2007-02-10 GnuCash IRC logs
00:03:45 *** dbr has quit IRC
00:06:51 *** jpeach has joined #gnucash
00:55:27 *** bistra__ has joined #gnucash
00:57:09 <bistra__> Fedora Core 4, Running GNUCASH 1.8, program exits every time I try to 'print' and invoice (new or previous). Printer was working and works ok with open office
00:58:46 <bistra__> Tried to upgrade to 2.0.4. Need libofx. Configure cannot find ParseEventGenerator.h and neither can I. Downlowded latest and can see EventGenerator is this the same???
00:59:24 <bistra__> Sorry in front of ParseEventGenerator it should say
01:00:14 <bistra__> I need libofx to configure gnucash2.0.4. Its configure needs ParseEventGenerator.h. It cannot find it
01:00:48 <bistra__> can find EventGenerator is this the same thing???
01:07:17 <jsled> probably not. why are you using FC4?
01:07:59 <bistra__> small old computer used for office work, though fc6 might be too large
01:08:53 <jsled> define "small"
01:09:38 <jsled> I can't really help you, I don't think; at this hour (or any hour, probably) your best bet it to post to gnucash-user.
01:09:47 <bistra__> Dell Dimension433c (I'check the specs now and advise)
01:12:19 <bistra__> ok jsled I will struggle on and advise. Note I had trouble using the /boot record in FC6
01:12:30 <bistra__> ok using Master Record, weird!!!!
01:12:59 <bistra__> jsled, tks.
01:13:01 <jsled> Well, the only advice I can give is that software is a fast-paced thing.
01:13:08 <jsled> So, one must stay current.
01:13:55 <jsled> We targeted the dependency profile of FC3/RHEL4 for 2.0, but ... bitrot, and all.
01:13:59 <bistra__> OK doing my best, trying to load 2.0.4 onto another machine running fc6-64. Let you know how I go.
01:18:35 <bistra__> Installing on FC6-64 ./configure stops on XML::Parser. perl module
01:18:51 <bistra__> Off I go and look for that! :)))
01:20:32 <bistra__> will be in touch again. Goodnight
01:20:36 *** bistra__ has quit IRC
03:08:38 *** rlaager has joined #gnucash
03:37:30 *** Esaj has joined #gnucash
05:41:13 *** andi5 has joined #gnucash
05:41:13 *** gncbot sets mode: +o andi5
05:49:33 *** Esaj has quit IRC
05:51:28 *** |gunni| has joined #gnucash
06:43:47 *** rauch has joined #gnucash
06:44:00 <rauch> hi all
06:45:06 <rauch> I sent a mail to gnucash-user 40 minutes ago and wonder, why it did not come back yet...
06:45:47 <rauch> Did actually anyone receive it yet?
06:46:12 <andi5> http://lists.gnucash.org/pipermail/gnucash-user/2007-February/thread.html does not list you
06:46:43 <rauch> hm, maybe stuck in a queue somewhere...
06:48:40 <rauch> well, I'll wait and see, how long it takes ;)
06:51:32 <rauch> It's about the one having problem with gnucash-docs (just like me) and I filed a bug against opensuse some time ago with their susehelper script
06:51:48 <rauch> but no answer yet from one of their devs :/
06:53:28 *** rauch is now known as rauch|afk
07:41:55 *** nomeata has joined #gnucash
07:53:25 *** |gunni| has quit IRC
07:57:03 *** nomeata has quit IRC
08:05:25 *** |gunni| has joined #gnucash
08:20:26 *** |gunni| has quit IRC
08:23:25 *** |gunni| has joined #gnucash
08:24:09 *** |gunni| has joined #gnucash
08:44:49 *** jpeach has left #gnucash
09:24:05 *** mnoir has joined #gnucash
09:34:55 *** mnoir has quit IRC
09:41:11 *** warlord-afk is now known as warlord
09:55:19 *** ErKa has joined #gnucash
10:27:50 *** dbr has joined #gnucash
10:33:03 *** wavydavy has joined #gnucash
10:33:45 <wavydavy> running gnucash (unprivilaged user): ERROR: no code for module (g-wrap gw-glib)
10:34:11 <jsled> wavydavy: what distro?
10:34:11 <wavydavy> g-wrap-1.9.7 installed
10:34:19 <wavydavy> Linux from scratch.
10:34:30 <jsled> Ah. What version of slib and guile?
10:34:55 <jsled> Oh, doesn't matter. g-wrap-1.9.7 doesn't work. Use 1.9.6
10:35:10 <wavydavy> oh. ok will install. Ta
10:35:24 * jsled digs up version-compat note...
10:36:43 <jsled> http://lists.gnucash.org/pipermail/gnucash-devel/2007-January/019660.html
10:37:34 <wavydavy> rebuild gnucash! :( that took ages. Oh well, here goes. Thanks
10:39:09 <jsled> Oh, also...
10:39:15 <jsled> (before you get much further)
10:39:30 <wavydavy> :) yes jsled?
10:39:37 <jsled> That error is probably because g-wrap was built w/o glib (1.0) support. Do you have glib (1.0)?
10:39:46 <wavydavy> nope, 2
10:39:57 <jsled> Right. So you need to patch g-wrap-1.9.6 to look for glib2 instead.
10:40:10 <jsled> Then you'll get the 'gw-glib' library that it can't find.
10:40:20 <wavydavy> I had read that. I couldn't figure out how to point 1.9.7 to look for glib2. Perhaps I can't?
10:40:35 <wavydavy> ok cool. I'll try that.
10:40:41 <jsled> You can, but it doesn't matter, because (indepdently) g-wrap-1.9.7 doesn't work with gnucash.
10:41:51 <jsled> For g-wrap-1.9.6 the change is easy.. there's a patch out there, but just edit the "right place" in configure.in
10:42:01 <jsled> Change "glib" to "glib-2.0". That's it.
10:42:23 <wavydavy> There is a configure.ac, but a configure.in?
10:42:34 <jsled> Ah. configure.ac, I meant. :)
10:43:22 <jsled> So, why LFS? Every other distro has already encoded the solutions to all these problems in their package management... :)
10:44:22 <wavydavy> I like pain apparently :) Learning experience.
10:54:29 <andi5> wavydavy: maybe we do not like it...
10:55:42 <wavydavy> andi5: who's we and what's it?
10:56:23 <andi5> we = the class of people you have to ask just because you like pain ; it = pain
10:56:52 <wavydavy> ah, ok. fair doos.
11:01:19 *** andi5 has left #gnucash
11:06:30 *** nomeata has joined #gnucash
11:20:47 *** hampton|away is now known as hampton
11:22:06 <prock> http://paste.debian.net/21704
11:22:23 <prock> what do you suppose I'm missing? I haven't built trunk on this system (OSX) yet.
11:23:28 <jsled> Ah. It looks like that 'int' should be a 'gint'. :/
11:23:37 <prock> configure command was : ./configure --enable-compile-warnings --prefix=`pwd`/local
11:24:02 <jsled> Er, a "gsize", even.
11:24:34 <jsled> Sorry about that ... I should run with '--enable-compile-warnings', I guess ... I didn't see that.
11:26:58 <jsled> Hmm, though I didn't get an error with the flag just now. :/
11:27:25 <hampton> 32-bit vs 64-bit?
11:27:56 *** warlord is now known as warlord-afk
11:27:58 <prock> 10.4 only runs in 32 I think
11:28:09 <jsled> I'm 32, here.
11:28:16 <jsled> prock: checking fix in now.
11:29:08 *** prock has quit IRC
11:29:16 <jsled> :( !
11:29:21 <hampton> That's not good.
12:04:41 *** ErKa has quit IRC
12:10:22 *** rauch|afk is now known as rauch
12:11:29 <rauch> hm, still no mail after 6 hours, I'll resend...
12:12:50 <jsled> rauch: what address? I can take a look at the server-side logs.
12:12:54 <jsled> Er, from what address?
12:13:09 <rauch> from info@rauch-webdesign.de to gnucash-user
12:15:22 <hampton> jsled: fwiw I was able to compile (i686 system) without any errors at all.
12:15:54 <hampton> No warnings even, except about relinking .la files
12:16:04 <jsled> great.
12:18:57 *** prock has joined #gnucash
12:19:46 <jsled> rauch: that one got through; it looks like there was a greylisting callback at 6am that ended in a "450 4.7.1 [...] denied for 600 seconds". Then the recent attempt re-checked and did a "greylisting update", and is apparently happier.
12:20:05 <jsled> prock: cpu cooler, now?
12:20:20 <prock> jsled: it's off, so yeah.
12:21:59 <rauch> jsled: hm, I know that greylisting, a friend of mine uses that too and there it takes ages to deliver a mail...
12:23:07 <rauch> I think, my providers mailserver is acting a bit weird upon that, maybe I should use another one I have more control over...
12:24:51 <rauch> it seems like he does not resend rejected mail for hours :(
12:33:16 <rauch> anyway, thx jsled for looking, cu all
12:33:28 <jsled> cheers
12:33:31 *** rauch has left #gnucash
12:43:31 <dbr> my trunk version is r15476. Am I remembering correctly that I'll lose backward datafile compatibility if I catch up on trunk?
12:45:16 <jsled> dbr: not yet, and only for SXes.
12:46:45 <dbr> jsled: ? does that mean that extra data field was pulled, or that SX's will be strange?
12:47:42 *** gander has joined #gnucash
12:49:10 <jsled> "extra data field"? I've not verified, but if you try to open the migrated file with 1.8 or 2.0, it should error out.
12:52:13 <jsled> Yeah. It'll say "there was an error parsing the file [...]".
12:52:52 *** mnoir has joined #gnucash
12:53:58 <dbr> will it still open what's left after the parsing error?
12:54:46 <dbr> I test on live data by saving every couple minutes/entries. going back to 2.0.4 if something fails in trunk
12:55:08 <jsled> Right. Don't test on your real data file.
12:55:57 <dbr> but then testing is more limited and time consuming.
12:56:11 <jsled> No, it will not continue after the parsing error. I guess it could say "hey, I can't parse this SX, I can throw the whole thing away if you like...", but -- again -- more UI.
12:56:41 <jsled> I'm only suggesting using a copy of the datafile. Yes, it's more time consuming, but it's much safer.
12:57:04 <dbr> anybody guessing on a 2.0.5 timeframe yet?
12:58:34 <jsled> I'm going to need to do the same thing, too. the trunk version always defaults to opening my 2.0 data file, which -- while great data -- isn't what I want.
12:59:27 <dbr> re safer -- sure, but if I'm going to test more features, then I get to duplicate all the entry work again in the real data. Certainly at some point there will be breaks. I even segregated my 2.0 work from 1.8 when cstim said it was easy to go back and forth. I guess I'm just being snarky and lazy. sorry.
13:01:26 <jsled> Ah, I see your point. If you can't do data entry, then there's a whole set of stuff that can't be tested, or you need to do the entry twice.
13:02:29 <jsled> hmmm... /me tries something else.
13:05:21 <jsled> No, as expected, the XML layer fails to load if there's an unanticipated tag. I guess we could roll a 2.0.x that doesn't, and -- up until some near-end 2.1/2.2 point -- save both the FreqSpec and Recurrence out.
13:10:52 *** warlord-afk is now known as warlord
13:11:01 *** gander has quit IRC
13:11:17 <jsled> Huh. That's broken. If we don't have an command-line option (like, say, for --loglevel), it treats command-line args like filenames?
13:11:38 <jsled> When I run dave's command, it tries to open "--nofile" (or "--loglevel", whichever is first)
13:15:36 <jsled> Ah, in fact, we don't check the return of g_option_context_parse at all.
13:18:10 *** ErKa has joined #gnucash
13:19:22 *** gander has joined #gnucash
13:24:01 <prock> trunk doesn't use g-wrap anymore, correct? why does autogen.sh tell me: /opt/local/share/aclocal/glib.m4:8: warning: underquoted definition of AM_PATH_GLIB?
13:24:42 <jsled> because that's not about g-wrap?
13:24:55 <prock> whoops wrong paste: /opt/local/share/aclocal/g-wrap.m4:23: warning: underquoted definition of AM_PATH_GWRAP
13:25:26 <warlord> prock: because /opt/local/share/aclocal is in your aclocal path
13:25:57 <prock> I more meant the warning about AM_PATH_GWRAP... but ok.
13:26:32 <prock> I guess then aclocal reads in everything in the aclocal path?
13:26:40 <jsled> I think it's just loading everything it... yeah.
13:27:00 <warlord> yes, it loads everyting in the path, regardless of if you're using it
13:33:28 *** bonez39 has quit IRC
13:36:18 <jsled> Huh. Now I'm seeing the same thing david herman is.
13:36:26 <jsled> WTF?
13:37:06 <warlord> Interesting.
13:38:12 <jsled> Oh. Shit.
13:39:02 <warlord> :(
13:39:04 <warlord> ???
13:39:09 *** nomeata has quit IRC
13:41:24 <jsled> insufficient testing of my changes. We're opening uninitialized memory as the log-output filename.
13:42:20 <jsled> insufficient testing of my changes. We're opening uninitialized memory as the log-output filename.
13:42:36 <jsled> (er...) We do the same thing with `add_quotes_file`, actually. :/
13:42:59 <hampton> add_quotes_file?
13:43:41 <jsled> yeah. It's not initialized, but it's passed to qof_session_begin(..) with whatever it might be pointing at.
13:43:47 *** MrN has joined #gnucash
13:43:47 *** garpert has joined #gnucash
13:44:09 <warlord> Eww.
13:44:16 <warlord> Valgrind? Or compiler warning?
13:44:37 <jsled> Inspection, tracking down dvherman's issue.
13:45:37 <warlord> Oh.
13:46:46 <jsled> Or, maybe not. WTF?
13:51:34 <jsled> Huh. And now I can't reproduce it. :/
13:54:25 <prock> hmm another problem with my build, am I doing something wrong or is this something with gnucash?: http://paste.debian.net/21710
13:56:30 <MrN> hi
14:01:19 *** wavydavy has quit IRC
14:04:16 <warlord> prock: This is on Mac? MacPorts or Fink?
14:05:08 <prock> macports
14:05:28 <warlord> Why don't you "sudo port install gnucash"?
14:05:28 <prock> I had fink installed but I moved /sw out of th eway.
14:05:38 <prock> because I want to build trunk
14:05:50 <prock> I already have gnucash installed but I deactivated it.
14:06:37 <warlord> Just making sure you have all the deps built.
14:06:56 *** garpert has quit IRC
14:07:22 <warlord> from there, follow the instructions on the wiki.
14:07:34 <warlord> (oh, you'll also need to "sudo port install swig")
14:07:44 <prock> oh sorry I must have missed those. yeah I've already gotten swig
14:09:20 <warlord> Okay.. Then start from a clean checkout, set PATH, CPPFLAGS, LDFLAGS, and DYLD_LIBRARY_PATH as per the wiki, then run autogen, configure, make, and make install.
14:09:40 <prock> ok will do thanks
14:10:02 <prock> I take it I don't actually need g-wrap though...
14:13:10 <hampton> correct
14:13:41 <warlord> prock: not for trunk, no.
14:24:17 <hampton> HBCI adds 6 items to the import menu. :-( There's got to be a better approach than an ever increasing menu list.
14:24:41 <warlord> Such as?
14:25:39 <hampton> I don't know. A single "Import..." entry that pops up a dialog. The dialog has a pop-up list of all possible import mechanisms, and defaults to whatever you used last time?
14:26:54 <warlord> Why is that any better than multiple menu items?
14:27:09 <hampton> all the existing import commands take you to a dialog anyway to input the file name
14:27:13 <hampton> less cluttered
14:30:03 <warlord> wow... make check is VERBOSE now!
14:30:15 <jsled> for those backend tests.
14:30:55 <warlord> oooh, test-sx failed.
14:31:09 <warlord> PASS: test-print-parse-amount
14:31:09 <warlord> /bin/sh: line 1: 22684 Segmentation fault GNC_MODULE_PATH="../../../src/engine/.libs:../../../src/calculation/.libs:../../../src/app-utils/.libs:${GNC_MODULE_PATH}" GUILE_LOAD_PATH="../../../src/core-utils:../../../src/gnc-module:../../../src/engine:../../../src/scm:../../../src/app-utils:${GUILE_LOAD_PATH}" LD_LIBRARY_PATH="../../../src/app-utils/.libs:../../../src/calculation/.libs:../../../src/engine/.libs:../../../src/gnc-module/.libs:../../../src
14:31:09 <warlord> /core-utils/.libs:../../../lib/libqof/qof/.libs:${LD_LIBRARY_PATH}" DYLD_LIBRARY_PATH="../../../src/app-utils/.libs:../../../src/calculation/.libs:../../../src/engine/.libs:../../../src/gnc-module/.libs:../../../src/core-utils/.libs:../../../lib/libqof/qof/.libs:${DYLD_LIBRARY_PATH}" ${dir}$tst
14:31:11 <warlord> FAIL: test-sx
14:31:15 <warlord> ==========================================
14:31:17 <warlord> 1 of 6 tests failed
14:31:19 <warlord> Please report to gnucash-devel@gnucash.org
14:31:20 <jsled> awesome.
14:31:21 <warlord> ==========================================
14:31:23 <warlord> make[5]: *** [check-TESTS] Error 1
14:31:25 <warlord> make[5]: Leaving directory `/home/warlord/src/gnucash/gnucash-svk/build/src/app-utils/test'
14:31:39 <jsled> Yeah, something's broken right now.
14:32:52 <warlord> ok
14:38:17 <jsled> hmm. passes for me, but I think the problem's memory-corruption related. warlord, that was with 15555?
14:39:26 <warlord> Yeah
14:47:39 <jsled> warlord: hmm. Any chance you have time to try and get a backtrace on that?
14:47:47 <prock> ok, same error.
14:49:52 <warlord> Program received signal SIGSEGV, Segmentation fault.
14:49:52 <warlord> [Switching to Thread -1208994112 (LWP 22856)]
14:49:52 <warlord> 0x00fd54b1 in __gen_tempname () from /lib/libc.so.6
14:49:52 <warlord> (gdb) bt
14:49:52 <warlord> #0 0x00fd54b1 in __gen_tempname () from /lib/libc.so.6
14:49:53 <warlord> #1 0x01045475 in mkstemp64 () from /lib/libc.so.6
14:49:55 <warlord> #2 0x0061595d in g_mkstemp () from /usr/lib/libglib-2.0.so.0
14:49:57 <warlord> #3 0x0048dc70 in qof_log_init_filename (log_filename=0x0) at qoflog.c:149
14:49:59 <warlord> #4 0x0048da6c in qof_log_init () at qoflog.c:87
14:50:01 <warlord> #5 0x00496332 in qof_init () at qofutil.c:548
14:50:03 <warlord> #6 0x00f2cb60 in gnc_engine_init (argc=0, argv=0x0) at gnc-engine.c:93
14:50:05 <warlord> #7 0x0804c87c in main () at test-sx.c:184
14:50:13 <warlord> prock: ppc or intel?
14:50:16 <prock> intel
14:50:21 <warlord> Oh. Hmm...
14:50:38 <warlord> I'm afraid I can't help you there.. I've only got a MacPPC.
14:50:47 <warlord> and it works just fine for me...
14:51:31 * hampton has a macintel...
14:51:39 <hampton> but I've never tried to compile gnucash on it
14:52:23 <warlord> jsled: It's calling qof_log_init_filename() with a NULL log_filename/fname
14:52:31 <jsled> yeah. I screwed the test up.
14:52:41 <jsled> testing a fix now...
14:53:58 <prock> ah looks like this is a bug with the dependencies.
14:55:12 <warlord> jsled: why doesn't qof_log_init_filename() deal with a NULL filename?
14:55:36 <jsled> warlord: well, because it's buggy. :)
14:56:40 <jsled> This is part of what I meant by "I'm not thrilled with [initilization], but [...]" from the email.
14:57:01 <warlord> OH.. I see..
14:57:41 <warlord> biab
14:57:43 *** warlord is now known as warlord-afk
14:57:49 <jsled> re-initing the log destination wasn't really supported before, and it's still a bit too convoluted right now.
15:55:52 *** spel has joined #gnucash
15:57:13 *** mnoir has quit IRC
16:04:41 *** sjc has joined #gnucash
16:06:02 *** spel has quit IRC
16:14:56 *** gander has quit IRC
16:23:59 *** gander has joined #gnucash
17:09:55 *** ErKa has quit IRC
17:48:13 *** warlord-afk is now known as warlord
17:59:16 *** _gunni_ has joined #gnucash
18:08:07 *** |gunni| has quit IRC
18:14:38 *** Demitar_ has joined #gnucash
18:18:42 *** conrad has joined #gnucash
18:18:42 *** gncbot sets mode: +o conrad
18:36:43 *** bab has joined #gnucash
18:36:57 <bab> has anyone here had any luck getting gnucash to run on FreeBSD (6.2)?
18:37:39 <bab> i was running on gentoo, and it got terribly broken. for several reasons I reinstalled with freebsd and now it tries to start up for about 5-10 while guile eats the cpu, then dies
18:37:50 <bab> *5-10min
18:38:27 *** gander has quit IRC
18:39:29 <bab> (error is at http://bab.pastebin.co.uk/10250)
18:46:42 <warlord> jsled: you here?
18:47:03 <warlord> bab: Did you read the FAQ?
18:47:53 <jsled> warlord: for a bit.
18:47:55 <warlord> bab: what version of slib? Looks like you have a bad slib.
18:48:16 <warlord> jsled: I'm actually using the Mortgage Druid and can't figure out how to tell it my monthly escrow.
18:48:50 <warlord> I can tell it what my quarterly taxes and yearly insurance payments, but I can't figure out how to tell it the escrow amount the bank is pulling..
18:53:10 <jsled> warlord: yeah, there's no place to do it in the druid. The escrow is exactly as much as the payments made through the escrow account.
18:53:19 <jsled> I didn't really understand how it worked when I wrote it.
18:53:26 <jsled> Now, I have a mortgage, and I do. :)
18:53:53 <jsled> In any case, once the SXes are created, you can probably manipulate the formulæ created to accomodate.
18:54:20 <warlord> Okay.. the P/I and Payment amounts are correct.
18:54:31 <warlord> I'll adjust the others as necessary.
18:57:17 * jsled goes afk again.
19:17:37 <warlord> jsled: Hmm... Got this from 2.0.x:
19:17:38 <warlord> (gnucash:16964): GLib-CRITICAL **: g_date_update_dmy: assertion `g_date_valid_julian (d->julian_days)' failed
19:17:46 * rhowe just had a weird idea
19:17:58 <rhowe> I was wondering how to model credit lines
19:20:52 <rhowe> Could Equity accounts be used?
19:21:44 <rhowe> If not, I was thinking of a special type of account, which when set as a child of a "Bank" account or a credit card account would keep track of your credit line.. then you could run reports on available funds taking into account available credit
19:21:57 <rhowe> I would love a report which told me how much I could spend before hitting all my limits :)
19:26:52 <bab> warlord: sorry, i was afk
19:26:58 <bab> just a sec i'll figure out the slib version
19:27:24 <bab> warlord: 3a4_1
19:28:26 <warlord> bab: yeah, 3a4 doesn't work with gnucash prior to (the unreleased) 2.0.5
19:28:30 <warlord> Downgrade slib to 3a3
19:28:42 <bab> warlord: okay. any idea on a 2.0.5 timeline?
19:28:54 <warlord> rhowe: I dont think you'd need a new account type; probably just an extension to the exiting Credit Card account.
19:28:59 <warlord> bab: Nope.
19:29:02 <bab> ok
19:29:09 <warlord> Just downgrade to slib 3a3
19:29:18 <warlord> or 3a1
19:29:46 <rhowe> warlord: Well, I'm thinking bank accounts with overdraft limits too
19:30:40 <warlord> rhowe: so?
19:31:11 <rhowe> warlord: So I guess they'd need extending in a similar way?
19:33:32 <warlord> No..
19:34:04 <warlord> You just need to extend a single type (e.g. CC) to understand "credit limits". An overdraft account would then be a CC under your bank account with the specified limit.
19:34:14 <rhowe> Ah right
19:37:02 <rhowe> Hmm
19:38:07 <rhowe> Would using an Equity account work? Have a "Credit Limit" account as a subaccount, make it an Equity account and transfer from a "Credit" equity account?
19:40:51 <warlord> No, you dont want to us Equity...
19:40:57 <warlord> At least I dont think you would..
19:41:53 * rhowe gets confused whenever he thinks about Equity account
19:41:53 <rhowe> +s
19:42:11 <warlord> No doubt
19:42:43 <rhowe> And I know that whenever you think about using Equity accounts to solve something, it probably means you shouldn't :)
19:43:12 *** todd has joined #gnucash
19:43:31 <warlord> *nods*
19:43:43 <todd> has anyone noticed that when moving from gnucash 1.x -> 2.x things slow down considerably when scrolling etc?
19:44:42 <todd> hey warlord, 'tis been a while ;-)
19:44:48 <warlord> todd: I've not noticed that. (Hi!)
19:46:00 <todd> so its good to know its not just my wife complaining or her computer .. though I suspect a trident cyberblade at 1280x1024@24bpp on a 732Mhz system probably doesn't exactly help the situation either
19:46:19 <todd> warlord: is it a generic gnome 1.x vs 2.x thing or ??
19:46:59 <warlord> Yeah, I bet it's the the system.
19:48:15 <todd> http://developer.gnome.org/projects/gap/tech-docs/access-bugs.html trying the acessability key whatever that is
19:49:58 <warlord> Oh, accessability can be broken!
19:50:16 <todd> well someone claims it slows things down when its enabled
19:50:19 <todd> so I'm trying disbaling
19:50:20 <todd> it
19:50:42 <warlord> It probably does
19:51:01 <todd> ~/.gconf/desktop/gnome/interface/%gconf.xml
19:51:15 <todd> supposedly the line in there is <entry name="accessibility" that effects things
19:51:27 <warlord> could be.
19:51:30 <todd> but you can read ehre more details: http://bugzilla.gnome.org/show_bug.cgi?id=118657
19:51:39 <todd> will report if this fixes or not, should know soon
19:53:22 <todd> bleh, can't find that setting anywhere in here xml files, wonder if its in the global ones
19:54:15 <warlord> Hmm...
19:55:41 <todd> ok I'm convinced this is a dead end
19:56:42 <warlord> Are you running on a local machine or across remote-X
19:57:31 <todd> local system
19:57:52 <todd> its a dramatic change, from gnucash 1.x scrolling rapidly to gnucash 2.x doing one line every other second or so
19:58:03 <warlord> What OS?
19:58:17 <todd> OpenBSD
19:58:20 *** jpeach has joined #gnucash
19:58:22 <warlord> (I've not noticed this at all.. But I'm not running a 700MHz system)
19:58:23 *** jpeach has left #gnucash
20:00:22 <MrN> n8
20:00:48 *** MrN has quit IRC
20:01:08 <todd> well if I track it down to anything in particular I'll report back, over and out ;-)
20:01:17 <warlord> Later.
20:17:21 *** bab has quit IRC
20:53:18 *** _gunni_ has quit IRC
21:03:47 *** andi5 has joined #gnucash
21:03:47 *** gncbot sets mode: +o andi5
21:23:08 *** sjc has quit IRC
21:45:17 *** rlaager has left #gnucash
21:53:18 <todd> so, um, is there any way to get the 'find' button back on the toolbar for gnucash 2.x ?
21:53:50 <warlord> umm...
21:55:27 <todd> sorry to be asking, my crazy wife just realized its not there, it was for 1.x
21:55:49 <todd> I've told her about the control-f feature, she'll get used to it I'm sure, but lets just say I have much incentive to ask ;-)
21:57:02 <warlord> maybe hampton can help.
22:12:10 <hampton> I suppose it could be put back. I probably pulled it during the 2.0 redesign after reading the HIG line that ways "Place only the most commonly-used application functions on your toolbars. Don't just add buttons for every menu item."
22:12:17 <hampton> s/ways/says/
22:13:20 <andi5> hampton: we could remove "Refresh" from nearly everything except reports... guess what we do not necessarily need there ;-) (do we?)
22:13:49 <warlord> No, we dont
22:14:06 <warlord> And we don't need 'close' if we have the little [x] on the tabs.
22:15:31 <andi5> warlord: seconded
22:15:31 <hampton> The little [x] on the tabs is a user preference.
22:15:37 <todd> some apps permit customization of toolbars, not sure my wife should change the app, but she was not pleased to realize it had changed *grin* .. change is always hard no matter what you do ;-)
22:15:55 <andi5> yes... if disabled -> put back into toolbar
22:15:55 <warlord> true
22:16:08 <hampton> I wouldn't mind a customizable toolbar
22:16:18 *** twunder has quit IRC
22:16:36 <todd> hampton: I'm to relay the pleasure of my wife in hearing that you like the idea of a customizable toolbar *grin*
22:17:23 <hampton> I didn't say I was going to write one, but I would help someone else do it. Hint. Hint. :-)
22:17:36 <andi5> todd: tell your wife to press Ctrl-F... i guess that will not change too soon :-)
22:19:05 *** twunder has joined #gnucash
22:22:48 <todd> oh heh, understood ..
22:25:53 <andi5> oh... i am sorry, the toolbar does not show refresh everywhere, (it is only the menus), except for the reports, say norminal condition
22:47:17 <warlord> hampton: do you have gnucash configured to open a register in a new window or in a new tab?
22:48:45 <hampton> tab. why?
22:49:16 <warlord> Can you do me a favor?
22:49:24 <hampton> sure
22:49:29 <warlord> Can you change it to a 'new window' and then open a register window..
22:49:43 *** twunder has quit IRC
22:50:11 <hampton> what version of code? 2.0, trunk?
22:50:33 <warlord> Either. I'm using 2.0-branch at the moment but I'm fairly sure trunk has the same behavior that I'm trying to show you.
22:51:50 <hampton> ok. what am I looking for?
22:51:57 <warlord> Notice the size of the window..
22:52:22 <warlord> resize, make the register actually fully visible..
22:52:30 <warlord> then close the window and re-open the register again.
22:52:38 <warlord> notice the size..
22:53:21 <todd> warlord: heh, yeah, my wife just noticed that also .. that it always opens as small as possible
22:53:45 <hampton> yes, same here.
22:53:51 <warlord> Yep.
22:53:54 <andi5> i think we have some similar bug report about that.... asking for "remembering sizes"
22:54:21 <hampton> I never implemented code to remember "window" sizes unless they were open when the application closes. "Dialog" sizes are remembered.
22:54:44 <warlord> How hard would it be to remember register sizes?
22:54:50 <hampton> probably because I always use register-in-a-tab mode myself.
22:55:08 <warlord> probably, which is why I asked you to change for a moment ;)
22:56:11 <hampton> well, what do you want to remember and where? The default size of any register page? The default size of the 2nd register window, 3rd register window, etc.? A size for each individual register?
22:56:42 <andi5> what if there is a window with pages of mixed types? ... is that a "register window"?...
22:56:47 <hampton> I think the first would be the simplest and easiest for users to remember.
22:57:27 <hampton> andi5: can't be. The question is what happens when you open a new register in a new window.
22:57:28 <warlord> I'd be happy to remember the "most recent register-window" size
22:58:32 <hampton> "most recent register-window" size or "most recent window" size? Your question probably applies to reports-in-a-new-window as well.
22:59:15 <andi5> hampton: yes... my question is just about what _is_ the most recent register-window... do registers count that belong to the window "the" account tree view is in?
22:59:23 <hampton> We could probably hook something into the window resize code to remember the last setting.
22:59:26 <warlord> I think register-in-a-window and report-in-a-window should each have different sizes.
23:00:02 <warlord> e.g., whatever is the "first" tab in the window should control how the size is remembered..
23:00:18 <warlord> (e.g. Accounts, Register, Report, Invoice, ...)
23:01:41 <hampton> I think that's going to confuse people. It would make more sense if any new window opened at the size of the (last resized) existing windows.
23:02:10 <hampton> Or at the size of the window where you performed the operation that opened the new window.
23:02:41 <warlord> Really? I dont think it would confuse people..
23:02:57 <warlord> If you open a register, it opens to the "register" size. If you open a report, it opens to the "report" size..
23:03:15 <warlord> That seems pretty straightforward to me.
23:04:18 <hampton> And if you open a register in a new window, then open a report in that same window, and then resized the window, what would you expect the resize to apply to? You were looking at a report when you performed the resize.
23:04:37 <hampton> Why would you expect it to then apply to new register windows?
23:04:43 <warlord> That's a good question.
23:04:59 <andi5> changing the window size when switching tabs would be even worse, btw :)
23:05:15 <warlord> Maybe base it on the current tab when you do the resize?
23:05:16 <hampton> That's why I think that "open to the size of the current window" is a good solution.
23:05:37 <warlord> is that what you mean by "open to the size of the current window"?
23:08:12 <hampton> I mean that if you double click on a register in a 800x600 window, you get a new 800x600 window.
23:08:36 *** prock_ has joined #gnucash
23:09:25 <andi5> hm.... i guess hamptons idea is easy to implement, while warlords could probably do more what the user might like to have :)
23:09:46 <warlord> Yeah, I'm describing how I'd like it to behave.. ;)
23:10:12 <andi5> WHATTHE.... opensp and libofx cannot be built without optimizations... askdjf
23:10:26 <todd> on a positive note I'm told to pass on that you're to be commended that the titles are more user friendly and there is a '*' in the first char of the title when unsaved data exists ..
23:11:08 <andi5> todd: hampton is responsible for the asterisk, i think
23:11:13 <warlord> todd: hampton did that
23:14:55 <andi5> hampton: do you think that it will suffice to connect to configure_event, check for changes in width or height and save according to the type of the currenctly active page? ... oh, and set the size on new windows too, of cousre
23:17:43 *** prock has quit IRC
23:24:01 <hampton> Configure_event looks like the way to go.
23:26:04 <hampton> I *don't* like the idea of a per-page window size. What happens if you 1) open a register in a new window, 2) resize to 800x600, 3) open a report in that same window, 4) resize to 500x500, 5) switch back to the register page, 6) do something that opens a new window. What size should that new window be?
23:26:53 <warlord> What kind of new window do you open?
23:27:01 <hampton> 6) do something that opens a new register.
23:27:11 <andi5> 800x600
23:27:18 <warlord> 800x700
23:27:23 <warlord> er 800x600
23:27:23 <jsled> :)
23:27:27 <andi5> *g*
23:27:38 <jsled> Oh. I was going to say 854x233
23:27:38 <hampton> Why? You were looking at a 500x500 window, and the register page in that window is 500x500!
23:28:06 <warlord> because the last time you resized a register you resized it to 800x600.
23:28:31 <andi5> exactly... let us call us a "good size for a register"
23:29:07 <andi5> the second "us" is a "that", actually
23:29:24 <hampton> So. You're looking at a 500x500 window and suddenly an 800x600 window pops up? That's not intuitive, at least not to me.
23:29:52 <andi5> well, if you work with several windows, why would you want to open reports in tabs, btw?
23:29:54 <hampton> I understand what you, warlord want. I'm saying I think it will confuse most people.
23:30:06 <warlord> get todd to ask his wife.
23:30:11 <hampton> andi5: why would you ever want more than one window?
23:30:15 *** prock_ is now known as prock
23:30:18 <andi5> hampton: i do not know ;-)
23:30:27 * andi5 works maximized all time
23:30:30 <hampton> lol
23:30:33 <jsled> Yeah, I think we should get around all this issue by not supporting more than one window. :)
23:30:33 <warlord> Hehehe...
23:30:42 * hampton uses a single tabbed window
23:30:49 <warlord> Hey, I LIKE my registers in separate windows!
23:31:08 <andi5> warlord: have you thought about switching to ... say kmymoney? ;-)
23:31:18 <warlord> :-P
23:31:23 <warlord> That would require KDE.. EEEWwww
23:32:13 <andi5> i think messing tabs with windows is.... like messing
23:32:45 <jsled> I don't know. It does work for firefox ...
23:33:11 <jsled> But that's much less sensitive to tab-type specific sizing.
23:33:15 <warlord> firefox can open windows of different sizes.
23:33:42 <jsled> At the same time, when I do New Window, I would be surprised if it came up with a window of a different size.
23:33:55 <jsled> But, again, the tabs are generic, not typed.
23:34:05 <warlord> right.. gnucash windows are typed.
23:34:14 <warlord> If they weren't typed then I'd agree with you 100%
23:34:17 <andi5> jsled: firefox does just what hampton suggested.... but well, you know the difference (^^)
23:34:59 <hampton> gnucash windows aren't typed. You can have any tab in any window.
23:35:10 <jsled> I've got it! The six window interface. One each for Accounts, Registers, Reports, Budgets, SxList and Business. :)
23:35:23 <warlord> Hehehe
23:35:29 <hampton> They used to be typed in 1.8, but that's not true in 2.0.
23:36:07 <jsled> Well, they're typed in the sense that a register report has different size requirements than the SxList, or a report.
23:36:50 <jsled> (Or, can, anyways. A single large window does the trick for me as well.)
23:37:39 <jsled> But I side with hampton; new windows should inherit the size of the window they're opened from. It's stateless and simple, and we have better ways to spend time.
23:39:18 <warlord> Hmm... I suppose.. But I find it odd when I resize a window, close it, reopen it, and it's not the size I sete.
23:44:58 <hampton> maybe, but it would be obvious what gnucash was doing
23:45:51 <jsled> g'night all.
23:45:57 <andi5> hampton: does the HIG say anything about that?
23:50:30 <warlord> see ya jsled
23:52:56 <warlord> Honestly, anything would be better than the minimum size we've got now.. ;)
23:58:14 <prock> 2.0 build builds on on my macintel, but trunk does not. "ld: multiple definitions of symbol _gnc_module_current". register/register-core/gncmod-register-core.c and register/register-gnome/gncmod-register-gnome.c (on trunk) both define gnc_module_current (and others) outside of any block whereas 2.0 branch defines then with unique names (eg - "libgncmod_register_core_LTX_gnc_module_current")
23:59:26 <prock> I have similar errors for _gnc_module_init, _gnc_module_path, _gnc_module_revision, _gnc_module_system_interface, _gnc_module_age, and _gnc_module_description
23:59:43 <warlord> prock: send mail to -devel
23:59:56 <prock> okies