2023-09-11 GnuCash IRC logs

00:15:43 <jralls> fell, are you able to connect to the wiki? I'm getting timeouts, just for the wiki. Other code stuff-IRC logs, list archives, git, etc. are quite responsive.
00:23:08 *** jwhite has quit IRC
00:30:25 *** chris has quit IRC
00:33:25 <jralls> warlord, I'm getting a 504 Gateway Timeout "The gateway did not receive a timely response from the upstream server or application." from the wiki. Everything else on code seems to be working.
00:41:53 <fell> same here, jralls.
01:14:06 *** fell has quit IRC
01:15:25 *** fell has joined #gnucash
01:15:25 *** ChanServ sets mode: +o fell
01:56:51 *** gandalf has joined #gnucash
02:02:02 *** NoobAlice has quit IRC
02:35:35 <gjanssens> jralls: I will check your question and suggestions tonight or tomorrow. But at first glance I think it's best to move the payment_lot in front instead of adding it again. More details later.
02:35:35 <gncbot> gjanssens: Sent 1 day, 5 hours, and 23 minutes ago: <jralls> Can you find a minute to look at https://bugs.gnucash.org/show_bug.cgi?id=798925?
04:00:51 *** chocoz has quit IRC
04:40:34 *** luna has joined #gnucash
05:01:45 *** gandalf1 has joined #gnucash
05:02:13 *** gandalf has quit IRC
05:04:14 *** gandalf has joined #gnucash
05:04:45 *** gandalf1 has quit IRC
07:11:11 *** Aussie_matt has joined #gnucash
07:22:09 *** ChanServ sets mode: +qo warlord warlord
07:22:11 <warlord> jralls, fell -- I'll take a look.
07:29:11 <warlord> Looks like something between Apache and PHP was wedged. I restarted both and now the wiki is back.
07:40:24 *** Luk6655 has joined #gnucash
07:42:57 *** ChanServ sets mode: +v Luk6655
07:44:21 *** Luk6655 has quit IRC
07:44:56 *** Luk6655 has joined #gnucash
07:44:56 *** ChanServ sets mode: +v Luk6655
07:46:01 <Luk6655> Hi, if I'd like to add some translations (Polish, for simple stuff like "New Customer" that's not there), do I need to build from the source, or is there some other way?
07:46:36 <warlord> Luk6655, I believe you can work through the translation project, but in general, yes, you will need to build from source in order to test your translation.
07:46:57 <Luk6655> ok, thanks
07:58:56 *** Aussie_matt has quit IRC
08:24:22 *** Luk6655 has quit IRC
08:31:56 *** Luk6655 has joined #gnucash
08:31:56 *** ChanServ sets mode: +v Luk6655
08:39:47 *** cproo12 has joined #gnucash
08:39:47 *** ChanServ sets mode: +v cproo12
08:48:46 <Luk6655> I wonder, is there some example gnucash file (with bogus/made up data) that shows VAT accounting(UK, or some EU country)?
08:51:11 <warlord> Luk6655, good question. I don't know if we have actual sample data files like that.
08:51:29 *** cproo12 has quit IRC
08:53:05 <Luk6655> I'm trying to figure out if gnucash is a good option for my simple business accounts (the main thing of importance is VAT for me)
08:54:07 <Luk6655> at least there are some example accounts for UK VAT, it's pretty similar so hopefully I can figure it out
09:12:05 *** jwhite has joined #gnucash
09:12:05 *** ChanServ sets mode: +v jwhite
09:58:27 <Luk6655> Is it possible to change the invoice print template, to show different fields, for example customer's tax number?
09:58:31 <AdrienM> Luk6655 there are a few folks using it for business, check the gnucash-user list for threads. I've seen several related topics over the last few years.
09:59:02 <Luk6655> I'll have a look
09:59:44 <AdrienM> If invoices can be customized without writing an entirely custom report, check into the report Options and the Stylesheets.
10:02:19 <Luk6655> Is an invoice printout a report?
10:05:13 <Luk6655> yes, it looks like it is, and one can customise fields by going to Reports->Business->Printable Invoice and clicking on Edit report options
10:05:30 <Luk6655> I think this should do it, but I'll have to play with this later
10:08:34 <Luk6655> I'm just not sure where are field names defined. For example this "Invoice date, due date, billing ID, terms, job details" is all one field...
10:08:50 <warlord> Luk6655, there are multiple invoice reports, and there are some flexibility in what is displayed.
10:12:05 <Luk6655> warlord, indeed, there are some. Are the fields defined in .scm files somewhere?
10:12:28 <Luk6655> I found a 15 yeard old thread that talks about modifying this file share/gnucash/guile-modules/gnucash/report/invoice.scm
10:12:45 <warlord> yes, all reports are scheme
10:12:53 <warlord> that is one of the invoice reports.
10:13:14 <warlord> You might also look at that Tax Invoice report
10:14:40 <Luk6655> thanks, I will, I know some guile/scheme so I don't mind editing it if necessary
10:18:12 *** Luk6655 has quit IRC
11:27:19 *** el_capitan2 has quit IRC
11:49:18 *** cproo12 has joined #gnucash
11:49:19 *** ChanServ sets mode: +v cproo12
12:15:14 *** cproo12 has quit IRC
12:48:41 <jralls> warlord, the translation project dumped us a couple of years ago. All our translations except documentation are now done through https://hosted.weblate.org/projects/gnucash.
12:50:31 <warlord> sorry, weblate...
12:50:40 <warlord> Yes
12:52:03 <jralls> warlord The wiki still isn't working for me. BTW, that 504 takes over 30 minutes to appear.
12:53:27 *** palerider has joined #gnucash
12:56:57 <warlord> Hmm.. .code has a load of 60..
12:57:16 <warlord> There are over 11 instances of "cmake"
12:58:36 <warlord> .. and a ton of pythons too.
12:59:35 <warlord> I'm killing httpd to try to figure out what's going on...
13:00:45 <warlord> Lots of: python /usr/share/mediawiki/extensions/SyntaxHighlight_GeSHi/pygments/pygmentize -l sh -f html -O cssclass=mw-highlight,encoding=utf-8
13:02:25 <jralls> Weird. Is it doing an update of some kind?
13:02:48 *** cproo12 has joined #gnucash
13:02:48 *** ChanServ sets mode: +v cproo12
13:03:56 <warlord> I don't know. I just stopped httpd and php-fpm services
13:04:04 <warlord> The load is now coming down...
13:04:14 <warlord> From 60 down to 9...
13:11:44 *** palerider has left #gnucash
13:20:43 *** el_capitan2 has joined #gnucash
13:20:43 *** ChanServ sets mode: +v el_capitan2
13:25:57 <jralls> warlord, I wonder if somebody was ddos-ing our wiki, maybe inadvertently. The httpd logs might shed some light.
13:26:12 <warlord> Maybe.
13:26:17 <warlord> It's shut down right now.
13:31:19 <warlord> I don't see anything in the logs that would indicate a ddos.
13:31:31 *** cproo12 has quit IRC
13:32:40 <jralls> Hmm. So if not connects to pages that need syntax highlighting what else might trigger all of those python jobs?
13:35:58 <warlord> I don't know...
13:36:03 <warlord> I'm cleaning up the database right now..
13:37:25 <warlord> OKay, restarting httpd
13:38:55 <warlord> Wiki is up right now...
13:39:40 <jralls> Yes, I can reach it.
13:40:15 <warlord> There were a bunch of "cmake" commands in the background ... some from the docs build!
13:41:48 <warlord> I'm rebuilding by hand...
13:43:31 <jralls> Interesting, but what about the not from the docs build cmake jobs? The flatpak build finished at 0100 and the win32 at 0333.
13:44:02 <warlord> This seems to be taking a long time from the gnucash-docs:
13:44:02 <warlord> cmake -DCMAKE_INSTALL_PREFIX=/home/gnucash-docs/gnucash-docs/output -DWITH_MOBI=ON /home/gnucash-docs/gnucash-docs
13:44:27 <warlord> win32 is on a separate vm.
13:44:42 <warlord> flatpak IS built here, but if it finished at 0100 then it didn't do anything.
13:47:39 <warlord> Aha. Here is why the docs are failing:
13:48:06 <warlord> cmake -DCMAKE_INSTALL_PREFIX=/home/gnucash-docs/gnucash-docs/output -DWITH_MOBI=ON /home/gnucash-docs/gnucash-docs |& more
13:48:06 <warlord> -- The C compiler identification is GNU 8.3.1
13:48:06 <warlord> -- Check for working C compiler: /usr/bin/cc
13:48:06 <warlord> -- Check for working C compiler: /usr/bin/cc -- works
13:48:06 <warlord> -- Detecting C compiler ABI info
13:48:08 <warlord> -- Detecting C compiler ABI info - done
13:48:10 <warlord> -- Detecting C compile features
13:48:12 <warlord> -- Detecting C compile features - done
13:48:14 <warlord> CMake Error at cmake/AddGncDocTargets.cmake:32 (list):
13:48:16 <warlord> list does not recognize sub-command POP_FRONT
13:48:18 <warlord> Call Stack (most recent call first):
13:48:20 <warlord> zh/guide/CMakeLists.txt:222 (add_gnc_doc_targets)
13:48:48 <jralls> What's cmake --version?
13:48:50 <warlord> ... and then it just loops, printing out those last 4 lines.
13:49:01 <warlord> cmake --version
13:49:01 <warlord> cmake version 3.14.5
13:51:23 <jralls> And POP_FRONT was introduced in 3.15. gjanssens must have missed that when he went through the upgrade-downgrade a few weeks ago.
13:53:05 <warlord> oops
13:55:38 <jralls> But that's been going on since 8/18. It's not what took out the wiki yesterday.
13:56:33 <warlord> What took out the wiki was the load of 60 -- these cmake's have been building.. and my restart this morning didn't fix that.
13:57:07 <warlord> I only fixed that just now when I saw the load...
13:58:22 <jralls> Oh, almost a month of doc builds. That's a bit weird, cmake doesn't usually retry.
13:59:51 <warlord> I dont know why it was looping like that.
14:00:22 <warlord> How hard is it to get the code to wotk with 3.14.5?
14:01:08 <jralls> I'm looking at that now.
14:01:10 <warlord> gjanssens, ??
14:11:02 <gjanssens> jralls, warlord: I don't remember doing something with POP_FRONT ? Where does that come from ?
14:11:23 <warlord> I think the zh tree
14:11:59 <jralls> Right, docs PR 323.
14:12:34 <gjanssens> Right, found that bit as well.
14:13:22 <gjanssens> I'm surprised this didn't generate warnings or errors earlier as our base line cmake version was 3.10
14:13:45 <jralls> So all we need is to iterate the list differently. Too bad, the POP_FRONT way elegantly retrieves the pair.
14:14:04 <gjanssens> Shouldn't that at least warn about a list command not existing in the requested compatibility version ?
14:15:04 <gjanssens> But yes, the immediate fix would be to iterate the list differently.
14:15:44 <jralls> It wasn't merged until 3 weeks ago. Yes, I'd think it would generate a warning, but probably only on code. Everything everyone else is using is considerably newer.
14:18:46 <warlord> And code runs cmake with >/dev/null 2>/dev/null
14:18:54 <warlord> So no errors are actually output...
14:19:11 <warlord> And in the case of code, it goes into a loop and the cmake never actually exits
14:19:13 <jralls> From cmake docs: "Lists are meant for simple use cases such as a list of source files and should not be used for complex data processing tasks."
14:38:17 <jralls> warlord, I pushed a fix. Please test it.
14:38:55 <warlord> Okay. Give me a minute.
14:41:36 <warlord> cmake passed.
14:44:30 <warlord> but... docs-build failed.
14:45:58 <warlord> I am running without the script to get actual output.
14:49:49 <jralls> What failed? I tested only pdf because pdf and chm are the only stylesheets that Xu replaces.
14:51:15 <warlord> Still trying to figure that out.
14:57:05 <jralls> Do you have cjkuni-ukai-fonts installed? Is it even available?
14:58:59 <warlord> Huh.. .build is still going...
15:00:56 <warlord> Right now it's at:
15:00:56 <warlord> [ 80%] Generating gnucash-manual.fo
15:00:57 <warlord> Making portrait pages on USletter paper (8.5inx11in)
15:09:02 *** zoid has joined #gnucash
15:09:40 *** ChanServ sets mode: +v zoid
15:10:55 <zoid> Is there a more accepted to way to account for credit card statement credits/cash back other than listing it as an income source?
15:11:34 <jralls> zoid, you can call it income or a negative expense, totally your choice.
15:13:08 <warlord> Build is still running....
15:16:50 <jralls> That's slow.
15:18:53 <warlord> Yeah. But it did finish...
15:19:06 <warlord> HOWEVER, when I run it from the script, it dies immediately and says the build failed.
15:19:10 <warlord> If I build from hand, it works.
15:19:47 <jralls> That's annoying. Can you change the redirects to a file to see why the script is dying?
15:20:24 <jralls> Or for a manual run just remove them and see what it says in the terminal.
15:21:10 <jralls> Maybe it's as simple as there's no zh directory to write to.
15:21:59 <warlord> [ 9%] Generating xdghelptrigger
15:21:59 <warlord> failed to create symbolic link '/home/gnucash-docs/gnucash-docs/build/share/doc/
15:21:59 <warlord> HTML/en/gnucash-manual' because existing path cannot be removed: Is a directory
15:21:59 <warlord> make[3]: *** [C/manual/CMakeFiles/C-manual-html.dir/build.make:398: C/manual/xdg
15:21:59 <warlord> helptrigger] Error 1
15:22:00 <warlord> make[2]: *** [CMakeFiles/Makefile2:987: C/manual/CMakeFiles/C-manual-html.dir/al
15:22:02 <warlord> l] Error 2
15:22:04 <warlord> make[1]: *** [CMakeFiles/Makefile2:149: CMakeFiles/html.dir/rule] Error 2
15:22:06 <warlord> make: *** [Makefile:190: html] Error 2
15:22:41 <warlord> Sorry about the embedded newlines
15:24:37 <jralls> Looks like your script needs to start with `rm -rf build/`.
15:25:45 <warlord> I just tried that, but cmake isn't actually creating anything.
15:26:09 <warlord> I just did "mv build build-bak" and then re-ran the script and it died with an empty "build" dir
15:26:21 <warlord> Then I tried: cd build && cmake ....
15:26:31 <jralls> That's make not cmake. Aside, why aren't you using ninja? No wonder it's slow.
15:26:40 <warlord> cmake -DCMAKE_INSTALL_PREFIX=/home/gnucash-docs/gnucash-docs/output -DWITH_MOBI=ON /home/gnucash-docs/gnucash-docs
15:26:40 <warlord> -- Configuring done
15:26:40 <warlord> -- Generating done
15:26:40 <warlord> -- Build files have been written to: /home/gnucash-docs/gnucash-docs
15:26:40 <warlord> [gnucash-docs@code build]$ ls -l
15:26:41 <warlord> total 0
15:27:13 <warlord> My point is that CMAKE isn't outputting anything.
15:30:54 <warlord> jralls, any idea why cmake is not creating anything in the CWD?
15:31:25 <jralls> No, but it said it put the build files in SRCDIR. Did it?
15:32:33 <warlord> What's the point of having a build-dir if it puts it in SRCDIR?
15:33:09 <warlord> But yes, there IS a makefile in .. (SRCDIR)
15:34:42 <warlord> Looks like I need to change that to be: cmake -DCMAKE_INSTALL_PREFIX=/home/gnucash-docs/gnucash-docs/output -DWITH_MOBI=ON -S /home/gnucash-docs/gnucash-docs -B .
15:35:53 <jralls> No, I think what's happening is that you did a test build in the source directory and now cmake is finding the CMakeCache.txt there. cd to srcdir and do a `git clean -fdx` then switch back to build and try again.
15:37:22 <warlord> The script runs:
15:37:23 <warlord> git checkout $branch -q >/dev/null 2>&1 || my_die "git checkout failed."
15:37:23 <warlord> git fetch -q >/dev/null 2>&1 || my_die "git fetch failed."
15:37:23 <warlord> git reset --hard origin/$branch -q >/dev/null 2>&1 || my_die "git reset failed."
15:37:40 <warlord> Anyways, I changed it to my method and the script is currently running.
15:38:42 <jralls> git reset has no effect on untracked files like CMakeCache.txt. That's why I said `git clean`.
15:39:38 <warlord> Understood...
15:39:54 <gjanssens> Do I read that correctly your script is building with srcdir==builddir ?
15:41:19 <gjanssens> (Hmm, no, you're doing a cd build && cmake ...)
15:41:20 <warlord> gjanssens, no, the script was originally runnning as ( cd builddir ; cmake .... SRCDIR )
15:42:28 <gjanssens> Ok I didn't see the SRCDIR on your commands above. I presume you skipped it for brevity/privacy. With srcdir it should work. No need for -B .
15:43:12 <warlord> It should, but it wrote the Makefile into SRCDIR.
15:43:41 <jralls> That's because there's a CMakeCache.txt in SRCDIR.
15:44:02 <gjanssens> Indeed as it then considers that directory to be the buid dir.
15:44:08 <jralls> You really do need to git-clean it.
15:45:11 <gjanssens> (you can run cmake on a pre-existing build dir to update cmake options for that build dir)
15:45:57 <warlord> I'll git clean once this build finishes.
15:54:33 <gjanssens> Got to go. Good luck debugging the docs build!
15:54:53 *** gjanssens has quit IRC
16:27:38 *** el_capitan2 has quit IRC
16:30:25 <warlord> Okay, the build finished. I just did the git clean and am rebuilding again.
16:39:24 *** gandalf has quit IRC
17:03:12 *** palerider has joined #gnucash
17:04:54 *** palerider has left #gnucash
17:31:20 <fell> Oh, my first sucessfull wiki edit after several weeks. Congrats, warlord!
17:50:37 <warlord> Thanks.
17:53:29 <warlord> rebuild finished, too. I removed the -S.. -B.. from the script (after the git clean). All is happy now.
17:55:01 <warlord> jralls, installed cjkuni-ukai-fonts
18:35:19 *** miklcct has quit IRC
18:37:14 *** miklcct has joined #gnucash
18:37:14 *** ChanServ sets mode: +v miklcct
19:34:36 *** AdrienM has quit IRC
19:59:19 *** AdrienM has joined #gnucash
19:59:19 *** ChanServ sets mode: +v AdrienM
20:33:04 *** chris has joined #gnucash
20:33:04 *** ChanServ sets mode: +v chris
20:33:04 *** gncbot sets mode: +o chris
20:44:03 *** chris has quit IRC
20:53:47 *** cproo12 has joined #gnucash
20:53:47 *** ChanServ sets mode: +v cproo12
21:28:44 *** cproo12 has quit IRC
22:15:56 <fell> jralls, shouldn't we copy cjkuni-ukai-fonts into fonts or make it a dependency?
22:27:00 *** el_capitan2 has joined #gnucash
22:27:00 *** ChanServ sets mode: +v el_capitan2
22:48:34 *** NoobAlice has joined #gnucash
22:48:35 *** ChanServ sets mode: +v NoobAlice
23:11:08 *** NoobAlice has quit IRC
23:15:40 *** cproo12 has joined #gnucash
23:15:40 *** ChanServ sets mode: +v cproo12
23:25:51 *** cproo12 has quit IRC