2020-12-15 GnuCash IRC logs

00:58:50 *** Mechtilde has joined #gnucash
01:37:15 *** AdrienM has quit IRC
01:37:59 *** AdrienM has joined #gnucash
01:37:59 *** ChanServ sets mode: +v AdrienM
01:45:46 <giuseppef_> Good morning, looking now at weblate. It seems it made some changes on po I didn't push
01:46:30 <giuseppef_> Can I push on git my current it.po before starting with weblate?
01:51:37 *** fell has quit IRC
01:52:56 *** fell has joined #gnucash
01:52:56 *** ChanServ sets mode: +o fell
02:01:17 <fell> jralls, an async example: https://hosted.weblate.org/projects/minetest/minetest/#alerts
02:08:04 <fell> Bon giorno, guiiseppef! Yes, I think you should create your PR and later we look in the weblate PR.
02:08:57 <fell> giuseppef_ ^
02:10:29 *** marusich has quit IRC
02:10:45 <giuseppef_> Ok, just give me a schedule. I am waiting form suggestions from gnucash-it, but I can make a PR and later working just on weblate. I don't understand if weblat is still a test or if you are going that way
02:14:33 *** Mechtilde has quit IRC
02:15:58 <fell> At first, weblate is a test, later it might become one official offer.
02:16:47 <fell> Finally the translators will decide, what is easiest for them.
02:21:17 <fell> The deadline: noon 26 December UTC-8.
02:23:38 *** marusich has joined #gnucash
02:23:38 *** ChanServ sets mode: +v marusich
02:23:45 <fell> But you can send the ready part, the day after merge announce the nighlies on gnucash-it and lATER SEN CORRECTION.
02:24:53 <fell> oops, fat fingers
02:30:58 <fell> Not much traffic on gnucash-it, giuseppef_
02:32:40 <giuseppef_> I know fell, it was my last attempt to find someone helping me in reviewing translations.
02:33:22 <giuseppef_> months ago I suggested weblast or GlotPress, because usually translators are not programmers and git is difficult to be used
02:33:33 <fell> The people from bolzano ask on gnucash-de. ;-)
02:34:22 <giuseppef_> No surprise. In Bolzano half people speaks German
02:34:45 <giuseppef_> In that area we have two official languages by law
02:36:06 <giuseppef_> If for 4.3 we will not use weblate, I will wait for PR a couple of days more. Just to avoid multiples PR on github
02:36:34 <fell> It needed some spare time to have a look on the features and docs of weblate.
02:38:41 *** bertbob has quit IRC
02:40:32 *** bertbob has joined #gnucash
02:40:32 *** ChanServ sets mode: +v bertbob
02:43:37 *** marusich has quit IRC
02:47:38 *** marusich has joined #gnucash
02:47:38 *** ChanServ sets mode: +v marusich
02:51:49 *** marusich has quit IRC
02:54:05 *** jervin has joined #gnucash
02:55:46 *** marusich has joined #gnucash
02:55:47 *** ChanServ sets mode: +v marusich
02:56:38 *** qwer has joined #gnucash
03:09:54 <giuseppef_> However I think you should create one or more "team leaders" or "language leaders" to approve translations on weblate. Right now it seems strings have been translated "automatically" or something, and it is a risk to allow any translation to be directly accepted. Later I will do some tests
03:12:32 *** jervin has quit IRC
03:14:56 *** giuseppef has quit IRC
03:15:19 *** giuseppef has joined #gnucash
03:15:19 *** ChanServ sets mode: +v giuseppef
03:16:48 *** giuseppef has quit IRC
03:17:30 *** giuseppef has joined #gnucash
03:17:30 *** ChanServ sets mode: +v giuseppef
03:26:29 *** giuseppef has quit IRC
03:28:38 *** bertbob has quit IRC
03:35:39 *** Mechtilde has joined #gnucash
03:43:08 *** bertbob has joined #gnucash
03:43:08 *** ChanServ sets mode: +v bertbob
04:21:21 *** marusich has quit IRC
04:28:03 *** gjanssens has joined #gnucash
04:28:03 *** ChanServ sets mode: +o gjanssens
04:36:31 *** Aussie_matt has quit IRC
05:03:03 *** User_ has joined #gnucash
05:06:14 *** field^Zzz4 has joined #gnucash
05:50:08 *** User_ has quit IRC
06:04:06 *** Herbert has joined #gnucash
07:14:22 *** storyjesse has joined #gnucash
07:52:33 <warlord> .
08:02:39 <chris> warlord you're supposed to be away
08:02:48 <chris> tut tut
08:03:16 <warlord> What do you mean?
08:04:40 <chris> maybe i misread your msg ... not yet
08:06:51 <warlord> Right. Dec 20
08:06:57 <warlord> Today is the 15th
08:16:55 <chris> latest .scm commit should be a nice assist to any report writer
08:17:33 *** KaiForce has joined #gnucash
08:17:56 <chris> forget to (use-modules (gnucash report)) and the traceback will say Unbound variable: gnc:define-report. Did you forget (use-modules (gnucash report report-core))?
08:19:05 *** qwer has quit IRC
08:26:38 <warlord> nice
08:39:12 *** Mechtilde has quit IRC
08:40:01 *** Mechtilde has joined #gnucash
08:46:00 *** jw4 has quit IRC
08:46:27 *** jw4 has joined #gnucash
08:52:16 *** Jimraehl1 has joined #gnucash
08:53:13 *** Jimraehl1 has quit IRC
09:11:10 *** Mechtilde has quit IRC
09:32:45 *** storyjesse has quit IRC
09:36:05 *** halipaul has quit IRC
10:24:59 *** qwer has joined #gnucash
10:54:39 *** AdrienM has quit IRC
10:55:02 *** AdrienM has joined #gnucash
10:55:02 *** ChanServ sets mode: +v AdrienM
10:55:25 *** qwer has quit IRC
10:55:59 *** jcarl43 has joined #gnucash
10:55:59 *** ChanServ sets mode: +v jcarl43
10:57:01 *** bertbob has quit IRC
11:00:03 *** bertbob has joined #gnucash
11:00:03 *** ChanServ sets mode: +v bertbob
11:25:58 <fell> giuseppef_, yu are the current team leader it. :-)
11:45:01 *** Herbert has joined #gnucash
11:45:54 <jralls> fell, I'm inclined to close https://github.com/Gnucash/gnucash-htdocs/pull/33 and tell the author to get lost. The title is spelling but the changes are more rewording.
11:45:59 *** Herbert has quit IRC
11:46:44 <jralls> What am I supposed to look at on the minetest weblate repo?
11:48:17 <fell> How it looks, when nobody syncs.
11:49:09 <jralls> I've already set up webhooks on the github repos to push any changes on to weblate.
11:49:58 *** Mechtilde has joined #gnucash
11:50:08 <jralls> So the issue is figuring out how to manage weblate's commits.
11:50:30 <fell> About PR 33, I was recently also thinking about using some spelling checks.
11:51:57 <jralls> No problem with spelling checks. The title of the PR is misleading.
11:52:44 <fell> I have modified glossary/de last night and couldn't find a push button. Iassume it is pushed at midnight?
11:55:14 <fell> I would prefer to delay PR33. 1, to see, if we get active website translators now. 2. Master ist still frozen odr did that change, gjanssens?
11:56:40 <jralls> I think the website will stay broken until linas can attend to it. That means the 4.3 release announcement likely will be on github only. :-(
11:57:07 <jralls> Though I guess I could push it to beta.
11:57:18 <fell> and on https:…beta
11:58:41 <fell> Perhaps we should suggest in the wiki main page to use beta.
11:59:17 <fell> Do we not have a tempate?
12:00:43 <jralls> You mean a wiki template for the website address? I don't think so, it's never been a problem before.
12:02:36 <fell> WebURL
12:03:27 <fell> The Idea came when we switched to https
12:06:11 <fell> The full list is https://wiki.gnucash.org/wiki/index.php?title=Special%3APrefixIndex&prefix=&namespace=10
12:18:39 *** ArtGravity has joined #gnucash
12:18:39 *** ChanServ sets mode: +v ArtGravity
12:26:20 *** guak has joined #gnucash
12:27:31 <warlord> I'm still trying to get linas to respond. I got an email and a linked-in response from him, but it's just been promises to get back to me ...
12:45:23 *** angel has joined #gnucash
12:52:13 *** gjanssens has quit IRC
12:56:46 <jralls> ebassi just announced that Gtk 4.0 will release on the 16th.
13:06:03 *** giuseppef has joined #gnucash
13:06:03 *** ChanServ sets mode: +v giuseppef
13:20:11 *** giuseppef has quit IRC
13:20:24 *** giuseppef has joined #gnucash
13:20:24 *** ChanServ sets mode: +v giuseppef
13:26:31 <warlord> Nice christmas present.
13:35:32 <jralls> warlord, where are we with bug-writing.html.tmpl? IIRC we were trying to figure out how to move files from a new gitolite repo to BZ.
13:41:37 <warlord> Sorry, I haven't put much thought into it. Easiest thing would be to have a checkout with a hook (or cron job) to copy the contents into place.
13:42:00 <warlord> Or, arguably, just make that directory the git checkout and have a hook to run a pull.
13:48:53 <jralls> So I have a gitolite commit ready with a new fragment in post-recieve "if [ "$reponame" = "gnucash-gitolite" ]; then /usr/local/bin/pull-bz-template-update; fi".
13:50:00 <warlord> is that the reponame?
13:50:11 <warlord> Does the new repo exist?
13:50:31 <jralls> Not yet, creating it is part of the commit I have staged.
13:51:06 <jralls> Shall I commit and push?
13:52:09 <warlord> Don't have it be /usr/local/bin -- use the same path as the website update
13:53:28 <jralls> That would be $HOME/bin.
13:54:13 <warlord> Indeed.
13:54:33 <warlord> Let me crate pull-bz-template-update there... Just as a placeholder.
13:55:26 <warlord> done.
13:55:44 <warlord> If you push and it runs it should print out "pulling update"
13:59:42 <jralls> OK. So the repo I have set up to push to it is rooted at the base of BZ, so the file will be templates/en/custom/bug-writing.html.tmpl. Would it work better without the path so you can create the custom dir and init a repo in it?
14:01:43 <warlord> Are we going to want templates/XX/custom for XX != en ?
14:03:39 <jralls> Maybe. Are the non-custom templates localized?
14:04:27 <warlord> Well, right now I only have templates/en/default -- so, no, they are not (currently) localized.
14:09:55 <jralls> Hmm. Making custom a repo is safe, BZ upgrades won't touch it. Making any parent a repo risks either it being blown away by an upgrade or breaking the upgrade and requiring manual intervention.
14:12:02 <jralls> OTOH, I could leave the repo as-is and pull-bz-template-update could `cp $1 /path/to/BZ/$1`.
14:13:51 *** Mechtilde has quit IRC
14:14:56 *** Mechtilde has joined #gnucash
14:21:06 <Mechtilde> .
14:25:06 <warlord> Well, if it's just custom then I can just to a git clone directly into /usr/share/bugzilla/templates/en/custom
14:25:18 <warlord> And the update script just does a git pull.
14:25:44 <warlord> What I do not know is if I need to restart BZ (httpd) to get template updates to take
14:26:14 *** angel has quit IRC
14:26:38 <Mechtilde> fell, do I read it right at the German list? The flicker Code works for GnuCash?
14:26:40 <warlord> So I think the top-level of the repo should be the contents of "custom", so that I can git clone <repo> custom and it'll be "right"
14:27:12 <jralls> I guess we'll find out shortly after I push the first change...
14:29:32 <warlord> Indeed..
14:33:14 <warlord> And update script is in place.
14:33:34 <warlord> So just keep in mind that the structure of custom/ needs to match the structure of default/
14:40:45 <jralls> IIRC https://bugzilla.readthedocs.io/en/latest/integrating/templates.html files in custom must be in the same place as in default for BZ to notice them. We don't need to replicate the whole directory.
14:41:30 <jralls> That said, it doesn't say where bug-writing.html.tmpl lives. I thought it was directly in the custom directory. Is that not the case?
14:46:26 <jralls> Ah, I see from https://github.com/bugzilla/bugzilla/ that it's in pages.
14:47:09 <warlord> Correct, it is in pages/
14:47:27 <warlord> And also correct, we don't need to replicate the whole directory, just the structure of replacements
14:49:47 <jralls> OK, the new one is pushed. Did the hook work?
14:51:47 <jralls> The old one still comes up, so either I didn't get it in the right place or a restart is required.
14:51:59 <warlord> what did your push say?
14:53:21 <warlord> It doesn't look like the git pull succeeded
14:53:27 <jralls> remote: *** for refs/heads/master update 0000000000000000000000000000000000000000->1748649167e78fcb68c6c4cdeb654d5a3cab7e49
14:53:28 <warlord> The custom directory is still empty.
14:53:58 <warlord> Did you see "pulling bz template update"?
14:54:05 <jralls> No.
14:54:41 <jralls> I did see also remote: *** hooks.mailinglist is not set so no email will be sent
14:55:14 <warlord> That's okay -- I still did get an email.
14:56:38 <warlord> Hmm. I wonder if it's a permission issue.
14:56:49 <warlord> I did a git pull by hand and it worked, but I have different permissions.
14:57:20 <warlord> I suspect I might need a different way to escape out. First, check BZ and see if your changes made it out?
14:57:41 <jralls> Did you remember to chmod +x the script and chown it to the owner of the bugzilla directory?
14:59:34 <jralls> The new template works now.
14:59:43 <warlord> yes
15:00:10 <warlord> But I think SELinux might be blocking the git owner from access to the BZ directory (even though it is "owned" by git)
15:00:40 <jralls> SELinux can be a pain.
15:01:15 <warlord> yep. Can you push another change and copy-paste the full output of your push?
15:04:43 <jralls> https://gist.github.com/jralls/27ed4ab378211b45ab3b0cdba5fde649
15:05:11 <jralls> Pretty much the same as last time.
15:07:52 <warlord> The trigger isn't firing.
15:09:17 <warlord> At least, I don't THINK it is firing.
15:09:34 <warlord> The email is going off okay, but the script isn't being run
15:11:56 <warlord> Aha, This line is wrong:
15:11:57 <warlord> if [ "$reponame" = "gnucash-gitolite" ]
15:12:01 <warlord> it's gnucash-bugzilla
15:12:05 <warlord> not gnucash-gitolite
15:12:12 <warlord> So.. the trigger isn't being called!!
15:12:37 <jralls> Yup. Saw that and was fixing as you typed. Pushed now.
15:13:16 <warlord> Now try an update to bz
15:14:15 <jralls> Did: remote: pulling bz template update "\n" remote: fatal: not a git repository: '.'
15:17:55 <warlord> hmm...
15:19:02 <warlord> Okay, try one more thing?
15:19:40 <warlord> Er, rather, try again
15:21:03 <giuseppef_> Hi all, I have a sugestion on subaccount renumeration
15:21:34 <jralls> warlord, same fatal error.
15:22:22 <giuseppef_> If I try to renumerate sub-accounts, using a step of 10, and the accounts has more then 9 subaccounts, the 10th is numbered prefix-100, The 1st is prefix-10.
15:23:01 <giuseppef_> This is ok, but ordering accounts by numbers, brings prefix-100 before prefix-20
15:23:06 <warlord> Okay, I've got another option. Give me a minute.
15:23:30 <giuseppef_> I think numbers should be padded with zeroes
15:26:34 <jralls> giuseppef_, That seems reasonable, but the request will be quickly forgotten here. You can write up an enhancement bug report. Or if you can write C you could do a pull request.
15:35:24 <warlord> jralls, okay, I think I got it.
15:35:57 <jralls> warlord, "remote: scheduling bz template update"
15:36:30 <warlord> Yep..
15:36:42 <jralls> And it went through.
15:36:59 <jralls> Meaning that the page has the change.
15:37:02 <warlord> Yep. There might be a 1-minute delay
15:37:24 <jralls> I can live with a minute. ;-)
15:39:12 <warlord> Excellent. Annoyingly I have to do with via a file marker and a cron job. But it works, even if it's a little expensive.
15:40:06 <jralls> Eww. That was easier than beating SELinux into submission?
15:40:33 *** jw4 has quit IRC
15:42:12 *** jw4 has joined #gnucash
15:42:12 *** ChanServ sets mode: +v jw4
15:51:01 <warlord> yes
16:08:00 *** codesmythe has quit IRC
16:13:46 *** Jonmaikel has joined #gnucash
16:13:51 *** Mechtilde has quit IRC
16:14:30 *** Jonmaikel has quit IRC
16:28:31 *** bertbob has quit IRC
16:34:52 *** jervin has joined #gnucash
16:36:50 *** jervin has quit IRC
16:36:57 *** jervin has joined #gnucash
16:40:07 *** jervin has quit IRC
16:42:18 *** bertbob has joined #gnucash
16:42:18 *** ChanServ sets mode: +v bertbob
16:47:13 *** bertbob has quit IRC
16:52:46 *** bertbob has joined #gnucash
16:52:46 *** ChanServ sets mode: +v bertbob
17:29:33 *** jervin has joined #gnucash
17:43:57 *** Aussie_matt has joined #gnucash
18:54:25 <chris> jralls: insert https://fukuchi.org/works/qrencode/ into borrowed?
19:08:29 <chris> what will be gtk4 features
19:19:39 *** jcarl43 has quit IRC
19:24:22 <chris> https://old.reddit.com/r/GnuCash/comments/kdcvjy/expenses_with_negative_values_in_pesos_currency/ -- exchange-rate should NOT be negative
19:25:12 <chris> ^ these restrictions should be handled in the engine, no?
19:31:33 *** chris has quit IRC
20:24:22 <fell> Ah, about QRencode. Chris, did you check the license? For what do you need it?
20:28:02 <fell> @tell Mechtilde I can not test it, but we got confirmation from a flatpak and a windows user. So only macOS is untested.
20:28:02 <gncbot> fell: The operation succeeded.
20:43:24 *** guak has quit IRC
21:04:42 *** ArtGravity has quit IRC
21:28:17 *** jervin has quit IRC
22:03:48 <fell> chris: That opens the question: should GC5 switch GTK4? Perhaps put it on -devel.
22:06:15 *** Agfarmer18 has joined #gnucash
22:10:21 *** Agfarmer18 has quit IRC
22:48:27 *** field^Mop has joined #gnucash
22:50:14 *** field^Zzz4 has quit IRC
23:08:43 *** bertbob has quit IRC
23:10:56 *** bertbob has joined #gnucash
23:10:56 *** ChanServ sets mode: +v bertbob
23:36:40 *** prg has quit IRC
23:37:56 *** prg has joined #gnucash