2019-09-17 GnuCash IRC logs
00:02:30 *** tzhuang has quit IRC
00:05:36 *** tzhuang has joined #gnucash
00:27:30 *** omnireq_ has quit IRC
00:28:23 *** omnireq_ has joined #gnucash
00:29:25 *** JayC has joined #gnucash
00:29:25 *** ChanServ sets mode: +v JayC
00:32:42 *** Mechtilde has joined #gnucash
00:43:50 *** omnireq_ has quit IRC
00:46:22 *** omnireq has joined #gnucash
00:47:09 *** IslandAlpha has joined #gnucash
01:40:20 *** IslandAlpha has quit IRC
01:42:39 *** IslandAlpha has joined #gnucash
01:47:24 *** Mechtilde has quit IRC
01:48:37 *** fell has quit IRC
01:48:44 *** fell has joined #gnucash
01:48:45 *** ChanServ sets mode: +o fell
01:56:57 *** fell_laptop has joined #gnucash
01:56:57 *** ChanServ sets mode: +o fell_laptop
01:56:57 *** fell has quit IRC
01:58:23 *** fell has joined #gnucash
01:58:24 *** ChanServ sets mode: +o fell
02:00:59 *** fell_laptop has quit IRC
02:06:04 *** omnireq_ has joined #gnucash
02:07:17 *** omnireq has quit IRC
02:17:38 *** fabior has joined #gnucash
02:27:11 *** boldstripe has joined #gnucash
02:31:07 *** fabior has quit IRC
02:45:26 *** fell has quit IRC
02:45:41 *** Mechtilde has joined #gnucash
02:46:46 *** gour has joined #gnucash
02:46:46 *** ChanServ sets mode: +v gour
02:52:40 *** Aussie_matt has quit IRC
03:30:07 *** CarwynNelson has joined #gnucash
03:30:07 *** ChanServ sets mode: +v CarwynNelson
03:32:51 *** gjanssens has joined #gnucash
03:32:51 *** ChanServ sets mode: +o gjanssens
03:42:21 *** gour has quit IRC
03:44:20 *** gour has joined #gnucash
03:44:20 *** ChanServ sets mode: +v gour
03:50:16 *** boldstripe has quit IRC
04:00:05 *** CarwynNelson has quit IRC
04:03:12 *** CarwynNelson has joined #gnucash
04:03:12 *** ChanServ sets mode: +v CarwynNelson
04:20:47 *** IslandAlpha has quit IRC
04:27:09 *** CarwynNelson has quit IRC
04:32:13 *** IslandAlpha has joined #gnucash
04:49:56 *** CarwynNelson has joined #gnucash
04:49:56 *** ChanServ sets mode: +v CarwynNelson
04:52:57 *** CarwynNelson has quit IRC
04:53:10 *** CarwynNelson has joined #gnucash
04:53:10 *** ChanServ sets mode: +v CarwynNelson
04:55:28 *** fabior has joined #gnucash
05:01:40 <gjanssens> .
05:05:32 *** michielu has joined #gnucash
05:05:32 *** ChanServ sets mode: +v michielu
05:14:02 *** fabior has quit IRC
05:28:24 *** fabior has joined #gnucash
05:29:29 *** IslandAlpha has quit IRC
05:32:42 *** fabior has quit IRC
05:42:01 <michielu> Hi, I am encountering a bug in Ubuntu 16.04 and it seems to be related to https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=690270
05:42:01 <gncbot> michielu: Sent 17 hours and 45 minutes ago: <jralls> That's https://bugs.gnucash.org/show_bug.cgi?id=795014, there's a workaround in the comments.
05:43:02 <michielu> the workaround seems to be 'unset GTK_IM_MODULE' and that appears to work fine.
05:45:23 <michielu> apparently the X11-Input-Method (xim) is set as input method for Gnome by default. The linked page has further links to fixes in the network-manager-applet, which has a resolution stating 'It showed up that the client_window is NULL because nm-connection-editor handling of Tab key resulted in calling gtk_entry_unrealize() (which set client window to NULL) before the IM context machinery was done."
05:46:17 <michielu> Do I need to report this anywher?
05:48:24 *** storyjesse has joined #gnucash
05:55:26 *** oozer has joined #gnucash
05:55:48 <CarwynNelson> michielu: is this a bug you are encountering with the packaged version of gnucash on 16.04?
05:56:43 <michielu> as well. Also with builds from tarballs of 3.6, 3.7, git builds using tag 3.6, 2.8 and 3.7 and using the HEAD of maint
05:57:24 <michielu> they all crash when at some point I 'tab' from one field in the register overview to another and it seems to be related, according to the debian bug report, to the input-method resolution
05:57:40 <CarwynNelson> I'm not a contributor, but if it's happening on the version of gnucash available through the package manager then that sounds to me like it's something that needs to be reported.
05:58:08 <CarwynNelson> Is the 16.04 package maintained by the folk in this channel?
05:58:17 <CarwynNelson> Or rather, by the gnucash project?
05:58:38 <michielu> come to think of it, I am not sure about the packaged version. It was too old to read my file, so I had to update it. I guess I did not properly check the packaged 2.6.4 version
05:58:59 <michielu> or is it 2.4.6, I forgot
06:02:26 <CarwynNelson> My knowledge is limited, but my interpretation of what you were saying is that there is a bug in 16.04 (from debian?) that causes gnucash to crash, and that the workaround is at build level?
06:03:11 <CarwynNelson> If that's the case then it seems to me that the workaround needs to be included in the packaging
06:03:18 <CarwynNelson> but again, limited knowledge on this topic.
06:11:59 *** CarwynNelson has quit IRC
06:12:09 *** CarwynNelson has joined #gnucash
06:12:10 *** ChanServ sets mode: +v CarwynNelson
06:20:20 *** CarwynNelson has quit IRC
06:20:27 *** CarwynNelson has joined #gnucash
06:20:27 *** ChanServ sets mode: +v CarwynNelson
06:24:19 *** Mechtilde has quit IRC
06:39:56 *** Jimraehl1 has joined #gnucash
06:41:18 *** Jimraehl1 has left #gnucash
07:01:26 *** CarwynNelson has quit IRC
07:14:21 *** CDB-Man_ has joined #gnucash
07:14:22 *** ChanServ sets mode: +v CDB-Man_
07:15:04 *** oozer has quit IRC
07:15:08 *** michielu has left #gnucash
07:16:48 *** CDB-Man has quit IRC
07:26:51 *** boldstripe has joined #gnucash
07:38:37 <ElonSatoshi[m]> I have multiple unnamed budgets, now do I delete them?
07:39:57 *** CarwynNelson has joined #gnucash
07:39:57 *** ChanServ sets mode: +v CarwynNelson
07:43:30 <CarwynNelson> ElonSatoshi[m]: there should be a delete budget button, maybe under the edit menu?
07:43:34 <CarwynNelson> I don't have gnucash in front of me to check
07:46:00 <ElonSatoshi[m]> I just found out I can open each budget (Actions > Budget > Open budget) and there's a button to delete them in the row of buttons between the menu bar and the tabs.
07:46:35 <CarwynNelson> Are the docs at https://wiki.gnucash.org/docs/MAINT kept up to date / worth using as a reference for understanding how things work? I have been taking the sort of brute-force method at the moment of reading through the code, but it looks like that website would be a much better way of getting an overview.
07:46:45 <CarwynNelson> ElonSatoshi[m]: Glad you found it! :)
07:50:33 *** User__ has joined #gnucash
07:55:07 *** User__ has quit IRC
08:17:09 <chris> CarwynNelson: doxygen docs are worth keeping in mind, but reading code is better
08:21:11 <chris> plus, reading code mean you can spot bugs :)
08:23:26 <CarwynNelson> I guess I was more looking at it as a nice overview of what a subsystem is composed of. One of the things I was struggling with over the past few days was fitting what I learned into a slightly bigger picture
08:23:44 <CarwynNelson> So it's good to know that they are a decent source, but of course I won't give up reading the code :)
08:37:29 <sbluhm> Hello Gnucash Team, I am currently on the latest CentOS 7 and selfcompiled GnuCash 3.7 with AqBanking 5.99. Unfortunately, when I access the online funtionality, gnucash crashes with Memory Access Error. That's it. Can anyone assist me in debugging this? I am not really sure where to start
08:39:44 *** storyjesse has quit IRC
08:39:52 *** storyjesse has joined #gnucash
08:41:47 *** warlord has joined #gnucash
08:41:47 *** gncbot sets mode: +o warlord
08:42:38 <CarwynNelson> I can't promise much, and I am in work right now, but does it literally just say "Memory Access Error"?
08:43:22 <sbluhm> That was translated from German "Speicherzugriffsfehler (Speicherabzug geschrieben)
08:43:38 <warlord> .
08:44:56 <sbluhm> I will see if I can get this reproduced on an english environment
08:45:54 <CarwynNelson> It might be worth raising a bug for it too.
08:46:34 <CarwynNelson> If you could provide specifics like what versions of dependencies you have, what commands you used to build it, what exact buttons you pressed in the ui before the crash, any logs that gnucash generated
08:46:57 <CarwynNelson> that would likely be a big help to anyone trying to reproduce it on their end
08:47:29 <sbluhm> I would love to raise a abug. I am actually just not sure what to provide (apart from "Memory Access Error".
08:47:41 <sbluhm> let me see what I can do. Thanks
08:48:03 <CarwynNelson> also a commit hash of when you built from might also be helpful
08:48:04 <sbluhm> is there a way to get more output to at least identify where the crash occures? like a more verbose output
08:48:18 <sbluhm> ok, i can give an install script
08:50:05 <CarwynNelson> I've not really looked at gnucash logs before, but maybe this page might help? https://wiki.gnucash.org/wiki/Logging
08:50:27 <sbluhm> THANK YOU!
08:50:56 <CarwynNelson> lol, that's ok
09:05:27 *** fabior has joined #gnucash
09:22:46 *** fabior has quit IRC
09:27:09 *** fabior has joined #gnucash
09:28:03 *** fabior has quit IRC
09:58:17 <chris> is 'ninja test' an unusable relic from 'make test'?
09:59:40 *** oozer has joined #gnucash
09:59:47 <chris> lots of 'not run' and 'failed'
10:00:07 <chris> on the other hand, it shows tests in progress more interactively
10:05:02 <chris> after 'ninja check' which compiles dependencies, 'ninja test' now passes
10:07:07 <warlord> ninja is weird.
10:10:20 <CarwynNelson> is ninja weird in general? or is it just the ninja files that cmake outputs for gnucash?
10:20:16 *** omnireq_ has quit IRC
10:25:47 <warlord> Dunno. I've never actually used ninja. I've always been happy with make
10:25:56 <warlord> But I'm Old Skool
10:31:51 <CarwynNelson> I only started using it a few days ago, and it made my build faster so I was happy
10:32:03 <CarwynNelson> But i'm not really sure of the merits of it over standard make
10:32:30 <chris> multicore :)
10:33:01 <warlord> make -j can do multicore
10:33:31 <chris> oh
10:44:56 *** fell has joined #gnucash
10:44:56 *** ChanServ sets mode: +o fell
10:55:41 <sbluhm> i keep forgetting about make -j. Just always been too patient, I guess (or takes too long to type).
10:59:38 *** storyjesse1 has joined #gnucash
11:00:00 *** storyjesse has quit IRC
11:00:00 *** storyjesse1 is now known as storyjesse
11:03:29 <warlord> heh
11:09:54 *** CarwynNelson has quit IRC
11:13:34 <gjanssens> warlord, chris: make test would fail just like ninja test if you didn't run make check/ninja check beforehand
11:13:56 <gjanssens> It's not due to ninja vs make but how cmake configures them
11:14:17 <gjanssens> check and test are two different targets though I forgot the details
11:15:46 <gjanssens> As for performance, ninja appears to be slightly faster still even when using the same multicore settings on make
11:16:17 <gjanssens> It's nice though ninja does multicore by default
11:17:21 <gjanssens> sbluhm: for a crash it would be very helpful if you could also post a stack trace https://wiki.gnucash.org/wiki/Stack_Trace
11:17:44 <gjanssens> Please do so in a bug report, not here on irc
11:19:24 *** Mechtilde has joined #gnucash
11:20:30 <warlord> One would think that "test" should depend on "check"
11:22:59 *** omnireq has joined #gnucash
11:22:59 *** ChanServ sets mode: +v omnireq
11:26:14 *** PeterAxup has joined #gnucash
11:33:13 *** ChanServ sets mode: +v PeterAxup
11:35:14 <gjanssens> I'm sure there's a reason why the difference exists, I just don't know that reason :)
11:36:36 <chris> lots of these in code
11:36:55 <chris> gjanssens: if you have time, quick chat late here,
11:37:11 <chris> what's the status of the dangling PRs --
11:37:38 <chris> I merge safe ones (eg spelling mistakes)?
11:37:55 <chris> shall I kill deprecated scheme code?
11:38:01 <chris> (on master)
11:38:34 <gjanssens> chris: you can certainly merge the safe ones
11:39:55 <chris> ok
11:44:58 <gjanssens> as for deprecated scheme code, you can remove the deprecations that happened on current maint.
11:45:24 <gjanssens> The deprecations I did due to relocation of the source files should wait for 5.x as they only happened on master
11:46:45 *** fabior has joined #gnucash
11:50:09 *** fabior has quit IRC
11:53:59 <sbluhm> @gjanssens, sure thing. I will try to reproduce it on a clean machine and post all steps and traces.
11:53:59 <gncbot> sbluhm: Error: "gjanssens," is not a valid command.
11:54:13 <sbluhm> gjanssens, sure thing. I will try to reproduce it on a clean machine and post all steps and traces.
11:54:24 <gjanssens> tx
11:59:15 *** CarwynNelson has joined #gnucash
11:59:15 *** ChanServ sets mode: +v CarwynNelson
12:09:44 <CarwynNelson> I'm sure there was some documentation somehwhere about this, but what is a QofBook? I guess it would also be useful to know what libqof is
12:10:07 <CarwynNelson> or rather just qof, since it seems the libqof folder is now gone :)
12:11:39 <warlord> QOF == Query Object Framework. It's a generic infrastructure for queriable objects.
12:11:48 <warlord> A QofBook is a collection of them.
12:13:19 *** tzhuang has quit IRC
12:19:53 <CarwynNelson> cheers :)
12:27:36 *** storyjesse has quit IRC
12:32:44 *** PeterAxup has quit IRC
12:33:03 *** PeterAxup has joined #gnucash
12:33:07 <jralls> CarwynNelson: The issue michielu was complaining about already has multiple bug reports, two of which were referenced about 10 lines before you told him to file another one.
12:34:06 *** ChanServ sets mode: +v PeterAxup
12:34:57 *** PeterAxup has quit IRC
12:38:36 *** PeterAxup has joined #gnucash
12:39:13 *** ChanServ sets mode: +v PeterAxup
12:48:20 <PeterAxup> I'm new. How long does it take for the confirmation email to show up so that I can confirm my account?
13:04:04 <jralls> PeterAxup: What account would that be?
13:04:22 <PeterAxup> This chat account.
13:06:01 <jralls> I don't think there is any confirmation email. The fact that you have voice means that NickServ has authenticated you. Be sure to keep your password in your password manager, you'll need to re-identify with it from time to time
13:07:26 <CarwynNelson> jralls: apologies
13:07:39 <PeterAxup> NickServe keeps sending me this msg: 11:39:12 AM - NickServ: Password accepted - you are now recognized.
13:07:39 <PeterAxup> 11:39:13 AM - NickServ: Your email address is not confirmed. To confirm it, follow the instructions that were emailed to you.
13:07:39 <PeterAxup> 11:39:13 AM - NickServ: Your account will expire, if not confirmed, in 6 days, 22 hours, 54 minutes.
13:07:48 <CarwynNelson> Check your spam or ask nickserv to resend it
13:08:12 <CarwynNelson> Mine didn't go into spam, but I did need to ask nickserv to resend and as soon as I did it appeared in my inbox
13:08:29 <PeterAxup> How do I ask NickServ to resend. Nothing in spam box.
13:09:02 <CarwynNelson> maybe "/msg nickserv resend" might do the trick?
13:09:19 <CarwynNelson> if you msg nickserv with help it should list out all of the commands that it offers
13:10:18 <PeterAxup> Thank you--worked like a charm!
13:18:51 *** fabior has joined #gnucash
13:22:02 *** fabior has quit IRC
13:22:56 <CarwynNelson> Glad I could help
13:53:07 *** calvinct has joined #gnucash
13:56:58 <CarwynNelson> jralls: I am trying to compile your options branch, but it looks like test-trial-balance.scm hasn't been committed and cmake isn't happy about that. I'm sure I can get around it, but I just figured I would let you know
13:57:54 <jralls> CarwynNelson, yes, it's very much a work in progress. What I pushed yesterday didn't compile. I now have it compiling but the unit tests don't yet work.
13:58:14 <jralls> Aside from that are you able to make sense of the code?
13:58:55 *** PeterAxup has quit IRC
14:03:31 *** frakturfreak has joined #gnucash
14:03:31 *** ChanServ sets mode: +v frakturfreak
14:08:50 *** guak has joined #gnucash
14:11:19 <CarwynNelson> it took me a bit to wrap my head around the crtp pattern, but on the whole i think i understand what is the code is doing
14:13:32 <CarwynNelson> I might have a few questions but I didn't want to spam you in irc with them
14:15:26 <jralls> It's not spam, just be aware that it might yet blow up on me and need a restart. But it's good that it mostly makes sense. I worry about going too far down the template rabbit hole and winding up with code that no one (including me in 6 months) will be able to understand.
14:17:25 <CarwynNelson> That being said, what is the best/preferred medium for discussing work in progress like this?
14:18:42 *** Agfarmer18 has joined #gnucash
14:19:53 <jralls> If you have comments on code putting them inline on Github works really well.
14:21:11 <warlord> jralls, actually, I turned off moderation.
14:21:12 <jralls> Once the changes are ready to merge I'll make a PR for gjanssens (and anyone else who's so inclined) to look over.
14:22:39 <jralls> warlord: Huh. So PeterAxup didn't need to register after all. Still he showed in my client's user list as registered.
14:23:36 <jralls> CarwynNelson: But at this point I'm not even sure that the design direction is going to work out.
14:27:03 <CarwynNelson> is there anything I can do to help with that in any way? even as a sort of "outsider" sounding board.
14:28:07 <warlord> because he DID register, but no, he didn't have to.
14:28:53 <jralls> CarwynNelson: Well, does the approach look feasible to you so far?
14:36:36 <CarwynNelson> I'm not particularly well versed in what the options code has to stand up to / what its requirements are, so I would be curious to know. That being said, if it boils down to "the scheme code but implemented in c++" then so far it makes sense to me.
14:37:17 <CarwynNelson> I like that it's all typesafe too, but with a flexible inteface in the variant of all possible option types
14:37:26 *** calvinct has quit IRC
14:41:02 <jralls> Scheme is actually type-safe too, though it's less explicit about it. That can produce exceptions if one isn't careful with side effects. Of course Scheme is supposed to be functional without side-effects but that's not always feasible when blending Scheme and C.
14:44:18 <jralls> The existing option system has two separate roles: The obvious one is storing parameters about reports and the counters for generating invoice/bill numbers. The other is creating the GUI for setting those options from Scheme as part of generating the report.
14:47:09 <jralls> An extra wrinkle is that report options are saved as strings of scheme code instead of as key-value pairs. That's an obvious security hole.
14:50:38 *** jervin has joined #gnucash
14:50:46 *** KevinDB has quit IRC
14:52:03 *** KevinDB has joined #gnucash
14:52:03 *** ChanServ sets mode: +v KevinDB
14:53:57 <jralls> CarwynNelson: Just force-pushed an updated c++options. It now builds and the C++ unittests pass. Scheme doesn't work because gnc-make-text-option operates on optiondb but doesn't know how to attach a widget yet.
14:54:52 <jralls> Back in a couple of hours.
15:01:04 *** User__ has joined #gnucash
15:01:32 *** Mechtilde has quit IRC
15:03:09 *** calvinct has joined #gnucash
15:10:50 *** Mechtilde has joined #gnucash
15:13:52 *** Mechtilde has quit IRC
15:14:12 *** User__ has quit IRC
15:16:35 <fell> sbluhm: Beobachte bitte die Entwicklung auf https://lists.aquamaniac.de/listinfo/aqbanking-user
15:18:53 <gjanssens> fell: my change to the gnucash appdata file earlier today seems to have fixed the flathub publishing step. We now have version detatils on the flathub page :)
15:19:51 <fell> CarwynNelson: Users with AqBanking 5.99 should watch https://lists.aquamaniac.de/listinfo/aqbanking-user and report errors there.
15:20:47 <fell> Hoi, Geert!
15:24:40 <fell> While the mandatory changes of FinTS 4.1 are in aqbanking 4.8, we got all surprised by the optional changes and had no opportunity to test them before they were applied by the german banks.
15:25:37 <fell> There testing happens now in AqBanking 5.99 and users should report there.
15:29:50 *** Mechtilde has joined #gnucash
15:32:49 *** Mechtilde has quit IRC
15:33:17 <CarwynNelson> not that i use it, but just out of curiosity, will the latest aqbanking version be backported to older versions of gnucash?
15:42:35 <fell> CarwynNelson: Because of the complexity Martin is not planing to backport the changes in AqB 4.x. So we are awayting Aqb 6.x soon.
15:43:34 <fell> That again seems to have changes, which now cause Gnucash crashes.
15:44:30 <fell> I am hoping cstim will fix them soon, too.
15:45:09 <fell> In short, I fear, there will be no backports.
15:45:37 *** oozer has quit IRC
15:47:04 *** Mechtilde has joined #gnucash
15:48:11 <fell> ganssens: Yes, it looks nice now. And I assume it is not only an improvement @flathub.
15:54:57 <gjanssens> fell: no, we're now properly adhering to the appdata spec so possibly other tools that consume appdata may like our appdata file better now.
15:55:55 <gjanssens> fell: is our current maint branch not aqbanking 6 ready ? Or were these changes only done on master ?
15:56:12 <fell> I was just wondering, where to link https://www.freedesktop.org/software/appstream/docs/chap-Metadata.html
15:56:42 * gjanssens hasn't followed that aqbanking 6 bit too closely as he's not using aqbanking at all
15:57:48 *** Mechtilde has quit IRC
15:58:07 <fell> There was a mail by cstim some time ago, complaining that also master was not adjusted for recent aqb versions
15:58:31 <fell> Then he started the adjustments.
15:58:55 <gjanssens> Oh ok.
15:59:27 * gjanssens is going afk for the night...
15:59:32 <gjanssens> See you all later
16:00:13 *** gjanssens has quit IRC
16:01:03 *** oozer has joined #gnucash
16:04:59 *** kael has joined #gnucash
16:04:59 *** ChanServ sets mode: +v kael
16:05:42 <fell> ... and finally the next gnucash release (maint) should be able to use aqb6 without crashing.
16:06:19 *** calvinct has quit IRC
16:13:00 *** kael has quit IRC
16:21:06 <jralls> fell: I understood from cstim
16:21:57 <jralls> 's email yesterday that he's waiting for aqb6 to stabilize a bit before he takes on the GnuCash changes.
16:22:41 <jralls> His other changes will be in master now, maint hadn't merged up until the 3.7 release.
16:30:03 *** tzhuang has joined #gnucash
16:35:50 *** JayC has quit IRC
16:37:30 *** jervin has quit IRC
16:40:18 *** Agfarmer18 has quit IRC
16:44:50 <fell> jralls, yes it is too early now. aqbanking-user got this month more traffic than the whole year before. But he started already: https://github.com/Gnucash/gnucash/commit/50e59077d7160ba2d9eb1cb587c3b0c5597fee31
16:48:54 <jralls> That's just a spelling change. OTOH it's poor practice on Martin's part to be changing API this late in the dev cycle.
16:50:31 <fell> I would say, it was bad practise of Bafin, Bundestag, Kreditwirtschaft to set such a short frame whithout opportunity for tewsting.
16:51:38 <jralls> That's a separate issue, but I bet the commercial software devs had plenty of notice and testing opportunities including sandboxes.
16:52:07 <jralls> But Martin's not allowed in their clubhouse so he had no way of knowing.
16:53:44 *** guak has quit IRC
16:54:14 <fell> Not sure, at least BAFin lifted the obligatory date recenly.
16:54:55 <fell> They saw already the banks would not fulfill the requirements.
16:56:05 <fell> The good new: It seems Martin got a few new contributors.
16:59:05 <jralls> Now he's gone and p**d me off: https://github.com/Gnucash/gnucash/commit/ccaaa14187d5aa37e8ed24669fbba6ad055f9df7
16:59:41 <jralls> Were there any reports about cmake failures on the de list?
17:03:52 <fell> I would assume on aqbanking-user, where I still have many unread mails, thanks to telekoms need of 4 days to repair my DSL.
17:10:58 *** frakturfreak has quit IRC
17:11:55 <fell> jralls, I asked him to appear here - more often.
17:13:48 <jralls> Ok. I'm not sure that will make a difference.
17:15:05 <fell> You can find a german overview about the PSD2-Chaos at https://www.finanz-szene.de/tag/psd2/
17:17:04 <CarwynNelson> Is there a difference between the PSD2 api defined by the german regulator, and PSD2 in the UK?
17:24:07 *** gour has quit IRC
17:26:42 *** PeterAxup has joined #gnucash
17:26:42 *** ChanServ sets mode: +v PeterAxup
17:27:03 <jralls> CarwynNelson: There's the EU's PSD2 regulation that's the same throughout the EU and there's the effect of PSD2 on banking software. Germany has its own national-standard banking protocol called FinTS and it's the changes in that brought about by the PSD2 regulation taking effect that are causing trouble to German GnuCash users.
17:27:45 *** PeterAxup1 has joined #gnucash
17:27:45 *** ChanServ sets mode: +v PeterAxup1
17:29:01 *** PeterAxup1 has quit IRC
17:30:36 *** PeterAxup has quit IRC
17:30:56 <fell> CarwynNelson: the german "online banking" started already 1976 with 9-track tapes and got continous evolution.
17:37:12 <CarwynNelson> That's really cool! From what I can tell PSD2 is a big step forward in the UK. Unfortunately it seems that banks only have an obligation to allow third parties access to your data via the APIs so there is no guarantee that a consumer using something like gnucash would be able to connect to their bank.
17:44:51 *** warlord has quit IRC
17:48:30 <CarwynNelson> jralls: i'm using gcc 7.4 to compile gnucash and it compiles maint just fine, but it's falling over on the options code in your branch. Has the minimum gcc version requirement gone up?
17:50:49 <jralls> CarwynNelson: Not yet. We're planning to go for C++17 but for now it's C++11. Did you get a fresh force pull from my push 3 hours ago? Yesterday's version had a bunch of build issues.
17:54:29 <CarwynNelson> Yeah, I actually did a clean checkout like 30 minutes ago
17:59:05 <CarwynNelson> gotta go to bed, catch up tomorrow. Thanks for all the help today :)
17:59:17 *** CarwynNelson has quit IRC
18:10:34 <fell> jralls: cstims commit is the follow-up of André Colomb's aqbanking mail from 17 Sep 2019 19:47:31 +0200
18:12:16 <jralls> fell: Can you give an archive pointer?
18:12:34 <fell> The archive is not public
18:12:56 <fell> Re: [Aqbanking-user] Erfahrungsbericht PSD2 mit GnuCash 3.7
18:14:21 <jralls> Ok. Does Colomb provide any technical detail about why the cmake function test failed?
18:15:37 <fell> Change Dir:
18:15:38 <fell> /usr/local/src/gnucash-3.7/build-gnucash-3.7_amc/CMakeFiles/CMakeTmp
18:15:40 <fell> Run Build Command:"/usr/bin/ninja" "cmTC_81a88"
18:15:41 <fell> [1/2] Building C object CMakeFiles/cmTC_81a88.dir/CheckSymbolExists.c.o
18:15:43 <fell> [2/2] Linking C executable cmTC_81a88
18:15:44 <fell> FAILED: cmTC_81a88
18:16:13 <fell> : && /usr/bin/cc CMakeFiles/cmTC_81a88.dir/CheckSymbolExists.c.o -o
18:16:14 <fell> cmTC_81a88 -laqbanking && :
18:16:16 <fell> CMakeFiles/cmTC_81a88.dir/CheckSymbolExists.c.o: In function `main':
18:16:17 <fell> CheckSymbolExists.c:(.text+0x1b): undefined reference to
18:16:19 <fell> `AB_Banking_RuntimeConfig_SetCharValue'
18:16:20 <fell> collect2: error: ld returned 1 exit status
18:16:22 <fell> ninja: build stopped: subcommand failed.
18:16:23 <fell> F
18:16:41 <fell> File
18:16:43 <fell> /usr/local/src/gnucash-3.7/build-gnucash-3.7_amc/CMakeFiles/CMakeTmp/CheckSymbolExists.c:
18:16:44 <fell> /* */
18:16:46 <fell> #include <aqbanking/banking.h>
18:16:47 <fell> int main(int argc, char** argv)
18:16:49 <fell> {
18:16:50 <fell> (void)argv;
18:16:52 <fell> #ifndef AB_Banking_RuntimeConfig_SetCharValue
18:16:53 <fell> return ((int*)(&AB_Banking_RuntimeConfig_SetCharValue))[argc];
18:16:55 <fell> #else
18:16:56 <fell> (void)argc;
18:16:58 <fell> return 0;
18:16:59 <fell> #endif
18:17:01 <fell> }
18:24:22 <fell> #ifndef AB... return AB_
18:24:35 *** mohave has joined #gnucash
18:24:40 <fell> Shouldn't it be the other way?
18:24:45 *** mohave_ has joined #gnucash
18:25:46 <fell> jralls ^
18:25:58 <jralls> No, it's to protect agains a macro.
18:27:44 <mohave_> I compiled gnucash on MacOS about a year ago and am now at it again. Thought I'd make sure the dependencies were up to date first, so tried "unbuild update". This resulted in the following error: jhbuild update: failed to parse /Users/chris/Source/jhbuild/modulesets/https:/gitlab.gnome.org/GNOME/gtk-osx/raw/master/modulesets-stable/gtk-osx.modules: [Errno 2] No such file or directory: u'/Users/chris/Source/jhbuild/modulesets/https:/g
18:27:44 <mohave_> itlab.gnome.org/GNOME/gtk-osx/raw/master/modulesets-stable/gtk-osx.modules'
18:27:55 <mohave_> Any ideas on how to proceed?
18:28:54 <jralls> What the heck is "unbuild update"?
18:28:56 <mohave_> s/unbuild/jhbuild
18:29:24 <mohave_> wonderful spellcheck helping out
18:29:41 <jralls> :-) Are you on MacOS?
18:29:53 <mohave_> yes
18:31:50 <jralls> I don't know that jhbuild update works with gtk-osx. I've never tried it.
18:32:11 <jralls> Have you migrated jhbuild to the new pyenv version?
18:32:59 <mohave_> no, if it's newer than September of last year
18:33:25 <mohave_> maybe I should just start fresh
18:33:42 <jralls> You'll need to do that, it's necessary to build with meson and several of the Gnome modules have switched to that.
18:34:55 <mohave_> Thanks John, think I will delete all and start fresh.
18:35:07 <jralls> Note that the new script is gtk-osx-setup.sh instead of gtk-osx-setup-build.sh and that it puts everything in ~/.new_local by default. That's overridable with environment variables you'll find near the top of gtk-osx-setup.sh
18:35:55 <mohave_> thanks for the tip
18:38:01 <jralls> fell: I just did a fresh pull of gwen and aqb master, built and installed, then ran cmake from before cstim's change. The AB_Banking_RuntimeConfig_SetCharValue test works fine.
18:43:00 <jralls> Since Colomb's got a linker error I suspect that he failed to correctly install libaqbanking so that the header matched but the library was missing the symbol. Did cstim follow up on that by asking him to `nm /path/to/libaqbanking.so | grep AB_Banking_RuntimeConfig_SetCharValue`?
18:55:11 *** Aussie_matt has joined #gnucash
19:07:12 *** omnireq has quit IRC
19:17:53 *** mohave has quit IRC
19:17:53 *** mohave_ has quit IRC
19:24:39 <jralls> @tell CarwynNelson I'll be AFK today, but I built c++options on ArchLinux. There were some missing includes and a missing swig decl that clang didn't care about and I hadn't included PkgConfig::GTK3 in the test program specs, again something clang didn't care about.
19:24:39 <gncbot> jralls: The operation succeeded.
19:27:01 <jralls> @tell CarwynNelson So it builds but the tests fail because there's something wrong with the ArchLinux Googletest: It doesn't run SetUpTestSuite so gtk_init() isn't getting called.
19:27:01 <gncbot> jralls: The operation succeeded.
19:29:59 *** mohave has joined #gnucash
19:30:13 *** mohave_ has joined #gnucash
19:42:38 *** mohave_ has quit IRC
19:42:38 *** mohave has quit IRC
19:51:06 <fell> jralls: I sent your answer to aqbanking-user.
20:12:15 *** oozer has quit IRC
20:16:59 *** fell has quit IRC
20:17:03 *** fell_laptop has joined #gnucash
20:17:04 *** ChanServ sets mode: +o fell_laptop
20:17:09 *** fell_laptop is now known as fell
20:17:14 <fell> in another branch of the thread, André watches, the file gschemas.compiled (and a few directories) are created with a wrong umask, leading to a crash. He suspects Cmake.
20:17:23 *** omnireq has joined #gnucash
20:29:32 <fell> jralls, Andrè suspects
20:29:34 <fell> CMakeLists.txt:387:
20:29:35 <fell> set(CMAKE_REQUIRED_LIBRARIES "-laqbanking")
20:30:46 <fell> because of the missing -L Flags
20:38:28 <jralls> fell: Does he have libaqb installed somewhere not on CMAKE_PREFIX_PATH?
20:38:49 <jralls> -L shouldn't normally be necessary.
20:39:43 <fell> Yes, check is finding aqb5, but gnucash finds aqb6
20:41:02 <fell> It seems he has his distri's aqb5 and the self built aq6
21:27:05 *** KevinDB has quit IRC
21:37:31 *** warlord has joined #gnucash
21:37:31 *** gncbot sets mode: +o warlord
21:40:27 *** warlord has quit IRC
21:40:30 *** warlord has joined #gnucash
21:40:30 *** gncbot sets mode: +o warlord
21:46:47 *** badger92 has joined #gnucash
21:48:42 *** warlord has quit IRC
21:48:52 *** warlord has joined #gnucash
21:48:52 *** gncbot sets mode: +o warlord
21:54:27 *** warlord has quit IRC
22:01:13 *** warlord has joined #gnucash
22:01:13 *** gncbot sets mode: +o warlord
22:04:14 *** fell has quit IRC
22:04:17 *** fell_laptop has joined #gnucash
22:04:18 *** ChanServ sets mode: +o fell_laptop
22:12:57 *** warlord has quit IRC
22:13:54 *** warlord has joined #gnucash
22:13:54 *** gncbot sets mode: +o warlord
22:15:27 *** fell_laptop has quit IRC
22:15:30 *** fell has joined #gnucash
22:15:30 *** ChanServ sets mode: +o fell
22:22:00 *** KevinDB has joined #gnucash
22:22:00 *** ChanServ sets mode: +v KevinDB
22:31:07 <jralls> fell: Got it, makes sense as no distro will have aqb6.
22:43:39 <fell> ... for now
23:11:13 *** fell has quit IRC
23:12:57 *** warlord has quit IRC
23:14:22 *** fell has joined #gnucash
23:14:22 *** ChanServ sets mode: +o fell
23:15:09 *** warlord has joined #gnucash
23:15:09 *** gncbot sets mode: +o warlord
23:22:22 *** warlord has quit IRC
23:22:26 *** warlord has joined #gnucash
23:22:26 *** gncbot sets mode: +o warlord
23:58:35 *** omnireq_ has joined #gnucash
23:59:47 *** omnireq has quit IRC