2019-04-19 GnuCash IRC logs

00:07:22 *** boldstripe has quit IRC
00:08:18 *** boldstripe has joined #gnucash
00:53:06 *** kapil___ has quit IRC
01:49:26 *** nicoroumeau has joined #gnucash
01:55:38 *** tienne has joined #gnucash
02:00:29 *** fell has quit IRC
02:01:48 *** fell has joined #gnucash
02:01:48 *** ChanServ sets mode: +o fell
02:03:37 *** nicoroumeau has quit IRC
02:08:03 *** boldstripe has quit IRC
02:08:58 *** boldstripe has joined #gnucash
02:32:28 *** Aussie_matt has quit IRC
02:32:47 *** Mechtilde has joined #gnucash
02:33:21 *** Aussie_matt has joined #gnucash
02:34:09 *** gjanssens has joined #gnucash
02:34:09 *** ChanServ sets mode: +o gjanssens
02:41:07 *** Mechtilde has quit IRC
02:41:28 *** Mechtilde has joined #gnucash
03:17:49 *** Aussie_matt has quit IRC
03:19:16 *** Aussie_matt has joined #gnucash
03:25:17 *** nimish2711 has quit IRC
03:26:52 *** tienne has quit IRC
03:44:55 *** Aussie_matt has quit IRC
03:49:46 *** Aussie_matt has joined #gnucash
03:53:54 *** Aussie_matt has quit IRC
04:08:43 *** boldstripe has quit IRC
04:08:51 *** Mechtilde has quit IRC
04:09:38 *** boldstripe has joined #gnucash
04:12:03 *** Mechtilde has joined #gnucash
04:13:57 *** nimish2711 has joined #gnucash
04:19:52 *** kapil___ has joined #gnucash
04:21:46 *** Mechtilde has quit IRC
04:23:45 *** nimish2711 has quit IRC
04:29:47 *** jcamos has joined #gnucash
04:34:20 *** fabior has joined #gnucash
04:36:15 *** nimish2711 has joined #gnucash
04:41:56 *** tienne has joined #gnucash
04:54:52 *** tienne has quit IRC
04:58:59 *** bertbob has quit IRC
05:14:53 *** bertbob has joined #gnucash
05:14:54 *** ChanServ sets mode: +v bertbob
05:22:49 *** ChanServ sets mode: +v jcamos
05:24:30 *** marusich has quit IRC
05:32:49 *** jcamos has quit IRC
05:32:59 *** jcamos has joined #gnucash
05:33:27 *** jcamos has quit IRC
05:33:34 *** jcamos has joined #gnucash
05:33:34 *** ChanServ sets mode: +v jcamos
05:44:25 <chris> jralls if I apply monkey-patch it fixes the TYR unicode display on reports, but messes up serialization to/from saved-reports.
05:44:34 <chris> I can understand the frustration it causes.
05:45:05 <chris> I wonder if it means, upgrading the Windows build to guile-2.2 causes similar issues. I suspect it will.
05:45:20 <chris> If it will, it would be useful to start fixing serialization now.
05:45:34 * chris cannot generate windows build yet
05:50:57 *** tienne has joined #gnucash
05:51:11 <chris> I think that the patch to string-ports functions is a cleaner fix than hunting for individual functions and converting to non-string-ports...
05:56:23 *** luwum[m] has quit IRC
05:56:24 *** mmkodali[m] has quit IRC
05:56:24 *** peter-butler[m] has quit IRC
05:56:34 *** Couto[m] has quit IRC
06:02:13 *** Jimraehl1 has joined #gnucash
06:02:48 *** Jimraehl1 has left #gnucash
06:06:06 *** Couto[m] has joined #gnucash
06:09:23 *** boldstripe has quit IRC
06:10:19 *** boldstripe has joined #gnucash
06:13:06 *** jcamos has quit IRC
06:23:21 *** Yotson has quit IRC
06:24:37 *** oozer has joined #gnucash
06:24:40 *** frakturfreak has joined #gnucash
06:24:40 *** ChanServ sets mode: +v frakturfreak
06:24:41 <chris> jralls with the patch the code can display TRY symbol, saves unicode into book.gcm and saved-reports-2.8
06:25:10 <chris> reload from book.gcm munges unicode again however loading from saved-reports-2.8 is fine.
06:25:17 <chris> interesting.
06:25:57 * chris would be grateful for an idiot-proof windows build generator.
06:33:46 <chris> jralls/gjanssens: gnc-plugin-page-report.c:935 will call eval_string(option_string) directly and I suspect something like scm_from_utf8_string is required to understand unicode from SchemeOptions in book.gcm
06:39:21 <chris> overall IMO the monkey-patch *is* a positive step to fix this
06:44:16 *** frakturfreak has quit IRC
06:44:56 *** Yotson has joined #gnucash
06:52:29 <fell> Chris, what is wrong with Glade?
06:54:13 <fell> or are you talking abut the OS windowsß
06:54:21 <fell> ?
06:55:59 *** omnireq has quit IRC
06:57:12 *** tienne has quit IRC
07:00:55 *** nimish2711 has quit IRC
07:24:11 <chris> trying to debug 796728
07:28:01 *** jcamos has joined #gnucash
07:28:01 *** ChanServ sets mode: +v jcamos
07:30:51 *** trying2Simplify has joined #gnucash
07:37:29 <gjanssens> .
07:37:29 <gncbot> gjanssens: Sent 1 week, 3 days, 20 hours, and 6 minutes ago: <fell> I think better than editing the old new would be to send a supplement of the release announcement .
07:37:30 <gncbot> gjanssens: Sent 1 week, 3 days, 17 hours, and 55 minutes ago: <jralls> fell I agree. I've also fixed the problem on Win32 with the FQ updater CMD script and I'll build gnucash-3.5-1.setup.exe today, then do a supplemental announcement that includes the bug closures.
07:37:31 <gncbot> gjanssens: Sent 1 week, 2 days, 19 hours, and 46 minutes ago: <warlord> FYI, https://bugs.gnucash.org/show_bug.cgi?id=797174 is asking for Tags . I think we already have an open RFE.
07:37:32 <gncbot> gjanssens: Sent 1 week, 2 days, 19 hours, and 46 minutes ago: <warlord> oh, looks like jralls already noticed that. :)
07:37:33 <gncbot> gjanssens: Sent 1 week, 0 days, 20 hours, and 12 minutes ago: <chris> I still think we should modularise transaction.scm so that standard-reports can be loaded with (load) calls. let me know what you think.
07:39:06 <gjanssens> chris: what's monkey-patch ?
07:39:44 * gjanssens is a bit out of context having been away for more than a week
07:39:51 <gjanssens> The project has been busy it seems
07:39:58 <chris> gjanssens long story
07:40:18 <chris> issue is Turkish lira is selectively displaying as ? in report
07:40:34 <chris> this is extended-unicode symbol
07:41:02 <chris> but is only occasionally being mangled by reports
07:41:31 <chris> I've debugged to 2 functions: (format #f "~a bla" str) and (with-output-to-string) both use string-ports to build string
07:42:28 <chris> if I remove format, and use another function, the unicode is displayed correctly
07:42:47 <chris> conclusion: string-ports mangle unicode in guile-2.0
07:42:51 <chris> I've confirmed with guile-user
07:43:07 <chris> see thread in http://lists.gnu.org/archive/html/guile-user/2019-04/msg00017.html
07:43:52 <chris> there's a patch being suggested which 'upgrade' guile-2.0 to handle strings similarly to guile-2.2
07:43:56 <chris> see last email
07:44:19 <chris> this *does* fix unicode-Turkish-lira showing in reports
07:44:23 <chris> but has other side effects
07:44:45 <chris> that helps?
07:47:47 <chris> perhaps better summarize all findings in bugzilla
07:52:40 *** trying2Simplify has quit IRC
08:03:08 *** nimish2711 has joined #gnucash
08:07:27 *** kapil___ has quit IRC
08:22:06 <gjanssens> chris: thanks for the recap
08:22:21 <gjanssens> I think it would be helpful to add this info in the bug as well indeed
08:23:20 <chris> Ok I will, however
08:23:47 <chris> there are separate issues at play
08:24:58 <chris> some bugs are about loading string from slot
08:25:14 <chris> some bugs are about loading string from slot eg 797127
08:25:49 <chris> https://bugs.gnucash.org/show_bug.cgi?id=797065 is a different one i think
08:26:06 <chris> so perhaps i deserve my own bug :)
08:28:43 <gjanssens> :)
08:29:12 <gjanssens> And so what's the story behind modularizing transaction.scm ?
08:34:42 *** jervin has joined #gnucash
08:40:04 *** luwum[m] has joined #gnucash
08:40:08 *** mmkodali[m] has joined #gnucash
08:40:09 *** MatrixTraveler[m] has joined #gnucash
08:40:12 *** peter-butler[m] has joined #gnucash
09:17:52 *** jcamos has quit IRC
09:23:29 *** jervin has quit IRC
09:23:50 *** jervin has joined #gnucash
09:24:09 <warlord> gjanssens, welcome back
09:34:04 <gjanssens> thanks warlord
09:34:20 *** jervin has quit IRC
09:45:18 *** Mechtilde has joined #gnucash
10:05:23 <chris> gjanssens from your efforts to allow (load) reports... I tried to modularize but quite difficult to do it well
10:07:50 *** nimish2711 has quit IRC
10:07:55 <fell> Have Windows and MacOS stderr and stdout?
10:08:41 <fell> from gnucash/gnucash-bin.c:125
10:09:16 *** nimish2711 has joined #gnucash
10:12:52 <gjanssens> fell: I believe MacOS does, but I'm not sure it also would work on Windows
10:13:17 <fell> Hi, Geert! Nice you are back!
10:13:44 <gjanssens> chris: oh, was that a reference to the load issues I had and in which you suggested transaction report is really a reporting engine?
10:14:11 <chris> yes I tried to modularise the options-generator and the renderer
10:14:18 <chris> but failed
10:14:27 <gjanssens> Hi fell, I'm trying to catch up, but there's over 400 mails waiting for me in gnucash-user...
10:14:54 <gjanssens> chris: was it not possible to simply put these two functions in a separate file ?
10:15:16 <chris> no they need to import modules
10:15:56 <gjanssens> well, I meant in an independent module in a separate file...
10:16:11 <gjanssens> that way they should be able to import modules themselves, no ?
10:16:40 <chris> well yeah that was the attempt... but standard-reports/* are imported using resolve-module which seems to offer a different environment to the reports
10:17:05 <chris> anyway i didn't look too much into it. i can try again if we think modularizing standard-reports so that they can be (load)ed is worthwhile
10:17:41 <chris> (aside I'd be grateful if the build/compile/run script could be shared... I can build but the charts are failing to find jqplot.js)
10:17:51 <gjanssens> ah, that's probably a matter of telling the build system to create links for the new module
10:18:41 <gjanssens> as for the build/compile/run script, I don't have one
10:19:13 <gjanssens> I create a build directory somewhere (outside of the source tree in my case)
10:19:35 <gjanssens> Inside that build directory I run cmake with a few tweaks (to enable python and debug mode)
10:19:50 <gjanssens> (and ninja as build system)
10:20:01 <gjanssens> Then ninja-build && ninja-build install
10:20:19 <gjanssens> Oh, I do set an install prefix in the cmake command line as not not get the default /usr/local
10:20:35 <gjanssens> To run I switch to the installation directory and then type ./bin/gnucash
10:21:11 <gjanssens> In theory the install step shouldn't be necessary, but as it only takes 1s on my nvme ssd, I have kept the habit
10:21:16 <chris> well since my partition-overwrite incident i haven't been able to install
10:21:19 <chris> output: https://pastebin.com/raw/ixXyERsq
10:22:13 <gjanssens> Is that from a clean build directory ?
10:22:29 *** nimish2711 has quit IRC
10:22:33 <chris> yes
10:23:03 <gjanssens> Which exact commands did you use to run cmake, build and install ?
10:23:26 <chris> https://pastebin.com/raw/YQvC572h
10:25:59 <warlord> Hmm, my ovirt server is getting to 64% storage usage...
10:27:48 <gjanssens> chris: you're not supposed to set the install prefix the same as your build directory
10:28:34 *** nimish2711 has joined #gnucash
10:28:46 <gjanssens> also I can't tell if /home/chris/sources/gnucash/maint exists
10:29:00 <chris> ^ thats the source
10:29:13 <gjanssens> I assume it does though, other wise that cmake invocation would already have failed
10:29:59 * chris knows nothing other than (parens)...
10:30:57 <gjanssens> fwiw on my current report cleanup branch the charts fail to render when running from the build directory, while they render properly when running from the installation prefix
10:31:27 <gjanssens> So it seems our build scripts have a bug causing the fake install in the build directory to be incomplete or incorrectly set up
10:32:00 <gjanssens> Try rerunning the cmake command with a different install prefix, then add a ninja-install to your command list and run from the prefix
10:32:04 <gjanssens> That may fix it.
10:32:38 <gjanssens> (Different from your build directory, so for example /home/chris/sources/gnucash/install)
10:34:38 <gjanssens> Here's an updated version of your scripts: https://pastebin.com/raw/rhMCMyQd
10:36:06 <gjanssens> sorry a few more tweaks in here: https://pastebin.com/raw/LqYMKG52
10:36:25 <gjanssens> (remove the install directory as well, and effectively run from the install directory)
10:39:11 <gjanssens> gtg...
10:49:43 <chris> cool that works thanks
10:51:10 *** nimish2711 has quit IRC
10:52:00 *** nimish2711 has joined #gnucash
10:52:35 *** omnireq has joined #gnucash
10:52:35 *** ChanServ sets mode: +v omnireq
11:01:33 *** ArtGravity has joined #gnucash
11:01:33 *** ChanServ sets mode: +v ArtGravity
11:02:31 <chris> part of my current effort is to tweak charts so that they import jqplot into headers rather than body, which means there will be no need to substring/replace the duplicate jqplot.js in report.scm
11:02:57 <chris> the substring-replace function in utilities.scm is btw bad and buggy
11:04:22 <chris> *gnc:substring-replace-from-to
11:24:40 *** Mechtilde has quit IRC
11:36:19 *** guak has joined #gnucash
11:38:19 *** ctjctj has joined #gnucash
11:40:58 *** chris has quit IRC
11:42:57 *** ChanServ sets mode: +v ctjctj
11:44:03 <ctjctj> I'm using GnuCash 3.3. When I have a printable invoice showing and click on the "export to PDF" it takes me to the print page. Last month it was doing an export. What should I do/change/check in order to get the old "export" behavior back. (sorry if this multiposts, I had to register to get voice)
11:45:55 *** Mechtilde has joined #gnucash
11:53:02 *** Mechtilde has quit IRC
11:53:39 *** Mechtilde has joined #gnucash
12:15:33 *** Mechtilde has quit IRC
12:17:27 *** Mechtilde has joined #gnucash
12:30:47 *** nicoroumeau has joined #gnucash
12:32:42 <fell> ctjctj: your OSß Same version last month?
12:33:10 <ctjctj> I can't tell. It looks like I got an upgrade.
12:33:38 <ctjctj> Rephrase: I have no idea what last months version was. I do automatic updates via ubuntu. This looks like an updated version.
12:34:01 *** nicoroumeau has quit IRC
12:34:43 <fell> And in the Print dialog is no "To file: ...pdf"?
12:36:14 <ctjctj> It does have a print to pdf.
12:36:46 <fell> Does that work for you?
12:38:01 <ctjctj> I consider that a "work around" With "export to pdf" i got a save as pdf option with a good file name that went to my "invoices" folder. With print to file I get "output.pdf"
12:38:21 <ctjctj> Old "export to pdf" gave me a save dialog with a good file name.
12:38:47 <ctjctj> Right now, clicking on "export to pdf" gives me a print dialog.
12:39:51 <fell> I believe, you have run in https://bugs.gnucash.org/show_bug.cgi?id=787018
12:42:06 <ctjctj> fell, yep. That looks like a hit. The biggest thing is that print to file needs a good default output file name. "output.pdf" is no where near as useful as "Invoice-123445-04_19_2019.pdf"
12:44:54 <ctjctj> fell, good google foo. Thanks. That is my bug. It looks like they are nearing a fix to the issue to address everything I've just mentioned. In the mean time, I can use the work around. Thank you.
12:45:13 *** boldstripe has quit IRC
12:46:10 *** boldstripe has joined #gnucash
12:47:26 <fell> So the remaining question is: who has time to review MikeE' s patch https://bugs.gnucash.org/attachment.cgi?id=371868 , gjanssens & jralls. ;-)
12:54:13 <fell> warlord, if you have some time: The gnome extensions for bugzilla had a browse page, where we could see pending patches.
12:55:16 <warlord> Yeah, but I feel like it was more than just a .cgi file -- I feel like there were additional BZ core changes required to implement that.
12:58:35 <fell> i wonder, if installing that package fully, would break something.
13:01:32 <warlord> I don't know, but I'm not sure I want to try it on the production BZ server.
13:03:10 <fell> How hard is it to mirror him?
13:31:47 *** guak1 has joined #gnucash
13:32:13 *** guak has quit IRC
13:32:13 *** guak1 is now known as guak
13:36:21 *** fabior has quit IRC
14:09:19 *** Mechtilde has quit IRC
14:18:20 *** oozer has quit IRC
14:21:59 <jralls> gjanssens: Welcome back!
14:24:14 <jralls> @tell chris there's more to the string port problem than a few currency symbols. In some locales Guile crashes at https://github.com/Gnucash/gnucash/blob/maint/gnucash/gnucash-bin.c#L534, https://bugs.gnucash.org/show_bug.cgi?id=796728
14:24:14 <gncbot> jralls: The operation succeeded.
14:25:43 *** jervin has joined #gnucash
14:25:46 <jralls> gjanssens, fell: `gnucash --logto=stderr` works fine in an MSYS2 shell, haven't tried from a CMD or Powershell one.
14:25:49 *** oozer has joined #gnucash
14:32:12 *** frakturfreak has joined #gnucash
14:33:57 *** oozer has quit IRC
14:35:11 *** eric_ has quit IRC
14:42:01 *** eric_ has joined #gnucash
14:45:52 *** jervin has quit IRC
14:45:53 *** boldstripe has quit IRC
14:58:13 <jralls> fell: I reviewed mikee's patch on 787018. Just cosmetic issues with the code and a plea for better commit messages.
14:58:43 <jralls> mikee: You said the other day that you're a bit rusty with git. If you want help just ask!
15:01:42 *** nimish2711 has quit IRC
15:03:03 <fell> I fear, Mikee is in the pub at this time. ;-)
15:06:48 <jralls> fell: No matter, he'll see it later. About checking for patches, https://bugs.gnucash.org/buglist.cgi?bug_status=NEW&bug_status=ASSIGNED&bug_status=NEEDINFO&bug_status=REOPENED&f1=attachments.ispatch&f2=flagtypes.name&f3=flagtypes.name&list_id=3717&o1=equals&o2=notequals&o3=notequals&query_format=advanced&v1=1&v2=rejected%2B&v3=committed%2B
15:08:36 <fell> Cool, jralls!
15:18:00 *** oozer has joined #gnucash
15:22:02 *** oozer has quit IRC
15:24:56 *** oozer has joined #gnucash
15:37:19 <gjanssens> fell, jralls: I have added my additional thoughts on the bug
15:38:00 <gjanssens> jralls: that's a nice query. I have saved it for future use as I was missing this functionality as well
16:01:55 *** jervin has joined #gnucash
16:05:49 *** jervin has quit IRC
16:18:03 <fell> chris, do we really need all the hard linebreaks in gnucash/report/standard-reports/income-gst-statement.scm:43?
16:19:38 <fell> IMHO at the end of a para \n or \n\n is ok, but inside a sentence?
16:20:52 <fell> @tell chris Do we really need all the hard linebreaks in gnucash/report/standard-reports/income-gst-statement.scm:43? IMHO at the end of a para \n or \n\n is ok, but inside a sentence?
16:20:52 <gncbot> fell: The operation succeeded.
16:36:32 *** ctjctj has quit IRC
16:51:54 *** gjanssens has quit IRC
17:22:48 *** nicoroumeau has joined #gnucash
18:49:17 *** jcamos has joined #gnucash
18:51:00 *** jcamos has quit IRC
18:51:07 *** jcamos has joined #gnucash
18:53:00 *** jcamos has quit IRC
18:53:09 *** jcamos has joined #gnucash
18:53:53 *** jcamos has joined #gnucash
18:54:17 *** jcamos has quit IRC
18:54:27 *** jcamos has joined #gnucash
18:54:27 *** ChanServ sets mode: +v jcamos
19:00:36 *** jcamos has quit IRC
19:00:52 *** nicoroumeau has quit IRC
19:14:38 *** frakturfreak has quit IRC
19:59:30 *** ArtGravity has quit IRC
20:09:09 *** fell has quit IRC
20:13:29 *** fell has joined #gnucash
20:13:29 *** ChanServ sets mode: +o fell
20:33:16 *** chris has joined #gnucash
20:33:16 *** ChanServ sets mode: +v chris
20:52:31 *** fell has quit IRC
21:18:26 *** guak has quit IRC
21:50:27 *** fell has joined #gnucash
21:50:27 *** ChanServ sets mode: +o fell
22:13:07 *** jethrogb has quit IRC
22:29:55 *** oozer has quit IRC
23:31:23 *** nimish2711 has joined #gnucash