2022-08-14 GnuCash IRC logs
00:26:05 *** Wourroon has quit IRC
00:37:05 *** Wourroon has joined #gnucash
00:37:05 *** ChanServ sets mode: +v Wourroon
01:04:28 *** field^Mop has joined #gnucash
01:39:29 *** jchonig has quit IRC
01:41:35 *** jchonig has joined #gnucash
01:43:04 *** FH_thecat has joined #gnucash
01:52:52 *** FH_thecat has quit IRC
01:59:21 *** fell has quit IRC
02:00:34 *** fell has joined #gnucash
02:00:35 *** ChanServ sets mode: +o fell
02:24:32 *** field^Mop has quit IRC
02:27:28 *** field^Mop has joined #gnucash
02:30:30 *** field^Mop has quit IRC
02:30:50 *** Wourroon has quit IRC
02:32:29 *** NoobAlice has quit IRC
02:41:45 *** Wourroon has joined #gnucash
02:41:45 *** ChanServ sets mode: +v Wourroon
02:47:27 *** Gandalf has joined #gnucash
03:39:40 *** field^Mop has joined #gnucash
03:41:04 *** Wourroon has quit IRC
03:41:30 *** Wourroon has joined #gnucash
03:41:30 *** ChanServ sets mode: +v Wourroon
04:10:27 *** bertbob has quit IRC
04:10:40 *** field^Mop has quit IRC
04:33:31 *** bertbob has joined #gnucash
04:33:31 *** ChanServ sets mode: +v bertbob
04:50:16 *** raeburn has quit IRC
04:52:51 *** Aussie_matt has joined #gnucash
06:02:42 *** Wourroon has quit IRC
06:03:20 *** Wourroon has joined #gnucash
06:03:21 *** ChanServ sets mode: +v Wourroon
06:55:22 *** FH_thecat has joined #gnucash
06:56:11 *** field^Mop has joined #gnucash
06:56:23 *** FH_thecat has quit IRC
07:01:10 *** field^Mop has quit IRC
07:30:53 *** field^Mop has joined #gnucash
07:33:56 *** field^Mop has quit IRC
08:07:04 *** field^Mop has joined #gnucash
08:33:24 *** field^Mop has quit IRC
08:38:41 *** jmdaemon has quit IRC
08:48:22 *** field^Mop has joined #gnucash
08:52:23 *** field^Mop has quit IRC
09:18:44 *** FH_thecat has joined #gnucash
09:21:43 *** FH_thecat has quit IRC
09:23:07 *** field^Mop has joined #gnucash
09:27:07 *** field^Mop has quit IRC
09:56:42 *** field^Mop has joined #gnucash
10:05:42 *** field^Mop has quit IRC
10:19:25 *** field^Mop has joined #gnucash
10:43:11 *** FH_thecat has joined #gnucash
11:40:44 *** field^Mop has quit IRC
11:52:09 *** NoobAlice has joined #gnucash
12:26:28 *** paul has joined #gnucash
12:27:09 *** field^Mop has joined #gnucash
12:31:38 *** FH_thecat has quit IRC
12:35:10 *** field^Mop has quit IRC
12:59:18 *** FH_thecat has joined #gnucash
13:27:00 *** jralls has quit IRC
13:28:54 *** jralls has joined #gnucash
13:28:54 *** ChanServ sets mode: +o jralls
14:10:53 *** CDB-Man has joined #gnucash
14:10:53 *** ChanServ sets mode: +v CDB-Man
14:18:47 <jralls> Simon, any thoughts on the patch I posted to https://bugs.gnucash.org/show_bug.cgi?id=798262?
14:36:27 *** field^Mop has joined #gnucash
14:39:28 *** field^Mop has quit IRC
14:51:19 <Simon> jralls: as long as the currency of the transaction reflects what the user would get if they had entered it into the normal register in the same order, it would be correct
14:51:41 <Simon> alternatively (although obviously much more complex) it could ask the user what base currency they want if it's ambiguous
14:53:07 <Simon> I don't have any scheduled transactions in different currencies or any involving shares so it doesn't directly affect me
14:53:42 <Simon> you could instead tell them what the currency will be if it's ambiguous and let them reorder splits to get the desired currency, which is less UI work
14:53:46 *** Cork has quit IRC
14:54:22 <Simon> as long as it creates the transaction at all (which was my problem when using empty amounts) I don't think it matters what currency they get as long as it's consistent
14:54:22 <jralls> Simon, that is in itself ambiguous. When you use a normal register *that register's* currency sets the transaction currency. You could switch to split mode and enter two splits for accounts in a different currency and you'd have to enter an exchange rate for both splits to the transaction currency.
14:54:41 *** Cork has joined #gnucash
14:55:19 <Simon> ok, so there's no register to determine the currency
14:55:37 <Simon> I think the first split should determine the currency, which I think is what's proposed?
14:55:47 <jralls> Effectively. It's really a G/L register that doesn't have a currency.
14:56:27 <Simon> do you know why it was changed to be more restrictive in the first place?
14:56:47 <jralls> Yes, that's mostly what the code will do. If there are splits with amounts then it will be the first of those that determine the currency.
14:57:29 <Simon> and when none of them have amounts it's the first split?
14:57:37 <jralls> Right.
14:57:50 <jralls> As for why the change, https://bugs.gnucash.org/show_bug.cgi?id=797928.
14:58:22 <jralls> The case of no splits with an amount was an oversight.
14:59:19 <Simon> so in the case of https://bugs.gnucash.org/attachment.cgi?id=373857 what is it doing differently?
14:59:31 <Simon> first ignoring the zero-value split when determining the currency?
15:00:23 <Simon> my unit tests were an entirely new thing, so it would be worth extending it to add a test case for bug 797928
15:01:22 *** field^Mop has joined #gnucash
15:01:31 <Simon> it's especially suited to that since the expectation is that the SX is created without any prompting... which I think is all the tests can do
15:02:50 <Simon> is a zero-value amount treated the same as an empty amount?
15:03:25 <Simon> while it would allow for 3 levels of currency lookup (non-zero amount, zero amount, empty amount) I think it's unintuitive to do consider the empty amount different from a zero amount
15:04:23 *** field^Mop has quit IRC
15:04:30 <Simon> the alternative would be to let them explicitly set the transaction currency, with an "Auto" that tells you what the current decision is
15:05:18 <Simon> but I assume people are able to enter SX successfully today, it's just zero value amounts that cause problems
15:05:22 <jralls> Depends. When you're creating a transaction in the register having a 0 in the amount or price field will trigger the amount-price-value calculation.
15:05:52 *** FH_thecat has quit IRC
15:05:59 <Simon> ok
15:06:30 *** Cork has quit IRC
15:06:40 <Simon> how easy is it to add new things to the register UI for SX editing?
15:07:15 <Simon> if it had a column to tell you "this is going to prompt for a price" that might help people
15:07:25 <Simon> but then again they won't know how to stop it doing that
15:07:25 <jralls> Adding anything to the register UI is hard. Adding new things to the SX Editor is the same as for any other dialog.
15:08:12 <Simon> if the situation is that zero amount and empty amount are necessarily already different, then I'd go with the three-tier currency determination
15:08:20 <Simon> first split with: non-zero amount, zero amount, empty amount
15:08:41 <Simon> i.e. the more you enter the higher priority it is for amounts
15:08:55 *** Cork has joined #gnucash
15:09:12 <Simon> because entering an amount is going to trigger a calculation anyway?
15:10:10 <Simon> presumably it would be possible to just run the SX when you save it and find out if it's going to prompt for something?
15:10:42 <Simon> that would let people know they've made an SX that is going to prompt when run so that they can try to fix it immediately
15:11:01 <jralls> It's more complicated than that because of the differences between stock, currency, and SX editor registers and the different ways you can create SXes.
15:12:37 <jralls> Interesting idea to run the SX in the editor, maybe with a preview button that pops a dialog showing what the realized transaction would look like.
15:34:10 *** field^Mop has joined #gnucash
15:37:12 *** field^Mop has quit IRC
15:47:56 *** fell has quit IRC
15:51:39 *** fell has joined #gnucash
15:51:39 *** ChanServ sets mode: +o fell
15:55:55 *** field^Mop has joined #gnucash
15:58:57 *** field^Mop has quit IRC
17:06:41 *** Gandalf has quit IRC
17:09:00 *** field^Mop has joined #gnucash
17:14:00 *** field^Mop has quit IRC
17:26:36 *** NoobAlice has quit IRC
17:28:16 *** NoobAlice has joined #gnucash
17:28:16 *** ChanServ sets mode: +v NoobAlice
19:10:46 *** ericdm has quit IRC
19:11:19 *** ericdm has joined #gnucash
19:42:37 *** jmdaemon has joined #gnucash