2019-02-09 GnuCash IRC logs
00:16:07 *** Mechtilde has joined #gnucash
00:19:11 *** Mechtilde has quit IRC
00:28:08 <chris> wonder if jralls has opinion on book's accounting-period - I've amended to set start/end pair default as "today-12months+1day" to "today", and these absolute time64 dates are stored in book's slots.
00:29:26 <chris> this allows any date specifier to choose from relative dates - today / start-this-month / start-this-year / start-accounting-period / start-book-accounting-period
00:30:23 <chris> if we want to super-expand the concept of
00:30:48 <chris> 'book-accounting-period' then it must allow relative-dates and absolute-dates
00:31:03 *** Mechtilde has joined #gnucash
00:34:03 <chris> perhaps the book-accounting-period relative-dates list must then be from a limited list of 'this calendar year', 'last calendar year', 'Last UK Fiscal year ending 5/April' 'Last Fiscal Year ending 31/3'
00:34:07 *** Mechtilde has quit IRC
00:35:51 <chris> if Transaction Report selects dates 'start-book-accounting-period' to 'end-book-accounting-period' then the scheme magic will seek the appropriate period.
00:38:54 *** Aussie_matt has quit IRC
00:50:37 *** Mechtilde has joined #gnucash
00:53:41 *** Mechtilde has quit IRC
00:58:46 *** Mechtilde has joined #gnucash
01:01:50 *** Mechtilde has quit IRC
01:07:42 *** fell has quit IRC
01:09:00 *** fell has joined #gnucash
01:24:01 *** philippirrip has quit IRC
01:24:09 *** philippirrip has joined #gnucash
02:05:30 *** Mechtilde has joined #gnucash
02:08:34 *** Mechtilde has quit IRC
02:24:13 *** gncbot sets mode: +o fell
02:38:27 *** Mechtilde has joined #gnucash
02:43:10 *** gjanssens has joined #gnucash
02:43:10 *** ChanServ sets mode: +o gjanssens
03:07:07 <chris> gjanssens - see latest commits - warlord made me do it. allows preset list - compiled from CIA factbook, also custom start/end dates. if choosing from preset list, will always by default find the *last* FY.
03:19:32 <gjanssens> "warlord made me do it" LOL
03:39:38 * chris still not convinced. this approach does absolutely NOT make it any easier to do multi-FY reports.
03:40:02 *** boldstripe has joined #gnucash
03:50:01 *** boldstripe has quit IRC
03:53:02 *** boldstripe has joined #gnucash
04:03:38 *** O01eg has quit IRC
04:20:52 *** nimish has joined #gnucash
05:38:20 *** chris has quit IRC
06:04:33 *** nimish has quit IRC
06:11:42 *** chris has joined #gnucash
06:28:23 *** nimish has joined #gnucash
06:33:52 *** kapil___ has joined #gnucash
06:49:14 *** User has joined #gnucash
06:56:48 *** User has quit IRC
07:18:17 *** oozer has joined #gnucash
07:23:07 *** monkeyjuice has joined #gnucash
07:24:30 *** monkeyjuice has quit IRC
07:51:33 *** User has joined #gnucash
08:03:54 *** gour has quit IRC
08:04:23 *** gour has joined #gnucash
08:10:29 *** Jimraehl1 has joined #gnucash
08:13:18 *** Jimraehl1 has left #gnucash
08:27:06 <warlord> chris, it would be easier if there were an option "this FY" and "previous FY" in the list...
08:27:32 <chris> well exactly my point
08:28:13 <warlord> chris, the reports should be extended.. They currently have start/end current year, and start/end previous year. There should be a similar option for start/end current/previous FY.
08:29:11 <chris> well feature creepitis
08:29:34 <chris> let's say my book-accounting-period has options: absolute start/end or relative uk/oz/calendar
08:30:36 <chris> and report dates are also extended - (existing) start year/month/quarter/global-accounting-period must then be extended to (new) BOOK-accounting-period/global-PREV-accounting-period/BOOK-PREV-global-accounting-period
08:31:19 <chris> how many new options will gjanssens tolerate
08:33:17 <chris> plus, if the book-accounting-period are the set to absolute dates,
08:33:40 <chris> what would the meaning of start/end-book-accounting-period vs start/end-book-previous-accounting-period?
08:33:44 <warlord> I still do NOT believe that there should be fixed start/end dates.
08:34:07 <warlord> s/fixed/absolute
08:34:08 <chris> hmm
08:35:04 <chris> if that's the case, it resolves the question above
08:35:18 <chris> but it does reduce the utility of what i was aiming for
08:35:27 <warlord> How?
08:36:18 <chris> when I was starting out as contractor inUK, the accountant said I can choose any fy-end date I wish... 4/april 10/april 20/october... as long as i'm consistent the tax office don't mind
08:36:36 <warlord> Yes.....
08:36:43 <warlord> That still doesn't explain why you need absolute dates.
08:37:01 <warlord> It only means you need to be able to choose a month and day
08:37:13 <warlord> (which is what I've been saying all along)
08:37:15 <chris> ah
08:38:20 <chris> then the book-accounting-fy-end will be coded/stored as a time64 absolute day but only the d/m are useful?
08:38:43 <warlord> chris, yes
08:39:24 <chris> well as a coder i don't mind but it must be highlighted in bright RED that the year is ignored
08:39:36 <warlord> (or you can choose to store it as a d/m value... that doesn't really matter to me.. but only the d and m should be the only useful values.
08:40:15 <chris> the date control must handle 29/feb with a special case...
08:40:16 <warlord> Label: FY Start Month/Day FY End Month/Day
08:40:25 <warlord> Yes, indeed.
08:40:52 <warlord> And the law (in the US) says that if your end is 2/29, then on non-leap-years it is 2/28
08:41:08 <warlord> Which seems the obvious thing to do.
08:41:35 <chris> (obvious != globally acceptable)
08:42:23 <warlord> Oh? So if the end date is 2/29, what does YOUR locale say to do on a non-leap-year?
08:42:33 <warlord> (there really are only 2 options)
08:42:43 *** monkeyjuice has joined #gnucash
08:42:55 <chris> ok don't know but coding these assumptions is always fraught with danger
08:43:25 <warlord> Eh, it's always a start. Adding options for absolutely every choice before you know that the choice is required is also fraught with perils.
08:43:26 *** monkeyjuice has quit IRC
08:43:43 <warlord> Sometimes it's better to hardcode a reasonable choice and then only add the option if there is pushback.
08:43:48 <warlord> It really is a tiny corner case.
08:44:29 <chris> ok let's say we want to accept a custom d/m specifier... i'd be keen to have ONE control to ask for start or end, and infer the other
08:44:51 <chris> and the ONE control will be a full d/m/y and warn only the d/m is actually used
08:45:35 <warlord> I'm fine with that.
08:45:57 <warlord> (I think)
08:46:13 <warlord> I think short or long years are exceptional..
08:46:55 <warlord> (and if they are the start or end of a business, then it probably doesn't matter -- at least for a short year.. I'm not sure how we would handle a long year, but then again I don't know if you are allowed to have a long year.
08:47:22 <chris> gah if they have a short/long year then they should use absolute report dates in the report itself!!
08:47:55 <warlord> I think the "warning" is easily accomplished in the widget label and hover popup. E.g.: Choose your year end month/day:
08:48:02 <warlord> That's fair.
08:48:49 <chris> the biggest question is why no one has stepped up to fix this years ago? (and who could?)
08:49:48 <chris> in my current book I would usually use absolute global accounting periods, even when reporting for my UK and Australian tax returns
08:51:25 <warlord> Probably because most users use calendar years, so it's not an issue. I know I do, so it's never been something that's itched me.
09:02:41 *** User has quit IRC
09:03:31 *** jervin has joined #gnucash
09:18:35 <chris> warlord done
09:24:44 *** boldstripe has quit IRC
09:27:20 *** boldstripe has joined #gnucash
09:37:06 *** trying2Simplify has joined #gnucash
09:40:15 *** fabior has joined #gnucash
10:00:43 *** jervin has quit IRC
10:02:32 *** kapil___ has quit IRC
10:14:22 <chris> result: https://imgur.com/a/h8TbeXI
10:40:23 *** jervin has joined #gnucash
10:41:02 *** jervin has quit IRC
11:00:01 *** bertbob has quit IRC
11:06:55 *** bertbob has joined #gnucash
11:39:28 *** guiloma has joined #gnucash
11:50:31 *** nimish has quit IRC
11:54:38 *** nimish has joined #gnucash
11:56:31 *** boldstripe_ has joined #gnucash
11:56:46 *** boldstripe has quit IRC
11:56:46 *** boldstripe_ is now known as boldstripe
11:58:02 *** trying2Simplify has quit IRC
12:06:19 *** O01eg has joined #gnucash
12:07:10 *** boldstripe_ has joined #gnucash
12:07:25 *** boldstripe has quit IRC
12:07:25 *** boldstripe_ is now known as boldstripe
12:18:06 *** guiloma has quit IRC
12:37:14 *** trying2Simplify has joined #gnucash
12:38:24 *** trying2Simplify has quit IRC
12:45:34 *** gjanssens has quit IRC
12:52:18 *** gjanssens has joined #gnucash
12:52:18 *** ChanServ sets mode: +o gjanssens
13:01:32 *** gjanssens has quit IRC
13:05:58 *** gour has quit IRC
13:07:12 *** gour has joined #gnucash
13:17:45 *** fabior has quit IRC
13:20:43 *** mdf has joined #gnucash
13:34:52 *** gjanssens has joined #gnucash
13:34:52 *** ChanServ sets mode: +o gjanssens
13:58:22 *** boldstripe has quit IRC
14:04:49 *** frakturfreak has joined #gnucash
14:25:02 *** Mechtilde has quit IRC
15:01:32 <warlord> chris, thanks!
15:02:19 <jralls> chris: xaccAccountGetReconciledLastDate is wrapped, otherwise you wouldn't be able to get to it from Scheme. But it gets the default wrapper. You need a Swig override to have a Scheme version that does what you want, i.e. return #f if there's no KVP reconciled date or the time64 if there is.
17:53:55 <chris> jralls i'll have to watch you do this for me once
17:55:16 *** tienne has joined #gnucash
17:55:56 <jralls> chris: OK. Do I have what you want right?
17:58:07 <chris> yes
17:59:05 <chris> thx for fixing gnc-date.... will be out for a few hours
18:03:42 <jralls> It's only partly fixed. Or more correctly the locale is getting reset somewhere early in inner_main on Windows so that it doesn't use the right strings.
18:05:48 <jralls> So it's fixed for you but maybe not for cstim.
19:04:53 *** gour has quit IRC
20:19:19 *** nimish has quit IRC
21:12:40 *** jonas_ has joined #gnucash
21:14:55 *** jonas has quit IRC
21:22:06 *** nimish has joined #gnucash
21:28:00 *** frakturfreak has quit IRC
21:32:56 *** oozer has quit IRC
21:57:58 *** tienne has quit IRC
22:51:33 <chris> jralls: wrt wrapping xaccAccountGetReconciledLastDate - is it worth it? I'm quite comfortable with the two-step approach. are there other wrapped swigged functions? I think you're aware I don't want special treatment :)
23:58:28 <chris> warlord I've refined book-accounting-period... if user chooses the awkward 29/2 as fy-end, they must choose a suitable absolute date eg 29/02/2020 and the fy-end calculator will adjust accordingly according to 'common-sense logic'