2018-04-22 GnuCash IRC logs

00:23:34 *** shakes808 has quit IRC
00:26:05 <CDB-Man> Hmm, if I open my gnucash file on another computer, my opened reports don't carry over. Is there a way to bring them over?
00:47:48 <chris> check saved-reports-X.X in your gnucash profile folder
00:47:48 <gncbot> chris: Sent 20 hours and 59 minutes ago: <fell> Yep, while modern Win versions use utf-16 instead of utf-8 they tend to fall back to the pre-ISO ANSI CP1252.
00:54:31 *** wget has quit IRC
00:54:31 *** ArtGravity has quit IRC
00:54:31 *** fekepp has quit IRC
00:54:31 *** xmaka has quit IRC
00:54:31 *** Aelius has quit IRC
00:54:31 *** cyphase has quit IRC
00:54:31 *** luwum[m] has quit IRC
00:54:31 *** DiogoGomes[m] has quit IRC
00:54:31 *** thardin_ has quit IRC
00:54:31 *** luc14n0 has quit IRC
00:55:22 *** ArtGravity has joined #gnucash
00:55:22 *** wget has joined #gnucash
00:55:22 *** fekepp has joined #gnucash
00:55:22 *** xmaka has joined #gnucash
00:55:22 *** thardin_ has joined #gnucash
00:55:22 *** Aelius has joined #gnucash
00:55:22 *** cyphase has joined #gnucash
00:55:22 *** luwum[m] has joined #gnucash
00:55:22 *** DiogoGomes[m] has joined #gnucash
00:55:22 *** luc14n0 has joined #gnucash
01:03:35 <CDB-Man> chris: which profiles folder? A search of the root directory reveals multiple
01:30:28 *** fell has quit IRC
01:31:40 *** fell has joined #gnucash
01:41:51 *** boldstripe has quit IRC
01:50:17 *** gour has joined #gnucash
02:13:26 *** Mechtilde has joined #gnucash
03:41:27 *** vxp_ has joined #gnucash
03:43:25 *** vxp has quit IRC
04:26:22 *** jotrago1 has joined #gnucash
04:46:42 *** gour_ has joined #gnucash
04:48:18 *** Aussie_matt has joined #gnucash
04:48:47 *** gour has quit IRC
04:56:11 *** chris has quit IRC
05:13:46 *** fabior has joined #gnucash
05:34:03 *** harshitaneja has joined #gnucash
05:52:13 *** Aussie_matt has quit IRC
05:52:49 *** Mechtilde has quit IRC
05:54:09 *** Mechtilde has joined #gnucash
06:03:51 *** gncbot sets mode: +o fell
06:05:41 <fell> CDB-Man: https://wiki.gnucash.org/wiki/FAQ#Q:_How_do_I_backup_my_GC_environment.2C_including_preferences.3F
06:45:17 *** gour_ is now known as gour
07:08:22 *** chris has joined #gnucash
07:25:35 *** fabior has quit IRC
07:37:47 *** moomba has joined #gnucash
07:41:46 <moomba> Hi everyone, I'm new to gnucash. I'm using version 2.6.12, have set everything up and entered a few invoices and clients. Problem is, every time I edit my invoice layout, it doesn't apply next time I hit print. I've saved a custom report configuration for an invoice and I'm trying to apply it to a new invoice, but it always applies the details of the invoice I used to construct the custom config in the first place. Am I going about this the right way? I need t
07:41:55 <moomba> Apologies for the text dump!
07:43:09 *** harshitaneja has quit IRC
07:44:06 *** harshitaneja has joined #gnucash
07:48:01 *** harshitaneja has quit IRC
07:49:16 *** gjanssens has joined #gnucash
07:49:16 *** ChanServ sets mode: +o gjanssens
07:49:40 <gjanssens> .
07:52:22 <moomba> Nevermind everyone, I figured it out thanks
07:52:28 *** moomba has left #gnucash
07:55:51 *** boldstripe has joined #gnucash
07:59:33 *** ncv__ has joined #gnucash
08:24:46 *** oozer has joined #gnucash
08:59:06 *** catalin has joined #gnucash
09:00:05 <catalin> On the budget planning, why if I set a negative value on an asset, is not substracted from Total?
09:01:32 <catalin> For example, on the next month I expect to sell an asset and transfer the money into another asset (like the Checking Account). When planning this in the budget, only the money decrease from an asset is not substracted from the total...
09:09:51 *** Jimraehl1 has joined #gnucash
09:10:54 *** Jimraehl1 has left #gnucash
09:21:10 *** kus has joined #gnucash
09:27:52 <CDB-Man> fell: thank you
10:05:07 *** fabior has joined #gnucash
10:11:54 *** frakturfreak has joined #gnucash
10:19:21 *** jotrago1 has quit IRC
10:21:26 *** frakturfreak has quit IRC
10:24:07 *** jotrago1 has joined #gnucash
10:42:20 *** Mechtilde has quit IRC
10:43:22 *** Mechtilde has joined #gnucash
10:49:22 <catalin> Regarding my previous question, I calculated something wrong. Everything works as expected.
10:49:28 *** catalin has quit IRC
11:17:24 *** boldstripe has quit IRC
11:19:20 *** boldstripe has joined #gnucash
11:36:26 <jralls> gjanssens: I discovered that the swig bug affects a lot of places that I can't easily work around. I
11:37:04 <jralls> gjanssens: Darn. I'm trying to build guile-2.2 to finesse the problem but it's not going too well.
11:57:46 <chris> I read guile is notorious to being very long to build... takes hours apparently
12:02:57 *** kus has quit IRC
12:05:11 <jralls> chris: Depends on the machine. It's a bit under an hour on mine. WebKit is much worse.
12:06:21 <gour> just wonder how much embedabble is Racket and then have some custom DSL in Racket to write custom reports
12:08:26 <chris> :-o nooo
12:08:27 <jralls> gour: We need more than just embeddable, we need SWIG to be able to write conversion wrappers.
12:08:46 <chris> gour: DSL can be done in any lisp-like
12:09:37 <gour> jralls: swig? hmm, that's not pretty then
12:10:16 <jralls> Indeed, but it's how we make the C library available to Scheme and Python.
12:10:21 <gour> chris: yeah, i just heard that racket is really powerful for that...was reading about Hackett language recently
12:11:06 <jralls> The alternative would be gobject-introspection. That would be a major course reversal.
12:11:41 <chris> I wish I could help with swig... I can't help feeling responsible for guile-2.0 headaches
12:12:25 <jralls> chris: Not your fault. gjanssens owned up to a bug he wrote in SWIG being the source of the problem.
12:13:05 <jralls> A guile-1.8 workaround got $ifdeffed to guile<=2.0 instead of guile<2.0.
12:13:19 <gour> in case it would be nice to get rid of swig deps
12:13:56 <jralls> gour: Pure C++, no Scheme, no Python bindings? I'd like that but you wouldn't.
12:13:56 *** harshitaneja has joined #gnucash
12:15:55 <chris> swig ----->----- RESTful api ;-)
12:16:29 <jralls> We might be able to get the Python bindings back with boost::python.
12:16:34 <jralls> chris: Ewww.
12:19:27 *** omidmnz has joined #gnucash
12:21:08 <jralls> GTG, back in ~4 hours.
12:21:13 *** jralls is now known as jralls_afk
12:21:57 <chris> gour how's lisp going
12:24:49 *** warlord has joined #gnucash
12:25:09 <omidmnz> Hi all. I have trouble bringing my data from GnuCash on Android to GnuCash Desktop. I have tried exporting to XML, which exports all transactions in the order they are entered, not the time and date. (At least that's what I see after I import the file.)
12:25:17 <omidmnz> I have also tried exporting the accounts in XML, and the transactions in QIF, which does export the transactions in date order, but after importing, they are not in order according to the clock time I have entered, which makes my log ugly, since my wallet for example, goes negative sometmies. (not after the day though)
12:25:56 <omidmnz> TL;DR: How can I export my transactions on mobile, so that they appear in order in date and time after I import on desktop.
12:30:47 *** vxp_ has quit IRC
12:50:03 *** kkknewton has quit IRC
12:52:02 *** fabior has quit IRC
12:58:22 *** omidmnz has quit IRC
12:59:36 *** omidmnz has joined #gnucash
13:21:28 <gour> chris: i've started with the Learn Schem in fixnum days, but was too busy with other non-computer-related things to go deeper...in any case, i consider it's worth to learn it, although probably learning Racket could be more generally useful
13:36:06 <gjanssens> jralls_afk: do you need guile 2.2 ? My bug already manifests with guile 2.0.
13:36:51 <gjanssens> Perhaps I should give it a name, like gnucash-dev-bleed or something... That would fit in the current trend of hyping bugs :)
13:38:54 <gjanssens> I you want to test whether my fix solves it, you can simply copy Lib/guile/guile_scm_run.swg over to your installed version of guile. It doesn't need recompilation of guile in itself
13:48:18 *** omidmnz has quit IRC
13:50:58 *** omidmnz has joined #gnucash
14:00:23 *** frakturfreak has joined #gnucash
14:03:38 *** omidmnz has quit IRC
14:31:07 *** vxp has joined #gnucash
15:21:21 *** gjanssens has quit IRC
15:23:28 *** gjanssens has joined #gnucash
15:23:28 *** ChanServ sets mode: +o gjanssens
15:50:04 *** gour has quit IRC
15:58:36 *** boldstripe has quit IRC
15:59:08 *** boldstripe has joined #gnucash
15:59:22 *** ncv__ has quit IRC
16:03:19 *** fabior has joined #gnucash
16:23:57 *** oozer has quit IRC
16:25:10 *** oozer has joined #gnucash
16:30:18 *** frakturfreak has quit IRC
16:35:45 *** Mechtilde has quit IRC
16:39:09 <jralls_afk> gjanssens: IIUC the bug is that scm_from_utf8_string is converted to scm_from_locale_string for guile <= 2.0. guile 2.2 > 2.0, so it shouldn't trigger the bug.
16:40:23 <gjanssens> jralls_afk: Unfortunately no. The check is only on SCM_MAJOR_VERSION
16:40:45 <gjanssens> So the check is really guile <= 2
16:41:32 <jralls_afk> gjanssens: Ah, I thought it required patching and rebuilding swig. We're currently using the MSYS2 swig so I went in the guile direction instead.
16:42:32 <gjanssens> The proper fix would indeed require patching and rebuilding swig, as the file I referred to is distributed with swig.
16:42:54 <gjanssens> But it's distributed unchanged from how it's in the swig sources.
16:43:42 <gjanssens> So for a test you can copy the fixed file over to the installed swig location
16:43:44 <jralls_afk> Oh. OK. Someone replied to me on #guile that they're working on getting 2.2 ready for Mingw, so better that I should use my time for something else.
16:45:10 <gjanssens> I also talked with the swig maintainers about a release time frame, but they couldn't give me any
16:45:38 <gjanssens> They're apparently working up to a 4.0 release and don't know whether a 3.0.x bugfix release will still happen.
16:45:54 <gjanssens> So on that front we'll have to be creative I'm afraid;
16:47:02 <jralls_afk> Sounds familiar...
16:48:10 <jralls_afk> Anyway, in guile_scm_run.swg, the change is just SCM_MAJOR_VERSION <= 2 to SCM_MAJOR_VERSION < 2, right?
16:50:05 *** jralls_afk is now known as jralls
16:50:40 <gjanssens> jralls: yes. I did clean it up a bit while I was at it but that's essentially it.
16:51:10 <jralls> Well, I just want to do the edit, build, and test. Pretty can wait a bit.
16:56:21 <gjanssens> I've been reading through the swig sources and it looks like we could provide our own guile_scm_run.swg
16:56:33 <gjanssens> so it would use ours instead of the one shipped with guile
16:57:07 <gjanssens> jralls: http://www.swig.org/Doc3.0/SWIGDocumentation.html#Library_nn2
16:57:33 <gjanssens> So if the my bugfix helps we could go that route
16:58:17 <jralls> Sounds like a better plan than patching the msys2 swig. In separate news, Alex reports that he's gotten webkit built for icu-61, so it's time to remove the "ignores" from pacman.conf.
17:13:29 <gjanssens> Ah good. I was just about to ask you which exact versions of ICU and boost you have on your box.
17:14:19 <gjanssens> I get cmake warnings about mismatches between my boost and its dependencies. I suspect this is related to boost, but cmake doesn't report additional detals
17:14:52 <gjanssens> Will a jhbuild run update the dependencies, or should I rerun setup-mingw64 for it ?
17:16:32 <jralls> Jhbuild will update the dependencies it builds, e.g. aqbanking or guile. setup-mingw64.ps1 and buildserver/build_package.ps1 will both upgrade mingw; setup-mingw64.ps1 will also regenerate jhbuildrc.
17:16:54 <gjanssens> Ok
17:20:43 *** fabior has quit IRC
17:21:57 <gjanssens> Hmm, I removed the Ignore lines from pacman, reran setup-mingw64.ps1 but I'm still seeing the cmake warnings related to boost and its dependencies.
17:24:34 <jralls> Right, the boost dependencies issue is another matter entirely. It has to do with the minimum version in CMakeList.txt being "too old" as far as FindBoost is concerned and it's warning that we might have problems if we don't update our code. It's completely ignorant of the fact that the bits of boost that we're using are very mature and not likely to have changed much.
17:24:47 <gjanssens> I simply reran the gnucash-git step (I was still in a jhbuild shell for it) after first choosing wipe directory and start over.
17:25:54 <gjanssens> ... and it passed the 30% (compilation of unittest-support.scm) so the updated dependencies solved that. Yay!
17:26:25 <gjanssens> I'll let it run for the night so I can hopefully make some progress during the next days...
17:26:36 <gjanssens> Now it's time to go to bed.
17:26:40 <gjanssens> See you later!
17:26:58 <gjanssens> Ok and ack on the boost explanation.
17:27:09 <jralls> G'night!
17:27:30 *** gjanssens has quit IRC
17:35:32 *** harshitaneja has quit IRC
17:54:29 <jralls> @tell gjanssens The swig fix fixes all of the remaining errors for 795134: expressions-2.0 and stylesheets-2.0 are where they belong.
17:54:29 <gncbot> jralls: The operation succeeded.
18:11:09 <jralls> chris: I think it also fixes 794965.
18:43:33 <jralls> @tell gjanssens Unfortunately copying guile_scm_run.swg to common (already on the include list) doesn't get it called. It's not %included anywhere so that paragraph in the docs apparently doesn't apply.
18:43:33 <gncbot> jralls: The operation succeeded.
18:50:27 *** pilotauto has joined #gnucash
19:08:27 <jralls> @tell gjanssens But undeffing scm_to_utf8_string and scm_from_utf8_string in guile-mappings.h does.
19:08:27 <gncbot> jralls: The operation succeeded.
19:32:36 *** Trel has joined #gnucash
21:52:37 *** tuxd00d has joined #gnucash
21:58:11 *** boldstripe has quit IRC
22:10:52 *** oozer has quit IRC
23:25:50 *** cyphase has quit IRC