Hey,
I could use some help with a fairly straightforward packaging task.
The yorba team has made 0.5.90 snapshots available here: http://yorba.org/download/shotwell/0.5/unstable/
These depend on a little library called gexiv2, which can be found here: http://yorba.org/download/gexiv2/0.0/unstable/
It would be great if somebody could package gexiv2 and get it into rawhide, so that we can update shotwell, I currently lack the spare cycles to do this.
Thanks,
Matthias
On 1 June 2010 23:44, Matthias Clasen mclasen@redhat.com wrote:
It would be great if somebody could package gexiv2 and get it into rawhide, so that we can update shotwell, I currently lack the spare cycles to do this.
How is gexiv2 different to exiv2? We already package the latter.
Richard.
On Wed, 2010-06-02 at 08:07 +0100, Richard Hughes wrote:
On 1 June 2010 23:44, Matthias Clasen mclasen@redhat.com wrote:
It would be great if somebody could package gexiv2 and get it into rawhide, so that we can update shotwell, I currently lack the spare cycles to do this.
How is gexiv2 different to exiv2? We already package the latter.
Richard.
hey,
From the project homepage[1]
"gexiv2 is a GObject-based wrapper around the Exiv2 library. It makes the basic features of Exiv2 available to GNOME applications."
Matthias: I'm going to try and package it today. Will keep the list posted.
regards, Ankur
On Tue, 2010-06-01 at 18:44 -0400, Matthias Clasen wrote:
Hey,
I could use some help with a fairly straightforward packaging task.
The yorba team has made 0.5.90 snapshots available here: http://yorba.org/download/shotwell/0.5/unstable/
These depend on a little library called gexiv2, which can be found here: http://yorba.org/download/gexiv2/0.0/unstable/
It would be great if somebody could package gexiv2 and get it into rawhide, so that we can update shotwell, I currently lack the spare cycles to do this.
Thanks,
Matthias
hi,
An update:
The configure script provided in the package is a non standard one which only supports the --prefix option. (try ./configure -h). Therefore, a packager, in this case me, cannot use the %configure macro for the build process. This also implies that the CFLAGS etc. will need to be exported manually, separately for each architecture.
That is way too much messing around in one spec file for such a simple package. I've filed a ticket[1] upstream requesting them to autotoolize the package. The package will have to wait until the ticket is resolved.
Thanks & regards, Ankur
On Wed, 2010-06-02 at 16:34 +0530, Ankur Sinha wrote:
An update:
The configure script provided in the package is a non standard one which only supports the --prefix option. (try ./configure -h). Therefore, a packager, in this case me, cannot use the %configure macro for the build process. This also implies that the CFLAGS etc. will need to be exported manually, separately for each architecture.
That is way too much messing around in one spec file for such a simple package. I've filed a ticket[1] upstream requesting them to autotoolize the package. The package will have to wait until the ticket is resolved.
Guess that means I will have to package it myself after all :-(
Too bad. Thanks for your efforts anyway.
On Wed, Jun 2, 2010 at 5:50 PM, Matthias Clasen mclasen@redhat.com wrote:
Guess that means I will have to package it myself after all :-(
Too bad. Thanks for your efforts anyway.
What about LibRaw ? Isn't that also required for shotwell ?
Kushal
On Wed, 2010-06-02 at 08:20 -0400, Matthias Clasen wrote:
Guess that means I will have to package it myself after all :-(
Too bad. Thanks for your efforts anyway.
hey,
I can go ahead and package it if you need it that badly. Should I?
I thought it would be best to wait for upstream to clean it up though. Hacking around the problem might not pass a review?
regards, Ankur
On Wed, 2010-06-02 at 18:16 +0530, Ankur Sinha wrote:
I can go ahead and package it if you need it that badly. Should I?
It would certainly help me.
I thought it would be best to wait for upstream to clean it up though.
If you think the current state of affairs is unacceptable, you could submit a patch to make things better. Waiting for somebody else is rarely a winning position in FOSS...
Hacking around the problem might not pass a review?
Shotwell had a similar setup when it was first packaged. I don't think 'upstream doesn't do everything perfectly to our liking' is a valid reason to reject a package. Of course, you might end up with a more picky reviewer than me...
On Wed, 2010-06-02 at 08:52 -0400, Matthias Clasen wrote:
On Wed, 2010-06-02 at 18:16 +0530, Ankur Sinha wrote:
I can go ahead and package it if you need it that badly. Should I?
It would certainly help me.
I thought it would be best to wait for upstream to clean it up though.
If you think the current state of affairs is unacceptable, you could submit a patch to make things better. Waiting for somebody else is rarely a winning position in FOSS...
Hacking around the problem might not pass a review?
Shotwell had a similar setup when it was first packaged. I don't think 'upstream doesn't do everything perfectly to our liking' is a valid reason to reject a package. Of course, you might end up with a more picky reviewer than me...
hey,
Okay, I'll work on it tonight then.
regards, Ankur
On Wed, 2010-06-02 at 08:52 -0400, Matthias Clasen wrote:
On Wed, 2010-06-02 at 18:16 +0530, Ankur Sinha wrote:
I can go ahead and package it if you need it that badly. Should I?
It would certainly help me.
I thought it would be best to wait for upstream to clean it up though.
If you think the current state of affairs is unacceptable, you could submit a patch to make things better. Waiting for somebody else is rarely a winning position in FOSS...
Hacking around the problem might not pass a review?
Shotwell had a similar setup when it was first packaged. I don't think 'upstream doesn't do everything perfectly to our liking' is a valid reason to reject a package. Of course, you might end up with a more picky reviewer than me...
hi,
review request:
regards, Ankur
On Wed, 2010-06-02 at 08:52 -0400, Matthias Clasen wrote:
On Wed, 2010-06-02 at 18:16 +0530, Ankur Sinha wrote:
I can go ahead and package it if you need it that badly. Should I?
It would certainly help me.
I thought it would be best to wait for upstream to clean it up though.
If you think the current state of affairs is unacceptable, you could submit a patch to make things better. Waiting for somebody else is rarely a winning position in FOSS...
Hacking around the problem might not pass a review?
Shotwell had a similar setup when it was first packaged. I don't think 'upstream doesn't do everything perfectly to our liking' is a valid reason to reject a package. Of course, you might end up with a more picky reviewer than me...
hello,
http://koji.fedoraproject.org/koji/packageinfo?packageID=10508
The link speaks for itself :)
It isn't building on F-12 yet, exiv2 needs to be updated[1]. I've built it for F-14 and F-13 in the meantime
regards, Ankur
[1]
http://koji.fedoraproject.org/koji/getfile?taskID=2261133&name=build.log...
desktop@lists.fedoraproject.org