2017-09-10 GnuCash IRC logs
00:23:36 *** Groan has quit IRC
01:10:01 *** kael has joined #gnucash
01:10:35 *** pilotauto has joined #gnucash
01:11:12 *** kael has quit IRC
02:22:44 *** Mechtilde has joined #gnucash
02:28:40 *** Mechtilde has quit IRC
02:29:55 *** Mechtilde has joined #gnucash
03:45:59 *** hoijui has joined #gnucash
04:20:50 *** karelk has quit IRC
04:31:55 *** gjanssens has joined #gnucash
04:31:55 *** ChanServ sets mode: +o gjanssens
04:33:51 <gjanssens> .
04:34:51 *** gour has joined #gnucash
04:38:40 *** fabior has joined #gnucash
05:24:37 *** karelk has joined #gnucash
05:26:13 *** To7 has quit IRC
05:49:06 *** fell has joined #gnucash
05:56:49 *** pilotauto has quit IRC
06:22:48 *** hoijui has quit IRC
06:39:03 *** kisko has joined #gnucash
06:57:38 *** fabior has quit IRC
07:28:14 *** User has joined #gnucash
08:11:26 *** bertbob has quit IRC
08:15:12 *** Groan has joined #gnucash
08:28:19 *** Groan has quit IRC
08:31:40 *** bertbob has joined #gnucash
08:34:57 *** bertbob has quit IRC
08:35:26 <gour> is it allowed to post to the users list via gmane?
08:46:46 <gour> forget it...i see my post in archives, but not in gmane
08:49:32 *** bertbob has joined #gnucash
09:04:21 *** O01eg has joined #gnucash
09:07:21 *** fabior has joined #gnucash
09:41:33 *** Mechtilde has quit IRC
09:41:47 *** Mechtilde has joined #gnucash
09:54:18 *** fabior has quit IRC
09:54:38 *** fabior has joined #gnucash
11:19:10 *** gncbot sets mode: +o fell
11:38:20 *** kael has joined #gnucash
11:53:44 *** kael has quit IRC
12:11:00 *** Cuare has joined #gnucash
13:29:21 *** User has quit IRC
13:29:28 *** User has joined #gnucash
13:30:17 *** Cuare has quit IRC
13:31:02 *** jralls has joined #gnucash
13:31:02 *** ChanServ sets mode: +o jralls
13:50:58 *** klobuczek has joined #gnucash
13:57:10 *** fabior has quit IRC
14:12:05 *** fabior has joined #gnucash
14:13:16 *** frakturfreak has joined #gnucash
14:16:42 <jralls> gjanssens: I think I've gotten all of your issues from yesterday resolved, plus I've set up to build three separate flavors, 'master', 'unstable', and 'release'.
14:18:36 *** Mechtilde has quit IRC
14:19:18 <gjanssens> jralls: great!
14:19:40 <jralls> Now to try it on the build VM...
14:23:55 <jralls> There's still one oddity, though: `tee` is separating the build output from the jhbuild output and putting the latter at the end.
14:26:28 *** kael has joined #gnucash
14:27:07 <jralls> Hmm, that blew up. Back to my local VM...
14:42:41 <jralls> There are also some issues with cmake make dist: autogenning the un-tarred tarball fails for "circular dependency of AC_LANG_COMPILER(C++)". That might be better handled by autogenning the source dir first so that configure and the Makefiles.in are in the tarball. OTOH if we just drop autotools then both problems go away.
14:42:47 <gjanssens> jralls: ISTR to solve the 'tee' problem one first has to redirect stderr to stdout and only then pipe into tee.
14:43:13 <jralls> I tried that, it didn't make any difference.
14:43:29 <gjanssens> Huh, don't know then...
14:43:59 <gjanssens> I'm more and more inclined to dropping autotools :)
14:44:33 <gjanssens> I believe we almost got everything working well with cmake
14:44:54 <gjanssens> The only thing I still see is an error at gnucash startup that init.py can't be found
14:45:06 <gjanssens> (that is when built with python bindings enabled)
14:45:18 <gjanssens> We probably don't install it yet.
14:45:29 <jralls> Hmm, the problem may be in jhbuild. I think it uses python logging; maybe it doesn't do quite the right thing with the make output.
14:46:21 <jralls> That init.py issue seems vaguely familiar. Wasn't there a crash from that a couple of years ago?
14:47:16 <gjanssens> Another thing I noticed specifically on the Windows build: cmake regenerates POTFILES.in (it does on all platforms), but what it generates it not correct. A number of prefixes are missing. You can look at the diff in the source dir to see what I mean
14:48:07 <gjanssens> From what I understand this init.py allows external python code to be run from within gnucash.
14:48:33 <gjanssens> I always build with python bindings enabled and haven't seen crashes in that area for a long time.
14:49:05 <gjanssens> I think the cmake build just forgets to install it or installs it in the wrong location. I haven't had time to check that yet.
14:49:10 *** fekepp has joined #gnucash
14:50:07 <gjanssens> However, that can be fixed and I don't think that's reason to keep the autotools around.
14:50:58 *** gjanssens is now known as gjanssens_afk
14:55:19 *** klobuczek has quit IRC
14:56:42 <jralls> The issue I see with POTFILES.in has to do with re-sorting. After generating the list there's a bit that adds [type: gettext/gsettings] in front of gschema paths unless WIN32. I left codesmythe a message about it last night.
15:39:08 <lmat> @tell codesmythe I'll give WITH_PYTHON=ON cmake a go soon and see how it goes.
15:39:08 <gncbot> lmat: The operation succeeded.
15:51:33 *** duga has joined #gnucash
15:51:38 <duga> hi
15:52:08 <duga> just installed gnucash and tried to import a MT940 file but gnucash freezes
15:53:40 <lmat> @tell codesmythe I need to go, but if you're curious before I get back, here's the build log: https://travis-ci.org/limitedAtonement/gnucash/builds/273931570 (it's not complete now, but perhaps will be by the time you see it.)
15:53:40 <gncbot> lmat: The operation succeeded.
15:55:55 <jralls> duga: What OS & GnuCash version?
15:56:28 <duga> @jralls latest table 2.6.17 on win10 and macos sierra
15:56:28 <gncbot> duga: Error: "jralls" is not a valid command.
15:57:05 <duga> jralls: latest stable 2.6.17 on win10 and macos sierra
15:57:50 <jralls> duga: Can you open a bug and attach the MT940 file?
15:58:19 <duga> jralls: ok
15:59:28 *** duga has quit IRC
16:14:10 *** gour_ has joined #gnucash
16:14:50 *** frakturfreak has quit IRC
16:16:24 *** gour has quit IRC
16:26:00 *** gour_ has quit IRC
16:50:58 *** kael has quit IRC
16:51:45 *** Unhammer has quit IRC
16:53:29 *** fekepp has quit IRC
16:53:31 *** fekepp has joined #gnucash
17:03:15 *** kael has joined #gnucash
17:04:18 *** Cuare has joined #gnucash
17:14:08 *** kael has quit IRC
17:15:11 *** meb has joined #gnucash
17:21:26 *** codesmythe has joined #gnucash
17:21:30 <codesmythe> .
17:21:30 <gncbot> codesmythe: Sent 20 hours and 22 minutes ago: <jralls> At https://github.com/Gnucash/gnucash/blob/master/po/CMakeLists.txt#L126 you prepend [type: gettext/gsettings] to gschema lines in POTFILES.in unless it's WIN32. Why do you exclude WIN32?
17:21:31 <gncbot> codesmythe: Sent 20 hours and 22 minutes ago: <jralls> The comments suggest the whole operation is diagnostic. Is it still necessary?
17:21:32 <gncbot> codesmythe: Sent 1 hour and 42 minutes ago: <lmat> I'll give WITH_PYTHON=ON cmake a go soon and see how it goes.
17:21:33 <gncbot> codesmythe: Sent 1 hour and 27 minutes ago: <lmat> I need to go, but if you're curious before I get back, here's the build log: https://travis-ci.org/limitedAtonement/gnucash/builds/273931570 (it's not complete now, but perhaps will be by the time you see it.)
17:22:18 *** meb has quit IRC
17:24:38 <codesmythe> jralls: As I recall, it was that (a) I was getting build errors on Windows, (b) this blog post indicated that the tags weren't necessary (but it looks like I am wrong about that as we don't use a version of getttext that recent): https://blogs.gnome.org/mclasen/2016/07/21/using-modern-gettext and (c) I could not discern any difference whether the tag was there or not, though I might not have understood where to look.
17:26:03 <jralls> codesmythe: OK, I'll try removing it and see what happens. It dirties the repo and messes with jhbuild's "do I need to rebuild" logic so I'd like to not do it.
17:27:12 <jralls> codesmythe: But we try to use the latest gettext because fell keeps noticing bugs in older ones. ;)
17:28:20 *** meb has joined #gnucash
17:30:45 <fell> Sorry, I don't have the full context of your conversation.
17:31:24 <jralls> fell: Note the end of the line.
17:33:47 <fell> :-)
17:57:19 <fell> Warlord: while it is nice to have the redirection of wiki.gnucash.org to /wiki, it is confusing for code.gnucash.org.
18:00:23 *** gjanssens_afk has quit IRC
18:06:04 *** fabior_ has joined #gnucash
18:06:09 *** fabior has quit IRC
18:06:29 *** fabior_ has quit IRC
18:08:46 <codesmythe> lmat: Looks like WITH_PYTHON=ON went through okay.
18:09:07 <codesmythe> Is there any reason we shouldn't turn on WITH_PYTHON=ON in Travis/
18:17:17 *** frank has joined #gnucash
18:18:18 *** frank has left #gnucash
18:21:18 *** fbruetting has joined #gnucash
18:36:55 *** fbruetting has quit IRC
18:37:37 *** fbruetting has joined #gnucash
18:39:39 *** muts has joined #gnucash
18:40:04 <muts> hello, regarding "advanced portfolio" report in gnucash, why does "rate of return" use money in rather than the basis?
18:40:30 <muts> if i sold shares of a fund to exchange them for another fund during a rebalance, why should this transaction be "double counted"
18:46:40 <jralls> muts: IIRC it's money in to prevent double-counting reinvested dividends.
18:47:14 <jralls> muts: As for the rebalance txn, what do you mean "double counted"?
18:48:32 <jralls> muts: Bear in mind that I don't use the APR and its designer, Mike Alexander, doesn't come here. He does monitor the user list, so that might be a better place to ask.
18:51:18 <muts> for example, if i buy $10,000 worth of A, then sell $1,000 to buy B, money in becomes $11,000 when only $10,000 was ever actually transacted
18:52:48 <muts> jralls: reinvested dividends would impact basis?
18:55:06 *** pilotauto has joined #gnucash
18:56:03 <jralls> muts: Doesn't the sale of $1000 also create a money out amount? And yes, of course reinvested divs affect basis: Even though you never see the cash you're still getting income (and paying tax on it if it's not in a sheltered account) and then buying shares with it.
19:14:40 *** fekepp has quit IRC
19:15:41 <muts> jralls: yes it does create the "money out" amount, but it seems gnucash doesnt take this into account when calculating returns
19:17:46 <jralls> muts: When you sell a security you book a capital gain. If you use the proceeds to buy another security that's a new investment. It would appear that you're trying to use GnuCash as a portfolio management tool, and it's not one of those.
19:18:19 *** bertbob has quit IRC
19:23:39 <muts> jralls: it almost seems like it would be capable lol
19:23:52 <muts> jralls: well thanks, i suppose at the least i can keep track of minor details
19:30:32 *** fbruetting has quit IRC
19:31:59 *** bertbob has joined #gnucash
19:32:06 *** fbruetting has joined #gnucash
19:35:35 *** fbruetting has quit IRC
19:36:24 *** fbruetting has joined #gnucash
19:36:31 <fell> I wonder if we should not split src/gnome-utils/gnc-main-window.c *copyright = in a formatstr. Else translators have to adjust it every year.
19:37:59 *** fbruetting has quit IRC
19:38:13 <fell> and use a derivate from GNUCASH_BUILD_DATE
19:38:49 *** fbruetting has joined #gnucash
19:43:01 *** fbruetting has quit IRC
19:45:18 *** fbruetting has joined #gnucash
19:48:31 *** fbruetting has quit IRC
19:49:21 *** fbruetting has joined #gnucash
19:49:24 <jralls> fell: That sounds like a good idea.
19:50:06 *** fbruetting has quit IRC
19:50:41 *** puck has quit IRC
19:50:55 *** fbruetting has joined #gnucash
19:51:57 *** fbruetting has joined #gnucash
19:52:08 *** puck has joined #gnucash
19:58:49 *** puck has quit IRC
20:01:28 *** puck has joined #gnucash
20:10:20 *** muts has quit IRC
20:15:05 *** fbruetting has quit IRC
20:19:04 *** fbruetting has joined #gnucash
20:39:13 *** fbruetting has quit IRC
20:40:38 <fell> jralls, can you review my last commit?
20:45:42 *** tuxd00d has quit IRC
20:46:21 <jralls> fell: Where is it?
20:47:23 <fell> in maint
20:52:51 <jralls> Looks like it will work. Does it?
20:54:00 <fell> It works, but I didn't touch code in the last years.
21:00:26 <jralls> Time to get back in practice! ;-)
21:13:31 <codesmythe> fell: Looks like a copy/paste error here: the second GNUCASH_BUILD_DATE should be GNUCASH_BUILD_YEAR. https://github.com/Gnucash/gnucash/blob/maint/src/core-utils/CMakeLists.txt#L70
21:16:33 <jralls> Good catch.
21:16:57 *** Artefact2 has quit IRC
21:22:56 *** Artefact2 has joined #gnucash
21:30:16 <lmat> codesmythe: I'll create the pull request.
21:31:45 <lmat> codesmythe: https://github.com/Gnucash/gnucash/pull/198
21:39:37 <codesmythe> lmat: Looks good to me. I notice that we're doing different configs for cmake-make vs cmake-ninja. I don't think that was intentional, but I think it is probably good. In general, I think we should eventually pick a set of configs we want to test and 'randomly' assign which generator to use.
21:40:07 <codesmythe> Testing ninja vs make per se doesn't really buy us anything other than testing cmake's ability to generate correctly.
21:40:48 <codesmythe> There's nothing in our cmake files that varies based on which generator is used (other than some small xcodebuilder stuff)
21:43:00 <lmat> codesmythe: That debug stuff in the cmake build is lifted from my debugging environment, and has nothing to do with the travis build.
21:45:18 <lmat> codesmythe: It can be removed. I believe you that testing the ninja generator and the default make generator is superfluous, but I cannot independently confirm that.
22:34:52 <fell> Thanks, codesmythe!
23:17:16 *** User has quit IRC
23:29:31 *** Groan has joined #gnucash