[Evolution-hackers] newcomers: How should I compile Evolution?
Howdy!
I can bet that (almost) every contributor has a different way to setup the
environment, compile and use the fresh compiled Evolution and I'm here to
describe the way I do my setup (based on Matthew's setup) :-)
Firstly, jhbuild or not jhbuild?
--
I like to use jhbuild for a few projects that I use and contribute. But
that's not the case for Evolution. As Evolution has a small set of
dependencies and usually none of them are bleeding edge, I'd say do *not*
go for jhbuild in this case.
So, what are you using, Fabiano?
-
I'm using these 2 scripts: common and unstable (
http://rachacuca.org/~fidencio/evolution/newcomers/).
Both of them are in my $HOME/.local/bin and this folder is part of my $PATH.
Then when I open a terminal I just do source unstable and I'm all set to
build from the master branch :-).
When configuring evolution-data-server or evolution I use:
./configure --prefix=$PREFIX, which the unstable script defines to be
$HOME/local/unstable -- so that's where evolution-data-server/evolution get
installed.
*If* I need to build a base library like GTK+, then I use:
./configure --prefix=$COMMON, which the common script defines to be
$HOME/local/common and that way I keep base libraries separated from
evolution-data-server/evolution.
And before running Evolution, what I do is start manually the
evolution-source-registry and the factories processes
(evolution-{addressbook,calendar}-factory) located in
$HOME/local/unstable/libexec/
I do believe this is the simplest way for compiling, using and debugging
Evolution Data Server/Evolution.
Best Regards,
--
Fabiano Fidêncio
___
evolution-hackers mailing list
evolution-hackers@gnome.org
To change your list options or unsubscribe, visit ...
https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] newcomers: How should I compile Evolution?
On Thu, 2014-03-06 at 09:58 +0100, Fabiano Fidêncio wrote: I can bet that (almost) every contributor has a different way to setup the environment, compile and use the fresh compiled Evolution and I'm here to describe the way I do my setup (based on Matthew's setup) :-) Nice instructions! I should transcribe this to a wiki page. Let me amend this with a couple more tricks... I also have three scripts (or just aliases would work too) named: autogen-eds autogen-evo autogen-ews Each of these is basically just... ./autogen.sh --prefix=$PREFIX (yadda, yadda, yadda) This is where I keep the configure options I routinely use for building evolution-data-server, evolution, and evolution-ews, respectively. It's mainly just options like --enable-this or --disable-that, etc. That way I don't have to remember them all or keep them typing them all. Also, a newcomer may not need this but just for completeness, I also have a script named 'stable', which is almost the same as 'unstable', but uses a different PREFIX ($HOME/local/stable). It too references 'common', which is why 'common' is a separate script. I use 'stable' for building our latest stable branch, currently gnome-3-10. It uses a different install prefix so I don't mix files from the two branches. That would be bad. Also, speaking of branches, generic git trick: For me the 'git-new-workdir' script that comes with git is a life saver! It allows you have multiple working directories for the same repo without cloning the whole repo, each checked out to a different branch. That was my biggest complaint with git when I first started using it -- that to switch branches I first had to pack up whatever I was doing in the current working directory, switch branches, and wipe the working directory clean so I don't pick up build artifacts from the other branch. But that increases the build time and slows me down. The 'git-new-workdir' script solves this, but it's not installed in /usr/bin for some reason. You have to dig it out of: /usr/share/doc/git/contrib/workdir or some similar place on your distro. Just copy the script to your $HOME/.local/bin, and do git-new-workdir --help to see how it works. That's my bag of tricks. Matthew Barnes ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] Proposal: Evolution 3.13 Release Schedule
On Mon, 2014-03-03 at 15:51 -0500, Matthew Barnes wrote: Been thinking about our release schedule after 3.12.0, when we break with GNOME and embark on our own annual development/support cycle. I'd like to release updates at a more predictable cadence than GNOME. With the GNOME 3.12.0 and 3.12.1 release dates in mind, I propose the following: - Release stable updates on the 2nd Monday of each month. - Release development updates on the 4th Monday of each month. Another thing I've been thinking about is what to call our stable branch names, starting with 3.12. The usual gnome-3-x branch name is going to be confusing, especially later this year when our version number starts to diverge from GNOME's version number. I suggest just naming the branches after the module name, similar to how GLib and GTK+ do it. e.g. evolution-3-12 evolution-data-server-3-12 evolution-ews-3-12 evolution-mapi-3-12 That sound okay? I don't have a strong preference for the name, other than not using 'gnome' anymore. Matt ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
[Evolution-hackers] When talking about branches...
Hi, when talking about branches, what about deleting all those obsolete for years? I mean the development branches, not those release related. I believe they are useless, definitely those not touched for years (even for months, like mine where I initiated the EClient stuff couple years ago). Bye, Milan ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] When talking about branches...
On Thu, 2014-03-06 at 17:06 +0100, Milan Crha wrote: Right, that's the only being active currently from my point of view too (and according to commit log). Should I figure out the right git command and just get rid of those in eds/evo/ews/ema? Yeah, have at it. I think the commands you need are: git branch -r -- list all the remote branches git push origin :branch-name -- deletes a remote branch (I don't really understand the colon, but that's how it works.) Obviously be careful if you try to automate this or use grep or whatever. Matt ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] When talking about branches...
On Thu, 2014-03-06 at 11:40 -0500, Paul Smith wrote: If you have a new-ish git you can use the --delete flag instead of the colon (exactly the same result, but more readable): git push --delete origin branch-name Thanks for the tip, that's much easier to remember. Matt ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
Re: [Evolution-hackers] When talking about branches...
On Čt, 2014-03-06 at 17:06 +0100, Milan Crha wrote: On Thu, 2014-03-06 at 10:55 -0500, Matthew Barnes wrote: I think the wip/webkit-composer branch might still be alive (ask Tomas). Right, that's the only being active currently from my point of view too (and according to commit log). I'm perfectly fine with removing all the unused branches, but please don't remove wip/webkit-composer and wip-webkit2 branches. Tom ___ evolution-hackers mailing list evolution-hackers@gnome.org To change your list options or unsubscribe, visit ... https://mail.gnome.org/mailman/listinfo/evolution-hackers
