2007-06-02 GnuCash IRC logs
00:25:30 *** Geot has quit IRC
00:37:50 *** xai has quit IRC
01:22:49 *** benoitg has joined #gnucash
01:31:26 *** fell has quit IRC
03:00:48 *** nbinont has quit IRC
03:02:37 *** kam has quit IRC
03:26:37 *** Zoolooc has joined #gnucash
05:12:45 *** chemaja has joined #gnucash
05:18:00 *** chemaja has quit IRC
05:23:07 *** tpfennig has joined #gnucash
05:48:53 *** roxy_ has joined #gnucash
06:22:34 *** tpfennig has quit IRC
06:40:56 *** tpfennig has joined #gnucash
06:42:48 *** nomeata has joined #gnucash
06:52:32 *** ErKa has joined #gnucash
07:06:51 *** _gunni_ has joined #gnucash
07:27:43 *** roxy_ has quit IRC
07:27:43 *** cparker has quit IRC
07:27:43 *** ryjax has quit IRC
07:28:53 *** cparker has joined #gnucash
07:29:23 *** roxy_ has joined #gnucash
07:29:25 *** ryjax has joined #gnucash
07:56:56 *** twunder has joined #gnucash
08:09:04 *** jsled has quit IRC
08:29:35 *** nomeata has quit IRC
08:31:33 *** roxy_ has quit IRC
08:37:14 *** twunder has quit IRC
08:44:29 *** BlackBsd has joined #GnuCash
08:47:43 *** roxy_ has joined #gnucash
09:04:47 *** warlord-afk is now known as warlord
09:06:13 *** BlackBsd has quit IRC
09:13:02 *** |gunni| has joined #gnucash
09:22:05 *** _gunni_ has quit IRC
09:43:21 *** roxy_ has quit IRC
09:43:55 *** ErKa has quit IRC
10:27:36 *** ErKa has joined #gnucash
10:32:43 *** warlord is now known as warlord-afk
10:43:16 <lasindi> Question for CSV importer: I think I understand that in order to add a transaction, say a deposit into an account, I need to add two splits to the transaction: one transferring into the account and another transferring out. However, if I don't know the account that the money is coming from, I want to use the "Imbalance" account. How can I get a pointer to that account?
10:52:47 *** Rol1 has joined #gnucash
10:53:37 *** tpfennig has quit IRC
10:54:14 *** ErKa has quit IRC
11:04:06 *** nomeata has joined #gnucash
11:31:16 *** Geot has joined #gnucash
11:41:37 *** Wilddev has joined #gnucash
11:41:38 *** gncbot sets mode: +o Wilddev
11:41:48 <Wilddev> morning
11:52:27 *** andi5 has joined #gnucash
11:52:28 *** gncbot sets mode: +o andi5
12:00:49 <Wilddev> hey andi5
12:06:50 *** jsled has joined #gnucash
12:06:51 *** gncbot sets mode: +o jsled
12:10:10 <Wilddev> hey jsled
12:10:24 <jsled> hey
12:11:16 <Wilddev> 'nother day 'nother gnucash release :P
12:23:17 <andi5> hi
12:24:36 <jsled> Oh ... G8 stuff?
12:24:43 <andi5> yep
12:25:19 <Wilddev> sometimes I'm glad I dont have a TV
12:25:21 <Wilddev> heh
12:26:21 *** Demitar has quit IRC
12:51:28 *** IanL has quit IRC
12:53:05 <andi5> jsled: you said you thought about fixing these "element gnc-v2: Schemas validity error : Element 'gnc-v2': No matching global declaration available for the validation root" messages.... are you still interested? i think i remember what the problem is.
12:53:37 <Wilddev> andi5: did u see in 433156 he was seeing it still in r16142?
12:54:18 <jsled> andi5: I never really got anywhere with that, and you seem to be a bug-fixing machine ... go right ahead.
12:54:31 <andi5> i think he just verified that the bug has been fixed and his rev is 16142
12:54:38 *** IanL has joined #gnucash
12:54:51 <andi5> jsled: i need you advice then :)
12:54:54 <andi5> +r
12:55:11 <Wilddev> ohhh yes I think ur right
12:55:14 <Wilddev> thx
12:55:19 <andi5> :)
12:55:45 <Wilddev> hmm lots of SX fixes for 2.1.3
12:55:56 <Wilddev> how to put it in the announcement...
12:57:44 <andi5> jsled: in lib/libqof/backend/file/qsf-backend.c : g_module_check_init , the qsf backend registers the qsf qof backend... this happens while gnucash crawls some directories to find gnucash modules... well it is no gnucash modules then, but the backend is registered before the file backend nonetheless
12:58:47 <andi5> so either one tries to make these messages go away or change the order those two backends get registered
13:00:22 *** IanL has quit IRC
13:01:57 <jsled> Then I'd just make the QSF backend register after the gnucash one.
13:02:34 <jsled> (Or remove it entirely ... is the QSF backend used for anything?)
13:05:26 <Wilddev> andi5: what are the %s-NULL-printf fixes doing?
13:06:17 <Wilddev> Please can people check http://wiki.gnucash.org/wiki/Announcement_2.1.3 for me to see if its good?
13:08:55 *** nomeata has quit IRC
13:14:51 *** tpfennig has joined #gnucash
13:14:56 <jsled> Damnit. I really should have gotten the FreqSpec code out before this release.
13:15:04 <jsled> gah. Need more time.
13:15:54 * jsled adds a bit more color about the SX work.
13:18:01 <jsled> (done) Hmm. So Swig isn't actually a depend for the ebuild....
13:18:39 <Wilddev> does it need it?
13:19:55 <jsled> No. It's listed in there presently, but it doesn't need to be.
13:19:56 *** ErKa has joined #gnucash
13:20:06 <jsled> It's not either a build or runtime dependency, from the release tarballs.
13:20:08 <jsled> Only for svn.
13:20:20 <Wilddev> aha ok
13:21:19 <jsled> updated.
13:22:47 <jsled> Wilddev: announce looks fine.
13:22:57 <Wilddev> kewl
13:23:19 <jsled> I'm afk. Biaw.
13:24:14 *** tpfennig_ has joined #gnucash
13:24:31 <andi5> Wilddev: each INFO, WARN, DEBUG & friends all call printf in the end, but some systems, like solaris and windows, do not support parameters to be NULL for %s format specifiers ... that is why we need to wrap all those places where param!=NULL is not guaranteed ... currently this work is done by using the following unsatisfactory algorithm:
13:24:31 <andi5> (a) make gnucash crash
13:24:31 <andi5> (b) get stacktrace and fix it, go back to (a)
13:25:30 <Wilddev> aha good to know
13:27:37 <andi5> approx. 5 km to the city centre
13:29:21 *** tpfennig has quit IRC
13:29:27 <andi5> Wilddev: maybe the improvements to the windows packaging are not overly important.... iirc they all deal with cross-compiling and very few people care about that :)
13:29:50 <andi5> but hey, keep it in :)
13:30:16 <Wilddev> ok
13:30:34 <Wilddev> any more commits before I make the 2.1.3 tarball?
13:30:44 <andi5> very early today :)
13:31:07 <andi5> not from me, no
13:31:27 <Wilddev> andi5: yeah I might be going to a poker game later, wanted to get it done
13:31:46 <andi5> i see
13:37:44 *** Rol1 has quit IRC
14:30:35 <andi5> Wilddev: how is the releasing process going? i will switch off rather soonish :)
14:31:11 <Wilddev> andi5: so far seems fine, the first test build got thru distcheck and I'm compiling the release now
14:31:28 <andi5> great... thanks and bye
14:31:47 <Wilddev> ok ciao
14:31:50 *** andi5 has quit IRC
14:43:26 *** Wilddev has quit IRC
14:44:31 *** Wilddev has joined #gnucash
14:44:31 *** gncbot sets mode: +o Wilddev
15:26:57 *** wizkid239 has joined #gnucash
15:34:34 *** wizkid238 has quit IRC
15:42:10 *** BlackBsd has joined #GnuCash
15:54:42 *** BlackBsd has quit IRC
16:12:51 *** Wilddev has quit IRC
16:29:44 *** sjc has joined #gnucash
16:47:47 <jsled> damnit. power's out again.
16:47:50 <jsled> later.
16:51:10 *** sjc has quit IRC
16:58:21 *** BlackBsd has joined #GnuCash
16:58:37 *** jsled has quit IRC
17:00:46 *** wizkid239 is now known as wizkid238
17:07:19 *** nbinont has joined #gnucash
17:19:36 *** sjc has joined #gnucash
17:48:09 *** hampton has quit IRC
17:52:44 *** hampton has joined #gnucash
17:52:45 *** gncbot sets mode: +o hampton
17:54:42 *** BlackBsd has quit IRC
18:18:26 *** ErKa has quit IRC
18:50:33 *** Wilddev has joined #gnucash
18:50:33 *** gncbot sets mode: +o Wilddev
18:50:39 *** phil__m has joined #gnucash
18:52:15 *** Wilddev has quit IRC
18:53:08 *** benoitg has left #gnucash
18:53:46 <phil__m> Can someone please help me with setting up auto stock retrieval in gc2? gnc-fq-dump works fine, just not the GUI
19:05:53 *** nomeata has joined #gnucash
19:15:56 *** benoitg has joined #gnucash
19:16:15 *** benoitg has left #gnucash
19:55:55 *** warlord-afk is now known as warlord
19:57:53 <warlord> lasindi: you never want to use the imbalance account. You want to use the Generic Importer and force the user to choose a source or destination account.
20:01:00 <imperito> I get stuff in the imbalance account all the time importing QIFs from my 401k
20:01:00 <warlord> phil__m: If gnc-fq-dump works then all you need to do is turn on the right choice in the UI.
20:04:52 *** phil__m has quit IRC
20:05:02 *** phil__m has joined #gnucash
20:07:54 <phil__m> Ok, got the auto stock retrieval working now, but it was a real PITA (hint: error msgs like "unknown error" are not very helpful)
20:09:03 *** phil__m has quit IRC
20:17:59 <warlord> imperito: You shouldn't get it from QIF. You could get it from OFX. But the QIF importer should always force balanced transactions. OFX doesn't (which I consider a bug)
20:18:30 <warlord> anyways, dinner. BIAB
20:18:31 *** warlord is now known as warlord-afk
20:30:28 *** jml has joined #gnucash
20:52:51 *** twunder has joined #gnucash
20:56:51 *** benoitg has joined #gnucash
21:10:14 *** hampton has quit IRC
21:10:39 *** sjc has quit IRC
21:16:34 *** twunder has quit IRC
21:22:21 *** IanL has joined #gnucash
21:31:54 *** BlackBsd has joined #GnuCash
21:58:39 *** nomeata has quit IRC
22:07:29 *** hampton has joined #gnucash
22:07:29 *** gncbot sets mode: +o hampton
22:10:50 *** tpfennig_ has quit IRC
22:11:37 *** BlackBsd has quit IRC
22:12:47 *** _gunni_ has joined #gnucash
22:21:42 *** |gunni| has quit IRC
22:33:13 *** jml has quit IRC
22:34:46 *** Zoolooc_ has joined #gnucash
22:35:41 *** warlord-afk is now known as warlord
22:40:14 <warlord> @tell wilddev nice job on 2.1.3
22:40:14 <gncbot> warlord: The operation succeeded.
22:42:28 *** twunder has joined #gnucash
22:44:13 *** Zoolooc has quit IRC
22:53:57 <lasindi> warlord: When it goes through the generic importer, the transactions appear yellow and I can double click on them and select another account. However, I can also just click "OK" and let the transactions go through; then they appear with a little "X" in the account. Is there a way to keep the user from clicking OK before selecting other accounts?
22:57:55 <warlord> I dont know, but I think that THAT is a bug in the generic importer.
22:58:13 <warlord> It shouldn't let you just click through.. (or if nothing else it should auto-select a default)
23:00:06 <lasindi> Okay
23:00:52 *** benoitg has left #gnucash
23:02:18 <lasindi> The reason why I thought that the generic importer or something should choose "Imbalance" as a default is because that's what the OFX importer does, but you say even that is a bug?
23:09:43 *** benoitg has joined #gnucash
23:10:31 <warlord> Yes, IMHO even that is a bug.
23:11:02 <warlord> Although I suppose that "Imbalance" is just as good as "Unspecified" (which is the QIF importer default)
23:15:44 <warlord> Note that the transaction scrub routines will add the Imbalance splits so you shouldn't have to.. BUT, IMHO the importer UI should strongly encourage balanced transactions.
23:16:32 <lasindi> Are you talking specifically about the xaccTransScrubImbalance function?
23:17:06 <lasindi> I tried using it and it failed, because its call to xaccScrubUtilityGetOrMakeAccount couldn't get a balancing account apparently.
23:17:57 *** benoitg has left #gnucash
23:19:05 <warlord> you shouldn't have to call it yourself. It's called by xaccTransCommitEdit()
23:19:43 <lasindi> Oh
23:19:59 <lasindi> But isn't that what's called by the importer?
23:23:27 <warlord> which?
23:23:35 <lasindi> xaccTransCommitEdit
23:23:42 <warlord> Yes, everythingn has to call that.
23:23:57 <warlord> But that's what calls the scrubber. You shouldn't.
23:26:39 <lasindi> Okay, I'll take a look there and see if I can figure out why the importer isn't adding imbalance splits.
23:29:44 <warlord> Unless I misunderstand the gneeric importer framework, your importer creates the set of template transactions and then you hand it off to the generic importer which asks the user to map to the source/dest accounts and then converts it all to GNC transactions.
23:31:37 <lasindi> I think so. At the moment, I'm creating transactions with just one split (I'm assuming that I don't know the destination account), and then giving it to the generic importer.
23:32:36 <warlord> define what you mean by "creating transactions with just one split"?
23:33:43 *** twunder has quit IRC
23:35:59 <lasindi> I am testing right now with CSV files that have just three columns per row: date, description, amount. First the user selects an account to import to, then for each row I create a transaction and add one split. I set the split's account to the selected account and its amount/value to the amount in the file. Then I hand these transactions to the importer.
23:36:33 <warlord> Are you creating a Transaction*, or a template transaction?
23:36:52 <lasindi> A Transaction* (this is probably the problem?).
23:37:49 <warlord> * nods *
23:37:58 <warlord> the importer wants a template transaction
23:40:48 <lasindi> Okay, could you recommend the source file to look at for a template transaction's definition? (The gnc_gen_trans_list_add_trans function accepts a Transaction*, so I'm wondering if a template transaction has a different type or if it's somehow disguised.)
23:41:10 <warlord> I think it's "somehow disguised".
23:41:15 <warlord> Take a look at the OFX importer.
23:41:34 <warlord> (IMHO the use of Transaction* in the generic importer was a bad design decision)
23:42:19 <warlord> e.g., see import-main-matcher.h, gnc_gen_trans_list_add_trans()
23:42:38 <warlord> Looks like it DOES take a Transaction* BoO!
23:43:47 <lasindi> Hehe, yep
23:49:09 <warlord> Are you calling CommitEdit() before you pass the txn to the generic importer?
23:53:18 <warlord> your answer should be "no", but I want to make sure -- according to that comment you should pass in the uncommitted transaction.
23:55:26 <lasindi> No, I learned that that was a mistake earlier. :-)
23:55:47 <warlord> :-D
23:55:49 <warlord> Okay.
23:56:04 <warlord> So remind me: what's the problem now?
23:56:58 <lasindi> The problem is that these transactions go through the importer into the source account with apparently single splits, so there are no destination accounts for the transactions, not even Imbalance.
23:57:18 <warlord> Even if you choose a destination account in the importer?
23:57:31 <lasindi> No, if I choose an account it works fine.
23:57:38 <warlord> This is an importer bug..
23:57:48 <warlord> (that sounds like what I remembered it doing)
23:58:02 <lasindi> Even if I send plain transactions (i.e. not templates)?
23:58:06 <warlord> Yeah.
23:58:21 <lasindi> Okay.
23:58:44 <warlord> As I recall, the importer just doesn't add a second (balancing) split in the 'default' case. I consider that a bug. I fyou want to fix that you have my complete support ;)
23:59:52 <lasindi> Okay, I'll try but I can't guarantee anything.
23:59:56 <lasindi> I'm reading about this data type "GNCImportTransInfo". Do I need to do anything with that?
23:59:57 <warlord> of course.