2015-04-03 GnuCash IRC logs
00:00:18 *** fuzzybunny69y has joined #gnucash
00:00:35 <fuzzybunny69y> hey guys how do I copy and paste in gnucash for mac?
00:01:06 <fuzzybunny69y> I did a search for a certain description and am trying to copy and paste the same description to multiple entries and it isn’t working when I hit command + c or command + v
00:27:44 *** botzko has quit IRC
00:40:55 *** ErKa has joined #gnucash
01:28:21 *** jimvideo has quit IRC
01:48:31 *** O01eg has quit IRC
02:08:52 *** ErKa has quit IRC
02:15:43 *** MechtiIde has joined #gnucash
02:17:11 *** uXus has quit IRC
02:19:47 *** uXus has joined #gnucash
02:27:26 *** lwells has joined #gnucash
02:30:48 *** lwells_ has quit IRC
02:32:19 *** pacon has quit IRC
02:33:03 *** pacon has joined #gnucash
02:41:41 *** botzko has joined #gnucash
02:53:51 *** Krzysiek_K has joined #gnucash
03:00:36 *** gjanssens has joined #gnucash
03:00:36 *** gncbot sets mode: +o gjanssens
03:25:20 *** GabrieleV_ has joined #gnucash
03:26:04 *** GabrieleV has quit IRC
03:26:05 *** GabrieleV_ is now known as GabrieleV
03:49:59 *** GabrieleV_ has joined #gnucash
03:50:17 *** GabrieleV has quit IRC
03:50:17 *** GabrieleV_ is now known as GabrieleV
05:00:18 *** fabior has joined #gnucash
05:53:15 *** Krzysiek_K has quit IRC
06:38:19 *** Jimraehl1 has left #gnucash
06:56:43 *** Jimraehl1 has joined #gnucash
07:00:37 *** cartsoftware has quit IRC
07:36:29 *** aqua___ has joined #gnucash
07:57:32 *** rickoehn has joined #gnucash
08:00:34 *** himaxx has joined #gnucash
08:02:32 *** himaxx has quit IRC
08:38:03 *** fuzzybunny69y has quit IRC
08:40:24 *** fuzzybunny69y has joined #gnucash
08:43:52 *** StuM has quit IRC
08:48:11 *** StuM has joined #gnucash
09:08:51 *** pacon has quit IRC
09:31:43 *** StuM has quit IRC
09:39:33 *** botzko has quit IRC
09:55:54 *** aqua___ has quit IRC
10:00:03 *** ErKa has joined #gnucash
10:03:29 *** aqua___ has joined #gnucash
10:16:20 <warlord> jralls: did you see the cut&paste question from fuzzybunny69y ?
10:29:31 *** ErKa has quit IRC
10:46:21 *** fabior has quit IRC
11:00:20 *** fell_ is now known as fell
11:02:21 <fell> I never worked on gnucash-on-windows before. Should I run some tests after adding a missing translation line to gnucash.iss?
11:04:32 <fell> gjanssenss, jralls?
11:06:53 <warlord> I dont know what kind of tests you could run?
11:07:19 <fell> Me too! :-)
11:11:35 *** andy has quit IRC
11:21:02 *** ErKa has joined #gnucash
11:25:32 *** andy has joined #gnucash
11:41:35 <gjanssens> Fell, the only test you can run is to make the installer and check
11:41:38 <gjanssens> a
11:41:51 <gjanssens> a. the installer build doesn't fail
11:42:08 <gjanssens> b. the tranlated string is used when running the installer
11:42:37 <fell> Then I have to reactivate a Windos machine...
11:42:54 *** fuzzybunny69y has quit IRC
11:42:55 <gjanssens> That would be a prerequisite indeed...
11:44:04 <fell> push url is code.gnucash...?
11:44:17 <gjanssens> yes
11:44:39 *** aqua___ has quit IRC
11:44:49 <gjanssens> like gnucash and gnucash-docs but replace the last component with gnucash-on-windows
11:45:15 <fell> Today I only want to answer the de Mails, found a missing string, ...
11:46:29 <fell> GTK theme had none
11:46:38 <fell> in de
11:58:48 *** O01eg has joined #gnucash
11:59:28 <gjanssens> fell: interesting - that string apparently was never translated in DE
11:59:33 <gjanssens> thanks for adding it
12:00:36 <fell> Idea for translation checks: compare number of strings.
12:03:44 <gjanssens> running make check on a minimal debian install fails - missing additonal locales for the qof date-time checks
12:04:01 <gjanssens> this has come up before, but I wasn't paying attention...
12:04:18 <gjanssens> anyone can tell me how to install these missing locales ?
12:11:27 <jralls> @tell fuzzybunny6y cmd-c and cmd-p work on Mac *as long as you're using the official Gnucash.app from our downloads section.* Other builds may not enable the necessary adaptations, in which case ctrl-c and ctrl-v should work.
12:11:27 <gncbot> jralls: The operation succeeded.
12:32:25 <fell> gjanssens: which are missing?
12:32:57 <gjanssens> No idea. This is in a highly automated environment (travis-ci)
12:33:50 <gjanssens> Probably fr_FR and en_GB
12:34:33 <gjanssens> fell: other than en_US these are the two locales the test suite checks
12:35:37 <jralls> gjanssens: You need to edit /etc/locales to uncomment fr_FR, en_GB, and en_US, then run locale-gen.
12:36:35 <gjanssens> jralls: thanks, I'll investigate how I can do that via the travis configuration file
12:37:21 <jralls> I doubt that you can, at least in a secure way: locale-gen must be run as root.
12:38:47 <jralls> OTOH, it shouldn't be necessary: It's an OS-level change.
12:42:00 <gjanssens> jralls: /etc/locales doesn't seem to exist
12:42:05 <gjanssens> can I just create it ?
12:42:18 <gjanssens> (provided I have permissions of course)
12:42:32 <gjanssens> Or is it installed by a certain packag
12:42:41 <gjanssens> s/packag/package/
12:43:02 <jralls> gjanssens: Sorry, /etc/locale.gen.
12:43:15 <gjanssens> oh, ok
12:43:53 <jralls> Do you not have root on this machine?
12:46:45 <gjanssens> It's not my machine: https://travis-ci.org/gjanssens/gnucash/builds
12:47:24 <gjanssens> I'm experimenting with travis to enable automatic "make check" on each push to my repository
12:48:45 <gjanssens> if it works well we could enable it on the main gnucash repository too and get regular qa tests for free
12:48:54 <gjanssens> and automatically
12:49:43 <gjanssens> Bummer... I don't have permission to write to /etc/locale.gen on the system
12:50:07 <gjanssens> I can install new packages, but not directly manipulate config files in /etc (makes kind of sense)
12:50:54 <gjanssens> Oh well, I'll return to this later. Have other things to attend to also...
12:53:14 <jralls> I guess you could talk to the travis folks and see about getting those locales enabled or we could test for their availability and skip the tests if they're not.
12:56:10 <gjanssens> There was a mention of this on http://docs.travis-ci.com/user/common-build-problems/
12:56:21 <gjanssens> Trying their method now...
13:08:50 <gjanssens> Great! sudo apt-get --reinstall install -qq language-pack-en language-pack-fr does the trick
13:09:03 <gjanssens> The locale dependent test has passed
13:09:08 <gjanssens> Waiting for the rest to finish
13:09:32 <jralls> language-pack, huh? I thought that was a Fedora thing, but googling suggests that Ubuntu has picked it up. Real Debian uses dpkg-reconfigure locale, but that seems to run an interactive script. Anyeay, good that it works.
13:10:09 <gjanssens> All tests passed meanwhile
13:10:30 <gjanssens> It may be travis is running ubuntu vm's, I don't really know
13:10:42 <gjanssens> But the proof of concept is ok
13:11:15 <gjanssens> I'll play with it some more in the next couple of weeks and then we can decide if we want to activate this on our main repository
13:11:31 <gjanssens> travis can even run os x tests btw
13:12:52 <jralls> Hey, it's passing, let's turn it on now!
13:13:04 <gjanssens> the swig project test on os x
13:13:22 <gjanssens> Ok, I'll do so later this evening
13:13:40 <gjanssens> (or tomorrow)
13:14:03 <jralls> I don't think we have any tests that are osx-specific, and precious little code that is, mostly setting the environment and fiddling the menus.
13:15:24 <jralls> Speaking of the environment, the GNC_DBD_DIR setting in /etc/gnucash/environment bit me in the butt again. Is there a way to make it respect externally-set environment variables?
14:05:38 *** fell_ has joined #gnucash
14:05:39 *** gncbot sets mode: +o fell_
14:08:20 <gjanssens> jralls: what do you want to achieve exactly ?
14:09:23 <jralls> gjanssens: I want the GNC_DBD_DIR setting in the launcher script to over-ride the one in etc/gnucash/environment instead of the other way around.
14:09:35 *** fell_ has quit IRC
14:10:33 *** fell_ has joined #gnucash
14:10:33 *** gncbot sets mode: +o fell_
14:13:18 <gjanssens> Hmm, GNC_DBD_DIR is determined at configure time
14:13:40 <gjanssens> Is the location on OS X different durint build and after installation ?
14:14:13 *** fell has quit IRC
14:16:06 <jralls> gjanssens: A bit. Build is /Users/john/Development/Gnucash-Build/Gnucash-2.6/Leopard/inst/lib/dbd. "Install" means a relocatable directory structure called a bundle, so the current path needs to be discovered at runtime.
14:18:02 <gjanssens> The current environment file doesn't really use environment variable syntax, while the configure script does
14:18:10 <gjanssens> maybe you can abuse this?
14:18:59 <gjanssens> Add a component like {FOO_VAR} in GNC_DBD_DIR
14:19:16 <gjanssens> On the other hand, it's not a path, it's a directory that gets set
14:19:56 *** gjanssens is now known as gjanssens_
14:21:07 <jralls> So in configure set GNC_DBD_DIR to {FOO_VAR} and then set $FOO_VAR in the launcher script? Seems a bit opaque.
14:23:02 <jralls> I was thinking more along the lines of special-casing it in gnucash-bin and using the etc/gnucash/environment version only if $GNC_DBD_DIR isn't set. That's kind of ugly, too, I admit.
14:24:19 <jralls> Another alternative would be to use gnc_path_get_lib() in gnc-backend-dbd.c and only use the environment setting as a fallback.
14:26:21 <jralls> I mean gnv_path_get_libdir().
14:26:39 <jralls> Sigh. s/gnv/gnc/
14:30:38 <gjanssens_> that's also an option
14:30:45 *** gjanssens_ is now known as gjanssens
14:31:05 <gjanssens> jralls: and have you tried specifying a relative path instead ?
14:31:29 <jralls> Where?
14:32:37 <gjanssens> at configure time for --with-dbi-dbd-dir ?
14:34:30 <jralls> Not that I remember, but I think that that it would turn out to be relative to $PWD, which wouldn't be very useful.
14:35:15 <gjanssens> Yes, that's probably so
14:37:14 <jralls> Heh, in order to even configure it would need to be relative to the builddir, so it would still need to include too much path.
14:37:35 <gjanssens> Right...
14:38:52 <gjanssens> So far I like using gnc_path_get_libdir best
14:39:10 <gjanssens> Are there ever reasons when that is not desired ?
14:39:37 <gjanssens> For example when trying to test a version of dbi that's different from system-supplied ?
14:42:32 <jralls> Yeah, we've boxed ourselves in a bit by requiring GNC_DBD_DIR to be set by default.
14:45:57 <gjanssens> Does it ?
14:46:33 <gjanssens> It will only be added to the environment file if CUSTOM_GNC_DBD_DIR is set
14:47:50 <gjanssens> And that is only the case if you use --with-dbi-dbd-dir
14:50:41 <jralls> Which apparently I need to do. So first I should fix the dbd detection code in configure so that I don't and see what happens when running the test.
15:01:45 <gjanssens> ok
15:17:49 <jralls> Problem was wrong brackets: ${prefix} vs. $(prefix). Must have been something that autoconf tightened up. With that fixed make test passes and no GNC_DBD_LIB in environment.
15:26:31 <jralls> But I think I'll add the gnc_file_get_libdir() phrase anyway and make using $GNC_DBD_DIR extraordinary.
15:30:37 <gjanssens> Good catch
15:31:05 <gjanssens> With your second sentence do you mean that it should only rarely be needed to set GNC_DBD_DIR
15:31:12 <gjanssens> but if it's set it takes precendence ?
15:31:22 <gjanssens> s/precendence/precedence/
15:31:56 <gjanssens> Uh, never mind
15:32:00 <gjanssens> I was thinking backwards
15:32:20 <gjanssens> Or not...
15:32:40 *** fell_ has quit IRC
15:32:45 <gjanssens> Hmm, I'm getting tired it seems
15:32:49 *** fell_ has joined #gnucash
15:32:49 *** gncbot sets mode: +o fell_
15:34:14 <jralls> gjanssens: Yes, that's what I mean. It shouldn't be needed if binreloc can find the modules. That mostly applies in Mac bundles, I think.
15:38:44 <gjanssens> Ok
15:39:11 <gjanssens> travis integration is added to maint
15:39:16 <gjanssens> we can extend it further still if we like
15:39:21 <gjanssens> for example let it send mails or an irc notification if a build fails
15:39:23 <gjanssens> and have it test both gcc and clang
15:39:54 <gjanssens> I have disabled clang for now as it complains about a self-assignment somewhere in swig generated code
15:40:36 <gjanssens> That's with guile 1.8 but I suspect the same would happen with guile 2.0
15:43:51 <jralls> That's odd. I use apple clang all the time and it doesn't complain. Maybe the Travis server is using something very new?
15:44:18 <jralls> Anyway, an email to gnucash-devel would be nice, I think.
15:48:20 *** MechtiIde has quit IRC
16:14:10 *** aqua___ has joined #gnucash
17:07:15 <gjanssens> I'll look into sending a mail to gnucash-devel
17:07:26 <gjanssens> Later though. Time to log off...
17:07:27 <gjanssens> Bye
17:07:54 <jralls> Gnight.
17:08:22 *** gjanssens has quit IRC
17:46:25 *** rickoehn has quit IRC
18:12:40 *** aqua___ has quit IRC
18:16:00 *** jralls is now known as jralls-afk
18:23:24 *** aqua___ has joined #gnucash
18:24:33 *** fell_ has quit IRC
18:25:22 *** fell_ has joined #gnucash
18:25:22 *** gncbot sets mode: +o fell_
19:13:09 *** fell_ has quit IRC
19:13:50 *** fell_ has joined #gnucash
19:13:51 *** gncbot sets mode: +o fell_
19:18:57 *** fuzzybunny69y has joined #gnucash
19:19:11 *** fell_ has quit IRC
19:25:58 *** ErKa has quit IRC
20:29:00 *** ErKa has joined #gnucash
20:31:51 *** himaxx has joined #gnucash
20:33:07 *** himaxx has quit IRC
21:17:23 *** fell has joined #gnucash
21:17:24 *** gncbot sets mode: +o fell
22:01:44 *** pacon has joined #gnucash
22:24:56 *** fell has quit IRC
23:21:00 *** GabrieleV has quit IRC
23:21:00 *** GabrieleV_ has joined #gnucash
23:21:18 *** GabrieleV_ is now known as GabrieleV
23:34:28 *** fuzzybunny69y has quit IRC
23:36:35 *** fuzzybunny69y has joined #gnucash
23:37:13 *** fuzzybunny69y has quit IRC
23:40:33 *** aqua_ has joined #gnucash
23:49:56 *** aqua___ has quit IRC