *filter;
+ GFile *dir;
GList *l, *tmp;
gboolean ret = FALSE;
GError *error = NULL;
@@ -209,10 +210,10 @@ static gboolean generate_script(OsinfoOs *os)
goto cleanup;
}
+ dir = g_file_new_for_commandline_arg(output_dir ? output_dir : ".");
+
for (tmp = l; tmp != NULL; tmp = tmp->next) {
OsinfoInstallScript *script = tmp->data;
- GFile *dir = g_file_new_for_commandline_arg(output_dir ?
- output_dir : ".");
if (prefix)
osinfo_install_script_set_output_prefix(script, prefix);
@@ -237,6 +238,7 @@ static gboolean generate_script(OsinfoOs *os)
g_object_unref(scripts);
g_object_unref(filter);
g_object_unref(profile_scripts);
+ g_object_unref(dir);
return ret;
}
--
1.8.1
From berrange at redhat.com Thu Jan 31 15:03:24 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Thu, 31 Jan 2013 15:03:24 +0000
Subject: [Libosinfo] [PATCH] install-script tool: Display names of
generated files
In-Reply-To: <1359642788-12597-1-git-send-email-zeeshanak@gnome.org>
References: <1359642788-12597-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130131150324.GM23534@redhat.com>
On Thu, Jan 31, 2013 at 03:33:08PM +0100, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> ---
> tools/osinfo-install-script.c | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/tools/osinfo-install-script.c b/tools/osinfo-install-script.c
> index fabce96..0f5656b 100644
> --- a/tools/osinfo-install-script.c
> +++ b/tools/osinfo-install-script.c
> @@ -228,6 +228,7 @@ static gboolean generate_script(OsinfoOs *os)
> error->message ? error->message : "unknown");
> goto cleanup;
> }
> + g_print ("%s\n", osinfo_install_script_get_output_filename(script));
> }
> ret = TRUE;
If you're going todo this, can you also add a '-q' / '--quiet' command
line flag to allow it to be surpressed.
Regards,
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
From berrange at redhat.com Thu Jan 31 15:04:14 2013
From: berrange at redhat.com (Daniel P. Berrange)
Date: Thu, 31 Jan 2013 15:04:14 +0000
Subject: [Libosinfo] [PATCH] install-script tool: Take GFile creation
out of the loop
In-Reply-To: <1359644184-13735-1-git-send-email-zeeshanak@gnome.org>
References: <1359644184-13735-1-git-send-email-zeeshanak@gnome.org>
Message-ID: <20130131150414.GN23534@redhat.com>
On Thu, Jan 31, 2013 at 03:56:24PM +0100, Zeeshan Ali (Khattak) wrote:
> From: "Zeeshan Ali (Khattak)"
>
> Also, unref the GFile when done.
> ---
> tools/osinfo-install-script.c | 6 ++++--
> 1 file changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/tools/osinfo-install-script.c b/tools/osinfo-install-script.c
> index 0f5656b..9eabea6 100644
> --- a/tools/osinfo-install-script.c
> +++ b/tools/osinfo-install-script.c
> @@ -190,6 +190,7 @@ static gboolean generate_script(OsinfoOs *os)
> OsinfoInstallScriptList *scripts = osinfo_os_get_install_script_list(os);
> OsinfoInstallScriptList *profile_scripts;
> OsinfoFilter *filter;
> + GFile *dir;
> GList *l, *tmp;
> gboolean ret = FALSE;
> GError *error = NULL;
> @@ -209,10 +210,10 @@ static gboolean generate_script(OsinfoOs *os)
> goto cleanup;
> }
>
> + dir = g_file_new_for_commandline_arg(output_dir ? output_dir : ".");
> +
> for (tmp = l; tmp != NULL; tmp = tmp->next) {
> OsinfoInstallScript *script = tmp->data;
> - GFile *dir = g_file_new_for_commandline_arg(output_dir ?
> - output_dir : ".");
>
> if (prefix)
> osinfo_install_script_set_output_prefix(script, prefix);
> @@ -237,6 +238,7 @@ static gboolean generate_script(OsinfoOs *os)
> g_object_unref(scripts);
> g_object_unref(filter);
> g_object_unref(profile_scripts);
> + g_object_unref(dir);
> return ret;
> }
ACK
Daniel
--
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org -o- http://virt-manager.org :|
|: http://autobuild.org -o- http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|