2023-06-17 GnuCash IRC logs
00:23:01 *** gour has joined #gnucash
00:23:01 *** ChanServ sets mode: +v gour
00:52:46 *** NoobAlice has quit IRC
01:26:23 *** fell has quit IRC
01:27:41 *** fell has joined #gnucash
01:27:41 *** ChanServ sets mode: +o fell
01:40:48 *** gour has quit IRC
02:07:26 *** gour has joined #gnucash
02:07:26 *** ChanServ sets mode: +v gour
02:16:10 *** gjanssens has joined #gnucash
02:16:11 *** ChanServ sets mode: +o gjanssens
02:31:16 *** gandalf has joined #gnucash
03:28:57 *** gandalf has quit IRC
04:02:29 *** sunyibo has quit IRC
04:09:48 *** gour has quit IRC
04:14:35 *** gour has joined #gnucash
04:14:35 *** ChanServ sets mode: +v gour
04:52:08 *** gandalf has joined #gnucash
04:55:12 *** puck has quit IRC
04:55:20 *** puck has joined #gnucash
05:36:51 *** sunyibo has joined #gnucash
05:36:51 *** ChanServ sets mode: +v sunyibo
05:53:21 <Simon> the transction description feature still isn't up to 4.x standard
05:53:49 <Simon> if the column is too narrow to show the full description then there's a horizontal scroll bar and you can't see the whole description
05:54:27 <Simon> so if I'm editing a description where the last 3-4 characters vary then I can't see what the most recent one was
05:55:16 <Simon> previously it would have shown the auto-completion option inline
05:58:50 <Simon> in one case with shorter descriptions it is pre-scrolled to the right...
05:59:07 <Simon> in both cases the scroll area is much wider than the maximum visible transaction
06:22:33 *** gandalf1 has joined #gnucash
06:23:00 *** gandalf has quit IRC
06:23:00 *** gandalf1 is now known as gandalf
06:59:42 <chris> Simon: bob-IT isn't on IRC so it'll be better to comment on the PR or a bug report
09:46:56 *** NoobAlice has joined #gnucash
09:46:56 *** ChanServ sets mode: +v NoobAlice
10:39:12 <Simon> it's not directly related to Bob's PR
10:39:22 <Simon> also, 4.x is *worse*
11:00:48 <Simon> https://bugs.gnucash.org/show_bug.cgi?id=798960
11:02:42 *** Cork has quit IRC
12:18:58 *** palerider has joined #gnucash
12:25:56 *** palerider has left #gnucash
13:41:02 *** miklcct has quit IRC
13:46:10 *** miklcct has joined #gnucash
13:46:10 *** ChanServ sets mode: +v miklcct
14:00:28 *** ns12 has quit IRC
14:01:45 *** ns12 has joined #gnucash
14:49:58 *** CDB-Man has joined #gnucash
14:49:58 *** ChanServ sets mode: +v CDB-Man
15:34:57 *** Cork has joined #gnucash
16:18:29 *** gandalf has quit IRC
16:37:56 <CDB-Man> every time I try and create or update an expense barchart, gnucash crashes... what's the best way to diagnose?
16:37:56 <gncbot> CDB-Man: Sent 5 days, 21 hours, and 19 minutes ago: <chris> the upgraded stock_transaction assistant from tomorrow's build will be able to handle the ROC(reclassification) and ND(dividend). I'll upgrade IFRS report soon.
16:38:05 <CDB-Man> updating the config of an existing chart crashes
16:38:13 <CDB-Man> creating a new expense barchart also crashes
16:39:25 <Simon> I'm not familiar with debugging on Windows outside of cygwin
16:39:29 <CDB-Man> the initial load of the chart, ie after launching my gnucahs file, then going to my existing expense barchart, that works fine
16:39:44 <Simon> can you reproduce it with a file that you can share?
16:40:37 <CDB-Man> i can try
16:41:02 <Simon> is the the "Income & Expense Barchart" report or something else?
16:41:58 <CDB-Man> even using a simpler file crashes, so I dont think its do do with any data. its the "Expense Chart" report
16:42:59 <CDB-Man> ... okay it seems to be all reports
16:43:14 <CDB-Man> just tried to configure a differnet chart
16:43:23 <CDB-Man> gnucash-5.1-2023-06-17-git-5.1-275-g2c4a096b9c+.setup.exe -- it may be a build issue
16:47:53 <CDB-Man> I have filed bug 798961 https://bugs.gnucash.org/show_bug.cgi?id=798961
16:48:20 <Simon> can you try previous nightlies?
16:49:05 <CDB-Man> I can do a rewind sure
16:49:15 <Simon> do you know which one you were using before?
16:49:33 <Simon> you should check that works and then try the build in the middle
16:49:38 <CDB-Man> gnucash-5.1-2023-06-04-git-5.1-137-g4beb382c16+.setup.exe -- but dont recall if i made any report config edits that week
16:49:41 <CDB-Man> will rewind to then
16:49:45 <Simon> repeat until you know which one breaks it
16:50:12 <CDB-Man> was hoping to avoid the iterative testing if someone happened to know of a recent change in the config screen....
16:50:34 <CDB-Man> chris: https://bugs.gnucash.org/show_bug.cgi?id=798961
16:51:22 <Simon> there's nothing obvious relating to reports
16:51:37 <Simon> there's a load of refactoring stuff
16:52:04 <CDB-Man> gnucash-5.1-2023-06-17-git-5.1-275-g2c4a096b9c+.setup.exe -- my prior build has no issue
16:52:32 <CDB-Man> err
16:52:35 <CDB-Man> I mean this one: gnucash-5.1-2023-06-04-git-5.1-137-g4beb382c16+.setup.exe
16:53:06 <CDB-Man> just reinstalled that build
16:53:35 <Simon> if you try 2023-06-15 it won't have the refactoring commits, but dividing the builds by half is the most efficient if you're searching all of them
16:56:15 <CDB-Man> downloading 2023-06-15
16:56:21 <Simon> https://wiki.gnucash.org/wiki/Windows_Debugging
16:56:46 <Simon> I think the logging is unlikely to be useful
16:56:58 <Simon> there may be something in the trace file
17:01:27 <CDB-Man> the last few lines are just account balance lookups
17:01:28 <CDB-Man> * 16:37:24 INFO <gnc.account> [xaccAccountGetBalanceInCurrency] baln=3514935/100
17:01:28 <CDB-Man> * 16:37:24 INFO <gnc.account> [xaccAccountGetBalanceInCurrency] baln=3514935/100
17:01:28 <CDB-Man> * 16:37:24 INFO <gnc.account> [xaccAccountGetBalanceInCurrency] baln=0/1
17:01:28 <CDB-Man> * 16:37:24 INFO <gnc.account> [xaccAccountGetBalanceInCurrency] baln=0/1
17:01:29 <CDB-Man> * 16:37:24 INFO <gnc.account> [xaccAccountGetBalanceInCurrency] baln=0/1
17:01:31 <CDB-Man> * 16:37:24 INFO <gnc.account> [xaccAccountGetBalanceInCurrency] baln=-175794/100
17:01:33 <CDB-Man> * 16:37:24 INFO <gnc.account> [xaccAccountGetBalanceInCurrency] baln=-175794/100
17:01:35 <CDB-Man> * 16:37:24 INFO <gnc.account> [xaccAccountGetBalanceInCurrency] baln=-175794/100
17:04:41 *** gour has quit IRC
17:09:05 <CDB-Man> well, the 2023-06-15 version works
17:09:10 <CDB-Man> so the refactoring is the issue
17:13:48 <jralls> CDB-Man, I think I just reproduced your crash: Does it crash for you when you press the apply or ok button on the options dialog?
17:14:48 <CDB-Man> did not explicitly press the OK button, confirming apply definitely crashes
17:14:55 <CDB-Man> the "child accounts" button also crashes
17:15:08 <CDB-Man> so now that you mention it, to seems a button press is what triggers the issue
17:15:21 *** gjanssens has quit IRC
17:57:34 <Simon> it doesn't crash creating a report for me, but it does for "OK" on the options
17:57:50 <Simon> 0x00007ffff76b351e in get_selected_accounts_helper (s_model=0x55555d9926e0, s_path=0x555557e15090, s_iter=0x7fffffffa6c0, data=0x7fffffffa750)
17:57:52 <Simon> at /home/simon/src/gnucash/gnucash/gnome-utils/gnc-tree-view-account.c:1531
18:01:48 <Simon> not quite sure why it's crashing because nothing on that line is invalid until executed, unless filter_fn is invalid
18:03:11 <Simon> gtvsi->view->filter_data is invalid
18:05:11 <Simon> gtvsi->view looks invalid too
18:06:22 <jralls> Yup, it's https://github.com/Gnucash/gnucash/commit/5ea2e4e78e5310c5739905f15ca7647b68061677. A bunch of variables need to be nulled in the GObject init function.
18:07:32 <jralls> I guess G_DEFINE_TYPE_WITH_PRIVATE takes care of that for the private values, though it isn't documented.
18:30:07 <jralls> Hah! That's not it, it's https://github.com/Gnucash/gnucash/commit/5ea2e4e78e5310c5739905f15ca7647b68061677#diff-148ca67bcc574626cbd2ebe129c0fccdb32376976bce22a4e2228320ef016c5fL1587: He removed instead of replacing the initialization of gtvsi->view.
18:33:30 <jralls> Fixed, will be in tomorrow's nightlies.
18:33:48 *** fell has quit IRC
18:34:08 *** fell has joined #gnucash
18:34:08 *** ChanServ sets mode: +o fell
20:03:47 <CDB-Man> (y)
20:39:18 <CDB-Man> found a UX issue on the June 15 build...
20:39:55 <CDB-Man> when editing the transaction description (the green line), if I highlight text to replace and type over it, it deletes both the highlighted text and the character after it
20:40:55 <CDB-Man> example: 04/13/23 -- if I highlight "04" to type in "05", the resulting string is "053/23" -- the "/1" was also consumed
20:41:13 <CDB-Man> this only happens on the green line and not on the splits
20:41:25 <CDB-Man> i double checked to ensure the INS key is off
20:41:56 <CDB-Man> gnucash-5.1-2023-06-15-git-5.1-209-g4a60e01fcd+.setup.exe on this build
20:42:57 <jralls> CDB-Man: Can you put that on a comment on https://github.com/Gnucash/gnucash/pull/1625 so that Bob Fewell will see it?
20:43:39 <CDB-Man> will do
20:45:00 <CDB-Man> done
20:49:24 <jralls> Thanks
23:36:24 *** bertbob has quit IRC
23:37:46 *** bertbob has joined #gnucash
23:37:46 *** ChanServ sets mode: +v bertbob