2014-09-04 GnuCash IRC logs

00:12:16 *** Unhammer has quit IRC
00:17:52 *** Unhammer has joined #gnucash
01:03:35 *** GabrieleV has quit IRC
01:03:35 *** GabrieleV_ has joined #gnucash
01:03:36 *** GabrieleV_ is now known as GabrieleV
01:07:58 *** MechtiIde has joined #gnucash
01:50:57 *** MechtiIde has quit IRC
02:30:53 *** floh1111 has joined #gnucash
02:35:21 *** O01eg has quit IRC
03:11:24 *** fell_ has joined #gnucash
03:11:25 *** gncbot sets mode: +o fell_
03:25:44 *** gjanssens has joined #gnucash
03:25:44 *** gncbot sets mode: +o gjanssens
03:38:03 *** Mogget has quit IRC
03:38:06 *** Mogget has joined #gnucash
03:57:40 *** GabrieleV_ has joined #gnucash
03:57:43 *** GabrieleV has quit IRC
03:57:44 *** GabrieleV_ is now known as GabrieleV
04:17:53 <mikee> gjanssens: Good morning.
04:20:33 *** pinkman has joined #gnucash
04:20:50 <pinkman> Good morning everybody
04:21:38 <mikee> pinkman: Good morning.
04:23:22 <pinkman> oh hi, could you answer a question of mine?
04:24:50 <gjanssens> Hi mikee
04:31:32 <mikee> pinkman: Ask and find out.
04:32:38 <mikee> gjanssens: I've just update bug 735596. How do we encourage people to test stuff?
04:33:15 <gjanssens> mikee: I saw your patch apearing in the bug mails.
04:33:22 <gjanssens> I intend to check it out
04:33:45 <gjanssens> As to how to encourage others, I don't really know
04:34:03 <pinkman> thank you, I am new to gnu cash, so far it seems to be very good indeed, though I've encountered one problem:
04:34:15 <gjanssens> Your bugzilla comment talks about master and using the git command to apply the patch
04:34:50 <gjanssens> This implies the tester already needs to know how to use git, and know how to build gnucash from source
04:35:07 <mikee> It does.
04:35:09 <pinkman> I am used to print out all my transactions in chronological order at the end of the year and I thought that this is what the transactions register is there for.
04:35:22 <gjanssens> IMO that essentially limits the target audience to members of the gnucash-devel list
04:35:32 <gjanssens> So you may advertise your patch there.
04:35:42 <mikee> gjanssens: I'm open to suggestions
04:37:02 <mikee> gjanssens: I'll do that in a moment.
04:37:10 <gjanssens> Good
04:37:51 <gjanssens> Another alternative would be to just apply your patch to master and then ask people for feedback
04:38:13 <gjanssens> It may well break master, but you'll hear soon enough.
04:38:26 <mikee> :)
04:38:32 <gjanssens> The advantage of this is that you can also ask windows users to download the latest nightly build
04:38:37 <pinkman> But in Gnu Cash what happens after I've entered my transaction in the transaction register is it is also entered in the concerning account (which is fine) BUT the transaction doesn't stay in the register but disappears, which is useless to me if I want to have them in a list as mentioned.
04:38:55 <gjanssens> So you have a small addional audience that doesn't need to build before testing
04:38:58 <mikee> pinkman: The transaction report should do that.
04:39:13 <gjanssens> I'm not sure how large that audience would be though...
04:39:30 <gjanssens> So I don't know whether it really brings more testers or not
04:39:44 <pinkman> Is there a possibility to show me all the transactions I've made in a list?
04:39:58 <pinkman> in chronological order?
04:40:18 <mikee> It's worth a try though. I'll apply it to master later.
04:40:36 <gjanssens> mikee: ok
04:41:03 <gjanssens> pinkman: what do you mean your transactions disappear from the transaction register ?
04:41:14 <gjanssens> What do you mean exactly with "transaction register" ?
04:41:15 <mikee> pinkman: Ditto
04:41:30 <pinkman> after I've pressed enter it disappears from the register
04:41:50 <fell_> pinkman: View->Filter by ->Show all
04:42:09 *** fell_ is now known as fell
04:42:24 <gjanssens> pinkman: are you referring to the "General Ledger" ?
04:42:26 <pinkman> ty, I will try that out
04:42:49 <pinkman> yes to the general ledger, also called Journal in my language
04:43:09 <gjanssens> Ok, then probably fell's suggestion is what you are looking for.
04:43:32 <gjanssens> By default it shows only a month's worth of history
04:44:25 <pinkman> Wow that worked like a treat, THANK YOU very much
04:45:00 <gjanssens> Yay, fell scores :)
04:45:12 <gjanssens> You're welcome pinkman
04:45:24 <mikee> :)
04:45:42 <pinkman> Gnu Cash really is quite a good accounting program.
04:46:46 <pinkman> a revelation even
04:47:04 <mikee> Glad you like it.
04:47:16 * fell is wondering if we shouldn't a) make view all the default, because an active filter is confusing newbies,
04:47:32 <fell> b) make the filter setting persistent.
04:48:13 <mikee> It confused me too. I don't use it much/ever.
04:48:38 <gjanssens> fell: b) is an enhancement request, which I once looked at but for various reasons never got around to actually fix
04:49:11 <gjanssens> I have no idea why only a month is displayed. Performance perhaps ?
04:49:39 <gjanssens> The general ledger will show you ALL transactions you have entered. In a larger book that can be overwhelming
04:49:58 <fell> I assume. But OTOH a beginner has not much entries and later you can set the filter.
04:50:32 <gjanssens> True. I'm ok with changing the default if someone actually does it.
04:50:53 * gjanssens is looking at mikee's patch
04:51:29 <gjanssens> mikee: which editor do you use for the scm files ?
04:51:57 <gjanssens> I notice it adds/leaves a lot of trailing whitespace which clutters up the patch
04:56:36 <mikee> gjanssens: geany. I can tell it to remove trailing spaces.
04:56:56 <gjanssens> mikee: please do
04:57:44 <gjanssens> For example in patch 1/8, the taxinvoice.eguile.scm file has at least as much trailing whitespace changes as real changes.
04:57:50 <mikee> Not sure why I'd turned that option off.
04:58:18 <gjanssens> I have come across a few rare occasions where the trailing white space was important
04:58:52 <gjanssens> Usually the compile step tells me in those cases
04:59:06 <gjanssens> Obviously that won't happen with scm files :(
04:59:10 <mikee> Not in this case though. I'll fix the white space issue before I commit to master.
04:59:17 <mikee> x post
04:59:23 <fell> gjanssens: can't you move that spaces in the next line?
04:59:40 <gjanssens> fell: that's what I was about to suggest
05:04:10 <mikee> I recall at some point being told that 'fixing' trailing white space was bad (in that specific case) as the patch contained too many white space changes.
05:04:36 <mikee> In that instance it was the original files that had the white spaces.
05:04:45 <gjanssens> It would be bad to mix whitespace fixes in with other changes
05:05:15 <gjanssens> The goal is to separate whitespace fixes from real code changes to keep patches readable
05:05:16 <mikee> Indeed.
05:06:00 <gjanssens> So you can aplly the whitespace fixes in a separat commit.
05:06:08 <gjanssens> s/aplly/apply/
05:06:56 <gjanssens> Usually when I apply patches from others, I use git apply --whitespace=fix <patch>
05:07:51 <gjanssens> Since you will be applying your own commits, I can't of course :)
05:08:49 <gjanssens> By the way for some odd reason I can't apply our patch
05:09:33 <gjanssens> I get this error:
05:09:36 <gjanssens> error: report/business-reports/taxinvoice.css: does not exist in index
05:09:37 <gjanssens> error: report/business-reports/taxinvoice.eguile.scm: does not exist in index
05:10:21 <mikee> Um...
05:10:38 * gjanssens has never seen that before
05:10:53 <gjanssens> I don't see anything obviously wrong with your patch though
05:11:09 * mikee normally likes new things
05:11:46 <mikee> Not in this case though.
05:11:50 <gjanssens> :)
05:16:49 <mikee> I've just read it may be due to... whitespace errors. I'll do another patch.
05:17:10 <gjanssens> Right
05:18:08 <gjanssens> mikee: while you're at it, can you undo changing the title comment in taxinvoice.eguile.scm ?
05:18:19 <gjanssens> It's currently set to ebayinvoice.eguile.scm
05:18:37 <gjanssens> No doubt a left over from your first work
05:18:47 <mikee> Too late! But will do that now.
05:18:59 *** Krzysiek_K has quit IRC
05:19:48 <gjanssens> Man you're fast...
05:20:32 <mikee> I was already prepping it. :)
05:28:02 <gjanssens> mikee: the first commit in your new patch still has plenty of whitespace errors :(
05:28:27 <mikee> Fast != Good. I seem to have regressed the code somehow. Next patch should be correct, but I'll have to compile and test it again.
05:28:35 <gjanssens> And the error I get already happens in that commit
05:28:48 <gjanssens> Ok
05:29:06 <mikee> I'll need tea, lots of tea.
05:29:08 <gjanssens> Take your time :)
05:29:18 <gjanssens> And tea :)
05:54:43 *** zarchne has quit IRC
05:54:47 *** zarchne has joined #gnucash
06:10:48 <gjanssens> Thanks fell for creating the bug report for the date filter on the general ledger (and digging up the related report)
06:11:30 <fell> I just added the dependencies. :-)
06:14:33 <fell> Tea in the harem of Archimedes?
06:19:24 <mikee> fell: I just had to look that up. Looks interesting.
06:28:26 <mikee> Near black background with black dots to show spaces is not the best colour scheme in geany.
06:28:32 <mikee> Fixing that.
06:29:52 *** pinkman has quit IRC
06:50:53 *** Jimraehl1 has left #gnucash
06:59:34 *** Jimraehl1 has joined #gnucash
07:03:17 *** mlncn-agaric has joined #gnucash
07:06:06 <mikee> gjanssens: I've uploaded another patch. Went back to using traditional diff. There are days when git makes my head hurt.
07:10:02 *** omnimirage has joined #gnucash
07:11:31 <omnimirage> I'm just starting out and wish to ask a newbie question: I have an expense, which was transfered to my asset account, which automatically deducted the money, but my equity hasn't been automatically reduced
07:11:53 <omnimirage> Since the expense has lowered my asset and in turn, my equity, how do I correct this?
07:12:01 <omnimirage> And shouldn't it be automatically done?
07:12:06 <omnimirage> Sorry if the question is stupid
07:15:03 *** Jimraehl1 has left #gnucash
07:15:25 *** Jimraehl1 has joined #gnucash
07:20:16 *** himaxx has joined #gnucash
07:22:39 *** himaxx has quit IRC
07:23:19 *** mikee is now known as mikee-afk
07:25:12 <warlord> jralls: but shouldn't the 'fix' only happen once?
07:30:54 <warlord> omnimirage: you are confusing Equity, the account type, with "Instantaneous Equity", the Eq in the Accounting formula.
07:31:12 <warlord> In particular, Income and Expense are forms of Instantaneous Equity
07:31:25 <warlord> To compute your total Equity, run a Balance Sheet Report.
07:31:50 <warlord> ... and at the end of th eperiod you can run Tools->Close Books to roll up your income and expense into an Equity Account.
07:34:23 <gjanssens> mikee: the simple diff applies fine thanks
07:34:24 <gjanssens> I'll look at it this afternoon.
07:34:25 <gjanssens> What went wrong in git ?
07:40:39 <omnimirage> I seee. So it doesn't matter that my taps report an incorrect equity figure then, and I can just proceed?
07:40:47 <omnimirage> Thank you for the quick reply, btw
07:41:46 <warlord> "taps"?
07:43:31 <omnimirage> tabs*
07:43:35 <omnimirage> Sorry, am quite tired
07:43:47 <omnimirage> To be more detailed
07:44:09 <omnimirage> The main screen with all the accounts on it
07:44:52 <warlord> Ah, correct, the CoA only displays your Equity at a fixed point in time... Usually the time you started your accounting
07:45:10 <warlord> (unless you close the books, which will reset the Equity in the CoA to that point in time)
07:45:27 <omnimirage> What does close the books mean?
07:46:20 <warlord> omnimirage: <warlord> ... and at the end of th eperiod you can run Tools->Close Books to roll up your income and expense into an Equity Account.
07:46:36 <warlord> I already answered that question
07:46:40 <omnimirage> ah, I understand now
07:46:44 <omnimirage> Thank you
07:48:36 <warlord> you're welcome
08:02:44 *** omnimirage has quit IRC
08:17:20 *** himaxx has joined #gnucash
08:24:20 *** himaxx has quit IRC
08:24:43 *** himaxx has joined #gnucash
08:31:46 *** himaxx has quit IRC
08:32:08 *** himaxx has joined #gnucash
08:44:12 *** himaxx has quit IRC
08:44:37 *** himaxx has joined #gnucash
08:51:43 *** himaxx has quit IRC
08:52:06 *** himaxx has joined #gnucash
09:21:41 *** jimvideo has joined #gnucash
09:53:50 *** GabrieleV_ has joined #gnucash
09:54:02 *** GabrieleV has quit IRC
09:54:03 *** GabrieleV_ is now known as GabrieleV
10:01:28 <gjanssens> jralls: I didn't get your message because you added a colon to my nick :(
10:01:33 <gjanssens> Just read it now in the logs
10:02:28 <gjanssens> Isn't the correction from template namespace to user namespace supposed to be saved ?
10:03:30 <gjanssens> So that the next time you open your data file the commodity is correctly in the user namespace to begin with ?
10:03:54 *** mikee-afk is now known as mikee
10:04:23 <warlord> gjanssens: I asked the same question a couple hours ago
10:05:27 <gjanssens> warlord: I see. I didn't realize your question was about my issue as well
10:05:38 <warlord> it was specificaly about that. ;)
10:06:34 <gjanssens> :) "As well" implied my question vs your question in this case
10:07:12 * gjanssens 's English can sound funny sometimes :)
10:08:09 <warlord> gjanssens: yes, I know. My query to jralls was specifically about your issue.
10:09:40 <gjanssens> I did understand that. Oh, never mind :)
10:09:58 <warlord> Sorry, I'm still tired this morning
10:10:31 <gjanssens> Trouble sleeping with the toddler ?
10:10:57 <warlord> no, the toddler sleeps; it's the newborn. woke up at 6
10:11:48 <gjanssens> (Yay, another misuse of English terms by yours truly - toddler vs newborn)
10:12:30 *** mikee is now known as mikee-afk
10:18:18 *** Bodhi-Baum has joined #gnucash
10:24:14 <warlord> "Toddling" is the way a 1-2yo walks.
10:25:03 <warlord> A newborn (sometimes called an infant) is usually pre-walking
10:26:44 <gjanssens> Hey, thanks, learned something.
10:28:31 *** Azog has left #gnucash
10:34:35 *** O01eg has joined #gnucash
10:45:09 *** floh1111 has quit IRC
11:15:09 *** jimvideo has quit IRC
11:22:18 <warlord> Happy to help.
11:50:36 *** aqua___ has joined #gnucash
11:56:26 *** Gbarr has joined #gnucash
12:21:10 *** floh1111 has joined #gnucash
12:32:42 *** jralls has quit IRC
12:33:14 *** jralls has joined #gnucash
12:33:14 *** gncbot sets mode: +o jralls
12:34:41 *** floh1111 has quit IRC
12:36:26 <jralls> warlord, gjanssens: Yes, the change is supposed to be persisted.
12:40:56 <gjanssens> jralls: right. That's why I suggested it's a case of missing mark_instance. It's easier to spot in sql because xml writes the whole file whenever *anything* changes. So the missing persistence on this fix can easily go unnoticed there.
12:48:14 <jralls> Yup. I'm looking at that now. It's in gnc_commodity_new, which is called from the sql backend when loading the file. I wonder if I can safely run commit_edit in that context. Only one way to find out, I guess. :-/
12:51:01 *** MechtiIde has joined #gnucash
12:55:21 <warlord> jralls: I'm not sure you can.
12:55:59 <jralls> Actually, it's already in one. It may just need to be marked dirty.
12:57:31 <warlord> hmm
12:57:33 <warlord> ok
13:12:23 *** GabrieleV_ has joined #gnucash
13:12:29 *** GabrieleV has quit IRC
13:12:30 *** GabrieleV_ is now known as GabrieleV
13:57:17 *** cigarshark has joined #gnucash
14:04:43 *** mlq has quit IRC
14:12:43 <lmat> How is cutecash doing?
14:12:54 <warlord> lmat: it's a toy project
14:13:01 <warlord> apparently it compiles
14:13:03 <lmat> warlord: How do I build it ?
14:13:32 <warlord> cmake
14:31:24 *** GabrieleV_ has joined #gnucash
14:31:57 *** GabrieleV has quit IRC
14:31:58 *** GabrieleV_ is now known as GabrieleV
14:34:44 *** mlncn-agaric has quit IRC
14:35:32 *** Bodhi-Baum has quit IRC
14:44:16 <lmat> hmm, seems like it needs a find_package(Qt4 REQUIRED) ?
14:44:29 *** mlncn-agaric has joined #gnucash
14:48:40 *** MechtiIde has quit IRC
14:50:49 <lmat> oh, found the instructions! http://wiki.gnucash.org/wiki/Cutecash
15:01:34 <warlord> why do you care about cutecash?
15:14:25 <lmat> warlord: Lots of reasons!
15:14:56 <lmat> warlord: I'm interested in gnucash working toward c++, and cutecash is aimed in that exact direction
15:15:05 <lmat> warlord: (although from the "top" rather than the "bottom")
15:15:31 <lmat> warlord: Also, the success of cutecash can help ensure other successes in the future: such as a web-based gnucash installation, etc.
15:15:54 <warlord> I dont see how cutecash can lead to a web-based app.
15:16:51 <warlord> anyways, it was just a toy, a test. I dont think it's the direction it will go
15:18:59 <lmat> warlord: The way it leads to a web-based app is that it's a separate front-end for gnucash. As soon as the business logic is sufficiently separated from the UI (enough to put a different front end on), that front end can be anything: ncurses, web app, qt, etc.
15:19:07 <lmat> that's the hope, anyway :)
15:19:20 *** aqua___ has quit IRC
15:19:33 <warlord> Except with Cutecash the biz logic hasn't been separated, just re-implemented
15:20:21 <lmat> warlord: :(
15:20:34 <lmat> warlord: Hopefully it can at least be identified? So that it can be placed intelligently...
15:20:47 <warlord> We dont need cutecash for that
15:21:22 <lmat> so we look at the overlap between cutecash and the gnucash gui layer and put the ...what's it called, cross-section?... into a separate layer
15:21:59 <warlord> lmat: we dont need to do that.. I think the developers already know quite clearly what needs to be refactored out.. That's not the hard part
15:22:11 <warlord> Identification of the biz logic isn't the hard part
15:22:19 <warlord> Actually separating it out into a clear API, *THAT* is the hard part.
15:22:23 <lmat> ah
15:22:27 <warlord> And I dont think we need cutecash for that
15:24:32 <lmat> sure, makes sense :)
15:32:40 <lmat> cutecash is only a proof of concept?
15:46:04 <warlord> yes
15:47:23 <lmat> What concept is being proved?
15:47:29 <lmat> (or proven? ^_^)
15:48:21 <warlord> i dont know -- you would ask to ask cstim
15:48:54 <lmat> After reading the wiki entry on it, it looks like an alternative development effort!
15:56:41 <jralls> Christian wanted to see if he could make a Qt version of GnuCash because he likes it better than Gtk. He got stuck because the MVC separation in GnuCash is poor; there's too much "business logic" mixed in with the GUI code.
15:57:13 <lmat> jralls: ahh, makes sense
15:59:19 <jralls> warlord: The edit/commit for gnc-commodity's namespace change doesn't work because the SQL backend (sensibly) sets a "loading" flag which turns commit into a no-op.
16:00:08 <jralls> Any thoughts on a good way to trigger a commit after the load is done?
16:04:08 *** mlncn-agaric has quit IRC
16:07:15 *** fell_ has joined #gnucash
16:07:16 *** gncbot sets mode: +o fell_
16:09:13 *** fell has quit IRC
16:15:48 <warlord> jralls: add it to a list that get post-processed after-load?
16:16:14 <warlord> e.g., leave them all in begin_edit() and add them to a list, then after load flush the list and commit_edit?
16:22:24 * jralls ponders. We don't virtualize commit and the qof_instance version needs to be called with callbacks, so it would have to be a list per-type.
16:23:37 *** mlncn-agaric has joined #gnucash
16:23:40 <jralls> So it will have to be list-per-class, of which there will be only one for now.
16:25:49 <warlord> Yes, it would
16:26:29 <jralls> Even so, hard to do without coupling the backend and GncCommodity. Maybe a signal?
16:27:18 <jralls> I'll have to think about that a bit. In the meantime, I have to go to an appointment.
16:28:38 <warlord> Yeah, I was thinking a callback of some sort.
16:29:49 <warlord> But can't you just call qof_commit_edit() on the objects?
16:47:24 *** jralls has quit IRC
16:48:16 *** jralls has joined #gnucash
16:48:17 *** gncbot sets mode: +o jralls
16:49:47 *** rpg has joined #gnucash
17:57:23 *** GabrieleV_ has joined #gnucash
17:57:38 *** GabrieleV has quit IRC
17:57:38 *** GabrieleV_ is now known as GabrieleV
18:04:08 *** lmat has quit IRC
18:07:49 *** jralls has quit IRC
18:07:50 *** fell_ has quit IRC
18:07:53 *** fell_ has joined #gnucash
18:07:54 *** gncbot sets mode: +o fell_
18:08:26 *** jralls has joined #gnucash
18:08:26 *** gncbot sets mode: +o jralls
18:17:15 <jralls> warlord: No, qof_commit_edit() just marks them clean. It's qof_commit_edit_part2() that calls the backend, and it takes as arguments three subclass-specific callbacks, so we really need to call the per-class commit_edit function.
19:01:26 *** rpg has quit IRC
19:20:14 *** gjanssens has quit IRC
19:24:11 *** mlncn-agaric has quit IRC
19:46:26 *** GabrieleV_ has joined #gnucash
19:46:26 *** GabrieleV has quit IRC
19:46:26 *** GabrieleV_ is now known as GabrieleV
20:01:26 *** GabrieleV_ has joined #gnucash
20:01:26 *** GabrieleV has quit IRC
20:01:26 *** GabrieleV_ is now known as GabrieleV
20:19:12 <warlord> jralls: okay
20:19:30 <warlord> then yes, we need pre-class lists.
20:19:47 <warlord> or a list of class+object tuples
20:21:34 *** jralls has quit IRC
20:22:16 *** jralls has joined #gnucash
20:22:17 *** gncbot sets mode: +o jralls
20:47:05 *** mlncn-agaric has joined #gnucash
20:49:43 *** fell_ has quit IRC
20:50:06 *** fell_ has joined #gnucash
20:50:06 *** gncbot sets mode: +o fell_
21:35:52 *** cigarshark has quit IRC
22:22:22 *** fell_ has quit IRC
23:16:50 *** todd has joined #gnucash
23:23:33 *** mlncn-agaric has quit IRC
23:53:33 *** Coderjoe has quit IRC
23:58:46 *** Coderjoe has joined #gnucash