2010-03-29 GnuCash IRC logs
00:13:23 *** lejonet has quit IRC
00:43:12 *** lejonet has joined #gnucash
01:34:21 *** lejonet has quit IRC
03:11:38 *** gour has joined #gnucash
03:17:41 *** bentob0x has joined #gnucash
03:18:01 *** jpetersen has joined #gnucash
04:13:59 *** message144 has joined #gnucash
04:22:28 *** cortana has joined #gnucash
04:49:14 *** cortana has quit IRC
05:33:29 *** message144 has quit IRC
06:19:58 *** SmokeyD has joined #gnucash
06:23:24 *** ErKa has joined #gnucash
06:26:55 <SmokeyD> Hey everyone. I am using gnucash 2.2.9. I want to start with an empty file for 2010, but keep the jobs, customers, tax tables, etc. So I copied the file from 2009 and removed the transaction,invoice and entry tags and adjusted gnc:count-data accordingly
06:26:55 <gncbot> SmokeyD: Sent 13 hours and 59 minutes ago: <warlord> no, there are no schema files for GNC. Certainly nothing normative. As for your parse error, don't know what to tell you. It might be easier to do File -> Export Accounts and then re-add the biz objects.
06:28:01 <SmokeyD> But now I get the following error when opening the file: element gnc-v2: Schemas validity error : Element 'gnc-v2': No matching global declaration available for the validation root.
06:28:57 <SmokeyD> In response of the answer of warlord: export->accounts only exports the accounts, not the customers, jobs and vendors
06:29:20 <SmokeyD> and I don't want to enter them all over again (that would be a lot of work)
06:34:48 *** bentob0x has quit IRC
06:52:46 <SmokeyD> ah, nevermind. I found it. It was a problem with my xml editor. It stored the UTF byte order mark in the xml file which causes the gnucash parse error
07:41:55 *** Jimraehl has left #gnucash
07:46:40 *** ErKa has quit IRC
07:46:41 *** fDavid has joined #gnucash
07:49:49 *** Jimraehl has joined #gnucash
07:56:10 *** ErKa has joined #gnucash
08:22:25 *** Jimraehl has quit IRC
08:22:54 *** Jimraehl has joined #gnucash
08:36:51 <SmokeyD> hey everyone. Is gnucash sensitive to the case of the xml tags (GncEntry vs gncentry) and whitespace (newlines+indentation) in the .gnc files? I was trying to write a small tool that copys one gnc file into a new one with removal of all GncEntry, transaction and GncInvoice tags. I managed it manually, but my tool (python using BeautifulSoup) creates files which causes parse errors when opened by gnucash
08:53:55 *** bentob0x has joined #gnucash
09:03:41 *** SmokeyD has left #gnucash
09:22:22 *** fDavid has quit IRC
09:23:52 *** fDavid has joined #gnucash
09:37:32 *** KaiForce has joined #gnucash
09:52:13 *** warlord-afk is now known as wrlord
09:52:17 *** wrlord is now known as warlord
09:52:50 <warlord> @tell SmokeyD It should not be sensitive to whitespace, but it IS sensitive to xml tag case
09:52:50 <gncbot> warlord: The operation succeeded.
11:02:28 *** trumpen has joined #gnucash
11:10:55 *** jpetersen has quit IRC
11:14:43 *** Linuturk has quit IRC
11:15:48 *** Linuturk has joined #gnucash
11:22:35 *** KaiForce has quit IRC
12:29:19 *** wingo has joined #gnucash
12:29:23 <wingo> greets!
12:30:04 <wingo> q: is there an up-to-date git mirror for gnucash?
12:31:43 * wingo checks out with bzr instead
12:32:02 <gour> wingo: bzr-svn?
12:32:22 <warlord> wingo: we only support SVN.
12:32:25 <warlord> (at this time)
12:32:53 <wingo> i know you're probably tired of hearing it, but bzr and git are much better ;-)
12:33:06 <gour> :-D
12:33:12 <jsled> well, one of those is true. :)
12:33:30 <wingo> :)
12:33:36 <jsled> bzr is better than svn, I do admit.
12:34:14 * gour cannot handle git's ui. bzr is nice, but darcs is on (our) top
12:37:38 <warlord> There are too many damn SCCSs
12:38:24 <warlord> And this is not the place to argue the merits of one vs the other (at least until we get into a discussion of which one we should migrate to).
12:39:56 <gour> bzr has nice bzr-svn, bzr-git & bzr-hg plugins...LP is also nice
12:41:27 <gour> ...but is seems that majority of projects are just fascinated by git's speed :-(
12:42:32 <wingo> warlord: so, dunno how this is going to go, but how would you feel about dropping the dependency on slib?
12:42:58 <wingo> i haven't investigated too much, but it might be possible to only rely on facilities present in guile >= 1.6
12:43:39 <wingo> if we were to contribute patches rewriting those bits of scheme code to use standard guile facilities, would that be ok?
12:44:12 <wingo> the potential negative would be if people have custom reports locally that rely on slib -- then they would have to install slib on their own
12:44:33 <wingo> but gnucash itself wouldn't require slib to do anything included in the gnucash tarball
12:47:02 <warlord> wingo: I think dropping the slib requirement would be excellent!
12:47:46 <wingo> cool :)
12:47:49 <warlord> I'm not worried about people with custom reports.. By definition they are already doing something custom/on-their-own so they can, as you say, install slib themselves.
12:47:57 <gour> :-)
12:48:12 <warlord> The question, of course, is how to TEST that gnucash is completely slib-free?
12:49:34 <wingo> well it's not that bad -- slib doesn't export anything by itself, only provide/require, and there are very few slib facilities that are in use
12:50:04 <wingo> so you just grep for (require, rewrite those files, and finally keep a list of all required packages and grep for things they provide
12:50:10 <wingo> in case something depended on another file doing the require
12:50:31 <wingo> will poke at it tonight, i have an overnight train ride :)
12:50:51 <warlord> Okay
12:54:05 <wingo> this might involve incorporating a couple of files into gnucash itself, like printf.scm
12:54:40 <wingo> but if it's only a couple files that should still be better than forcing users to install slib, which is a bit of a pain
13:02:51 <warlord> So long as there isn't a bad interaction... Or maybe we stop using (printf)
13:03:33 *** jpetersen has joined #gnucash
13:04:40 <wingo> do you use ChangeLogs or commit messages?
13:05:28 <warlord> Commit Messages
13:05:35 <warlord> .. to derive the changelog
13:06:20 <wingo> cool
13:15:58 *** suse has joined #gnucash
13:18:32 *** Zoolooc has joined #gnucash
13:20:02 *** fDavid has quit IRC
13:35:55 *** wingo has quit IRC
13:36:22 *** blizgerg has quit IRC
14:26:32 *** daedeloth has joined #gnucash
14:33:50 *** andyt has joined #gnucash
15:01:04 *** daedeloth has quit IRC
15:10:35 *** daedeloth has joined #gnucash
15:35:27 *** gour has left #gnucash
15:46:37 *** cortana has joined #gnucash
16:05:38 *** ErKa has quit IRC
16:16:40 *** suse has quit IRC
16:38:32 *** puck has quit IRC
16:38:37 *** puck has joined #gnucash
17:05:10 *** jpetersen has quit IRC
17:32:36 *** warlord is now known as warlord-afk
17:41:03 *** message144 has joined #gnucash
17:47:05 *** daedeloth has quit IRC
18:56:48 *** andyt has quit IRC
19:04:01 *** raleightile has joined #gnucash
19:17:49 *** tedcx2 has left #gnucash
19:47:14 *** bentob0x has quit IRC
19:50:42 *** thomas_ has quit IRC
19:51:14 *** thomas has joined #gnucash
19:53:33 *** fDavid has joined #gnucash
20:02:15 *** Zoolooc has quit IRC
20:10:58 *** fDavid has quit IRC
20:31:20 *** fDavid has joined #gnucash
21:22:55 *** blizgerg has joined #gnucash
21:44:47 <Mer|in> so i'm just now starting to look @ the tax features in gnucash
21:45:32 <Mer|in> i'm still paying pmi - anyone know which tax category that is?
22:06:05 *** fDavid has quit IRC
22:16:33 *** message144 has quit IRC
22:26:18 *** warlord-afk is now known as warlord
22:26:34 <warlord> Mer|in: I dont think PMI is a tax category per-se. I think it's just an expense.
22:28:21 <Mer|in> so if i wanted to have it marked for the tax/txf reporting on my 1040 - how would i do it?
22:28:42 <Mer|in> or am i misunderstanding?
22:32:30 <warlord> What do you mean? I'm pretty sure you cannot deduct PMI from your income tax
22:35:47 <Mer|in> i dunno - turbo tax asks for it
22:37:07 <warlord> Really?
22:37:09 <warlord> Huh.
22:37:44 <Mer|in> it's not on my 1098, but from what i read, they don't have to put it on the 1098 if they put it on ure yearend statement
22:37:47 <Mer|in> http://www.irs.gov/publications/p936/ar02.html#en_US_publink1000229966
22:38:08 <warlord> Oh, I guess it's new.
22:38:24 <warlord> Okay.. I guess you CAN deduct PMI. Wow. Um, okay.
22:38:33 <warlord> (I haven't paid PMI in over a decade)
22:39:29 <Mer|in> yeah i think this year is my 1st opportunity to get out of it
22:39:35 <Mer|in> without having to refinance
22:39:50 <Mer|in> fha bs
22:40:35 <warlord> Fun
22:40:41 <Mer|in> yea really
22:43:01 <Mer|in> how are state taxes handled?
22:43:08 <Mer|in> nm, will read up in the handbook
23:03:52 <warlord> k
23:09:20 *** cortana has quit IRC
23:14:07 <Mer|in> so in the postgresql backend, where do backups get stored?
23:18:14 <warlord> backups?
23:18:20 <warlord> there are no backups in the SQL backends
23:18:31 <Mer|in> what about logs?
23:27:22 <warlord> There should be .log files for transactional changes.
23:28:11 <Mer|in> any idea where they would be under win32?
23:28:37 <warlord> In the same dir as your data file
23:28:54 <Mer|in> no problem here, btw - just curious, something i had w/ the xml backend...was wondering if it went away w/ the sql backend
23:30:00 <warlord> .log files shouldn't.. But then again you don't *need* log files with the sql backend because your data is saved at every transactional commit.
23:30:13 <warlord> the lack of backup can be an issue..
23:30:47 <Mer|in> yea but i can schedule hotbackups
23:30:52 <Mer|in> just haven't gotten there yet
23:31:12 <Mer|in> also thinking about creating an audit table w/ triggers
23:31:45 <Mer|in> since i haven't checked whether postgres has anything like flashback
23:31:59 <warlord> Yeah, I mentioned that to Phil.. Please feel free to donate code that implements that... But keep in mind whatever you implement needs to be cross-platform (PG, MySQL, and SQLite)
23:32:40 <Mer|in> well in the case of triggers the syntax is going to be db-specific
23:34:03 <Mer|in> but if those are the only supported db drivers....one can always handle syntax quirks in case statements or db-specific scripts
23:35:46 <warlord> I dont know how Phil has it abstracted. I know we use DBI
23:36:12 <Mer|in> had a look around the postgres db - once 2.4.0 is released i'm hoping i have time to throw together a list of things that i think will help with performance / integrity
23:36:23 <Mer|in> like foreign-key constraints
23:37:23 <Mer|in> from reading the postgres logs it looks like all transactions are loaded @ startup
23:38:04 <warlord> Things like that should be done *BEFORE* 2.4.0
23:38:11 <warlord> No schema changes after 2.4.0
23:38:43 <warlord> yes.. GnuCash is not a DB App
23:38:49 <warlord> It just uses the DB to store data.
23:38:58 <warlord> Changing that would require *MAJOR* changes to gnucash.
23:39:11 <warlord> ... so it was just easier to load everything at startup for now.
23:39:14 <Mer|in> yea i've seen that mentioned in the mailing list
23:42:30 <Mer|in> i really only know a little about the supported DBs, i'm actually in a straight-oracle environment right now
23:43:04 <Mer|in> but things like foreign-key and audit tables rarely require code changes beyond the db itself
23:43:41 <Mer|in> especially if u use db transactions
23:43:59 <warlord> we should be using db txns
23:44:12 <Mer|in> i'm seeing them used in the pg backend
23:48:34 <Mer|in> but yea i can imagine about changing how the db is loaded into memory
23:49:33 <warlord> Yeah.. But that might require extra state be stored in the DB in order to compute things like account balances, etc.
23:52:37 <Mer|in> well as long as it stays in memory and the app doesn't creep into swap...
23:53:48 <Mer|in> but i'll bet if there was a long-term plan to allow the reporting engine to make direct db queries...
23:54:11 <warlord> we'd have to re-architect QOF to make that possible.
23:54:14 <Mer|in> i dunno about other users but some of my graphs can take up 20 seconds o load
23:54:15 <warlord> anyways, I need to head to bed.
23:54:21 <Mer|in> enjoy
23:54:43 <warlord> That has nothing to do with in-memory -- that's completely an algorithmic issue in the report implementation...
23:54:52 <Mer|in> ah
23:55:04 <warlord> many reports often obtain the list of splits and then perform various computations in scheme, running through the list over and over.
23:55:14 <warlord> Anyways, I'm off
23:55:15 *** warlord is now known as warlord-afk
23:55:17 <Mer|in> nite