2009-02-09 GnuCash IRC logs

00:00:55 <offby1> *gasp*
00:00:59 <offby1> What are you suggesting?!
00:01:14 <offby1> Clearly the sensible course of action is to get _her_ to drink _more_
00:01:39 <dbreiser> at least you don't have gout -- drinking makes my toes hurt something fierce
00:01:53 <goodger> how did you get gout in this day and age?
00:02:05 <offby1> I have a good friend -- who isn't overweight -- who has gout.
00:02:11 <offby1> Perfectly reasonable eater.
00:02:15 <dbreiser> it's an autoimmune disease, not vacinable
00:02:16 <offby1> exercises and all that too.
00:02:41 <dbreiser> our Wii says I'm overweight, but my wife says I'm not
00:03:01 <dbreiser> I still play soccer regularly, but in goal, so I don't do as much running...
00:03:02 <offby1> that's why you married your wife and not your Wii.
00:03:42 <dbreiser> I married my wife before most game computers existed
00:03:52 <goodger> dbreiser: only one way to find out --- divide your weight by the square of your height, in kilograms and centimetres --- 18-25 is a healthy estimate
00:04:53 <dbreiser> centimeters? really?
00:05:41 <dbreiser> if you meant meters, then I'm slightly beyond the upper end :(
00:05:56 <goodger> wait
00:06:13 <goodger> ah
00:06:15 <goodger> yes
00:06:17 <goodger> I meant metres
00:06:23 <goodger> BMI is in kilograms per square metre
00:06:28 <goodger> sorry, it's five am
00:06:36 <dbreiser> a bit late
00:06:53 <dbreiser> yeah, I could stand to lose 7 kg
00:07:00 <goodger> yeah
00:07:44 <goodger> lovely
00:07:58 <goodger> I'd like to gain about a kilo
00:08:04 <offby1> here, have some of mine
00:08:24 <goodger> not in fat, though
00:09:07 <goodger> I need to either lose 5kg fat or gain some muscle
00:09:12 <goodger> the latter is likely more sensible
00:24:30 <goodger> tumtitum
00:24:48 * goodger whistles the QI theme uncontrollably
01:43:10 *** MechtiIde has quit IRC
01:54:39 *** dbreiser has quit IRC
02:20:40 *** Shaggs has quit IRC
02:39:36 *** palatin has joined #gnucash
02:40:27 *** palatin has quit IRC
02:55:25 *** bentob0x has joined #gnucash
03:23:03 *** ErKa has joined #gnucash
03:29:30 *** MechtiIde has joined #gnucash
04:02:04 *** aindilis` has quit IRC
06:30:35 *** ErKa has quit IRC
06:54:15 *** gxti has quit IRC
06:56:43 *** gxti has joined #gnucash
07:00:39 *** bentob0x has quit IRC
07:56:25 *** Jimraehl has left #gnucash
07:56:53 *** JimRaehl2 has joined #gnucash
07:59:44 *** JimRaehl2 has left #gnucash
08:08:00 *** ErKa has joined #gnucash
08:11:19 *** JimRaehl2 has joined #gnucash
08:42:09 *** warlord-afk is now known as warlord
09:57:12 <djMax> QIF import for investment accounts has a problem. It doesn't move the securities "under" the main account. Is there a way to preprocess the QIF to make the accounts match up properly?
10:06:01 <warlord> What do you mean "under the main account"?
10:10:40 *** sjc has joined #gnucash
10:54:13 <djMax> so I have an account Assets:Investments:Foo
10:54:37 <djMax> I start the QIF import process, it asks for the account name, I say "Foo". It correctly identifies the account, but any stocks in it end up in Foo:Stock
10:54:48 <djMax> as opposed to Assets:Investments:Foo:Stock
10:55:39 <djMax> now I suppose a workaround is to temporarily move Foo to toplevel, but wondered if there was a way to help it figure it out properly
10:57:49 <djMax> (because there's also a secondary problem specific to me, which is that I want subaccounts of stocks for associated options)
10:58:41 <djMax> alternatively, perhaps there's a way to get into the "remembered options" of QIF imports?
10:59:36 <djMax> I heard mention last week of an XML backend, but my GnuCash file seems decidedly not XML. Am I missing something?
11:00:40 <warlord> it's compressed.
11:00:46 <djMax> simple gz?
11:01:19 <djMax> (yes)
11:01:25 <warlord> djMax: Ahh, see, you need to tell it "Assets:Investments:Foo" -- or, rather, you need to map the "Foo" QIF Account to the "A:I:F" GnuCash account.
11:01:31 <djMax> exacto
11:01:49 <warlord> ... and you do that mapping in the importer.
11:02:02 <djMax> right, and there are hundreds of securities.
11:02:11 <warlord> And you need to do that for each one.. once.
11:02:16 <warlord> gnucash will remember it from import to import.
11:02:17 <djMax> yeah.
11:02:21 <djMax> And I can write code.
11:02:52 <djMax> I assume the "I remember last time I saw this thing which account it went to" is somewhere in the XML
11:05:28 <warlord> Nope.
11:05:42 <warlord> It's in the qif-account-map in your ~/.gnucash file.. it's a scheme file
11:06:17 <djMax> yowsa.
11:06:40 <djMax> looks doable though, after I get emacs to format the thing right
11:08:10 <warlord> :)
11:08:11 <djMax> any docs on the array format here? Or a source pointer?
11:08:29 <warlord> ummm... src/import-export/qif-import/*.scm ?
11:08:32 <djMax> couple numeric and bool args that seem random
11:08:43 <djMax> ok, will check em out
11:08:46 <warlord> That code is older than time itself.
11:09:02 <djMax> yeah, I'm tempted to convert the QIF to OFX instead.
11:09:37 <djMax> though I dunno much about OFX. There are xactions in the QIF like "remove shares"/"add shares" which are likely to be painful.
11:09:46 <djMax> is the OFX importer "better enough" that that might be worth it?
11:10:20 <warlord> The OFX importer doesn't handle investments well, and DEFINITELY doesn't handle splits.
11:11:09 <djMax> I find some docs about it that were written by Derek Atkins, is that one of y'all?
11:12:49 <warlord> Um, you're looking in the wrong place. That's me.. The current running QIF code isn't mine.
11:13:03 <warlord> did you look in import-export/qif? or /qif-import?
11:13:34 <djMax> Yeah, just wondering, because I went to school with a Derek Atkins (this is Max Metral from MIT).
11:14:16 <warlord> And you didn't even question the name "warlord"? ;)
11:14:44 <warlord> I realize it's been 15 years....
11:14:53 <djMax> yeah, I figured, but you know, it's a common goal to be a warlord. :)
11:15:09 <warlord> Hahahahaha
11:15:12 <warlord> Ahh, but *I* was the first.
11:15:19 <warlord> (well, except for alt.fan.warlord)
11:15:22 <djMax> Indeed, and the most renowned.
11:15:34 <djMax> other than some Colombian friends of mine I suppose.
11:15:46 <djMax> anyhow, good to bump into you after all these years!!
11:17:51 <djMax> yeesh, reading this scheme is unpleasant.
11:17:59 <offby1> Is one of those import formats -- QIF, OFX, etc --
11:18:26 <offby1> "better" than the others in the sense that, if my bank offers both, then choosing one of them will make my life happy whereas the other choice will lead to sadness and despair?
11:18:53 <offby1> certainly the UI seems different between QIF and OFX; I get the sense that the latter is able to more easily guess which transactions are which
11:19:00 <djMax> I can only say that QIF has reasonably high levels of sadness and despair associations. Haven't gotten to try OFX yet.
11:19:07 <djMax> QIF seems to be very old code.
11:19:09 <offby1> fair enough.
11:20:37 <offby1> gotta say that OFX UI is baffling.
11:20:50 <offby1> Every transaction has a bunch of red, green, and orange rectangles next to it.
11:21:28 <offby1> no idea what they mean. Well, I know red means "unreconciled", and green means "easily reconciled", but I can't figure out why a given transaction would have more than one of those colors. I'd expect each to have _exactly_ one.
11:21:43 *** bentob0x has joined #gnucash
11:23:39 <warlord> i have no idea about those colors..
11:23:52 <warlord> ofx is supposed to be "better" in some ways...
11:23:57 <warlord> but there are limitations.
11:24:19 *** aindilis` has joined #gnucash
11:25:43 *** jpeach has joined #gnucash
11:26:07 <djMax> boy would it be nice if GnuCash was ported to a modern language. Though of course I realize that's mostly insanity
11:26:59 <gxti> just because?
11:27:17 <djMax> because it would be much easier to modify
11:27:36 <djMax> I think there is a scheme to .Net cross compiler, though I'm sure .Net would annoy some set of people.
11:27:48 <djMax> ^I'm not sure^
11:28:05 <gxti> the only mono apps i've seen are memory hogs
11:28:09 <gxti> and they're fairly simple, too
11:28:19 <djMax> yeah, though it's getting better.
11:28:28 <djMax> I'd be fine with Ruby/Python/whatever too.
11:28:45 <djMax> but Scheme,C, C++ god bless em, are just ancient
11:28:58 <gxti> there's a reason they're still in use
11:29:02 <djMax> fine for things where speed is king I suppose but this is hardly that case.
11:29:16 *** donfede1 has quit IRC
11:29:21 <djMax> yeah, we have some around here for math/graphics, but other than that not much
11:29:32 <gxti> i do python at my day job, but i still see the value in keeping those around
11:30:16 <djMax> I'm not saying they should die, just that GnuCash doesn't need them. Or put another way, if we were starting from scratch, we clearly wouldn't pick those.
11:30:58 <gxti> i'll admit i don't know anything about gnucash's architecture at the moment, but a c/c++ core with logic implemented via something more lightweight seems sensible
11:31:08 <gxti> i wouldn't go for a pure python/ruby implementation
11:31:33 *** jpeach has left #gnucash
11:31:33 <djMax> Honestly, it would almost certainly be best done in Javascript and HTML.
11:31:42 <djMax> i.e. inside Gecko.
11:32:05 <gxti> writing the entire app in javascript would be inane
11:32:23 <djMax> why? JS has apps way more complicated than GC.
11:32:33 <gxti> doesn't mean it's a good idea
11:32:41 <gxti> i'd much rather work with python than js
11:33:07 <djMax> it's true, me too, but in terms of ability to attract a vibrant community, JS/HTML wins.
11:33:14 <djMax> developer community that is
11:33:30 <djMax> anyhow I'd imagine it's all academic given the size of this codebase
11:33:39 *** MechtiIde has quit IRC
11:33:40 <warlord> there's nothing wrong with c
11:33:49 *** palatin has joined #gnucash
11:33:55 <warlord> and we're working to move more of scheme into c.
11:33:56 <djMax> sure, nothing wrong with assembly either but I'd rather not use it.
11:34:03 <warlord> it's down to only 17% scheme
11:34:17 <djMax> did it start 100%?
11:34:20 <warlord> no..
11:34:27 <warlord> it was always a mix
11:34:32 <djMax> how active would you say development is?
11:34:39 <warlord> depends on the month.
11:34:48 <djMax> 2 people, 10 people?
11:35:45 <warlord> again, depends on the month.
11:35:49 <djMax> this line: (simple-obj-from-list (cdr entry) <qif-map-entry>)
11:35:54 <warlord> we've over a dozen with commit access
11:36:10 <warlord> but #commits/week varies
11:36:37 <djMax> simple-obj-from-list, is that a "core" function? Can't seem to find it
11:37:30 <warlord> ./src/app-utils/simple-obj.scm
11:37:58 <warlord> ( found using: find . -name \*.scm | xargs grep simple-obj-from-list )
11:38:09 <djMax> yeah, was using google instead. :)
11:41:50 <warlord> ah
11:44:04 *** donfede has joined #gnucash
11:47:41 <djMax> yeah, so seemingly the quickest path to QIF import of this complexity is to just import it stupidly and then munge the XML afterwards.
11:50:20 <warlord> MAYBE... Or spend the extra few minutes during the import to adjust it appropriately.
11:50:54 <djMax> seriously, it'd be like hours. The UI for doing these changes is not good and there are probably 350 securities over the last 10+ years
11:51:32 <djMax> probably takes 20 minutes to write an XML script
11:51:44 <djMax> an hour if I want a UI for it.
11:55:35 <warlord> Well, modifying the XML by hand is "not supported"
11:56:23 <djMax> :) totally understood. But I'll never make GnuCash work for me as is, and would love to not learn another UI framework, so I bet I'll be in the XML more than once. Might as well start easy.
11:56:49 <djMax> and it's not for any shortcoming on GnuCash, because the other packages can't even come close
11:57:37 <warlord> I think the intention was that the import process is only done once.. So if you've got lots of data you should expect to take /some/ non-zero time configuring it..
11:57:53 <djMax> yeah, I'm on day 3 and I only have about half the accounts
11:59:08 <warlord> hmmmm
12:11:50 *** ErKa has quit IRC
12:16:16 *** palatin has quit IRC
12:24:21 <djMax> For example, I have to manually look up every symbol because there's no integrated symbol lookup. Instead, I could just import them all as the defaults and then use an existing service to look them up
12:24:51 <djMax> (another case where had it been in JS, adding that feature would be almost trivial)
12:25:45 <warlord> what do you mean, "manually look up every symbol"?
12:26:01 <djMax> find the symbol and exchange for "Apple" for example.
12:26:03 <warlord> You mean to find a working quote source?
12:26:09 <djMax> And I know some, but not all.
12:26:14 <djMax> and with options it's even nastier.
12:26:24 <warlord> Ummm... that would be a useful (external) tool...
12:26:34 <warlord> one that gnucash could call out to..
12:26:48 <djMax> Yeah, but then it has to present it in the UI in some reasonable way
12:26:59 <djMax> which implies I'd need to write it
12:30:41 <warlord> Adding a menu item to the tools to call out to it would be easy.. writing the tool to do the lookup .. that's more challenging..
13:25:29 *** MechtiIde has joined #gnucash
13:29:21 *** Zoolooc has joined #gnucash
13:39:53 *** gxti has quit IRC
13:59:43 *** hvx has joined #gnucash
14:06:51 *** gxti has joined #gnucash
14:10:55 *** Demitar has quit IRC
14:14:53 *** Demitar has joined #gnucash
14:23:05 <djMax> so I expand an account with lots of subaccounts, then I hit "M". It correctly goes to the first M subaccount, but then it freezes for like 5 seconds. Is this expected?
14:25:16 <warlord> I didn't even know you could do that!
14:25:24 <warlord> You're in the CoA?
14:25:38 <djMax> in this case in the CoA inside "Edit Account"
14:25:41 <djMax> but yes, there too.
14:25:54 <djMax> and have seen it elsewhere, like in QIF import
14:31:14 <warlord> Never seen it freeze..
14:32:27 <djMax> does it let you type more characters?
14:36:26 <warlord> I dunno... As I said, I didn't even know about that feature.
14:37:20 <djMax> I just meant if you're actually doing it in the UI, does it work for you
14:38:13 <warlord> dont know.. havent tried.
14:38:13 *** sjc has quit IRC
14:44:24 *** sjc has joined #gnucash
14:45:46 <djMax> is it possible to auto-hide stock accounts with no shares?
14:50:28 <warlord> Not specifically, but you can set gnucash to hide accounts with 0 balance.
15:56:11 *** hvx has quit IRC
16:05:20 *** MechtiIde has quit IRC
16:54:55 *** nomeata has joined #gnucash
16:55:08 <warlord> chris: ping?
17:10:35 *** jpeach has joined #gnucash
17:10:46 *** jpeach has left #gnucash
17:50:46 *** cortana has joined #gnucash
17:52:44 *** Tim_Abell has joined #gnucash
17:52:59 *** Tim_Abell has left #gnucash
18:11:01 *** nomeata has quit IRC
18:22:59 *** bentob0x has quit IRC
18:38:43 *** Tim_Abell has joined #gnucash
18:39:04 <Tim_Abell> my balance graph filters to date now. i'm on a scheme roll this week
18:40:33 <Tim_Abell> http://github.com/timabell/gnucash-account-balance-chart/commits/master/
18:40:54 <warlord> cool!
18:47:58 *** feiner has joined #gnucash
18:48:13 <Tim_Abell> oh dear, i seem to be pushing the boundaries of the current graphing stuff
18:49:51 <warlord> heh
19:03:13 *** aindilis` has quit IRC
19:05:02 *** aindilis` has joined #gnucash
19:05:23 *** cortana has quit IRC
19:13:59 *** Tim_Abell has quit IRC
19:15:22 *** Tim_Abell has joined #gnucash
19:16:09 <Tim_Abell> baaaah! ctrl alt pagedown = change tab in gedit. ctrl alt backspace = restart x. guess how close they are on my laptop
19:23:19 *** Zoolooc has quit IRC
19:32:29 *** Tim_Abell has quit IRC
19:33:01 *** blink has joined #gnucash
19:35:00 <blink> I have a few split transactions that are being shown in the imbalance-cad category despite the imbalance amount being 0.00 for the transaction (all the rest of it balances)... how can I remove the 0.00 entries?
19:35:49 *** sjc has quit IRC
19:38:32 *** claybustr has joined #gnucash
19:40:12 <claybustr> is there any way to re-reconcile old statements in gnucash? It seems there is no concept of tagging a reconciliation that would allow me to verify a 3 month old statement after subsequent statements have been done.
19:41:57 <claybustr> I have some old transactions that have really screwed up my books. Sometimes I think an entry has been edited accidently to move it back into the past. Other times it's something legitimate like a merchant losing a transaction (it does happen) or entering against the wrong account - (but those usually get caught at the next reconcile)
19:44:46 <warlord> blink: you probably didn't remove the imbalance split.
19:45:23 <warlord> claybustr: correct, you must reconcile in monotonically-increasing time. You cannot reconcile month M and then later try to reconcile month M-1
19:46:11 <blink> warlord, thanks. I'm trying to, it's telling me "This is the split anchoring this transaction to the register"
19:47:07 <claybustr> so I just have to pick a date and start over?
19:47:37 <warlord> blink: yes, do it from another register, not the imbalance. Try 'Jump'
19:47:38 *** sjc has joined #gnucash
19:48:23 <claybustr> if reconcile was a date tag and not just a boolean this could be handled
19:48:31 <warlord> claybustr: pick a correct ending date with the correct ending balance and include all those earlier transactions... I'm not sure how you reconciled month M without month M-1 unless balance was 0 at the time of reconciliation.
19:48:45 <warlord> No, it couldn't.. It would always throw off your balances.
19:49:06 <warlord> (and it *IS* a date-tag -- there is a reconciliation date in the txn)
19:49:09 <warlord> er, in the split.
19:53:00 <blink> warlord: thanks that worked... do you know how I can avoid that happening in the future? I believe the transactions affected are from my scheduled transactions
19:53:30 <warlord> Sure.. dont hit "enter" when you're entering a transaction.
19:53:37 <warlord> Use tab and the up/down arrows.
19:53:59 <warlord> That way it wont 'Scrub' the transaction until you're done, so it wont add the Imbalance account.
19:54:27 <blink> ahh ok, that's what I've been doing. I'll adjust my behavior.
19:54:34 <blink> thanks again, much appreciated :)
19:57:07 <warlord> youre welcome.
20:13:50 *** blink has left #gnucash
20:16:17 *** claybustr has quit IRC
20:37:55 *** sjc has quit IRC
20:52:17 *** jpeach has joined #gnucash
20:52:30 *** jpeach has left #gnucash
21:17:39 *** aindilis` has quit IRC
21:31:47 *** warlord is now known as warlord-afk
22:33:25 <Rolf1> good morning, warlord-afk
22:33:55 <Rolf1> for personal use, I just recompiled the latest trunk into a debian package
22:34:43 <Rolf1> I thought I should let you know that some errors were emitted (not sure if those are from gnucash or because something wrong in the packaging which is why I thought I should let you know)
22:36:20 <Rolf1> http://rafb.net/p/FIh5cH81.html
22:40:53 *** puck has quit IRC
22:45:23 *** JimRaehl2 has left #gnucash
22:46:27 *** Jimraehl has joined #gnucash
22:50:57 *** linas has quit IRC
22:52:38 *** puck has joined #gnucash
22:52:45 *** linas has joined #gnucash
22:53:38 *** linas has quit IRC
22:54:22 *** linas has joined #gnucash
22:55:25 *** Xtroubaritz_ has quit IRC