2007-01-24 GnuCash IRC logs

00:00:02 <warlord> I found encfs a year or two ago when CryptFS stopped working for me.
00:00:14 <warlord> heh
00:01:51 <fen> why is gnucash win32 installer 30mb? does it include GTK libs?
00:04:01 <warlord> Yes
00:06:53 <warlord> it includes everything you need.. every library..
00:09:09 * benoitg is looking for a general solution to OFX encoding problems
00:09:26 <warlord> benoitg: a general solution? Hmm.
00:09:41 <jsled> benoitg: what's the problem?
00:10:32 <warlord> doesn't OFX declare an encoding?
00:10:51 <benoitg> Basically, libofx will return the ofx file in whatever encoding it's originally in. In practice OpenSP provides absolutely no usable encoding conversion functions.
00:11:14 <warlord> benoitg: is there some encoding declaration in the OFX protocol?
00:11:25 <benoitg> Yes
00:11:53 <benoitg> Before we switched to UTF-8 everywhere, it happened to just work most of the time, and the bug we had masked the fact that we have a general problem.
00:12:13 <benoitg> So I'll probably have to link libofx with iconv and parse the headers myself.
00:12:26 <benoitg> Rewrite the file, and then parse it.
00:13:00 <warlord> well, if libofx can at least tell gnucash what encoding the file is, gnucash can do the conversion.
00:13:24 <warlord> unless you're having parse errors due to encoding issues?
00:13:57 <benoitg> No parse error, but ofxdump output is garbled in a UTF-8 locale, which is annoying.
00:14:18 <warlord> Oh, there is that..
00:14:35 <benoitg> Besides, it's really libofx's job to provide at least a known encoding.
00:14:38 <warlord> I'm not sure whether libofx or the application should be in charge of encoding conversion.
00:14:56 <warlord> true, but a libofx_get_encoding() API would suffice.
00:15:23 <warlord> not that I'm complaining about libofx deciding to do the conversion on its own.
00:16:06 <benoitg> The other problem is the size of the buffers.
00:16:14 <warlord> oh?
00:16:23 <warlord> you're not using std::string? ;)
00:16:49 <benoitg> Internally, yes, externally, no.
00:17:23 <warlord> So what's the issue with buffer size?
00:19:24 <benoitg> The API is C-style, and specifies the max size of each element. I can't find in the spec for 1.6 if the size limit applies to the number of characters, or the number of bytes.
00:20:04 <warlord> Oh. It's not just a NULL-terminated const string owned by libofx?
00:20:20 <warlord> (or even a NULL-terminated non-const string owned by the application)?
00:20:21 <benoitg> Yes. There may not even be UTF-8 files in the wild, so it's possible nobody knows for sure.
00:21:36 <warlord> okay, so the buffer sizes aren't in the API.
00:22:13 <benoitg> They are.
00:22:26 <warlord> Oh? It's not dynamic?
00:22:35 <benoitg> No
00:22:54 <warlord> Oh. Boo!
00:24:50 <benoitg> Oh well, at worst the lib will strib a few chars at the end of utterly long values.
00:25:36 *** fen has quit IRC
00:25:51 <benoitg> ?/?"/$%"?/?$&%&&
00:25:57 <benoitg> The spec does say
00:26:18 <warlord> Huh?
00:26:46 <benoitg> Size is the number of charecters, meaning I got it completely wrong 5 years ago.
00:27:56 <warlord> Oh. OOPS!
00:28:08 <warlord> so #chars, not #bytes?
00:30:15 <benoitg> Yep. Back then I didn't know how to write an API in plain C API that works with dynamic string lengths anyway, but now it's going to have to wait for an API redesign.
00:32:26 <warlord> Oh, the way I would've done it is just have the API be something like: foo_t foo(..., char** pString, ...) Then libofx could allocate the string.
00:32:44 <warlord> Or you could make it a const; part of the parse-context.
00:32:52 * warlord shrugs
00:32:58 <warlord> next API rev.
00:36:36 <benoitg> warlord: Back then I was fresh out of school, the initial API didn't even have a context pointer, so I didn't know how to deal with memory leaks other than by making the entire thing copyable as as struct ;)
00:38:24 <warlord> Hehehe...
00:38:28 <warlord> Live and learn.
01:57:31 <warlord> Okay, good night all
01:57:33 *** warlord is now known as warlord-afk
03:18:13 *** RallyU has joined #gnucash
03:40:53 *** foo has quit IRC
03:41:35 <benoitg> Ok, libofx will now "do the right thing" as far as gnucash is concerned.
04:20:26 *** ErKa has joined #gnucash
05:07:40 *** cstim has joined #gnucash
05:07:41 *** gncbot sets mode: +o cstim
05:07:57 * cstim adds http://lists.gnucash.org/logs/2007-01-23.html#T23:08:12 to FAQ...
05:09:09 <cstim> hampton: http://www.ironicsans.com/2007/01/the_astoria_notes.html some funny story about NYC appartment life.
05:47:04 *** bonez39 has quit IRC
06:35:24 *** twunder has joined #gnucash
06:58:16 *** twunder has quit IRC
06:58:49 <cstim> bugzilla crossed the line to numbers >= 400,000. Except that apparently nobody created a funny bug#400000, like so: http://bugzilla.gnome.org/show_bug.cgi?id=100000
06:59:13 <CSWookie> cstim: Sounds like a bug, to me.
06:59:49 <cstim> sure :-)
07:05:48 <cstim> can you reproduce it?
07:13:18 *** conrad has joined #gnucash
07:13:18 *** gncbot sets mode: +o conrad
07:23:11 *** terence has joined #gnucash
07:26:39 *** ErKa has quit IRC
07:30:08 *** conrad has quit IRC
07:50:58 *** andi5 has joined #gnucash
07:50:59 *** gncbot sets mode: +o andi5
07:57:37 <cstim> hi andi5. Quite early for you to show up in irc?
07:58:44 <andi5> hi... yes, but i will not really be here, it is just easier than reading the complete logs in one go
07:59:00 <cstim> :-)
07:59:16 <cstim> andi5 -> andi5_not_really_here
07:59:27 *** andi5 is now known as andi5|nrh
08:00:58 <cstim> heh
08:21:43 *** ErKa has joined #gnucash
08:25:42 <terence> just wonder if anyone has used GNUCash to manage inventory?
08:25:54 <terence> I don't think it has that feature.
08:27:19 <cstim> terence: no, it doesn't have that feature.
08:28:36 <terence> thanks cstim.
08:33:33 *** twunder has joined #gnucash
08:39:13 *** conrad has joined #gnucash
08:39:14 *** gncbot sets mode: +o conrad
08:43:47 *** terence has quit IRC
09:02:59 *** ceplma has quit IRC
09:03:42 *** ceplma has joined #gnucash
09:15:53 *** twunder has quit IRC
09:42:18 *** ErKa has quit IRC
09:48:24 *** warlord-afk is now known as warlord
10:04:34 *** prock has quit IRC
10:19:42 *** kielein has joined #gnucash
10:22:36 *** RallyU has quit IRC
10:35:00 *** Demitar_ has quit IRC
10:35:05 *** twunder has joined #gnucash
10:35:43 *** andi5|nrh has quit IRC
10:45:42 *** ErKa has joined #gnucash
10:50:07 <cstim> can someone reproduce http://bugzilla.gnome.org/show_bug.cgi?id=394074 ? Instructions seem to be sufficient to do this.
10:55:07 <cstim> warlord: I think the email from Peter McAlpine is exactly the same issue as what we've discussed here yesterday. I think it's also the same reporter that brought it up here.
10:56:14 <warlord> I think so, too, but it still belongs on -devel and not on -user ;)
10:56:36 *** twunder has quit IRC
10:57:56 <cstim> yes, absolutely
11:00:35 <warlord> I cannot reproduce 394074.
11:01:02 <cstim> you mean, it doesn't crash for you? That's bad. I mean, good.
11:01:03 <cstim> :-)
11:02:09 <hampton> cstim: Funny web site. Thanks.
11:07:35 <cstim> does anyone know how to get gcc-4.0.x to display only the first ~100 lines of error messages and not spill out 10,000 lines so that I cannot see the initial error :-/
11:08:19 <cstim> or how can I pipe stderr in bash?
11:08:25 <cstim> (instead of stdout)
11:09:08 <warlord> cstim: instead of stdout or in addition to? You can redirect stderr to a file! 2>/tmp/errors.out
11:09:25 <cstim> yes, I could do that.
11:32:38 *** wizkid238 has quit IRC
11:35:32 <hampton> you can also redirect it to stdout: gcc 2>&1 | head -1000
11:35:32 *** CSWookie has quit IRC
11:39:59 *** ErKa has quit IRC
11:52:15 *** wizkid238 has joined #gnucash
12:02:34 *** cstim has quit IRC
12:13:49 *** andi5 has joined #gnucash
12:13:49 *** gncbot sets mode: +o andi5
12:17:24 *** bonez39 has joined #gnucash
12:17:44 *** twunder has joined #gnucash
12:21:42 *** wizkid238 has quit IRC
12:30:29 *** wizkid238 has joined #gnucash
12:46:40 *** imork has joined #gnucash
12:48:15 <imork> Hi, this might be a Stupid Question, but are all these .xac and .log files important?
12:48:34 *** MrN has joined #gnucash
12:49:02 <MrN> hi
12:49:46 <andi5> imork: http://wiki.gnucash.org/wiki/FAQ
12:58:17 <andi5> btw, i can reproduce #394074 pretty easily (have not tried with trunk yet, because i updated slib again ;-))
13:04:15 <warlord> andi5: maybe it's a debian/ubuntu-specific thing?
13:05:10 <andi5> i do not really think so... you said that you can still see the dialog, right? use it, especially check treeviews
13:05:51 <andi5> sometimes it suffices just to move the mouse... well, i will try to fix it
13:06:20 *** prock has joined #gnucash
13:06:49 <warlord> "check treeviews"?
13:07:08 <warlord> Yeah, I can still see the dialog.. Really, the dialog should get closed on book-close.
13:08:38 <andi5> well, i just open the balance sheet, open another file and play a bit with the dialog... it used to crash then... hmmm
13:08:48 <prock> I'm sorry for posting my email re bug 394074 to -user instead of -devel... I guess I wasn't thinking too clearly last night
13:09:10 <prock> prock === Peter McAlpine
13:09:23 *** bonez39 has quit IRC
13:10:48 <andi5> warlord: http://pastebin.ca/raw/327058 is an example backtrace... i do not have it opened in gdb anymore, though
13:11:00 <prock> oh it wasn't 394074
13:11:48 <andi5> prock: those 6 numbers are not random and all combinations point to the same bug ;-)
13:12:03 <prock> 373809
13:12:18 <prock> my bug is the only bug that matters, right?
13:13:23 <andi5> prock: i am working on it ... but i also face a rhythmbox bug and i cannot hear NO MUSIC....
13:14:00 <prock> cannot hear no music means music never stops?
13:14:12 <prock> kill -9 now work on my bug ;)
13:14:14 <prock> j/k
13:14:37 <andi5> well, andi5 is a child process of rhythmbox...
13:16:45 <andi5> prock: sorry, i got distracted by the fact that gnome-terminal is damn slow (again?) and then i tried all sorts of emacs shells just to find out that what i used is best aligned with my habits... *back*
13:22:28 <jsled> no music!@??!
13:23:08 *** jpeach has joined #gnucash
13:23:39 <andi5> weird.. it works again.. *clicks into tray list and never touches again*
13:32:52 <jpeach> Hi
13:32:52 <gncbot> jpeach: Sent 5 days, 21 hours, and 9 minutes ago: <jsled> (re: private message) Yes, I do! And yes, it is!
13:33:48 <jpeach> Where can I get version 2.0.4 as a deb package. The one on http://www.gnucash.org/pub/gnucash/ is version 1.X
13:34:17 <warlord> andi5: sorry, I stepped AFK for a while. Im back now.I only tried the txn report,and I didn't try editing the accounts page.. So it's probably limited to certain reports and/or certain options.
13:35:19 <warlord> jpeach: from Debian. We don't distribute packages.
13:36:08 <jpeach> warlord: Oh... ok, that is the way that I installed it first. However, the latest one in the standard repositories is 2.0.2.
13:36:13 <warlord> andi5: the real fix is to make the options window(s) go away when the book is closed.
13:36:23 <andi5> warlord: i know
13:36:33 <jpeach> warlord: Is there a place where I can download the binary for 2.0.4?
13:36:34 <warlord> jpeach: then send mail to the debian gnucash maintainer.
13:36:44 <andi5> wtf.... can a guile object be a list and vector at the same time?
13:37:04 <warlord> jpeach: No clue. As I already said, we don't distribute binaries. (Well, except for Windows)
13:37:40 <jpeach> warlord: Where would I find the email address for debain gnucash maintainer?
13:38:41 <andi5> jpeach: http://packages.qa.debian.org/g/gnucash.html
13:39:15 <jpeach> andi5: thanks!
13:39:38 <andi5> oh, i mean list and array at the same time..
13:43:44 <warlord> andi5: what do you mean, "list and array"?
13:44:12 <andi5> well, i have a "seq" thing and (list? seq) and (array? seq) both return #t
13:45:04 <warlord> I'm not sure I understand the difference between a list and an array in guile.
13:46:27 <andi5> ok, arrays come from slib/array.scm.... i will just try to find out, the truth must be out there...
13:47:28 <warlord> :)
14:21:59 *** jpeach has left #gnucash
14:31:57 <andi5> boah, looks like a horrible nesting of bugs... wtf...
14:37:25 <warlord> LOL
14:54:47 <andi5> ok... here is the basic problem... prior to slib 3a4 there was only *features* ... guile initiated it, slib/guile.init added what it thought guile is able to do itself (e.g. array, but neither sort nor record) ... (require 'record) will test *features* -> use guile record support
14:54:47 <andi5> now we have *features* and slb:features... the primer is untouched, slib:features is supposed to contain what the implementation can do natively, but it still does NOT contain sort or record.... so (require 'record) will now load slib's record.... making boom if we mix both
14:56:58 <andi5> i guess we could just always (require 'record) or always not require it... but there is another problem with array/sort i am still trying to "sort" out ;-)
14:58:52 <warlord> Hmm...
15:09:51 *** jpeach has joined #gnucash
15:09:54 *** jpeach has left #gnucash
15:12:35 *** |gunni| has joined #gnucash
15:13:10 <andi5> warlord: can you test this one for me? guile -c '(let ((a (list 5 6))) (if (array? a) (display "Is an array\n")) (array-shape a))'
15:19:03 <warlord> [warlord@cliodev ~]$ guile -c '(let ((a (list 5 6))) (if (array? a) (display "Is an array\n")) (array-shape a))'
15:19:03 <warlord> Is an array
15:19:03 <warlord> ERROR: In procedure map:
15:19:03 <warlord> ERROR: Wrong type argument in position 2: #f
15:19:17 <andi5> does that look promising to you?
15:19:41 <warlord> promising how?
15:19:52 <andi5> well, s,promising,goo,
15:19:54 <andi5> +d
15:21:45 <warlord> I guess...
15:22:01 <warlord> It would be nice if we didn't need slib at all.
15:22:57 <andi5> maybe
15:29:09 <prock> "+d" means?
15:29:26 <andi5> goo + d = good
15:29:42 <prock> lol
15:29:57 <andi5> it is a simply chemical reaction
15:30:00 <warlord> You mean it's not a barely-passing goo?
15:30:15 <andi5> no, it is not
15:30:57 <warlord> a goo that received a grade of d+?
15:31:02 <andi5> [ot] seems like gnome gets a new control center
15:49:19 <jsled> andi5: yeah, it looks not too bad.
15:58:00 *** ErKa has joined #gnucash
16:07:12 *** andi5 has quit IRC
16:50:38 *** bonez39 has joined #gnucash
17:01:48 *** twunder has quit IRC
17:17:53 <prock> -c '(let ((a (list 5 6))) (if
17:17:54 *** |gunni| has quit IRC
17:18:08 <warlord> prock: huh?
17:18:21 *** ErKa has quit IRC
17:18:58 <prock> pasted wrong window
17:21:23 *** |gunni| has joined #gnucash
17:21:54 *** |gunni| has joined #gnucash
17:24:18 *** prock has quit IRC
17:29:25 *** prock has joined #gnucash
17:34:18 <prock> le gasp... If I'm to write a new report it must be done with scheme
17:35:02 <prock> all those parenthesis make my brain hurt
17:35:54 <warlord> prock: Eh, emacs makes it easy.
17:36:04 <prock> warlord: that's a loaded statement
17:36:21 * prock starts an editor war
17:36:38 *** blahdeblah has joined #gnucash
17:37:31 <blahdeblah> Does gnucash have a standard field to indicate the party a transaction was enacted with? e.g. the payee of an expense, or the source of income?
17:38:07 <blahdeblah> (I'm using GnuCash 2.0.2 r14936 on Debian etch.)
17:38:45 <warlord> blahdeblah: "Description"?
17:39:53 <blahdeblah> Wouldn't that be used as a more general description of what the transaction was about?
17:40:20 <warlord> Maybe.
17:40:38 <warlord> You've got three things. The TXN Desc, TXN Notes, and Split memos.
17:40:51 <warlord> Use them however you wish ;) But generally DESC == Payee.
17:41:11 <warlord> Notes is more of a general txn memo/description
17:41:22 <warlord> Memo is a per-split, "what does THIS split mean" entry.
17:41:26 <warlord> anyways, gotta run.
17:41:29 *** warlord is now known as warlord-afk
17:41:34 <blahdeblah> I was hoping for something that could be reported on easily.
17:41:48 <blahdeblah> e.g. Show me all my expenses at this place, or all my income from that source.
17:42:02 <prock> blahdeblah: accounts are what you want
17:42:19 <prock> blahdeblah: create an account for each expense or income source
17:42:20 <blahdeblah> prock: So i should have a separate account for each payee and client?
17:42:30 <blahdeblah> OK
17:42:42 <prock> blahdeblah: yes if you want to track on that level of detail
17:42:50 <blahdeblah> Seems a little tedious, especially since they can't be created in the middle of a transaction.
17:43:07 <prock> blahdeblah: they can't?
17:43:37 <blahdeblah> You must know something i don't. :-)
17:46:20 <prock> blahdeblah: just type the name of the account you want to transfer to and if it doesn't exist it will ask if you want to create it.
17:46:37 <blahdeblah> OK - i'll give that a shot.
17:47:43 <blahdeblah> Where would be the normal place to put clients of a business (as an income source)? I don't understand the "accounting equation" well enough to work out whether it should be income or assets.
17:49:54 <prock> blahdeblah: my experience is limited to personal use but I have top level accounts "Income" and "Expenses". Sub-accounts under "Income" include "Salary", "Rental Income", "Gifts". Under expenses: "Electric", "Gas", "Food" (which also has subaccounts)
17:50:25 <blahdeblah> Hmmm... Looks like i need to do a lot more reading in the manual. :-|
17:50:30 <blahdeblah> Thanks for your help
17:51:19 <prock> what version of glade is used for trunk/src/gnome/glade?
17:51:37 <prock> s/used/recommended/g
17:52:54 *** benoitg has left #gnucash
17:56:04 <jsled> prock: I've got 2.6.8 installed. I'm not sure that it matters, though I know hampton has used a script to normalize the .glade file content in the past; I'm not sure of its current status.
17:56:14 *** blahdeblah has left #gnucash
18:02:19 *** kielein has quit IRC
18:04:49 *** blahdeblah has joined #gnucash
18:06:53 <blahdeblah> I'm experiencing random lockups in GnuCash on my Debian etch 64-bit system. Where's the best place to report it, and what should i include?
18:09:47 <prock> blahdeblah: reportbug with debian... but first check to make sure one doesn't already exist for your problem (bugs.debian.org)
18:10:12 <blahdeblah> Will do
18:23:01 <prock> are the .glades actually edited with the glade GUI or do people hack them by hand?
18:25:01 <hampton> I've been known to hack them by hand on occasion, but generally the glade gui is what's used
18:25:57 <hampton> util/glade-fixup is the tool used to strip out attributes inserted by the glade gui that aren't supported on the targeted level of gtk.
18:26:12 <prock> ah thanks for the tip
18:26:30 <hampton> It needs to be updated since the base level has moved from gtk2.4 to gtk2.6
18:31:35 <hampton> Hmpf. Looks like all of the stripped attributes but one are supported in gtk2.6. I'll have to edit some glade files and see if there are new to 2.10 attributes that need stripping.
18:36:36 *** |gunni| has quit IRC
18:43:38 * hampton goes to install FC4 in vmware so he has a gtk2.6 based system...
18:57:52 <jsled> I think I've maybe changed a vbox identifier by hand. Or changing a dialog default size. I can't imagine doing anything more complicated.
19:03:09 <prock> jsled: if I were to add the ability to disable/enable an sx would you have an opinion where the option for that goes? my version adds a checkbox as the first row of "Options" in the "Edit Scheduled Transaction" dialog
19:04:11 <jsled> That sounds like a good place for it.
19:05:00 <jsled> I mean, the SX editor is a bit of a mess, so it doesn't make it much worse.
19:05:03 <jsled> s/so/but/
19:05:38 <jsled> But ... why not just set the SX to "none".
19:05:38 <jsled> ?
19:11:04 <prock> jsled: because when I write my account projection report I want to be able to select sx's that aren't active 'for real'
19:16:39 <jsled> Ah.
19:18:56 <jsled> if it's part of the report, it might be better to do it at that point.
19:19:10 <jsled> Similar to selecting the to-be-reported-over accounts in the options.
19:20:15 <prock> but then I'd have to create all my sx's every time I want to run the report
19:21:21 <jsled> hmm?
19:21:35 <jsled> No, it'd be a multi-select list, probably defaulting to "all" sxes.
19:21:45 <jsled> You'd just unselect the ones that shouldn't be a function of the report.
19:21:58 <jsled> s/a function of/in/
19:22:03 <prock> right... the issue isn't keeping things off the report it's keeping things off the SLR
19:22:23 <jsled> Ah, right.
19:22:38 <prock> like I want a report to see if I can afford a new $5000 camera but I don't want it scheduled for really real yet
19:22:51 <jsled> Not to disrupt your momentum, but you might want to do a quick writeup of your design to -devel.
19:23:22 <prock> don't worry there's very little momentum to disrupt =)
19:23:37 <prock> and yes I'll put something together with UI mockups, etc.
19:24:49 <jsled> cool. Unless you're going to do them anyways, we can probably get by w/o mockups unless it's not clear what it'll be. It sounds like it should be.
19:26:00 <prock> my hope is that it will be an in-obtrusive way to meet my personal (and hopefully other's) budgeting needs... I really don't want to start (another) huge thread about the subject.
19:32:57 <prock> I'd like to have some sort of working(-ish) enable/disable patch before I announce my intentions though.
19:35:15 *** jharold has joined #gnucash
19:41:32 <prock> jsled: can I ask what your thought on it are? what do you think the response to my suggestion will be?
19:42:56 *** bonez39 has quit IRC
19:43:14 <jsled> Well, even better would be to populate a budget from a set of SXes.
19:43:48 <prock> yes... although to be honest I don't like the budgets in gnucash
19:44:17 *** MrN has quit IRC
19:44:30 <jsled> But that's a bigger code cahnge. I'm pretty fine with the idea ... the SX "enabled" flag ... a report to actually do the work.
19:44:45 <jsled> The flag will become a new data-format item, so we might have to plan for that a bit.
19:46:25 <prock> I thought about that... everything defaults to enabled so loading old .xml should be fine and I'm assuming old versions reading new .xml files will just ignore the tag
19:47:00 <prock> mind you I haven't looked at the file writing code at all
19:50:39 <jsled> no, they will not ignore the tag.
19:51:02 <jsled> I believe we croak if we see unexpected tags.
19:51:20 <prock> :(
19:52:13 <jsled> indeed.
20:12:51 *** rauch has quit IRC
20:16:46 *** imork has quit IRC
20:50:00 *** jharold has quit IRC
20:57:41 *** jharold has joined #gnucash
21:12:24 <prock> does --loglevel affect stdout output or just output to the .log file?
21:14:38 *** twunder has joined #gnucash
21:14:47 *** warlord-afk is now known as warlord
21:14:57 <warlord> prock: neither. it affects output to /tmp/gnucash.trace
21:15:35 <prock> warlord: even better thanks
21:39:55 <blahdeblah> Is there a way to customise the headings and the like for invoices created in gnucash?
21:41:43 <warlord> nope.
21:42:01 <warlord> at least not without changing the scheme code or changing the translation.
21:42:12 <warlord> (I assume you mean the column headings)
21:44:32 <blahdeblah> No, the part where it says: Invoice #333333
21:44:57 <blahdeblah> In .au, we are legally required to have the words "Tax Invoice" prominent on all invoices.
21:46:16 <warlord> *sigh* So I've heard. What a stupid law. No, you can't change that without modifying the scheme.
21:46:35 <blahdeblah> Where is the scheme stored?
21:46:49 <blahdeblah> I'm happy with a quick hack if it gets me running. ;-)
21:46:56 <warlord> (or you could create a po/en_AU.po that changes "Invoice" to "Tax Invoice")
21:47:03 <warlord> invoice.scm
21:47:07 <blahdeblah> And stupid laws are standard the world over...
21:47:09 <blahdeblah> Thanks
21:47:15 <warlord> Not THAT stupid!
21:47:36 <warlord> "An invoice has to say Tax Invoice"? Like, are there different kinds of invoices?!?
21:47:44 <blahdeblah> One could argue that customisable headers and footers are an extremely important feature...
21:48:20 <warlord> Send in a patch.
21:48:44 <blahdeblah> Will investigate doing so - what's the best address to mail them to?
21:48:52 <warlord> Your other option is to use Edit -> Stylesheets and create a GnuCash stylesheet with your header.
21:48:57 <warlord> gnucash-devel
21:49:03 <blahdeblah> @?
21:49:03 <gncbot> blahdeblah: Error: "?" is not a valid command.
21:49:09 <hampton> or bugzilla.gnome.org
21:49:31 <blahdeblah> warlord: gnucash-devel@?
21:49:36 <blahdeblah> gnucash.org?
21:49:38 <blahdeblah> gnome.org?
21:49:52 <hampton> gnucash.org
21:49:58 <blahdeblah> OK
21:50:07 <blahdeblah> BTW, here in the great land of Oz, there are actually non-tax invoices as well as tax invoices. ;-)
21:51:10 <blahdeblah> On the stylesheet issue, GnuCash 2.0.2 doesn't seem to have any options for editing the fixed text. Is that stored in a file, too?
21:52:14 <warlord> the "fixed text" is in invoice.scm
21:53:06 <blahdeblah> And to create my own stylesheet for invoices i need to add blah-invoice.scm?
21:54:20 <blahdeblah> Or is invoice.scm the same as a stylesheet of default?
21:54:37 <warlord> invoice.scm is the report.
21:54:43 <warlord> a stylesheet is completely different.
21:55:01 <warlord> I was thinking that you could create a stylesheet that adds the text "Tax Invoice"
21:55:04 <warlord> and then it would say:
21:55:07 <warlord> Tax Invoice
21:55:10 <warlord> Invoice #XXXXXX
21:55:27 <blahdeblah> Yep
21:55:57 <warlord> Edit -> Stylesheets
21:56:02 <warlord> to create a stylesheet.
21:56:08 <warlord> then just apply it to your invoice rpeort.
21:57:19 <blahdeblah> I must be thick - where is the option to add text?
21:59:27 <warlord> Might have to add it as an image.
21:59:39 <warlord> I might be easier to just edit invoice.scm
22:00:04 <blahdeblah> OK - i'll have a look there.
22:01:20 <warlord> blahdeblah: http://wiki.gnucash.org/wiki/FAQ#Q:_How_do_I_change_.22Invoice.22_to_.22Tax_Invoice.22_as_required_in_Australia.3F
22:01:40 <blahdeblah> ah thanks
22:01:45 * warlord should've just pointed you there originally.
22:04:49 <blahdeblah> Yep - it has some good info.
22:05:19 <blahdeblah> One thing that confuses me there is that now there seems to be a 5-account naming scheme, not 3 or 4.
22:05:44 <blahdeblah> (I'm very much a newbie when it comes to accounting stuff & running my own business.)
22:16:08 *** hampto1 has joined #gnucash
22:16:09 *** gncbot sets mode: +o hampto1
22:18:42 *** hampton has quit IRC
22:18:55 *** hampto1 is now known as hampton
22:28:18 <blahdeblah> Hmmm... Is there anything special i need to do to make changes to invoice.scm activate?
22:30:23 <warlord> you need to restsrt gnucash.
22:30:59 <blahdeblah> Done that - no joy
22:32:16 <warlord> Where did you edit the report? Are you sure your edited report is installed in the proper location where GnuCash is looking for it?
22:34:12 *** conrad has quit IRC
22:36:58 <blahdeblah> I'm using the Debian etch package, which installs the invoice files in /usr/share/gnucash/guile-modules/gnucash/report/ AFAICT
22:41:26 <prock> grarg... so I need the glade-gnome package... of course. hours_left_to_live -= 3;
22:48:50 *** richard has joined #gnucash
23:09:05 *** twunder has quit IRC
23:10:38 <richard> Hello. I've tried unsuccessfully to run GNUCash 2.0.4 on Novell SLED 10. Can anyone point me to some up-to-date RPMs for SuSE, or perhaps suggest a way to make SLIB 3a4 play nicely with Guile 1.6.7 and G-Wrap 1.9.7? I've been able to cludge it together, but the best it can do is fail at runtime.
23:15:17 <prock> ooh looks like there is a required field for xml parsing... maybe I'm in luck for forward/backwards compatibility (?).
23:23:51 <prock> richard: I had to downgrade to an earlier slib
23:24:25 <richard> prock: which version? i read something about 3a3 being "known good" but still had runtime errors
23:24:41 <prock> richard: but it's interesting that you're having the problem too... could you pastbin the error and I'll update the bug to (hopefully) raise it's priority?
23:28:55 *** AlonzoTG has joined #gnucash
23:29:01 <AlonzoTG> =\
23:29:12 <richard> ok, it'll take me a little while to rebuild everything again. i *think* the source of the problem is actually documented in the slib code...
23:29:20 <AlonzoTG> The portage stream downgraded my slib stuff but gnucash is STILL broken.
23:29:33 *** Demitar_ has joined #gnucash
23:29:33 <AlonzoTG> atg@leenooks ~ $ gnucash
23:29:34 <AlonzoTG> ERROR: In procedure open-file:
23:29:36 <AlonzoTG> ERROR: No such file or directory: "/usr/lib/slib/require"
23:29:37 <AlonzoTG> atg@leenooks ~ $
23:29:59 <richard> AlonzoTG: environment variables :-(
23:31:26 <AlonzoTG> atg@leenooks ~ $ set | grep slib
23:31:27 <AlonzoTG> atg@leenooks ~ $ set | grep require
23:31:39 <AlonzoTG> -
23:31:42 <AlonzoTG> ziltch...
23:32:46 <AlonzoTG> =\
23:32:51 <AlonzoTG> how do I fix this?
23:33:03 <AlonzoTG> I created a bogus file,
23:33:10 <AlonzoTG> atg@leenooks ~ $ gnucash
23:33:12 <AlonzoTG> ERROR: Unbound variable: require
23:33:13 <AlonzoTG> atg@leenooks ~ $
23:33:22 <AlonzoTG> !!!!!!!!
23:33:36 <richard> AlonzoTG, I've only been trying to use GNUCash for 2 days ;-)
23:34:05 <AlonzoTG> I used it up to two weeks ago when the fucking update stream from gentoo broke it.
23:34:21 <richard> I think the env variable it wants is something like SCHEME_LIBRARY_PATH=/usr/share/guile/1.6/slib/
23:34:46 <richard> that sorted that issue out for me, but i don't know if that's related to what Gentoo might have changed
23:35:15 <blahdeblah> Whenever i try to make an invoice through reports, it opens the invoice window with the message "No valid invoice selected".
23:35:24 <blahdeblah> Where do i have to select it?
23:35:36 <blahdeblah> I thought opening the invoice itself would do enough to "select" it...
23:35:40 * richard hasn't got that far yet
23:36:07 <AlonzoTG> atg@leenooks ~ $ export SCHEME_LIBRARY_PATH=/usr/share/guile/slib/
23:36:09 <AlonzoTG> atg@leenooks ~ $ gnucash
23:36:10 <AlonzoTG> ERROR: In procedure open-file:
23:36:12 <AlonzoTG> ERROR: No such file or directory: "/usr/lib/slib/printf"
23:36:13 <AlonzoTG> atg@leenooks ~ $
23:36:24 <AlonzoTG> look, I've got 2 wks of recipts here.
23:36:27 <AlonzoTG> I need to do my accounting.
23:36:31 <AlonzoTG> !!!!!
23:53:02 <blahdeblah> warlord: ping - any more suggestions on the reason that invoice doesn't display as expected?
23:56:21 <warlord> blahdeblah: Run the options and select the invoice you want to print.
23:56:56 <blahdeblah> Yep - i found that eventually.
23:57:14 <blahdeblah> I mean the question above about why it doesn't show as updated after i edit the .scm file
23:57:28 <blahdeblah> ... and restarted gnucash
23:57:35 <warlord> What file did you edit?
23:57:42 <warlord> (full path)
23:58:08 <blahdeblah> adam:/usr/share/gnucash/guile-modules/gnucash/report/*invoice.scm
23:58:14 <blahdeblah> whoops
23:58:17 <blahdeblah> adam is my hostname
00:00:02 <warlord> I found encfs a year or two ago when CryptFS stopped working for me.
00:00:14 <warlord> heh
00:01:51 <fen> why is gnucash win32 installer 30mb? does it include GTK libs?
00:04:01 <warlord> Yes
00:06:53 <warlord> it includes everything you need.. every library..
00:09:09 * benoitg is looking for a general solution to OFX encoding problems
00:09:26 <warlord> benoitg: a general solution? Hmm.
00:09:41 <jsled> benoitg: what's the problem?
00:10:32 <warlord> doesn't OFX declare an encoding?
00:10:51 <benoitg> Basically, libofx will return the ofx file in whatever encoding it's originally in. In practice OpenSP provides absolutely no usable encoding conversion functions.
00:11:14 <warlord> benoitg: is there some encoding declaration in the OFX protocol?
00:11:25 <benoitg> Yes
00:11:53 <benoitg> Before we switched to UTF-8 everywhere, it happened to just work most of the time, and the bug we had masked the fact that we have a general problem.
00:12:13 <benoitg> So I'll probably have to link libofx with iconv and parse the headers myself.
00:12:26 <benoitg> Rewrite the file, and then parse it.
00:13:00 <warlord> well, if libofx can at least tell gnucash what encoding the file is, gnucash can do the conversion.
00:13:24 <warlord> unless you're having parse errors due to encoding issues?
00:13:57 <benoitg> No parse error, but ofxdump output is garbled in a UTF-8 locale, which is annoying.
00:14:18 <warlord> Oh, there is that..
00:14:35 <benoitg> Besides, it's really libofx's job to provide at least a known encoding.
00:14:38 <warlord> I'm not sure whether libofx or the application should be in charge of encoding conversion.
00:14:56 <warlord> true, but a libofx_get_encoding() API would suffice.
00:15:23 <warlord> not that I'm complaining about libofx deciding to do the conversion on its own.
00:16:06 <benoitg> The other problem is the size of the buffers.
00:16:14 <warlord> oh?
00:16:23 <warlord> you're not using std::string? ;)
00:16:49 <benoitg> Internally, yes, externally, no.
00:17:23 <warlord> So what's the issue with buffer size?
00:19:24 <benoitg> The API is C-style, and specifies the max size of each element. I can't find in the spec for 1.6 if the size limit applies to the number of characters, or the number of bytes.
00:20:04 <warlord> Oh. It's not just a NULL-terminated const string owned by libofx?
00:20:20 <warlord> (or even a NULL-terminated non-const string owned by the application)?
00:20:21 <benoitg> Yes. There may not even be UTF-8 files in the wild, so it's possible nobody knows for sure.
00:21:36 <warlord> okay, so the buffer sizes aren't in the API.
00:22:13 <benoitg> They are.
00:22:26 <warlord> Oh? It's not dynamic?
00:22:35 <benoitg> No
00:22:54 <warlord> Oh. Boo!
00:24:50 <benoitg> Oh well, at worst the lib will strib a few chars at the end of utterly long values.
00:25:36 *** fen has quit IRC
00:25:51 <benoitg> ?/?"/$%"?/?$&%&&
00:25:57 <benoitg> The spec does say
00:26:18 <warlord> Huh?
00:26:46 <benoitg> Size is the number of charecters, meaning I got it completely wrong 5 years ago.
00:27:56 <warlord> Oh. OOPS!
00:28:08 <warlord> so #chars, not #bytes?
00:30:15 <benoitg> Yep. Back then I didn't know how to write an API in plain C API that works with dynamic string lengths anyway, but now it's going to have to wait for an API redesign.
00:32:26 <warlord> Oh, the way I would've done it is just have the API be something like: foo_t foo(..., char** pString, ...) Then libofx could allocate the string.
00:32:44 <warlord> Or you could make it a const; part of the parse-context.
00:32:52 * warlord shrugs
00:32:58 <warlord> next API rev.
00:36:36 <benoitg> warlord: Back then I was fresh out of school, the initial API didn't even have a context pointer, so I didn't know how to deal with memory leaks other than by making the entire thing copyable as as struct ;)
00:38:24 <warlord> Hehehe...
00:38:28 <warlord> Live and learn.
01:57:31 <warlord> Okay, good night all
01:57:33 *** warlord is now known as warlord-afk
03:18:13 *** RallyU has joined #gnucash
03:40:53 *** foo has quit IRC
03:41:35 <benoitg> Ok, libofx will now "do the right thing" as far as gnucash is concerned.
04:20:26 *** ErKa has joined #gnucash
05:07:40 *** cstim has joined #gnucash
05:07:41 *** gncbot sets mode: +o cstim
05:07:57 * cstim adds http://lists.gnucash.org/logs/2007-01-23.html#T23:08:12 to FAQ...
05:09:09 <cstim> hampton: http://www.ironicsans.com/2007/01/the_astoria_notes.html some funny story about NYC appartment life.
05:47:04 *** bonez39 has quit IRC
06:35:24 *** twunder has joined #gnucash
06:58:16 *** twunder has quit IRC
06:58:49 <cstim> bugzilla crossed the line to numbers >= 400,000. Except that apparently nobody created a funny bug#400000, like so: http://bugzilla.gnome.org/show_bug.cgi?id=100000
06:59:13 <CSWookie> cstim: Sounds like a bug, to me.
06:59:49 <cstim> sure :-)
07:05:48 <cstim> can you reproduce it?
07:13:18 *** conrad has joined #gnucash
07:13:18 *** gncbot sets mode: +o conrad
07:23:11 *** terence has joined #gnucash
07:26:39 *** ErKa has quit IRC
07:30:08 *** conrad has quit IRC
07:50:58 *** andi5 has joined #gnucash
07:50:59 *** gncbot sets mode: +o andi5
07:57:37 <cstim> hi andi5. Quite early for you to show up in irc?
07:58:44 <andi5> hi... yes, but i will not really be here, it is just easier than reading the complete logs in one go
07:59:00 <cstim> :-)
07:59:16 <cstim> andi5 -> andi5_not_really_here
07:59:27 *** andi5 is now known as andi5|nrh
08:00:58 <cstim> heh
08:21:43 *** ErKa has joined #gnucash
08:25:42 <terence> just wonder if anyone has used GNUCash to manage inventory?
08:25:54 <terence> I don't think it has that feature.
08:27:19 <cstim> terence: no, it doesn't have that feature.
08:28:36 <terence> thanks cstim.
08:33:33 *** twunder has joined #gnucash
08:39:13 *** conrad has joined #gnucash
08:39:14 *** gncbot sets mode: +o conrad
08:43:47 *** terence has quit IRC
09:02:59 *** ceplma has quit IRC
09:03:42 *** ceplma has joined #gnucash
09:15:53 *** twunder has quit IRC
09:42:18 *** ErKa has quit IRC
09:48:24 *** warlord-afk is now known as warlord
10:04:34 *** prock has quit IRC
10:19:42 *** kielein has joined #gnucash
10:22:36 *** RallyU has quit IRC
10:35:00 *** Demitar_ has quit IRC
10:35:05 *** twunder has joined #gnucash
10:35:43 *** andi5|nrh has quit IRC
10:45:42 *** ErKa has joined #gnucash
10:50:07 <cstim> can someone reproduce http://bugzilla.gnome.org/show_bug.cgi?id=394074 ? Instructions seem to be sufficient to do this.
10:55:07 <cstim> warlord: I think the email from Peter McAlpine is exactly the same issue as what we've discussed here yesterday. I think it's also the same reporter that brought it up here.
10:56:14 <warlord> I think so, too, but it still belongs on -devel and not on -user ;)
10:56:36 *** twunder has quit IRC
10:57:56 <cstim> yes, absolutely
11:00:35 <warlord> I cannot reproduce 394074.
11:01:02 <cstim> you mean, it doesn't crash for you? That's bad. I mean, good.
11:01:03 <cstim> :-)
11:02:09 <hampton> cstim: Funny web site. Thanks.
11:07:35 <cstim> does anyone know how to get gcc-4.0.x to display only the first ~100 lines of error messages and not spill out 10,000 lines so that I cannot see the initial error :-/
11:08:19 <cstim> or how can I pipe stderr in bash?
11:08:25 <cstim> (instead of stdout)
11:09:08 <warlord> cstim: instead of stdout or in addition to? You can redirect stderr to a file! 2>/tmp/errors.out
11:09:25 <cstim> yes, I could do that.
11:32:38 *** wizkid238 has quit IRC
11:35:32 <hampton> you can also redirect it to stdout: gcc 2>&1 | head -1000
11:35:32 *** CSWookie has quit IRC
11:39:59 *** ErKa has quit IRC
11:52:15 *** wizkid238 has joined #gnucash
12:02:34 *** cstim has quit IRC
12:13:49 *** andi5 has joined #gnucash
12:13:49 *** gncbot sets mode: +o andi5
12:17:24 *** bonez39 has joined #gnucash
12:17:44 *** twunder has joined #gnucash
12:21:42 *** wizkid238 has quit IRC
12:30:29 *** wizkid238 has joined #gnucash
12:46:40 *** imork has joined #gnucash
12:48:15 <imork> Hi, this might be a Stupid Question, but are all these .xac and .log files important?
12:48:34 *** MrN has joined #gnucash
12:49:02 <MrN> hi
12:49:46 <andi5> imork: http://wiki.gnucash.org/wiki/FAQ
12:58:17 <andi5> btw, i can reproduce #394074 pretty easily (have not tried with trunk yet, because i updated slib again ;-))
13:04:15 <warlord> andi5: maybe it's a debian/ubuntu-specific thing?
13:05:10 <andi5> i do not really think so... you said that you can still see the dialog, right? use it, especially check treeviews
13:05:51 <andi5> sometimes it suffices just to move the mouse... well, i will try to fix it
13:06:20 *** prock has joined #gnucash
13:06:49 <warlord> "check treeviews"?
13:07:08 <warlord> Yeah, I can still see the dialog.. Really, the dialog should get closed on book-close.
13:08:38 <andi5> well, i just open the balance sheet, open another file and play a bit with the dialog... it used to crash then... hmmm
13:08:48 <prock> I'm sorry for posting my email re bug 394074 to -user instead of -devel... I guess I wasn't thinking too clearly last night
13:09:10 <prock> prock === Peter McAlpine
13:09:23 *** bonez39 has quit IRC
13:10:48 <andi5> warlord: http://pastebin.ca/raw/327058 is an example backtrace... i do not have it opened in gdb anymore, though
13:11:00 <prock> oh it wasn't 394074
13:11:48 <andi5> prock: those 6 numbers are not random and all combinations point to the same bug ;-)
13:12:03 <prock> 373809
13:12:18 <prock> my bug is the only bug that matters, right?
13:13:23 <andi5> prock: i am working on it ... but i also face a rhythmbox bug and i cannot hear NO MUSIC....
13:14:00 <prock> cannot hear no music means music never stops?
13:14:12 <prock> kill -9 now work on my bug ;)
13:14:14 <prock> j/k
13:14:37 <andi5> well, andi5 is a child process of rhythmbox...
13:16:45 <andi5> prock: sorry, i got distracted by the fact that gnome-terminal is damn slow (again?) and then i tried all sorts of emacs shells just to find out that what i used is best aligned with my habits... *back*
13:22:28 <jsled> no music!@??!
13:23:08 *** jpeach has joined #gnucash
13:23:39 <andi5> weird.. it works again.. *clicks into tray list and never touches again*
13:32:52 <jpeach> Hi
13:32:52 <gncbot> jpeach: Sent 5 days, 21 hours, and 9 minutes ago: <jsled> (re: private message) Yes, I do! And yes, it is!
13:33:48 <jpeach> Where can I get version 2.0.4 as a deb package. The one on http://www.gnucash.org/pub/gnucash/ is version 1.X
13:34:17 <warlord> andi5: sorry, I stepped AFK for a while. Im back now.I only tried the txn report,and I didn't try editing the accounts page.. So it's probably limited to certain reports and/or certain options.
13:35:19 <warlord> jpeach: from Debian. We don't distribute packages.
13:36:08 <jpeach> warlord: Oh... ok, that is the way that I installed it first. However, the latest one in the standard repositories is 2.0.2.
13:36:13 <warlord> andi5: the real fix is to make the options window(s) go away when the book is closed.
13:36:23 <andi5> warlord: i know
13:36:33 <jpeach> warlord: Is there a place where I can download the binary for 2.0.4?
13:36:34 <warlord> jpeach: then send mail to the debian gnucash maintainer.
13:36:44 <andi5> wtf.... can a guile object be a list and vector at the same time?
13:37:04 <warlord> jpeach: No clue. As I already said, we don't distribute binaries. (Well, except for Windows)
13:37:40 <jpeach> warlord: Where would I find the email address for debain gnucash maintainer?
13:38:41 <andi5> jpeach: http://packages.qa.debian.org/g/gnucash.html
13:39:15 <jpeach> andi5: thanks!
13:39:38 <andi5> oh, i mean list and array at the same time..
13:43:44 <warlord> andi5: what do you mean, "list and array"?
13:44:12 <andi5> well, i have a "seq" thing and (list? seq) and (array? seq) both return #t
13:45:04 <warlord> I'm not sure I understand the difference between a list and an array in guile.
13:46:27 <andi5> ok, arrays come from slib/array.scm.... i will just try to find out, the truth must be out there...
13:47:28 <warlord> :)
14:21:59 *** jpeach has left #gnucash
14:31:57 <andi5> boah, looks like a horrible nesting of bugs... wtf...
14:37:25 <warlord> LOL
14:54:47 <andi5> ok... here is the basic problem... prior to slib 3a4 there was only *features* ... guile initiated it, slib/guile.init added what it thought guile is able to do itself (e.g. array, but neither sort nor record) ... (require 'record) will test *features* -> use guile record support
14:54:47 <andi5> now we have *features* and slb:features... the primer is untouched, slib:features is supposed to contain what the implementation can do natively, but it still does NOT contain sort or record.... so (require 'record) will now load slib's record.... making boom if we mix both
14:56:58 <andi5> i guess we could just always (require 'record) or always not require it... but there is another problem with array/sort i am still trying to "sort" out ;-)
14:58:52 <warlord> Hmm...
15:09:51 *** jpeach has joined #gnucash
15:09:54 *** jpeach has left #gnucash
15:12:35 *** |gunni| has joined #gnucash
15:13:10 <andi5> warlord: can you test this one for me? guile -c '(let ((a (list 5 6))) (if (array? a) (display "Is an array\n")) (array-shape a))'
15:19:03 <warlord> [warlord@cliodev ~]$ guile -c '(let ((a (list 5 6))) (if (array? a) (display "Is an array\n")) (array-shape a))'
15:19:03 <warlord> Is an array
15:19:03 <warlord> ERROR: In procedure map:
15:19:03 <warlord> ERROR: Wrong type argument in position 2: #f
15:19:17 <andi5> does that look promising to you?
15:19:41 <warlord> promising how?
15:19:52 <andi5> well, s,promising,goo,
15:19:54 <andi5> +d
15:21:45 <warlord> I guess...
15:22:01 <warlord> It would be nice if we didn't need slib at all.
15:22:57 <andi5> maybe
15:29:09 <prock> "+d" means?
15:29:26 <andi5> goo + d = good
15:29:42 <prock> lol
15:29:57 <andi5> it is a simply chemical reaction
15:30:00 <warlord> You mean it's not a barely-passing goo?
15:30:15 <andi5> no, it is not
15:30:57 <warlord> a goo that received a grade of d+?
15:31:02 <andi5> [ot] seems like gnome gets a new control center
15:49:19 <jsled> andi5: yeah, it looks not too bad.
15:58:00 *** ErKa has joined #gnucash
16:07:12 *** andi5 has quit IRC
16:50:38 *** bonez39 has joined #gnucash
17:01:48 *** twunder has quit IRC
17:17:53 <prock> -c '(let ((a (list 5 6))) (if
17:17:54 *** |gunni| has quit IRC
17:18:08 <warlord> prock: huh?
17:18:21 *** ErKa has quit IRC
17:18:58 <prock> pasted wrong window
17:21:23 *** |gunni| has joined #gnucash
17:21:54 *** |gunni| has joined #gnucash
17:24:18 *** prock has quit IRC
17:29:25 *** prock has joined #gnucash
17:34:18 <prock> le gasp... If I'm to write a new report it must be done with scheme
17:35:02 <prock> all those parenthesis make my brain hurt
17:35:54 <warlord> prock: Eh, emacs makes it easy.
17:36:04 <prock> warlord: that's a loaded statement
17:36:21 * prock starts an editor war
17:36:38 *** blahdeblah has joined #gnucash
17:37:31 <blahdeblah> Does gnucash have a standard field to indicate the party a transaction was enacted with? e.g. the payee of an expense, or the source of income?
17:38:07 <blahdeblah> (I'm using GnuCash 2.0.2 r14936 on Debian etch.)
17:38:45 <warlord> blahdeblah: "Description"?
17:39:53 <blahdeblah> Wouldn't that be used as a more general description of what the transaction was about?
17:40:20 <warlord> Maybe.
17:40:38 <warlord> You've got three things. The TXN Desc, TXN Notes, and Split memos.
17:40:51 <warlord> Use them however you wish ;) But generally DESC == Payee.
17:41:11 <warlord> Notes is more of a general txn memo/description
17:41:22 <warlord> Memo is a per-split, "what does THIS split mean" entry.
17:41:26 <warlord> anyways, gotta run.
17:41:29 *** warlord is now known as warlord-afk
17:41:34 <blahdeblah> I was hoping for something that could be reported on easily.
17:41:48 <blahdeblah> e.g. Show me all my expenses at this place, or all my income from that source.
17:42:02 <prock> blahdeblah: accounts are what you want
17:42:19 <prock> blahdeblah: create an account for each expense or income source
17:42:20 <blahdeblah> prock: So i should have a separate account for each payee and client?
17:42:30 <blahdeblah> OK
17:42:42 <prock> blahdeblah: yes if you want to track on that level of detail
17:42:50 <blahdeblah> Seems a little tedious, especially since they can't be created in the middle of a transaction.
17:43:07 <prock> blahdeblah: they can't?
17:43:37 <blahdeblah> You must know something i don't. :-)
17:46:20 <prock> blahdeblah: just type the name of the account you want to transfer to and if it doesn't exist it will ask if you want to create it.
17:46:37 <blahdeblah> OK - i'll give that a shot.
17:47:43 <blahdeblah> Where would be the normal place to put clients of a business (as an income source)? I don't understand the "accounting equation" well enough to work out whether it should be income or assets.
17:49:54 <prock> blahdeblah: my experience is limited to personal use but I have top level accounts "Income" and "Expenses". Sub-accounts under "Income" include "Salary", "Rental Income", "Gifts". Under expenses: "Electric", "Gas", "Food" (which also has subaccounts)
17:50:25 <blahdeblah> Hmmm... Looks like i need to do a lot more reading in the manual. :-|
17:50:30 <blahdeblah> Thanks for your help
17:51:19 <prock> what version of glade is used for trunk/src/gnome/glade?
17:51:37 <prock> s/used/recommended/g
17:52:54 *** benoitg has left #gnucash
17:56:04 <jsled> prock: I've got 2.6.8 installed. I'm not sure that it matters, though I know hampton has used a script to normalize the .glade file content in the past; I'm not sure of its current status.
17:56:14 *** blahdeblah has left #gnucash
18:02:19 *** kielein has quit IRC
18:04:49 *** blahdeblah has joined #gnucash
18:06:53 <blahdeblah> I'm experiencing random lockups in GnuCash on my Debian etch 64-bit system. Where's the best place to report it, and what should i include?
18:09:47 <prock> blahdeblah: reportbug with debian... but first check to make sure one doesn't already exist for your problem (bugs.debian.org)
18:10:12 <blahdeblah> Will do
18:23:01 <prock> are the .glades actually edited with the glade GUI or do people hack them by hand?
18:25:01 <hampton> I've been known to hack them by hand on occasion, but generally the glade gui is what's used
18:25:57 <hampton> util/glade-fixup is the tool used to strip out attributes inserted by the glade gui that aren't supported on the targeted level of gtk.
18:26:12 <prock> ah thanks for the tip
18:26:30 <hampton> It needs to be updated since the base level has moved from gtk2.4 to gtk2.6
18:31:35 <hampton> Hmpf. Looks like all of the stripped attributes but one are supported in gtk2.6. I'll have to edit some glade files and see if there are new to 2.10 attributes that need stripping.
18:36:36 *** |gunni| has quit IRC
18:43:38 * hampton goes to install FC4 in vmware so he has a gtk2.6 based system...
18:57:52 <jsled> I think I've maybe changed a vbox identifier by hand. Or changing a dialog default size. I can't imagine doing anything more complicated.
19:03:09 <prock> jsled: if I were to add the ability to disable/enable an sx would you have an opinion where the option for that goes? my version adds a checkbox as the first row of "Options" in the "Edit Scheduled Transaction" dialog
19:04:11 <jsled> That sounds like a good place for it.
19:05:00 <jsled> I mean, the SX editor is a bit of a mess, so it doesn't make it much worse.
19:05:03 <jsled> s/so/but/
19:05:38 <jsled> But ... why not just set the SX to "none".
19:05:38 <jsled> ?
19:11:04 <prock> jsled: because when I write my account projection report I want to be able to select sx's that aren't active 'for real'
19:16:39 <jsled> Ah.
19:18:56 <jsled> if it's part of the report, it might be better to do it at that point.
19:19:10 <jsled> Similar to selecting the to-be-reported-over accounts in the options.
19:20:15 <prock> but then I'd have to create all my sx's every time I want to run the report
19:21:21 <jsled> hmm?
19:21:35 <jsled> No, it'd be a multi-select list, probably defaulting to "all" sxes.
19:21:45 <jsled> You'd just unselect the ones that shouldn't be a function of the report.
19:21:58 <jsled> s/a function of/in/
19:22:03 <prock> right... the issue isn't keeping things off the report it's keeping things off the SLR
19:22:23 <jsled> Ah, right.
19:22:38 <prock> like I want a report to see if I can afford a new $5000 camera but I don't want it scheduled for really real yet
19:22:51 <jsled> Not to disrupt your momentum, but you might want to do a quick writeup of your design to -devel.
19:23:22 <prock> don't worry there's very little momentum to disrupt =)
19:23:37 <prock> and yes I'll put something together with UI mockups, etc.
19:24:49 <jsled> cool. Unless you're going to do them anyways, we can probably get by w/o mockups unless it's not clear what it'll be. It sounds like it should be.
19:26:00 <prock> my hope is that it will be an in-obtrusive way to meet my personal (and hopefully other's) budgeting needs... I really don't want to start (another) huge thread about the subject.
19:32:57 <prock> I'd like to have some sort of working(-ish) enable/disable patch before I announce my intentions though.
19:35:15 *** jharold has joined #gnucash
19:41:32 <prock> jsled: can I ask what your thought on it are? what do you think the response to my suggestion will be?
19:42:56 *** bonez39 has quit IRC
19:43:14 <jsled> Well, even better would be to populate a budget from a set of SXes.
19:43:48 <prock> yes... although to be honest I don't like the budgets in gnucash
19:44:17 *** MrN has quit IRC
19:44:30 <jsled> But that's a bigger code cahnge. I'm pretty fine with the idea ... the SX "enabled" flag ... a report to actually do the work.
19:44:45 <jsled> The flag will become a new data-format item, so we might have to plan for that a bit.
19:46:25 <prock> I thought about that... everything defaults to enabled so loading old .xml should be fine and I'm assuming old versions reading new .xml files will just ignore the tag
19:47:00 <prock> mind you I haven't looked at the file writing code at all
19:50:39 <jsled> no, they will not ignore the tag.
19:51:02 <jsled> I believe we croak if we see unexpected tags.
19:51:20 <prock> :(
19:52:13 <jsled> indeed.
20:12:51 *** rauch has quit IRC
20:16:46 *** imork has quit IRC
20:50:00 *** jharold has quit IRC
20:57:41 *** jharold has joined #gnucash
21:12:24 <prock> does --loglevel affect stdout output or just output to the .log file?
21:14:38 *** twunder has joined #gnucash
21:14:47 *** warlord-afk is now known as warlord
21:14:57 <warlord> prock: neither. it affects output to /tmp/gnucash.trace
21:15:35 <prock> warlord: even better thanks
21:39:55 <blahdeblah> Is there a way to customise the headings and the like for invoices created in gnucash?
21:41:43 <warlord> nope.
21:42:01 <warlord> at least not without changing the scheme code or changing the translation.
21:42:12 <warlord> (I assume you mean the column headings)
21:44:32 <blahdeblah> No, the part where it says: Invoice #333333
21:44:57 <blahdeblah> In .au, we are legally required to have the words "Tax Invoice" prominent on all invoices.
21:46:16 <warlord> *sigh* So I've heard. What a stupid law. No, you can't change that without modifying the scheme.
21:46:35 <blahdeblah> Where is the scheme stored?
21:46:49 <blahdeblah> I'm happy with a quick hack if it gets me running. ;-)
21:46:56 <warlord> (or you could create a po/en_AU.po that changes "Invoice" to "Tax Invoice")
21:47:03 <warlord> invoice.scm
21:47:07 <blahdeblah> And stupid laws are standard the world over...
21:47:09 <blahdeblah> Thanks
21:47:15 <warlord> Not THAT stupid!
21:47:36 <warlord> "An invoice has to say Tax Invoice"? Like, are there different kinds of invoices?!?
21:47:44 <blahdeblah> One could argue that customisable headers and footers are an extremely important feature...
21:48:20 <warlord> Send in a patch.
21:48:44 <blahdeblah> Will investigate doing so - what's the best address to mail them to?
21:48:52 <warlord> Your other option is to use Edit -> Stylesheets and create a GnuCash stylesheet with your header.
21:48:57 <warlord> gnucash-devel
21:49:03 <blahdeblah> @?
21:49:03 <gncbot> blahdeblah: Error: "?" is not a valid command.
21:49:09 <hampton> or bugzilla.gnome.org
21:49:31 <blahdeblah> warlord: gnucash-devel@?
21:49:36 <blahdeblah> gnucash.org?
21:49:38 <blahdeblah> gnome.org?
21:49:52 <hampton> gnucash.org
21:49:58 <blahdeblah> OK
21:50:07 <blahdeblah> BTW, here in the great land of Oz, there are actually non-tax invoices as well as tax invoices. ;-)
21:51:10 <blahdeblah> On the stylesheet issue, GnuCash 2.0.2 doesn't seem to have any options for editing the fixed text. Is that stored in a file, too?
21:52:14 <warlord> the "fixed text" is in invoice.scm
21:53:06 <blahdeblah> And to create my own stylesheet for invoices i need to add blah-invoice.scm?
21:54:20 <blahdeblah> Or is invoice.scm the same as a stylesheet of default?
21:54:37 <warlord> invoice.scm is the report.
21:54:43 <warlord> a stylesheet is completely different.
21:55:01 <warlord> I was thinking that you could create a stylesheet that adds the text "Tax Invoice"
21:55:04 <warlord> and then it would say:
21:55:07 <warlord> Tax Invoice
21:55:10 <warlord> Invoice #XXXXXX
21:55:27 <blahdeblah> Yep
21:55:57 <warlord> Edit -> Stylesheets
21:56:02 <warlord> to create a stylesheet.
21:56:08 <warlord> then just apply it to your invoice rpeort.
21:57:19 <blahdeblah> I must be thick - where is the option to add text?
21:59:27 <warlord> Might have to add it as an image.
21:59:39 <warlord> I might be easier to just edit invoice.scm
22:00:04 <blahdeblah> OK - i'll have a look there.
22:01:20 <warlord> blahdeblah: http://wiki.gnucash.org/wiki/FAQ#Q:_How_do_I_change_.22Invoice.22_to_.22Tax_Invoice.22_as_required_in_Australia.3F
22:01:40 <blahdeblah> ah thanks
22:01:45 * warlord should've just pointed you there originally.
22:04:49 <blahdeblah> Yep - it has some good info.
22:05:19 <blahdeblah> One thing that confuses me there is that now there seems to be a 5-account naming scheme, not 3 or 4.
22:05:44 <blahdeblah> (I'm very much a newbie when it comes to accounting stuff & running my own business.)
22:16:08 *** hampto1 has joined #gnucash
22:16:09 *** gncbot sets mode: +o hampto1
22:18:42 *** hampton has quit IRC
22:18:55 *** hampto1 is now known as hampton
22:28:18 <blahdeblah> Hmmm... Is there anything special i need to do to make changes to invoice.scm activate?
22:30:23 <warlord> you need to restsrt gnucash.
22:30:59 <blahdeblah> Done that - no joy
22:32:16 <warlord> Where did you edit the report? Are you sure your edited report is installed in the proper location where GnuCash is looking for it?
22:34:12 *** conrad has quit IRC
22:36:58 <blahdeblah> I'm using the Debian etch package, which installs the invoice files in /usr/share/gnucash/guile-modules/gnucash/report/ AFAICT
22:41:26 <prock> grarg... so I need the glade-gnome package... of course. hours_left_to_live -= 3;
22:48:50 *** richard has joined #gnucash
23:09:05 *** twunder has quit IRC
23:10:38 <richard> Hello. I've tried unsuccessfully to run GNUCash 2.0.4 on Novell SLED 10. Can anyone point me to some up-to-date RPMs for SuSE, or perhaps suggest a way to make SLIB 3a4 play nicely with Guile 1.6.7 and G-Wrap 1.9.7? I've been able to cludge it together, but the best it can do is fail at runtime.
23:15:17 <prock> ooh looks like there is a required field for xml parsing... maybe I'm in luck for forward/backwards compatibility (?).
23:23:51 <prock> richard: I had to downgrade to an earlier slib
23:24:25 <richard> prock: which version? i read something about 3a3 being "known good" but still had runtime errors
23:24:41 <prock> richard: but it's interesting that you're having the problem too... could you pastbin the error and I'll update the bug to (hopefully) raise it's priority?
23:28:55 *** AlonzoTG has joined #gnucash
23:29:01 <AlonzoTG> =\
23:29:12 <richard> ok, it'll take me a little while to rebuild everything again. i *think* the source of the problem is actually documented in the slib code...
23:29:20 <AlonzoTG> The portage stream downgraded my slib stuff but gnucash is STILL broken.
23:29:33 *** Demitar_ has joined #gnucash
23:29:33 <AlonzoTG> atg@leenooks ~ $ gnucash
23:29:34 <AlonzoTG> ERROR: In procedure open-file:
23:29:36 <AlonzoTG> ERROR: No such file or directory: "/usr/lib/slib/require"
23:29:37 <AlonzoTG> atg@leenooks ~ $
23:29:59 <richard> AlonzoTG: environment variables :-(
23:31:26 <AlonzoTG> atg@leenooks ~ $ set | grep slib
23:31:27 <AlonzoTG> atg@leenooks ~ $ set | grep require
23:31:39 <AlonzoTG> -
23:31:42 <AlonzoTG> ziltch...
23:32:46 <AlonzoTG> =\
23:32:51 <AlonzoTG> how do I fix this?
23:33:03 <AlonzoTG> I created a bogus file,
23:33:10 <AlonzoTG> atg@leenooks ~ $ gnucash
23:33:12 <AlonzoTG> ERROR: Unbound variable: require
23:33:13 <AlonzoTG> atg@leenooks ~ $
23:33:22 <AlonzoTG> !!!!!!!!
23:33:36 <richard> AlonzoTG, I've only been trying to use GNUCash for 2 days ;-)
23:34:05 <AlonzoTG> I used it up to two weeks ago when the fucking update stream from gentoo broke it.
23:34:21 <richard> I think the env variable it wants is something like SCHEME_LIBRARY_PATH=/usr/share/guile/1.6/slib/
23:34:46 <richard> that sorted that issue out for me, but i don't know if that's related to what Gentoo might have changed
23:35:15 <blahdeblah> Whenever i try to make an invoice through reports, it opens the invoice window with the message "No valid invoice selected".
23:35:24 <blahdeblah> Where do i have to select it?
23:35:36 <blahdeblah> I thought opening the invoice itself would do enough to "select" it...
23:35:40 * richard hasn't got that far yet
23:36:07 <AlonzoTG> atg@leenooks ~ $ export SCHEME_LIBRARY_PATH=/usr/share/guile/slib/
23:36:09 <AlonzoTG> atg@leenooks ~ $ gnucash
23:36:10 <AlonzoTG> ERROR: In procedure open-file:
23:36:12 <AlonzoTG> ERROR: No such file or directory: "/usr/lib/slib/printf"
23:36:13 <AlonzoTG> atg@leenooks ~ $
23:36:24 <AlonzoTG> look, I've got 2 wks of recipts here.
23:36:27 <AlonzoTG> I need to do my accounting.
23:36:31 <AlonzoTG> !!!!!
23:53:02 <blahdeblah> warlord: ping - any more suggestions on the reason that invoice doesn't display as expected?
23:56:21 <warlord> blahdeblah: Run the options and select the invoice you want to print.
23:56:56 <blahdeblah> Yep - i found that eventually.
23:57:14 <blahdeblah> I mean the question above about why it doesn't show as updated after i edit the .scm file
23:57:28 <blahdeblah> ... and restarted gnucash
23:57:35 <warlord> What file did you edit?
23:57:42 <warlord> (full path)
23:58:08 <blahdeblah> adam:/usr/share/gnucash/guile-modules/gnucash/report/*invoice.scm
23:58:14 <blahdeblah> whoops
23:58:17 <blahdeblah> adam is my hostname