2015-03-13 GnuCash IRC logs

00:15:46 *** GabrieleV_ has joined #gnucash
00:16:19 *** GabrieleV has quit IRC
00:16:20 *** GabrieleV_ is now known as GabrieleV
00:28:44 *** Krzysiek_K has quit IRC
00:55:46 *** GabrieleV_ has joined #gnucash
00:56:17 *** GabrieleV has quit IRC
00:56:17 *** GabrieleV_ is now known as GabrieleV
01:01:48 <fell_> warlord: I am waiting minutes on the push of a minor wiki edit
01:02:11 <fell_> while ping seems OK: rtt min/avg/max/mdev = 211.723/255.618/331.040/53.569 ms
01:11:30 <fell_> It took 20'
01:19:02 *** O01eg has quit IRC
01:24:35 *** ErKa has quit IRC
01:48:23 <fell_> ... to answer. Together with the git push answer error, which I got, it seems your upstream is sometimes very lean.
02:07:13 *** MechtiIde has joined #gnucash
02:32:18 *** GabrieleV_ has joined #gnucash
02:32:48 *** GabrieleV has quit IRC
02:32:49 *** GabrieleV_ is now known as GabrieleV
02:43:52 *** MechtiIde has quit IRC
02:51:23 *** fell_ sets mode: +o fell
02:55:17 *** Krzysiek_K has joined #gnucash
02:55:56 *** Krzysiek_K has quit IRC
03:23:50 *** cartsoftware has joined #gnucash
03:24:00 *** andreacosta has joined #gnucash
04:51:14 *** len has quit IRC
05:31:34 *** fabior has joined #gnucash
05:34:45 *** cartsoftware has quit IRC
05:36:11 *** cartsoftware has joined #gnucash
05:47:00 *** cartsoftware has left #gnucash
06:05:21 *** StuM has quit IRC
06:05:34 *** aqua___ has joined #gnucash
06:31:37 *** StuM has joined #gnucash
06:59:35 *** andreacosta has quit IRC
07:00:08 *** andreacosta has joined #gnucash
07:04:23 *** cartsoftware has joined #gnucash
07:15:23 *** Jimraehl1 has joined #gnucash
07:16:10 *** aqua___ has quit IRC
07:51:33 *** rickoehn has joined #gnucash
08:06:15 *** fabior has quit IRC
08:17:34 <warlord> Jimraehl1: Hey, how do you like google fiber?
08:18:47 <warlord> fell: My upstream is supposed to be 10mbps
08:19:05 <warlord> But there could be other things making restrictions.
08:19:28 <warlord> (I really need to refresh my hardware -- it's all getting a bit long in the tooth)
08:21:48 *** jralls has joined #gnucash
08:31:28 *** andy has quit IRC
08:37:17 *** andy has joined #gnucash
09:09:42 *** ErKa has joined #gnucash
09:14:53 *** StuM has quit IRC
09:38:37 *** jralls has quit IRC
09:45:40 *** ErKa has quit IRC
09:55:33 *** fabior has joined #gnucash
10:07:46 *** cartsoftware has quit IRC
10:07:58 *** andreacosta has quit IRC
10:09:09 *** andreacosta has joined #gnucash
10:09:26 *** cartsoftware has joined #gnucash
10:09:50 *** himaxx has joined #gnucash
10:18:33 *** himaxx has quit IRC
10:39:16 *** jralls has joined #gnucash
10:53:29 *** jralls has quit IRC
10:58:57 *** tpo has joined #gnucash
11:08:35 <tpo> I'm imported some transactions via libaqbanking (File->Import->MT940). This usually works well. This time however one of the transactions: 1) doesn't figure in the account register window 2) has an empty transfer field 3) can only be found via the search dialog
11:08:45 <tpo> s/I'm/I've/
11:09:49 <tpo> 4) it does not influence the balance
11:10:01 <tpo> so there are a few problems here:
11:10:13 <tpo> 1) there is a transaction, but it's invisible
11:10:50 <tpo> 2) a transaction has been imported, but there are a few things wrong with it and the user does not get notified
11:11:02 <tpo> 3) the balance is wrong
11:11:30 <tpo> 4) the transfer field got left empty
11:12:33 <tpo> libaqbanking v5.4.3beta-2+b1, gnucash 2.6.4, running on Debian jessie
11:13:18 <tpo> I think this situation is not good at under multiple angles
11:14:08 <tpo> Is there someone here currently that could comment on this?
11:14:38 <warlord> tpo: is there something special about this txn versus others you've imported?
11:16:26 <tpo> no, not that I can see. It's on an account that has a lot of traffic. The transaction is a "standard one" in that it's a train ticket that I bought, which happens all the time.
11:17:18 <warlord> If you jump to the expense account, does it have a balance there?
11:17:36 <warlord> Could you possibly have multiple currencies in your data file?
11:19:22 <tpo> warlord, just a second, I'll post a screenshot, then you can see
11:23:26 <tpo> warlord, here you go: http://tpo.sourcepole.ch/opaque/gnucash_bug.jpg
11:25:48 <tpo> warlord, I'll have a look whether there are multiple currencies present (they should not, because my bank doesn't allow me to do transactions directly in foreign currencies. Foreign transactions get exchanged to swiss francs automatically by the bank)
11:27:14 *** himaxx has joined #gnucash
11:28:40 <warlord> Hmm, the fact that you have all those little checkboxes in there seem to imply that you have imbalanced transactions.
11:29:14 <warlord> .. and you have no transfer accounts...
11:29:51 <tpo> as far as I can see, the one with the arrow is the only one without transfer account that is not balanced
11:29:58 <warlord> You'll notice that every transaction that has a "Saldo" of 0.00 also has a little box in the amount columns..
11:31:17 <fell_> Du solltest auf jeden fall die Buchung ins Ausgleichskonto (Imbalance) korrigieren.
11:32:13 <tpo> Yes, I can see that. I came into this particular view, via searching, finding that weird transaction, then clicking on "jump to transfer account" (Gegenbuchung in german)
11:33:00 <tpo> in my normal view of the account I do not see neither all those identical transactions, that are all there twice, except for the one with the arrow, nor do I see those little boxes.
11:34:01 <tpo> The fact, that all those transactions are there twice, is that because they really are there twice or is that just some particular kind of view, that is showing them twice?
11:34:50 <warlord> tpo: no, they are really there twice.
11:35:27 <tpo> why is that so?
11:35:47 <tpo> Is it possible, that by accident, I've imported the same MT940 file twice?
11:36:08 <warlord> Could be...
11:36:31 <warlord> It looks like you have a LOT of duplicated transactions in there.
11:36:42 <tpo> Would that have such a weird result - that is all transactions twice, but the doublettes without a transfer account?
11:36:55 <warlord> Exceptfor that 16.85 and 18.30
11:37:25 <warlord> Oh, I see the 18.30 one -- different date.
11:37:28 <tpo> the 18,30 is there twice too, only not consecutively
11:37:55 <warlord> I dont know how you'd get such a weird result..
11:38:06 <warlord> Is this your Bank Account View?
11:38:25 <tpo> I've changed the date of the first one (the one with the tag "n"), that's why they're not consecutive
11:38:46 <fell_> Try Ansicht->Vollständig to see the complete transaction.
11:39:37 <tpo> warlord, no, it's not my Bank Account View. It's the view I got into when I clicked on "Go to Transfer Account" in the Context menu of the transaction that I found via the Search function.
11:41:30 <tpo> fell_ I need to find out, why things broke, because I need to be able to rely on Gnucash / libaqbanking import. I think I shoud be able to fix things manually, but that is not the goal I'm aiming for here. What I want is to find out what went wrong and then to fix it in libaqbanking or in Gnucash.
11:43:00 <fell_> 1. duplicate import of mt94x.
11:43:29 <fell_> 2. while importing an accidental mouseclick?
11:43:58 *** jralls has joined #gnucash
11:44:37 <fell_> some import formats recognize duplicates, others not.
11:45:07 <tpo> When I scroll back/forth in the Bank Account View, then I see it's only transactions in January that are duplicated, which would support the duplicate import theory (i.e. last month I made a mistake and imported twice)
11:45:28 <tpo> But given the duplicate import:
11:45:56 <tpo> 1) It should be impossible to have transactions that are invisible
11:46:25 <tpo> all those transactions with the check boxes are normally invisible when I open that account
11:47:05 <tpo> 2) Gnucash or whichever part is doing the import should not allow such malformed transactions or should warn the user or something
11:47:12 <fell_> check Ansicht->Filter (all tabs)
11:48:05 <fell_> shure
11:48:11 <tpo> 3) why is there the the lone 6.85 transaction
11:48:15 *** ErKa has joined #gnucash
11:49:08 <tpo> fell_ wrt to the filter: everything is checked in "Status", thus all transactions should be shown, in "Date" is "show all transactions"
11:49:34 <tpo> So I'd think there's something broken in the View as well (?)
11:50:17 <warlord> tpo: what is "Post 85-611451-7 KK"?
11:50:33 <tpo> warlord, that's the name of the account
11:50:52 <tpo> (freely set by myself)
11:51:10 <warlord> tpo: that's your Bank account that you imported into?
11:51:16 <tpo> yes
11:51:34 <warlord> But you DONT see those transactions if you just normally open that account up from the Chart of Accounts?
11:52:07 <tpo> exactly, I do not see those transactions with the check boxes and without the transfer accounts
11:52:20 <warlord> clearly these transactions are not invisible; they just aren't visible from where you were looking for them... And they are broken .... that's what the [x] and 0.00 mean.
11:53:19 <warlord> Do you possibly have multiple accounts named "Post 85-611451-7 KK" in different parts of your hierarchy?
11:53:36 <tpo> warlord, that's not entirely correct - they are not visible when I open the account from the chart of accounts. I do not know how I'd get to this particular view if I had to.
11:54:52 *** misha_ has joined #gnucash
11:55:23 <tpo> warlord, no, as far as I can see, there are no two accounts "Post 85-611451-7 KK"
11:56:04 *** misha has quit IRC
11:56:29 <warlord> Without closing this tab, go to your CoA, and then double-click on the "Post 85-611451-7 KK" account and give me a screen shot?
11:57:01 <tpo> warlord, to reply more precisely to your question about the import of the transactions: it's not me that has imported the transactions into that particular bank account. It's the Import->MT940 function that did it by itself. I only do Import->MT940 and the transactions get associated with that particular account automagically.
11:57:58 <warlord> I'm pretty sure you need to tell the importer the account you're importing... And you need to (usually) map your transactions appropriately to the income/expense acounts.
11:58:14 *** jralls has quit IRC
11:59:13 <tpo> warlord, when I go to the CoA and doubleclick on the "Post 85-611451-7 KK" account, then I get automatically transferred to that Bank Account View tab I sent the screenshot of. In other word, no new tab is opened, that existing, open tab that is showing the transactions with the checked boxes is shown instead.
12:00:17 <tpo> warlord, the mapping happens the first time I import transactions for some particular account only. All consecutive imports have the transactions inserted into the correct account automagically.
12:00:41 <warlord> tpo: every transaction includes TWO accounts..
12:00:51 <warlord> You can map one (e.g. the Bank account)
12:01:03 <warlord> But the second account (the income/expense account) still needs to be handled during import.
12:01:22 <warlord> tpo: okay, close that tab, then go back to the CoA and double-click again. Then show me a screen shot if it's different.
12:01:39 <fell_> Yes, that automagical is the bayesian algorithm.
12:01:57 *** c0dewiz has joined #gnucash
12:02:25 <warlord> which still needs human verification in the cases where it's wrong.
12:03:10 <c0dewiz> greetinx
12:04:11 <fell_> Today I watched while importing allrelevant fields were on the right outside of the screen.
12:04:52 <fell_> You must know: There is something and I can scroll.
12:05:29 <fell_> This has became worse since the introduction of SEPA in europe.
12:05:44 <tpo> warlord, when I import a MT940 file, what happens is that I get shown a list of all the transactions that the importer has seen. Every transaction is tagged as either unrecognized, recognized or ignored. When the transaction is recognized, then the second account gets assigned automatically. When the transaction is not "recognized" then it gets assigned to "Ausgleichskonto-CHF" (= Reconciliation Account CHF)
12:06:47 <tpo> That "Ausgleichskonto-CHF" was not created by myself. It's either the importer or Gnucash that has created it.
12:06:49 <fell_> and you should change it before you finish the import
12:07:11 <fell_> else bayes will become more and more dumb.
12:08:10 <fell_> Ausgleichskonto means: you mad an error!
12:09:11 <tpo> fell_ I rather think that Ausgleichskonto means that the importer's bayesan magic did not find an account and thus is assigning it to the "default"
12:09:33 <tpo> I can *not* choose an account in the importer
12:09:57 <warlord> tpo: you think wrong. When you import into it you tell the importer "Yes, this *should* go into the imbalance account", which pushes it there further and further.
12:10:07 <fell_> doubleclick on the transaction
12:10:10 <warlord> tpo: Sure you can!! Double-click
12:10:30 <tpo> warlord, oh, I'll check right now...
12:14:10 <tpo> Uhm folks, I'm sorry, no I can't. When I double click, I get a pop-up, but I can't choose a target account there, I can only choose among similar prior transactions.
12:14:47 *** andreacosta has quit IRC
12:15:11 *** andreacosta has joined #gnucash
12:15:37 <warlord> tpo: you're in the import window?
12:15:51 <warlord> Where did you double-click?
12:16:29 <tpo> warlord, wrt to your request to close the Bank Account View tab and to reopen it. I did that, and was shown the same view as before (with the checkboxes). I closed that tab again, quit (terminated) gnucash, and restarted gnucash, dobleclicked on the account in CoA, and this time those transactions were *not* shown. I'll post a screenshot...
12:17:26 <tpo> warlord, yes, I'm in the import window. I can either doubleclick on the check box, or on the comment to the right of the check box.
12:18:07 <tpo> when I doubleclick on the trabsaction data itself (date, account, value, description) then nothing happens.
12:20:55 <tpo> here's the screenshot of the Bank Account View tab after restarting gnucash: http://tpo.sourcepole.ch/opaque/gnucash_bug_2.png
12:22:50 <warlord> Interesting. And now if you search for those transactions do you see them?
12:24:20 <tpo> warlord, by "search" you mean via Edit->Find or CTRL-F?
12:24:39 <warlord> Yes
12:24:51 <warlord> (from the CoA)
12:25:50 *** andreacosta has quit IRC
12:27:57 <tpo> Urgh! :-( ! No, I can't find them any more! To be more precise: from the CoA I'm searching for "Value" "is equal to" 6.85
12:28:07 <tpo> and I get an empty result list
12:28:15 <tpo> I'm stomped
12:29:00 *** cartsoftware has quit IRC
12:29:14 <warlord> Did you save before you quit?
12:29:34 <warlord> Or... the import was only half-completed?
12:29:38 <tpo> same if I search for transactions on the 2015-01-26: no result
12:29:42 <warlord> ... so you had "proto transactions" in there?
12:29:42 *** cartsoftware has joined #gnucash
12:31:53 <tpo> I can't remember if I saved *before quitting*. But I *did* save before I got here on IRC, because I was doublechecking transactions and reordering them, so they would match the online account view (which weirdly has sometimes slightly different dates on transactions)
12:33:41 <warlord> tpo: sounds to me like you were seeing the importer proto-transactions... Not sure why you were seeing those.
12:34:17 <tpo> but, warlord the import of those transactions has happened a month or so ago. *Not* today.
12:35:02 <tpo> In other words, before I had started gnucash *today*, those weird transactions were in gnucash's .xac file *allready*.
12:35:32 <tpo> To me it looks like those transactions got deleted only by having them displayed in that "special view" with those check boxes.
12:35:50 *** josePHPagoda has joined #gnucash
12:35:58 <tpo> Or maybe by importing *other* transaction, which I did today.
12:36:25 <warlord> tpo: I dont know.
12:38:59 *** MechtiIde has joined #gnucash
12:42:54 <c0dewiz> oh who've we got there? ;)
12:43:28 <c0dewiz> mechtiIde: just saw that you are one of the contributors of the german .po file
12:43:56 <c0dewiz> there would just one line that could have a small correction
12:44:00 <c0dewiz> +be
12:44:21 <MechtiIde> do it
12:44:32 <c0dewiz> me?
12:44:37 <c0dewiz> dunno how
12:44:47 <MechtiIde> or send me an email
12:44:53 <c0dewiz> i guess the reason is that i've landed on a mirror of the .po not on the original one
12:45:03 <c0dewiz> this is so short i could tell you the mod here
12:45:35 <c0dewiz> https://github.com/Gnucash/gnucash/blob/master/po/de.po
12:45:37 <tpo> warlord, I've checked the .xac, the weird 6.85 transaction is gone from there
12:45:38 <c0dewiz> line 26770
12:46:39 <fell> dont use master for translation!
12:46:48 <c0dewiz> hey, hey
12:46:55 <c0dewiz> I am not planning to change anything there
12:47:00 <c0dewiz> just saying WHAT should be changed
12:47:07 <c0dewiz> so stay calm
12:47:30 <warlord> c0dewiz: is that entry in the 'maint' de.po file?
12:47:37 <warlord> tpo: Ummmm....
12:47:50 <c0dewiz> warlord: no idea
12:48:13 <c0dewiz> just grabbed the first one i could get hold of to look up something, or even look for untranslated parts
12:48:28 *** jralls has joined #gnucash
12:49:21 <c0dewiz> line 26769, msgid "Crash the report" (original)
12:49:22 <MechtiIde> I look for line 26770 in maint
12:49:31 <c0dewiz> line 26770 in correct German would either be
12:49:36 <c0dewiz> 1) Verwirf den Bericht
12:49:37 <c0dewiz> or
12:49:40 <c0dewiz> 2) Bericht verwerfen
12:49:57 <c0dewiz> second one would probably do as well, needs not be imperative in German
12:49:59 <MechtiIde> this isn't from maint
12:50:24 <c0dewiz> excuse me I don't understand github lingo :P
12:50:45 <warlord> c0dewiz: it's not github, it's "GIT"
12:50:52 <c0dewiz> ok ok
12:50:55 <tpo> I need to run - I'll be back in 10min
12:51:10 <c0dewiz> well, just to cut it short, found a quirk, wanna fix it, checked translation file, that's all.
12:51:22 <c0dewiz> as i assumed the "pros" will know where to look/change/modify etc. pp
12:51:24 *** O01eg has joined #gnucash
12:51:25 <MechtiIde> c0dewiz, I found this translationIt's line 26539
12:51:26 <warlord> c0dewiz: we dont update the master po files
12:51:37 <c0dewiz> yeeeeeeeeeah
12:51:48 <warlord> So we're asking you whether the bug exists in maint, which is where translation updates go.
12:51:54 <c0dewiz> how often still? I just took one random one that seemed latest to me
12:51:56 <warlord> So please check if the bug exists in maint.
12:51:59 <MechtiIde> I myself don't want to change the translation
12:52:13 <warlord> maint gets updates as they come in.
12:52:15 <c0dewiz> mechtiIde: ok, granted
12:52:23 <fell_> 26769 in maint:
12:52:24 <c0dewiz> can't force you to lol
12:52:24 <fell_> msgid "Crash the report"
12:52:26 <fell_> msgstr "Verwerfe den Bericht"
12:52:32 <c0dewiz> fell: yes
12:52:35 <c0dewiz> Verwerfe is wrong.
12:52:36 <warlord> master wont get a translation update for at least another year, when we prepare for the 1.7/1.8 release cycle
12:52:43 <c0dewiz> "Wirf den Ball" not "Werfe den Ball".
12:52:53 <c0dewiz> that's very informal and grammatically wrong
12:53:19 <c0dewiz> warlord: i'm beginning to understand
12:54:18 <c0dewiz> although I dunno what this will correspond to in my Linux version (2.6.4)
12:54:36 <warlord> c0dewiz: maint is the 2.6 branch.
12:54:41 <c0dewiz> thanks ok
12:54:43 <warlord> so what's in maint is 2.6.5+
12:54:51 <c0dewiz> I see
12:55:03 <MechtiIde> fell, c0dewiz yes I look into DUDEN and "verwirf" is right http://www.duden.de/rechtschreibung/verwerfen
12:55:10 <c0dewiz> I know
12:55:32 <c0dewiz> but instead of the imperative, just use "... verwerfen", the infinitive instead of the imperative
12:55:44 <c0dewiz> you did that many times in similar translations
12:56:39 <MechtiIde> I only translated some few strings
12:56:48 <c0dewiz> "Aenderungen verwerfen" or "Verwirf Aenderungen" is the same even though the former is more used
12:57:03 <c0dewiz> (from my experience of translated software)
12:57:10 <c0dewiz> quantity never matters :)
12:57:28 <c0dewiz> mechtiIde: just picked you because you just rushed in :D
12:57:53 <c0dewiz> to tell the truth, my pet peeve on github is that you cannot contact users/developers directly
12:58:13 <c0dewiz> directly means, _from the page_. not having to use an external mailer
12:58:32 <c0dewiz> sourceforge was great in there, but github doesn't provide that
12:58:34 *** jralls has quit IRC
12:59:43 <c0dewiz> > you did that many times in similar translations <- blame english, i meant you in plural
13:01:52 <fell_> We have over 4000 messages and one guy added a dot to 500 of them...
13:02:28 <c0dewiz> ?
13:02:47 <c0dewiz> 4000 messages where?
13:02:58 <fell_> in Gnucash
13:03:03 <c0dewiz> ah i see
13:03:19 <c0dewiz> and the person that "fixed" it just flooded every translation with a period?
13:03:26 <c0dewiz> haha
13:03:59 <fell_> Before it was an infinitive as tooltip.
13:04:17 <fell_> Now it has become an incomplete sentence.
13:04:33 <c0dewiz> what is "it"?
13:04:38 <c0dewiz> you're speaking in riddles
13:04:59 <fell_> Most of the 500 strings.
13:05:16 <c0dewiz> almost thought you were referring ironically to _my_ grammar fix suggestion
13:05:49 <c0dewiz> but you're apparently pointing at something else
13:06:57 <tpo> (back), so I've looked at some older backup gnucash file, and the weird 6.85 CHF transaction is in there
13:07:10 <c0dewiz> bravo
13:07:21 <c0dewiz> who seeks shall find (amen)
13:07:41 <tpo> now I'm not sure how to proceed from here.
13:08:50 *** emp-- has quit IRC
13:09:29 <tpo> From a strictly user perspective: all transactions in a MT940 file *MUST* end up in gnucash's balance (otherwise the real balance and gnucash's accounting view will differ, which is should not happen)
13:09:51 <tpo> s/is//
13:09:52 <fell_> Did you watch somthing of interest in your http://wiki.gnucash.org/wiki/Tracefile ?
13:10:39 * tpo is checking
13:12:18 <tpo> fell_, I've got plenty of "* 17:09:53 WARN <aqbanking> swift940.c: 872: Bad extra data, ignoring (20141231009500543233840000000002)"
13:12:29 <tpo> (the last number is differing in each)
13:13:04 <tpo> then I've got four of these: "17:24:21 WARN <gnc.engine> Got a NULL guid_list but the QofGuidMatch is not MATCH_NULL (but instead 2). In other words, the list of GUID matches is empty but it must contain something non-empty."
13:13:31 <fell_> do you have log files in ~/.aqbanking which tell more?
13:14:30 <fell_> That is interesting.
13:14:32 <tpo> no log files there
13:14:38 <c0dewiz> !
13:15:21 <c0dewiz> mine are in ~/.aqbanking/backends/aqhbci/data/banks/de/<BLZ>/logs/*.log
13:15:33 <c0dewiz> (never do that on windows, you will break the path length limit ;))
13:15:37 <fell_> The bad 6.85 CHF are from a recent import?
13:16:03 <tpo> (it's interesting that the /tmp/gnucash.trace will aparently survive restarts? First entry at 17:06:43, restart was around 17:25)
13:16:46 <tpo> c0dewiz, I did a "find ~/.aqbanking" - no logs there
13:16:57 <fell_> because they get random siffixes under windows
13:17:00 <c0dewiz> ah ok that makes sense
13:17:05 <tpo> fell_ no they're from around the end of Febrauray
13:17:08 <tpo> February
13:18:11 <c0dewiz> absolutely no logs? this is peculiar.
13:18:18 <fell_> an beginning dot means, the entry is usually hidden. Enable show hidden files.
13:18:32 <tpo> fell_ I did a "ls -latr ~$my_accounting_dir/" and looked through the few most recent ones
13:18:49 <tpo> fell_ I'm on unix
13:18:55 <fell_> OK
13:18:59 <tpo> er Linux
13:19:21 <fell_> POSIX is enough ;-)
13:19:36 <c0dewiz> hehe
13:20:09 <fell_> I think about a bug report:
13:20:46 <fell_> export the account structure
13:21:17 <c0dewiz> must be bug report about aqbanking. aqbanking is _external_ module ; and logging is done by this module in its own folder, independently from gnucash stuff
13:22:08 *** mikee has quit IRC
13:23:20 <fell_> yes, see http://wiki.gnucash.org/wiki/Logging, http://wiki.gnucash.org/wiki/Aqbanking#Debugging ...
13:23:22 *** ErKa has quit IRC
13:23:54 <tpo> c0dewiz, I'm not sure. AQB is transmitting data to Gnucash, but it's Gnucash that must make sure that, even if "bullshit" is coming in, no "bullshit" will be able to land in the accounting. It's Gnucash that need to do basic sanitizing. Which as far as I understood did not happen correctly here. (?)
13:24:12 <c0dewiz> hmmm
13:24:24 <fell_> Yes, but to find the error q
13:24:42 <fell_> we probably need both logfiles.
13:25:10 <c0dewiz> righto
13:25:26 <fell_> In http://wiki.gnucash.org/wiki/Tracefile are the common log options
13:26:09 *** Krzysiek_K has joined #gnucash
13:26:52 <fell_> Enable all log option while importing the mt file and attach all (data and both log) files to bugzilla,
13:27:40 <fell_> perhaps you might remove personal information, passwords etc. before.
13:28:44 <c0dewiz> yeah
13:29:07 <c0dewiz> i almost dare say tpo has _gladly_ no log files in .aqbanking, as they are VERY wordy. full name and all
13:29:29 <c0dewiz> just checked one of mine for the fun of it
13:30:03 <tpo> I have log files from Gnucash the do show the problematic transaction in question
13:31:18 <tpo> slightly different question: that Bank Account View with the check boxes. How can I open such a view? Or is there a way to search for "proto-transactions" or "transactions that have only one account associated with them"?
13:32:28 <fell_> I don't know, how you got that view.
13:32:30 <tpo> My goal would be to clean the crap out of my accounting or respectively, find faulty transactions or respectively transactions, that got saved, but do not contribute to the balance and so make gnucash drift away from reality.
13:33:03 <tpo> fell_ I search for the amount of the not shown transactions and then clicken on "Gegenbuchung" and there I was
13:33:23 <fell_> usually the are in "Ausgleichskonto" and similar.
13:33:30 <tpo> but evidently if I don't know whether a transaction is missing I can't search for the amount...
13:33:51 <tpo> usually yes, but not in this case, since they do not have the other account set
13:34:07 <warlord> tpo: those proto transactions are generally only around in the middle of an import process. Once you finish the import process they shouldn';t exist anymore.
13:34:30 <fell_> You can store your file un compressed and use a texteditor.
13:34:34 <warlord> The issue is that the importer uses the same underlying "transaction" object internally, even before it's a real transaction.
13:34:37 <tpo> warlord, so no way to search for them? Not in the .xac either?
13:35:01 <warlord> IF they exist you could search in the xac. but if they exist you'd see them via Find or in the account register.
13:40:25 <tpo> Hmm... the transaction is actually present in the backup xac, the its reconciled-state of it's second split is "c" and the account is actually set to something existing...
13:41:37 <fell_> C=cleared, you got a message by the bank.
13:41:43 <tpo> So I guess at this point I'm not able to find broken transaction via the xac file
13:41:56 <tpo> ok, thanks
13:44:12 <fell_> If you open your bak file, open the bank account and set "Ansicht:Vollständig" you should see the other account.
13:45:49 <tpo> ok, thanks
13:46:42 <tpo> I've just tried a re-import, with just that one transaction selected, and it worked. I can't reproduce this way.
13:47:26 <tpo> However there's one bug that creeps in that way, that's reproducible: not-selected transactions get imported as empty lines.
13:48:05 <tpo> and dissapear on first click. Weird.
13:49:29 *** jralls has joined #gnucash
13:51:01 *** cartsoftware has quit IRC
13:59:56 *** ErKa has joined #gnucash
14:01:53 *** jralls has quit IRC
14:04:29 *** fell has quit IRC
14:08:45 <tpo> OK, I seem to be able to reproduce something
14:10:10 *** fabior has quit IRC
14:10:33 <c0dewiz> congrats :)
14:26:55 <c0dewiz> moving on
14:27:10 <tpo> :)
14:27:10 <c0dewiz> would anyone of you guys like to test out the filter bug i just found in the 2.6.4 linux versionß
14:27:35 <tpo> I might, go ahead
14:27:43 <c0dewiz> ok
14:28:52 <c0dewiz> 0. tpo do you have any accounts in your set that have a 0,00 balance?
14:29:09 <fell_> gn8!
14:29:12 *** fell_ is now known as fell_afk
14:29:15 <tpo> ah, i'd be so nice, if one could select more than one transaction at once: manually deleting hundreds of transactions: click-click-click-click...
14:29:33 <warlord> tpo: quit without saving.
14:29:41 <tpo> good nicht fell_ and thanks!
14:30:03 <kimmo2> warlord, good one
14:30:15 <tpo> uh, that will not delete the transactions, which I want to delete ... :-o
14:30:47 <tpo> c0dewiz, yes
14:31:16 <c0dewiz> 1. okay, so for a test, set Filter to [ ] Show zero total accounts
14:31:24 <c0dewiz> that means, you should hide them
14:32:18 <tpo> done
14:32:20 <c0dewiz> 2. now with one of the zero-total accounts, change the balance to something above zero. 1,00 currency units is ok.
14:33:35 <tpo> done
14:33:36 <c0dewiz> 3. hit refresh (i think that's ctrl-R on the keyboard, don't have GC open right now)
14:33:47 <c0dewiz> your test account WAS zero, but now is NO MORE zero.
14:34:05 <c0dewiz> that means, the refresh should (re-)show the account, but it does not
14:34:08 <tpo> and it's still invisible :-)
14:34:12 <c0dewiz> right
14:34:22 <c0dewiz> but i tell you how to make it invisible. it's silly
14:34:45 * tpo is seeing c0dewiz typing a bugreport...
14:35:01 <c0dewiz> revert filter settings as before, and then check box again
14:35:07 <c0dewiz> now everything looks as it should look
14:35:20 <tpo> yup
14:35:24 <c0dewiz> ha!
14:35:45 <c0dewiz> test account shows additionally to your ordinary ones, because the 0,00 was changed to 1,00.
14:36:12 <c0dewiz> bug report filed on march 01, but it takes a little bit too long for my taste <kimmo2> warlord, good one
14:36:16 <c0dewiz> bleh
14:36:22 <c0dewiz> https://bugzilla.gnome.org/show_bug.cgi?id=745385
14:36:24 <c0dewiz> this was more like it
14:36:45 * c0dewiz is just too windows'ified with this ctrl-v shit
14:37:17 * c0dewiz remarks that true posix'ers will only use middle mouse button to copy-paste
14:39:00 <c0dewiz> > but i tell you how to make it invisible. it's silly <-- yes that is really silly. should read VISIBLE lol
14:52:58 *** jralls has joined #gnucash
15:03:04 *** jralls has quit IRC
15:04:14 *** len has joined #gnucash
15:14:24 *** mikee has joined #gnucash
15:17:39 *** jlo has joined #gnucash
15:32:40 *** TommyT has joined #gnucash
15:37:28 *** GabrieleV_ has joined #gnucash
15:37:38 *** GabrieleV has quit IRC
15:37:38 *** GabrieleV_ is now known as GabrieleV
15:37:50 *** MechtiIde has quit IRC
15:51:07 *** josePHPagoda has quit IRC
15:54:32 *** jralls has joined #gnucash
15:55:40 <tpo> OK, here you are gentlemen all inclusive with bug description and quite radically reduced test data set: https://bugzilla.gnome.org/show_bug.cgi?id=746179
15:56:29 <tpo> So I'll have a break. Thanks a lot warlord for your (previous) patient explaining!
15:59:58 *** TommyT has quit IRC
16:00:38 <c0dewiz> speaking of bug reports...maybe I should have described that issue with the filter when fell was still awake...
16:00:52 <c0dewiz> afaik fell is a programmer too
16:01:21 <tpo> c0dewiz, just write a bug report
16:01:41 <c0dewiz> aha, so you missed the line where i wrote i already did that on march 01 ;)
16:01:57 <c0dewiz> https://bugzilla.gnome.org/show_bug.cgi?id=745385
16:02:09 <tpo> ah, no actually not, sorry, I was just not attentive at this moment... :-/
16:02:17 <c0dewiz> np, just it took me too long!
16:02:21 <tpo> :)
16:02:30 <c0dewiz> total silence for 2 weeks
16:02:54 <tpo> c0dewiz, you don't think you could fix that problem?
16:03:11 <c0dewiz> you mean the filter one? no i don't
16:03:29 <c0dewiz> i am a rookie when it comes to gc code
16:03:30 <tpo> your nick is c0dewiz and this doesn't look like a hard problem to me...
16:03:35 <c0dewiz> blergh
16:03:41 <tpo> gc == gnucash ?
16:03:59 <warlord> tpo: happy to help
16:04:11 <c0dewiz> my nick...yes i am a programmer and have a high experience also in bash scripting (incl some 4.0 quirks)
16:04:22 <c0dewiz> plus javascript
16:04:43 <c0dewiz> but to understand and fix this problem i must know about the basic structure of the gc tree
16:04:51 <c0dewiz> this is no 5-minute job
16:05:01 <c0dewiz> gc = gnucash yes
16:05:03 * tpo bows his head before warlord: you Sir, are awesome!!!
16:05:24 <warlord> I'm not sure I'd go THAT far..... ;)
16:07:06 <warlord> c0dewiz: just looking at your bug now.. This is from the Chart of Accounts page?
16:07:10 <tpo> I'm sure I would, I've been using gnucash for, umm ... 13 years now? And I was here twice in that time and you were there to help each time. If there's something that I'd consider a noble human trait to aspire to, then this is certainly one.
16:07:24 <c0dewiz> warlord: overview, yes
16:07:28 <c0dewiz> tree view
16:07:29 <warlord> LOL. Well, thank you.
16:07:42 <kimmo2> these guys are devoted
16:07:49 <kimmo2> I'm just procrastinating
16:08:03 <c0dewiz> heheh
16:08:03 <warlord> c0dewiz: I just want to verify -- you're talking about the Accounts Tab (Chart of Accounts)
16:08:33 <c0dewiz> gimme a min, i'll start up this thing
16:08:51 <tpo> fwiw warlord I've verified c0dewiz' problem just before
16:09:09 *** jralls has quit IRC
16:09:14 <c0dewiz> yes this was great tpo, because this way i know i was not making this up
16:09:29 <c0dewiz> nor would anyone else think that ;)
16:09:53 <warlord> glad to know it's reproducible.
16:10:06 <c0dewiz> btw, just as a naughty remark, i think gnucash has the same startup behavior as photoshop. and takes about the same time.
16:10:10 <warlord> I'm sure someone will look at it.
16:10:15 <kimmo2> I'd say that's just about account selection
16:10:21 <c0dewiz> telling you about 10987 modules it is loading lol
16:10:32 <kimmo2> there's no "all" accounts
16:10:34 <warlord> c0dewiz: LOL. There are only about 10 minutes.
16:10:45 <kimmo2> just "select all which are available now"
16:11:01 <kimmo2> warlord, c0dewiz's bug that is
16:11:10 <c0dewiz> warlord would you like me to repost the 4 steps?
16:11:19 <c0dewiz> to reproduce
16:12:11 <kimmo2> c0dewiz, do you visit the account selection in report options after creating the new account?
16:12:40 <c0dewiz> report options ???
16:12:55 <c0dewiz> this is _nothing_ about reports. nada. mitaan.
16:13:09 <kimmo2> ah, I'm such a report guy
16:13:14 <c0dewiz> guess so :P
16:13:18 <kimmo2> alas, the filters are not dynamic
16:13:37 <kimmo2> "apply filter" means apply to current state
16:13:52 <c0dewiz> they need not! but a ctrl-R refresh should at least bring up non-zero accounts that were previously hidden because they were zero
16:14:09 <kimmo2> that's arguable
16:14:41 <c0dewiz> but i'd like to hear a 'yes' or 'no' from warlord
16:16:53 <c0dewiz> my fault, warlord can't hear me when firing missiles from his tank ;)
16:17:05 <c0dewiz> too noisy
16:18:13 <kimmo2> heh
16:18:17 <c0dewiz> :P
16:18:32 * c0dewiz must agree with tpo that picking on nicks can be fun
16:20:01 <warlord> er, 10 MODULES... *brain fart*
16:20:13 <c0dewiz> lol
16:20:41 <c0dewiz> brain farts do not smell that'S the advantage of them
16:20:43 <warlord> Sorry, I'm off doing other work.
16:20:48 <c0dewiz> ok ok
16:20:50 <warlord> Anyways, I agree this is a bug.
16:20:51 <c0dewiz> nvm
16:21:07 <warlord> The CoA should update when there is new data.
16:21:34 <warlord> Although I bet I know the issue.
16:21:41 <c0dewiz> wow
16:21:47 <warlord> I bet it's due to the way the Query functon works for Accounts.
16:22:24 <c0dewiz> query in the sense of db query? (sql)
16:22:34 <warlord> ASSUMING the tree is using a query (it may, it may not).. but if it does, the issue is that queries for "all accounts" actually enumerates all accounts when the query is created; it wont add new accounts created later.
16:22:44 <warlord> query in the sense of QofQuery
16:22:50 <c0dewiz> ah ok
16:23:14 <warlord> I'm not familiar enough with the CoA code to know offhand that's how it handles these preferences.
16:23:19 <warlord> So I may be off base.
16:23:48 <c0dewiz> ah that stands for query object framework ... (qof sounds somewhat arabic)
16:24:00 <kimmo2> that would be my havent-seen-the-code guess as well, as that follows the same logic as the account selection for reports
16:24:59 <warlord> kimmo2: yeah, this issue shows up there, too.
16:25:29 <c0dewiz> haha we really should put up a glossary in "Gnucash lingo"
16:25:33 <kimmo2> if you know of it, it's pretty much a non-issue for reports, but for the CoA it should be dynamic
16:25:45 <kimmo2> a static enumeration doesn't really make sense there
16:26:16 <c0dewiz> i repeat myself again, if it is not dynamic a refresh MUST check for it
16:26:44 <c0dewiz> and it must not happen that a refresh ignores previously-zero-turned-non-zero accounts
16:26:46 <kimmo2> then again, also in the reports there _should_ be a "all applicable" option since for custom reports, the account selection is fixed, as opposed to basic reports which go through the default selection first
16:27:29 *** rickoehn has quit IRC
16:27:31 <kimmo2> as soon as I add a new account, I have to take an extra step in most reports
16:27:52 *** rickoehn has joined #gnucash
16:27:54 <warlord> c0dewiz: unfortunately a refresh does not rebuild the query.
16:27:57 <c0dewiz> ...because the new account is not recognized/considered immediately.
16:28:02 <c0dewiz> (guessing)
16:28:09 <warlord> and honestly there is no way to rebuild the query except going back to the options page
16:28:58 <kimmo2> sounds to me like this is a bug in the patches-welcome priority
16:29:04 <c0dewiz> again, if you toggle the filter back and forth the prev-zero (now non-zero) account DOES show up
16:29:18 <c0dewiz> so this silly step must trigger a kind of recheck
16:29:51 <warlord> c0dewiz: the issue is that QofQuery does not understand Accounts. It doesn't have a concept of them at all, an Account is just an object (like everything else). So the "SearchAllAccounts" part of QofQuery is just a wrapper around a "foreach-account: searchAccount".. Except that happens at Query SETUP, not at Query RUN.
16:29:56 <warlord> Refresh does a Query RUN
16:30:17 <warlord> c0dewiz: yes, because when you toggle the filter it rebuilds the Query (rerunning Query Setup)
16:30:18 <c0dewiz> 'the filter' being short for [ ] Show zero-balance accounts (or similar)
16:30:36 <warlord> I bet if you toggle any other filter it would also appear
16:30:49 <c0dewiz> Qof works in an abstract kind of way.
16:30:54 <c0dewiz> Reminds me of abstract classes.
16:30:56 <c0dewiz> :)
16:31:01 <c0dewiz> i got it
16:31:53 <warlord> So a simple test is whether toggling some other flag does the same thing?
16:32:11 <c0dewiz> right. but i think that won't do.
16:32:13 <c0dewiz> checking...
16:34:38 <c0dewiz> I'LL BE DAMNED.
16:34:56 <c0dewiz> messed with the "show hidden accounts" filter
16:35:21 <c0dewiz> on/off and abracadabra the account that was previously zero appeared...
16:35:50 <warlord> There you go.
16:36:01 <c0dewiz> you are right, it is "fixed" by *whatever* filter. no specific one.
16:36:50 <warlord> Right, because that rebuilds the Query Setup.
16:37:13 <c0dewiz> btw I've just learned another thing
16:37:30 <warlord> To fix it properly would require QofQuery to gain a knowledge of Accounts as "special" objects.
16:37:41 <c0dewiz> removing this bogus/test account brings up the dialog "this acc contains transactions. delete?"
16:37:48 <c0dewiz> OK BUT
16:38:27 <tpo> c0dewiz, I was going to say: gnucash is a ver worthy project to dive into, for it's wealth (overkill? muahaha) of technologies. Last time I walked into its lisp code and my being was transported into another dimension. But the hints that warlord gave you together with some flexing and stretchin of your "grep" muscles should make you appear in the right code spot.
16:38:29 <c0dewiz> when you delete this acc, it's not that the transaction is removed from the source account as well. it is just set to Imbalance
16:38:40 <c0dewiz> lisp???!
16:38:51 <c0dewiz> NEVER. i hate this language.
16:39:37 <c0dewiz> that's the one with the parentheses overkill no? end)))))))
16:39:39 <warlord> I can tell you exactly what file(s) you want..
16:39:48 <warlord> You want the qof query files in lib/qof
16:39:53 <c0dewiz> no language is harder to read than this ugly lisp BS
16:39:55 <warlord> (and they are 100% in C)
16:40:20 <c0dewiz> that sounds comfortable. my C knowledge is solid enough.
16:41:00 <c0dewiz> just out of curiosity, where does gc use lisp ??
16:43:47 <warlord> c0dewiz: reports, qif import, options, SX functions, and a few other places.
16:43:53 <c0dewiz> urghhh h h h
16:44:05 <c0dewiz> hopefully this will all be ported to python someday
16:44:07 <tpo> too bad, the code in question is in C: a great oportunity for enlightenment of a needing soul has passed by
16:44:10 <warlord> c0dewiz: nope
16:44:15 <c0dewiz> Python? --- YES! http://norvig.com/python-lisp.html
16:44:17 <c0dewiz> hah
16:44:30 <c0dewiz> these two beasts are closer than some would believe they are
16:45:42 <tpo> c0dewiz, /usr/share/gnucash/guile-modules
16:46:21 <c0dewiz> and if Python is an Audi Q8 then Lisp is a '69 Chevy ;) Noble, yes, but antique nonetheless.
16:47:21 <tpo> a code wizzard without lisp knowledge is like a turtle without a rabbit friend
16:47:47 <c0dewiz> very funny
16:48:00 <c0dewiz> lisp was no more part of my education. we were on ibm's (rexx)
16:48:11 <c0dewiz> so you can't put the blame on me, silly
16:48:30 <tpo> oh, rexx must be interesing as well...
16:48:39 <c0dewiz> it was. and pretty versatile
16:48:52 <c0dewiz> much neater than this lisp crud for sure
16:49:26 <c0dewiz> so were the mainframes :) and real productive people, no way to browse the net to gain some major distraction :P
16:50:30 <c0dewiz> rexx worked both on the mainframes and on amigas (!) the latter of which most people thought were gaming machines.
16:50:39 *** fell has joined #gnucash
16:50:39 *** gncbot sets mode: +o fell
16:58:27 *** cartsoftware has joined #gnucash
16:59:29 *** fell_afk has quit IRC
16:59:31 *** jralls has joined #gnucash
17:00:07 <tpo> good night everybody!
17:00:11 *** tpo has quit IRC
17:09:44 *** jralls has quit IRC
17:11:12 *** pacon has joined #gnucash
17:14:30 *** c0dewiz has quit IRC
17:50:15 *** StuM has joined #gnucash
17:52:40 *** josePHPagoda has joined #gnucash
17:54:56 *** rickoehn has quit IRC
17:59:03 *** Krzysiek_K has quit IRC
18:00:33 *** jralls has joined #gnucash
18:10:43 *** jralls has quit IRC
18:26:33 *** jralls has joined #gnucash
18:30:56 *** Simon has quit IRC
18:33:36 *** lmat has joined #gnucash
18:44:59 *** Simon has joined #gnucash
18:46:36 *** aqua___ has joined #gnucash
19:00:33 *** cartsoftware has quit IRC
19:38:59 *** GabrieleV_ has joined #gnucash
19:39:04 *** GabrieleV has quit IRC
19:39:04 *** GabrieleV_ is now known as GabrieleV
20:07:50 <warlord> @op jralls mikee
20:07:51 *** gncbot sets mode: +oo jralls mikee
20:23:02 *** GabrieleV_ has joined #gnucash
20:23:11 *** GabrieleV has quit IRC
20:23:12 *** GabrieleV_ is now known as GabrieleV
20:31:25 *** pacon has quit IRC
20:41:37 *** ErKa has quit IRC
21:42:33 *** GabrieleV_ has joined #gnucash
21:42:39 *** GabrieleV has quit IRC
21:42:39 *** GabrieleV_ is now known as GabrieleV
22:11:32 *** lmat has quit IRC
23:38:03 *** jralls has quit IRC