On Wed, 2016-05-18 at 10:22 +0200, Ralf Mardorf wrote:
> If it was build against 3.18, nothing should break when 3.20 is
> provided at runtime.
Hi,
except of the selected color. As I said earlier, the code in WebKitGTK+
is conditionally compiled. It means, if the build is built against gtk+
3.18, then the code is not compiled, if it is built against gtk+ 3.20+,
then it is compiled.
> https://tracker.debian.org/pkg/webkit2gtk
>
> "libgtk-3-dev (>= 3.20.0)" -
> https://tracker.debian.org/media/packages/w/webkit2gtk/control-2.12.2
> -1
>
> IOW the current version in testing _is_ build against 3.20. The OP
> mentioned to use 2.4.11-1 yesterday, so the OP today should run
Please do not mix the two. You reference 2.12.2, then make decisions on
2.4.11 (2.4 is much less than 2.12). :)
In any case, the pointer to the tracker was helpful. The right package
is webkitgtk, not webkit2gtk, and even I didn't find x86_64 build here:
https://buildd.debian.org/status/package.php?p=webkitgtk
I picked the amd64 instead and it led me here:
https://buildd.debian.org/status/fetch.php?pkg=webkitgtk&arch=amd64&ver=2.4.11-1&stamp=1460381568
It's a very long file, no need to download it all, just few first KB,
to be able to search for "libgtk-3" (quotes for clarity only), which
says:
libgtk-3-dev (>= 3.10.0)
and later on:
Get:335 http://mirror.1und1.de/debian unstable/main amd64 libgtk-3-dev amd64
3.18.9-1 [913 kB]
That explains it, there was used gtk+ 3.18.9, not 3.20.x. The webkitgtk
package should be rebuild against gtk+ 3.20.0+ to have the selected
colors fixed.
Disclaimer: If I chose incorrect log file, architecture, whatever, then
please bare with me. Do your part and search the right logs yourself. :)
Bye,
Milan
_______________________________________________
evolution-list mailing list
evolution-list@gnome.org
To change your list options or unsubscribe, visit ...
https://mail.gnome.org/mailman/listinfo/evolution-list
