2018-06-07 GnuCash IRC logs
00:28:39 *** jotrago has quit IRC
00:40:31 *** jralls has joined #gnucash
00:40:31 *** ChanServ sets mode: +o jralls
00:43:37 *** jralls has quit IRC
01:05:33 *** fell has quit IRC
01:06:08 *** fell has joined #gnucash
01:07:30 *** frakturfreak has quit IRC
01:19:25 *** jralls has joined #gnucash
01:19:25 *** ChanServ sets mode: +o jralls
01:22:23 *** frakturfreak has joined #gnucash
01:22:24 *** eagles0513876 has joined #gnucash
01:22:30 *** jralls has quit IRC
01:22:50 *** eagles0513876 has quit IRC
01:23:16 *** eagles0513875 has quit IRC
02:12:04 *** gour has joined #gnucash
02:29:45 *** jralls has joined #gnucash
02:29:46 *** ChanServ sets mode: +o jralls
02:32:51 *** jralls has quit IRC
03:19:35 *** tms has quit IRC
03:19:48 *** tms has joined #gnucash
03:22:21 *** chf has quit IRC
03:31:02 *** jralls has joined #gnucash
03:31:02 *** ChanServ sets mode: +o jralls
03:34:55 *** jralls has quit IRC
03:48:22 *** gjanssens has joined #gnucash
03:48:22 *** ChanServ sets mode: +o gjanssens
03:49:13 *** fekepp has joined #gnucash
04:30:26 <gjanssens> .
04:41:32 *** jralls has joined #gnucash
04:41:32 *** ChanServ sets mode: +o jralls
04:44:42 *** jralls has quit IRC
05:06:02 *** ncv has joined #gnucash
05:08:15 *** ncv has quit IRC
05:14:49 *** gour_ has joined #gnucash
05:17:03 *** gour has quit IRC
05:27:26 *** jralls has joined #gnucash
05:27:26 *** ChanServ sets mode: +o jralls
05:30:30 *** jralls has quit IRC
06:11:37 *** pilotauto has quit IRC
06:24:45 *** chris has joined #gnucash
06:26:02 <chris> while creating html-anytag.scm I had a thought:
06:26:51 <chris> html-text.scm, html-table.scm and html-[line|bar|etc]chart.scm are the basic building blocks for reports
06:27:15 <chris> all aiming/trying to make html easy
06:27:39 <chris> but scheme now comes with its own XML generator
06:28:12 <chris> called sxml
06:29:02 *** Aussie_matt has joined #gnucash
06:29:18 <chris> so, creating a table becomes (define table '(table (thead (th (tr "head1" "head2"))) (tbody (tr (td "val1" "val2")))))
06:30:18 <chris> then (sxml->xml table) converts it to string "<table><thead><th><tr> (etc)" automatically
06:30:21 *** oozer has joined #gnucash
06:33:47 <chris> gjanssens/jralls the travis failure for #363 is likely a hiccup and is likely to pass on restart
06:37:40 *** jralls has joined #gnucash
06:37:40 *** ChanServ sets mode: +o jralls
06:40:50 *** jralls has quit IRC
06:44:21 *** gnomey has quit IRC
06:50:51 *** Jimraehl1 has joined #gnucash
06:52:05 *** Jimraehl1 has left #gnucash
06:55:26 <chris> CSS-made-easy: see how TR can be modified to style the subtotal-table https://imgur.com/a/2Br1Jph
06:57:41 *** chf has joined #gnucash
07:01:15 *** gnomey has joined #gnucash
07:07:57 *** Aussie_matt has quit IRC
07:09:32 <gjanssens> chris: nice :)
07:25:40 <chris> so, instead of converting .scm to eguile, I may just be able to upgrade invoice.scm to mimic invoice-eguile.scm ^_^
07:49:49 *** jralls has joined #gnucash
07:49:50 *** ChanServ sets mode: +o jralls
07:52:58 *** jralls has quit IRC
08:03:44 <gjanssens> chris: the ci failure was indeed a hickup. I have restarted the job and this time it worked fine
08:05:37 <gjanssens> chris: If similar flexibility gets into invoice.scm, that would be fine. But for me the handwritten mix of data and layout in the current report system is harder to make sense out than e-guile :)
08:06:40 <gjanssens> On the other hand I'm not an active (business) user of gnucash any more (I can't for legal reasons here) so it's less of a concern to me.
08:07:22 <gjanssens> Reconciling 4 invoice reports into one does bring another challenge though: migration
08:07:58 <gjanssens> If your code gets into gnucash, we need to provide migration paths for the obsolete reports to the new unified one.
08:08:29 <gjanssens> But that's a later step. First one is to get a unified report integrated and tested.
08:22:42 <chris> gjanssens the uuids will remain the same, the option names will be the same, so I really do not anticipate any migration challenges at all.
08:23:56 <chris> i'll any saved reports in any invoice variant will run exactly the same, with different html and some css to mimic older reports
08:24:31 *** boldstripe has joined #gnucash
08:24:33 <chris> but it'll remain to be seen how it pans out
08:25:45 <gjanssens> ok
08:26:05 <gjanssens> As for sxml, is it a default part of guile and if so starting from which version ?
08:27:11 <chris> looks like 2.0 onwards
08:30:33 <gjanssens> That's certainly an interesting intermediate format to look into.
08:33:00 *** oozer has quit IRC
08:54:08 *** jralls has joined #gnucash
08:54:08 *** ChanServ sets mode: +o jralls
08:54:23 *** oozer has joined #gnucash
08:57:03 *** Dee has joined #gnucash
08:57:17 *** jralls has quit IRC
09:02:54 *** fabior has joined #gnucash
09:04:00 *** oozer has quit IRC
09:15:21 *** Agfarmer18 has joined #gnucash
09:26:47 *** Agfarmer18 has quit IRC
09:31:35 *** harshitaneja has joined #gnucash
09:45:01 *** xmaka has quit IRC
09:45:39 *** xmaka has joined #gnucash
09:51:10 *** fekepp has quit IRC
09:57:03 *** fabior has quit IRC
10:03:10 *** Robert8471 has joined #gnucash
10:03:49 *** Mechtilde has joined #gnucash
10:04:38 *** Robert847 has quit IRC
10:09:47 *** jralls has joined #gnucash
10:09:47 *** ChanServ sets mode: +o jralls
10:13:07 *** fabior has joined #gnucash
10:13:33 *** jralls has quit IRC
10:16:54 *** bertbob has quit IRC
10:17:56 *** bertbob has joined #gnucash
10:22:39 *** fabior has quit IRC
10:50:09 *** fekepp has joined #gnucash
10:58:02 *** User__ has joined #gnucash
11:14:14 *** kael has joined #gnucash
11:19:04 *** User__ has quit IRC
11:20:02 *** ArtGravity has joined #gnucash
11:21:32 *** jralls has joined #gnucash
11:21:33 *** ChanServ sets mode: +o jralls
11:24:38 *** jralls has quit IRC
11:45:46 *** gncbot sets mode: +o fell
11:52:09 *** oozer has joined #gnucash
11:57:46 <fell> @tell jralls on the mailinglist in "GnuCash 3.1 Released" the section "Getting GnuCash for Windows and MacOS" exists twice.
11:57:46 <gncbot> fell: The operation succeeded.
12:00:03 *** Dee has quit IRC
12:02:11 *** kael has quit IRC
12:03:14 <fell> chris, can you write down your research results in a wiki page "Scheme". They might be of interest also for people not following IRC.
12:04:48 *** oozer has quit IRC
12:05:54 *** kael has joined #gnucash
12:13:00 *** gour_ is now known as gour
12:21:52 <fell> @tell jralls The recent path is "https://sourceforge.net/projects/gnucash/files/gnucash%20%28stable%29/3.1/" while you announce "https://downloads.sourceforge.net/sourceforge/gnucash/", which does not return the files.
12:21:52 <gncbot> fell: The operation succeeded.
12:24:28 *** kael has quit IRC
12:26:27 *** jralls has joined #gnucash
12:26:28 *** ChanServ sets mode: +o jralls
12:27:44 <gjanssens> jralls: fell left you a couple of messages
12:29:36 *** jralls has quit IRC
12:36:00 <fell> I still try to find out, why TP missed the last gnucash.pot files.
12:37:20 <gjanssens> fell: I believe because there were no changes compared to the previous version.
12:38:13 <fell> We are no longer on 2.7
12:38:59 <fell> and before they updated the statistics for each release.
12:39:08 <gjanssens> I had asked the TP maintainer about this and that's what he answerred me:
12:39:29 <gjanssens> "There were no string changes since 2.7.8. No need to bother the translators for this."
12:40:03 <fell> Benno or whomß
12:40:12 <gjanssens> Benno
12:40:13 <fell> ?
12:40:40 <gjanssens> That was in a private mail conversation on May 12
12:40:50 <fell> OK
12:40:58 <gjanssens> Because I also had noticed the pot file didn't change
12:41:49 <gjanssens> I probably should have mentioned this here or on gnucash-devel :(
12:41:55 <gjanssens> Did so now :(
12:42:01 <fell> NP
12:42:03 <gjanssens> eh :)
12:51:22 *** kael has joined #gnucash
12:53:03 *** jralls has joined #gnucash
12:53:04 *** ChanServ sets mode: +o jralls
12:55:59 *** kael has quit IRC
13:01:25 <jralls> hampton: Welcome back. The notes problem was https://bugzilla.gnome.org/show_bug.cgi?id=796369.
13:01:25 <gncbot> jralls: Sent 1 hour and 3 minutes ago: <fell> on the mailinglist in GnuCash 3.1 Released the section Getting GnuCash for Windows and MacOS exists twice.
13:01:26 <gncbot> jralls: Sent 39 minutes ago: <fell> The recent path is https://sourceforge.net/projects/gnucash/files/gnucash%20%28stable%29/3.1/ while you announce https://downloads.sourceforge.net/sourceforge/gnucash/ , which does not return the files.
13:03:48 <jralls> fell: I send Benno a link to every release tarball, it's up to him to maintain the TP website.
13:05:25 <fell> jralls: gjanssens asked him and there were no changes in gnucash.pot.
13:05:57 *** kael has joined #gnucash
13:08:51 <jralls> fell: It's correct that there haven't yet been any changes in gnucash.pot other than a few line numbers since 2.7.8. I dunno if that necessarily means that the website should show 2.7.8 as the latest release.
13:09:37 <fell> but a 3.1 tarball
13:10:27 <fell> And I missed the statistical data
13:11:40 <gjanssens> jralls: I did ask the translation project about this early May. I see I even cc'd you on that message and you should have received Bennos reply as well.
13:12:01 <jralls> gjanssens: Yes, I remember.
13:12:07 <gjanssens> (The one I repeated here 40 mins ago)
13:12:57 <gjanssens> So apparently for the TP that means 2.7.8 should be shown as latest release...
13:13:41 <gjanssens> Perhaps we should insert at least one translation change in each release to work around this ;-)
13:14:13 <fell> gjanssens, but it would had saved me a few hours, if you had cced me or devel,too-. ;-)
13:16:27 <gjanssens> True hence my :(
13:17:25 <gjanssens> I had cc'd jralls as he's release manager and wanted to be sure the missing potfiles was not due to a change or omission on our part. If it were jrall would no doubt have told me then.
13:18:23 <gjanssens> By the way, you could have asked around before starting a few hours of search ;p
13:18:32 <jralls> Yeah, more of a commission on my part as I told Benno when I sent him the links that there had been no string changes.
13:33:48 <jralls> gjanssens: Are you OK with merging 363 (enable div/span) into maint?
13:35:58 <gjanssens> jralls: considering chris is dutifully adding tests I'm ok with this.
13:36:13 <jralls> ;-)
13:36:36 <gjanssens> I haven't read through the PR myself so I trust you did.
13:37:19 <jralls> Yes, it's pretty mild. It adds the ability without actually using it yet, a nice separation.
13:39:22 <gjanssens> Indeed.
13:45:10 <jralls> As for mixing of data and layout, whether by doing the layout with hand-coded scheme or eguile-embedded html, it's still ugly and hard to maintain, never mind customize. We've discussed before having an intermediate data-only format that then gets laid out separately and not necessarily in html.
13:46:40 <jralls> For example, the recurring complaints about pagination in PDFs arise because html doesn't understand pagination.
13:55:47 <gjanssens> Indeed.
13:55:52 <gjanssens> Got to run now...
13:55:54 *** gjanssens is now known as gjanssens_afk
14:19:57 *** ArtGravity has quit IRC
14:24:19 *** badger92 has joined #gnucash
14:35:19 *** Mechtilde has quit IRC
14:41:07 *** Mechtilde has joined #gnucash
14:45:11 *** Mechtilde has quit IRC
15:31:05 <jralls> chris: I don't understand the status of https://github.com/Gnucash/gnucash/pull/316. Is it ready for review?
15:42:39 *** fabrice has joined #gnucash
15:55:35 *** GabrieleV has joined #gnucash
16:04:44 *** oozer has joined #gnucash
16:13:17 *** eagles0513875 has joined #gnucash
16:24:11 *** gjanssens_afk has quit IRC
16:47:16 *** fekepp has quit IRC
16:53:29 *** gour has quit IRC
16:55:58 *** fekepp has joined #gnucash
16:59:07 *** fekepp has quit IRC
17:01:07 *** fekepp has joined #gnucash
17:09:22 *** grammata has joined #gnucash
17:12:33 <grammata> Hi, I have a few invoices regarding purchases for my company. All these purchases I made without paying VAT. However, my accountant recommends I still add the VAT in Assets (to my advantage). How would you recommend I go about this task using a split transaction? Thanks in advance.
17:13:46 <grammata> I'm still a bit confused about split transactions, to be honest.
17:16:13 *** fell_laptop has joined #gnucash
17:25:44 *** grammata has quit IRC
17:33:12 *** mipesom has quit IRC
17:43:51 <chris> jralls #316 is a fairly big change according to gjanssens - it combines all charts into 1 and upgrades jqplot to chartjs. from my POV it's fine (just needs a rebase) but I think gjanssens has some reservations about upgrading chartjs.
17:44:11 <chris> I don't think I can test easily because result is html+js
17:45:12 <chris> well, rather, tests can become a silly "open html-chart object, add data, check data exists" type - ideally we'd want some tests to test resulting pixels
17:45:19 <chris> but this is out of my league
17:46:00 <jralls> Can you separate out changing to chartjs and consolidating reports into separate PRs? Does consolidating reports depend on chartjs features not in jquery?
17:46:55 <chris> in theory they can be separated out, but it's fairly hefty work because I didn't really analyze how jqplot expects its input... I just built up for chartjs incrementally until *ALL* of jqplot options were 'converted'
17:47:01 <jralls> As for testing for pixels, no, that makes for very fragile tests.
17:47:31 *** mipesom has joined #gnucash
17:47:47 <jralls> You can certainly test that the generated html and js are what you expect.
17:48:52 <chris> hmm doable, just annoying
17:49:37 <jralls> The testing or the splitting?
17:50:03 <jralls> Oh, sorry, missed the comment in the middle.
17:50:32 <chris> testing... needs extracting json options and analyzing... but given I tightly control input I guess I know what to expect
17:50:57 <jralls> Exactly. That's normally how unit tests work.
17:51:48 <chris> ok. it'll have to be over next few weeks. i'd be happy to not have to generate jqplot option generator tbh; this thing is evil
17:52:20 <chris> or rather assumes antiquated terminology just like gnucash lol
17:53:23 <jralls> gjanssens's issue may be that there are too many moving parts in the PR for him to keep track of what's doing what. That makes it really hard to review.
17:54:32 <chris> sure... it started as an experiment to merge line and bar chart together, and ended up as a complete chart generator and upgrade to chartjs
17:55:03 <chris> perhaps the best review is to use and experiment...
17:55:43 <chris> I'll add onto TODO in any case
17:56:03 <jralls> Looking over the commit list, ISTM the first 4 should stand on their own. Then 38a7083 could be a second. After that it looks like they bounce around a bit.
17:58:55 <jralls> Was html-barchart -> html-chart a simple rename or are there code changes in there too?
18:00:00 <chris> iirc only variable name changes, and occasionally something gjanssens changed in linechart and forgot to update in barchart
18:00:37 <chris> eg something about refreshing the xaxis tick interval was updated to either bar or line only by mistake
18:00:58 <chris> most of these were removed because chartjs does them automatically
18:07:56 <jralls> OK. Of the post-chartjs changes that's the only one that might be hard to review because of renaming-and-editing in one go so there isn't a usable diff to show the edits.
18:11:01 <jralls> jqplot->chartjs is enormous but that's only because it's replacing one pile of dependency code with another. There's not really anything to review there.
18:16:56 <jralls> It looks like the chartjs API is similar enough to the jqplot one that the changes in our code are pretty minimal. Is that right or am I missing the changes because they're buried?
18:17:43 <chris> overall it's the same - create an options object, send as a parameter to the js function.
18:18:01 <chris> if a facility requires customized string, send a callback
18:18:19 <chris> chartjs needs fewer callbacks because it's more sophisticated
18:18:47 <chris> the only part that worries me is testing on windows - i don't know if all features are usable on webkit1
18:19:28 <chris> ^^^^ options object specifies all chartjs input ie data + presentation choices
18:22:27 <jralls> So those options object changes are buried in 29ec994 because you reworked one of the charts to use chartjs and at the same time covered all of the chart types, then deleted the others. That's very hard to review.
18:23:31 <jralls> Is the reason for the different chart types with jqplot because each type needs its own callbacks?
18:25:55 <chris> not really... at a basic level they both are called similarly
18:26:47 <chris> in the past I guess there was complete aversion to renaming/refactoring, so when barchart was developed, and someone wished linechart, they didn't wish to stomp over prior dev work
18:27:36 <jralls> OK. Fortunately you have no such aversion. ;-)
18:29:26 <chris> I have aversion to wastage!
18:29:39 <jralls> So to move the ball forward on this I think you need two preliminary PRs, one for net-chart and one for html-chart, both still with jqplot. Maybe do the price-scatter cleanup as a third pre-chartjs PR.
18:30:42 *** fekepp has quit IRC
18:31:32 <chris> ok should be fine for this weekend
18:33:16 <jralls> Then with that done a three-commit PR: 1. Add chartjs. 2. Change the GC code to use that instead of jqplot. 3. Remove jqplot. That shows that chartjs simplifies our code, a clear benefit on top of chartjs being currently maintained while jqplot isn't.
18:35:37 <chris> back to invoices, I think I'll rebranch invoice-refactoring-and-merging (incomplete) on top of test-invoice (complete)
18:36:56 <chris> I'll do these in the weekend I think
18:37:29 <jralls> I don't see either of those branches in your github.
18:39:16 <chris> just pushed maint-test-invoice
18:43:30 <chris> I'll be off for short sleep now; if travis is happy then feel free to review/merge
18:43:50 <jralls> OK, good night.
18:44:01 *** pilotauto has joined #gnucash
19:50:16 *** warlord has joined #gnucash
19:54:26 <jralls> warlord: Home at last or still enroute?
20:03:56 *** harshitaneja has quit IRC
20:04:34 *** harshitaneja has joined #gnucash
20:08:46 *** harshitaneja has quit IRC
20:09:25 *** harshitaneja has joined #gnucash
20:10:50 *** g5pw has quit IRC
20:16:02 *** g5pw has joined #gnucash
20:19:11 *** g5pw has quit IRC
20:23:56 *** g5pw has joined #gnucash
20:29:10 *** harshitaneja has quit IRC
20:29:45 *** harshitaneja has joined #gnucash
20:33:46 *** harshitaneja has quit IRC
20:34:25 *** harshitaneja has joined #gnucash
20:34:49 *** g5pw has quit IRC
20:36:54 *** g5pw has joined #gnucash
20:39:57 *** g5pw has quit IRC
20:42:28 *** g5pw has joined #gnucash
21:04:54 *** kael has quit IRC
21:34:49 *** oozer has quit IRC
22:56:54 *** boldstripe has quit IRC
22:59:01 *** fell_laptop has quit IRC
23:30:42 *** harshitaneja has quit IRC
23:33:53 *** harshitaneja has joined #gnucash
23:35:56 *** kael has joined #gnucash