2007-03-15 GnuCash IRC logs

00:06:39 <foo> haha.
00:26:38 <chris> It's updated.
00:34:04 <warlord> Looks like we didn't make it.
00:57:29 <aj> eh, there's only five listed?
01:03:34 <warlord> Yeah... Weird..
01:17:39 <warlord> they've got 6 listed now... So I think they're still working on the list.
01:22:26 *** dbr has quit IRC
01:33:24 <hampton> night all
01:33:32 *** hampton is now known as hampton|away
01:44:29 <warlord> night hampton|away
01:51:48 *** warlord is now known as warlord-afk
01:58:31 <warlord-afk> GSoC changed their site again. No more list, and it says "The list of accepted orgs will be posted here soon."
03:16:38 *** wizkid238_ has joined #gnucash
03:22:45 *** wizkid238 has quit IRC
03:48:52 *** ghetek has quit IRC
04:29:45 *** ceplma has joined #gnucash
05:30:45 *** ErKa has joined #gnucash
06:41:13 *** twunder has joined #gnucash
07:18:29 *** twunder has quit IRC
07:48:38 *** Adri2000 has joined #gnucash
07:49:31 <Adri2000> hi
07:51:22 <Adri2000> "../../../src/gnome-utils/.libs/libgncmod-gnome-utils.so: undefined reference to `gtk_html_print'" when building gnucash 2.0.2, any idea?
08:11:22 <jsled> Adri2000: what distro?
08:11:46 <Adri2000> jsled: I'm building the package for ubuntu
08:11:47 <jsled> Whee. <http://code.google.com/soc/gnucash/about.html>
08:11:53 <jsled> What version?
08:13:17 <Adri2000> feisty, the developement version, actually I'm an ubuntu developer and just wanted to upload a rebuild of gnucash to fix a dependency problem (that makes the package uninstabllable), but now it fails to build :(
08:13:19 <aj> jsled: told ya :)
08:13:36 *** BlackBsd has joined #GnuCash
08:14:15 <jsled> Adri2000: I thought so. libgtkhtml-3.13 does not respect the libgtkhtml "3.8" API.
08:15:01 <jsled> Specifically, it switches from gnome print to gtk print.
08:15:51 <Adri2000> the current libgtkhtml build-dependency for the package is 3.8
08:16:05 <jsled> yup.
08:16:15 <Adri2000> should I try 3.13?
08:16:27 <jsled> Adri2000: re-read what I wrote.
08:16:58 <jsled> I'm under the impression that Feisty has libgtkhtml-3.13.
08:17:13 <jsled> 3.13.92 I heard last night, c.f. <http://lists.gnucash.org/logs/2007-03-14.html#T17:58:45>.
08:18:12 *** thkoch has joined #gnucash
08:19:05 <Adri2000> we have libgtkhtml2-0, libgtkhtml3.14-19 and libgtkhtml3.8-19
08:19:49 <jsled> Ah, so libgtkhtml3.14 looks new, and promising.
08:19:59 <jsled> What version if libgtkhtml3.8 ?
08:20:11 <jsled> Er, what release version?
08:20:26 <Adri2000> 3.13.92
08:20:30 <Adri2000> err
08:20:35 <Adri2000> no, that's it
08:20:39 * jsled nods.
08:20:52 <Adri2000> libgtkhtml3.8-19: 3.13.92-0ubuntu1
08:20:59 <Adri2000> is it right?
08:21:01 <jsled> What's libgtkhtml3.14, then?
08:21:12 <Adri2000> libgtkhtml3.14-19: 3.14.0-0ubuntu1
08:21:33 <jsled> Ah, of course, they released 3.14 for the gnome-2.18 release yesterday.
08:21:41 <jsled> I'd bet it's just 3.13.92.
08:21:57 <jsled> But their "3.8" release still isn't backwards compatible.
08:21:59 <Adri2000> does 3.14 respect the 3.8 api?
08:22:19 <jsled> No, or they'd have been able to call it 3.8. :)
08:24:18 <jsled> I don't know that for sure.
08:24:50 <jsled> But gtkhtml has a track record of bumping the API version all the time; we check for gtkhtml-3.1, -3.2, -3.6 and -3.8.
08:25:03 <jsled> It's really funny, because we build fine against them all...
08:25:13 <jsled> But the *one* time we don't, they *don't* upgrade the version number.
08:25:41 <jsled> I think Feisty should revert to gtkhtml3.8-3.12.3, and use gtkhtml3.14-3.14.0 for new stuff.
08:25:51 <jsled> Using this 3.13.92 development version seems broken.
08:26:41 <Adri2000> ok, thank you for your help
08:27:03 <Adri2000> I will ask the guy who takes care about gnome and gtkhtml if it possible to revert
08:27:10 <jsled> Adri2000: I think there's more understanding to be had about the situation.
08:27:36 <jsled> Adri2000: I'd be very helpful if you relayed his response.
08:27:44 <jsled> I'd imagine there might be good reason to have gone to gtkhtml3.8-3.13.92.
08:27:57 <jsled> (though it might be fulfilled by having libgtkhtml3.14-3.14.0, now.)
08:30:00 <Adri2000> I'm going to leave now, but will come back here this evening (CET)
08:30:34 <jsled> Adri2000: thanks!
08:32:41 *** change008 has joined #gnucash
08:32:45 <change008> hi all
08:33:11 <change008> can i host gnucash on a server and access it remotely?
08:33:22 <jsled> via remote X, sure.
08:33:43 <jsled> But it's a desktop app, not client/server.
08:33:56 <change008> oh ok i c :(
08:34:12 <jsled> Yes, it's mostly in C as well... ;)
08:34:46 <change008> what will tyhe lag b via remote X?
08:35:06 <jsled> It shouldn't be too bad, but we've heard varying reports.
08:35:30 <change008> oh ok kewl thanx xse
08:35:38 <jsled> I've also heard that it works over freenx
08:35:46 <change008> freenx?
08:35:59 <jsled> An alternative remote display technology
08:36:01 <jsled> http://freenx.berlios.de/
08:36:21 <jsled> I don't know we've heard reports about it over "low-bandwidth" X.
08:36:54 <change008> do u know of any other acc package that may be server/client/hosted?
08:37:22 <jsled> SQLLedger comes to mind. It's web-based.
08:38:17 <change008> yar ill look at that i need a webbased app
08:38:42 *** Adri2000 has quit IRC
08:49:52 *** hampton|away has left #gnucash
08:50:40 *** change008 has quit IRC
09:05:02 *** andi5 has joined #gnucash
09:06:58 *** twunder has joined #gnucash
09:10:42 <andi5> jsled: note that i only committed a fix to let trunk compile against newer goffice versions, i.e. those with pixbuf-renderer
09:11:25 <andi5> this one issue is not (yet) present on 2.0, because we do not test for goffice-0.3.pc, i.e. will always use the internal copy if goffice is too new
09:12:20 <andi5> if we want to build 2.0 against a newer gtkhtml library, we will have to allow to build against newer goffices and use the cairo-renderer
09:14:53 <jsled> s/pixbuf-/cairo-/ in your first line?
09:19:51 *** hampton2 has joined #gnucash
09:19:51 *** gncbot sets mode: +o hampton2
09:19:55 *** hampton2 is now known as hampton
09:22:33 <jsled> Hmm. Need to submit a W9 to GOOG as part of the SoC Mentoring agreement.
09:30:48 <andi5> yes, or s/with/without/
09:41:26 *** IanL has joined #gnucash
10:04:38 *** warlord-afk is now known as warlord
10:04:44 <warlord> w00t! We made GSoC!
10:05:11 <jsled> :)
10:05:27 <hampton> :-)
10:08:27 <warlord> andi5: I dont think we need to change 2.0 to build against newer goffice.
10:08:51 <warlord> jsled: for change008, I've also run gnucash over VNC and it works fine.
10:10:06 *** kielein has joined #gnucash
10:12:33 <warlord> Hmm, we DO need to fill out a W-9.
10:12:47 <warlord> I wonder how much it'll bite me to use my Tax-ID?
10:13:00 <aj> warlord: for the org or for individual mentors?
10:13:03 <warlord> Or, I suppose I could use the IHTFP Consulting Tax-ID ...
10:13:11 <warlord> aj: For the Org.
10:13:25 <aj> warlord: going through Software in the Public Interest or the SFLC Conservancy's a better bet imo, fwiw
10:14:28 <aj> warlord: (SPI handled Debian, PostgreSQL and X.org last year iirc and should again this year; SFLC also comes highly recommended)
10:15:44 <aj> warlord: (www.spi-inc.org and conservancy.softwarefreedom.org if you haven't already googled)
10:17:14 <warlord> Ok
10:23:02 <warlord> So I think everyone here needs to apply to be a mentor.
10:23:52 <jsled> I have
10:24:12 <warlord> Also, I'm willing to have IHTFP take the "tax hit" on the $500/student so that GnuCash could get some extra $$ for operating costs or new hardware. It would be worth it to me, I think, because I/we could use that money to, say, buy a MacIntel for developer use!
10:25:52 <warlord> jsled: are you sure you checked off Gnucash?
10:25:55 <warlord> I dont see you in the list.
10:26:21 <andi5> warlord: suppose you want to use 2.0.5 on a system like ubuntu feisty with recent gtkhtml and goffice, i.e. gtkhtml prints with the help of GtkPrint, whereas the internal goffice does not have a cairo renderer to add its graphs
10:26:48 <warlord> andi5: I think feisty needs to downgrade gtkhtml from 3.13.92 to 3.12.xx
10:27:11 <jsled> "GnuCash Your Mentoring Acceptance is Pending" on the mentor dashboard.
10:27:18 <hampton> agreed. what's the point of having 3.13.92 and 3.14 on the same system.
10:27:44 <warlord> andi5: see http://lists.gnucash.org/logs/2007-03-15.html#T08:25:41
10:28:42 <warlord> jsled: interesting. I dont see it in my list, yet.
10:29:01 <hampton> btw, I now have a macintel but haven't tried to compile gnucash yet.
10:29:37 <aj> warlord: yeah, they're not going through to the admin for debian yet either
10:29:58 <aj> warlord: based on last year, we'll probably have to be fairly patient with the SoC webapp :)
10:30:16 <warlord> aj: good to know.
10:30:16 <jsled> [ot] Geez. How can it take way longer to build the gtk docs than the sources...
10:30:39 <hampton> [ot] lol. I noticed that too.
10:30:46 <jsled> Ah. xsltproc is involved.
10:31:08 <hampton> quick. new SoC project. speed up xsltproc. :-)
10:31:14 <jsled> heh.
10:32:00 <andi5> warlord thanks, now i see
10:35:25 <warlord> hampton: FWIW I tested the new check printing technology on FC5 and the check was printed perfectly. Although it didn't remember my settings from 2.0 -> trunk (probably because the name changed)
10:59:19 <hampton> I think that's because the ordering in the menu is different. 2.0 was hard coded. trunk is whatever order they are read from the directory.
10:59:35 <warlord> Oh, not sorted?
10:59:44 <hampton> I thought about sorting them, and also grouping system vs. user. vs custom.
11:00:25 <warlord> I assume that the 'last used' is stored as a string, not an index?
11:03:24 <hampton> no, its an index. that code wasn't changed.
11:03:46 <warlord> Oh.. So if the directory order changes then the setting will break?
11:06:24 <hampton> yes. now that you put it that way.... ;-)
11:07:14 <hampton> even if its sorted, the code can't save an index. An upgrade might include new reports that sort to the beginning of the list
11:08:35 * hampton thinks he needs to compile gnucash on his macintel
11:09:08 <warlord> Right, so I think either formats should have GUIDs, or we should remember the name.
11:12:21 <hampton> either one has issues. what happens when a user copies a check format from /usr/share/gnucash/checks to ~/.gnucash/checks and starts making changes? Unless they change the magic identifier (name or guid) there can be duplication.
11:13:32 <hampton> At one point in development I was munging the names of user check formats by adding "(user)" to them. That didn't make the commit. (I'd rather use horizontal lines in the popup menu to differentiate format groups.)
11:13:59 *** hampton sets mode: +oo andi5 benoitg
11:14:45 <jsled> but we can detect and complain about the duplication.
11:14:55 <hampton> yup
11:16:09 <warlord> Or we can /internally/ have a <namespace>|<name> string.
11:16:30 <warlord> e.g. system|<name>, user|<name>, custom|<name>
11:19:49 *** ceplma has quit IRC
11:20:59 * hampton is thinking it would be easier to use guids and complain about duplicates. That way if a user changed a report title it would still be selected properly. Only question is where to complain (g_warning or a dialog).
11:22:04 <jsled> Dialog, if possible. though "why'd my format not show up?" "check the console output" "oh, 'CRIT duplicate guid'. fixed."
11:22:15 <warlord> dialog
11:22:30 <hampton> easy enoughh
11:22:56 <hampton> ok, /me now has tonight's project :-)
11:23:30 <warlord> :)
11:42:27 *** IanL has quit IRC
12:13:28 *** andi5 has quit IRC
12:32:12 <jsled> warlord: I suggest announcing to -devel (and maybe to -user) about the SoC acceptance.
12:32:23 <warlord> jsled: Yeah, I'll go do that.
12:38:19 *** benoitg has left #gnucash
12:39:16 *** benoitg has joined #gnucash
12:39:29 *** ceplma has joined #gnucash
12:42:13 *** rlaager has quit IRC
12:52:20 <chris> warlord: I think you can let the mentors for selected projects provide their own W9.
12:54:56 <warlord> Does the $500 go to the Mentor or the Project?
12:55:53 *** bonez39 has joined #gnucash
12:55:58 <jsled> The "mentoring organization", as per <http://code.google.com/support/bin/answer.py?answer=60322&topic=10731>
12:56:48 <jsled> I'd imagine the individual mentors need to handle the W9s; maybe they'll split the value on the 1099s amongst the mentors or something.
12:58:58 <warlord> Well, I've got a corporate tax-id I can use; and if we then use the proceeds to buy, say, computers or equipment or other expenses then it wont affect my 2007 taxes.
13:05:55 <chris> I'm just saying that I _think_ one way of handling it when there's no real organizational legal entity is to just defer to the mentors.
13:07:11 <chris> and in that case, the $500 would go to the mentor. Google can't pay an "organization" that doesn't exist as a legal entity.
13:08:03 <jsled> That's true.
13:08:54 *** andi5 has joined #gnucash
13:08:55 *** gncbot sets mode: +o andi5
13:09:05 <chris> hiho andi5
13:09:08 <warlord> Hmm
13:09:10 <warlord> Hiya andi5
13:09:14 <andi5> hiho
13:09:54 <warlord> Well, what's the difference of the organization going through, e.g. SPI or SFLC vs. e.g. IHTFP Consulting?
13:10:01 <chris> hey, andi5, you're still a student, no?
13:10:12 <jsled> Oooh! Mentor yourself! :)
13:10:44 <andi5> no, i am not a student any more, being a PhD student or so now
13:12:13 <andi5> jsled: too much differences... i do not think anyone can mentor himself ;-)
13:12:23 <andi5> or many or what
13:19:57 <warlord> basically, i'm happy to use IHTFP Consulting on the W-9 for the organization (even though that means that I, personally, would pay the taxes on the amount received).. But since I already manage the tip-jar I don't see that as being any different. This would just be "more tips" :)
13:58:03 *** ceplma has quit IRC
14:00:02 *** hampton has quit IRC
14:03:29 *** ErKa has quit IRC
14:07:07 *** bonez39 has quit IRC
14:12:21 *** Adri2000 has joined #gnucash
14:13:03 *** andi5 has quit IRC
14:14:14 *** bonez39 has joined #gnucash
14:39:56 *** hampton2 has joined #gnucash
14:39:57 *** gncbot sets mode: +o hampton2
14:40:04 *** hampton2 is now known as hampton
14:57:18 *** _gunni_ has joined #gnucash
15:02:47 *** ErKa has joined #gnucash
15:07:43 *** sjc has joined #gnucash
15:30:39 *** esodan has joined #gnucash
15:31:49 <esodan> Hi all, I can inform you that the gobject-engine-dev branch is near to compile completly
15:32:34 <warlord> nice.
15:34:27 <esodan> I have all libqof and engine compiling, I'm on gnome-search directory
15:34:58 <esodan> The use of GType as the type system has changed alot of things here and there...
15:35:05 *** _gunni_ has quit IRC
15:35:23 <warlord> esodan: I'm sure it has.
15:35:31 <warlord> Have you been committing as you get each file working?
15:35:41 <warlord> (and if not, why not?)
15:35:53 *** rlaager has joined #gnucash
15:36:39 <esodan> Well I'd done as soon as I can, in my work I can't access using ssh... :(
15:36:49 <jsled> Yeah, I don't understand why you're not dying to commit ... what?
15:37:25 <warlord> What do you mean, can't acces using ssh from work?
15:38:27 *** bonez39 has quit IRC
15:38:33 <chris> esodan probably works behind a firewall that blocks ssh.
15:38:38 <esodan> Yes, I'm behind a proxy and the policies dict I can't use ssh
15:39:14 <esodan> then I need to use a public or home internet connection...
15:40:15 <esodan> I'm at work now...
15:40:19 <warlord> do they actually block it? Would it help you if I set up ssh on an alternate port for you?
15:41:08 <esodan> No... simply in my organization they don't take that time for me...
15:41:58 <esodan> I'm a voluntary and my work is very diferent... not programing or so is my job...
15:43:18 <warlord> Ah.
15:43:25 <esodan> I have stoped in dialog-search.c file...
15:43:26 <warlord> So I guess you can commit once you get home?
15:43:48 <warlord> The dialog-search should mostly already be gobjects.
15:43:56 <esodan> Yes I'll do or may at lunch time...
15:44:48 *** bonez39 has joined #gnucash
15:45:25 <esodan> Well my problem isn't realated to gobject to dialog-search (this isn't gobject, but that is other thing)...
15:45:39 <warlord> what other thing?
15:45:41 <warlord> GDA?
15:47:34 <esodan> No, I found that there are a lot of objects that could be declared as GObject like dialog-search, but it isn't... I don't want to work here for now... I just have a problem in the creation of searchs...
15:48:48 <esodan> in get_params_list function you have a ganc_search_param_prepend method witch use SPLIT_ACCOUNT_GUID...
15:50:11 <esodan> Becouse I switched all the type system to GType you must declare any data type you use in order to asing a GType id... but SPLIT_ACCOUNT_GUID is a string and I'm no sure how it is used for...
15:50:32 <warlord> SPLIT_ACCOUNT_GUID is a parameter name
15:50:48 <warlord> The parameters are still named, not typed.
15:52:09 <esodan> I found that I can, and I'll do, use the GObject properties, thing about QofParameters = GObject Properties...
15:53:05 <warlord> And that's in pass #2..
15:53:26 <warlord> So why is that being an issue now?
15:53:27 <esodan> Yes of course...
15:53:35 <warlord> Right now it's still a string..
15:56:03 <esodan> well in /src/gnome-utils/search-param.h the function gnc_search_param_prepend takes a GNCIdTypeConst, and this is a QofIdType and this is a GType (using a typedef declaration)... but the paramenter assigned to this function is SPLIT_ACCOUNT_GUID... witch is a string... not a GType...
15:56:53 <warlord> Oh, I see...
15:56:55 <warlord> hmm.....
15:58:15 <esodan> Well I'll simply change the declaration for gnc_search_param_prepend and gnc_search_param_prepend_internal to just take a gchar pointer...
15:59:50 <warlord> Yeah, that should probably be (explicitly) a const ghcar*
16:00:50 *** bonez39 has quit IRC
16:01:08 <esodan> Well may be changing GNCIdTypeConst as a string and when needed convert a GType as string or a string to a GType...
16:02:37 <esodan> This problem will be resolved when we will use GObject properties and not QofParam's...
16:03:32 <esodan> This must be the second step, in order to avoid this problems...
16:05:06 <warlord> Yeah, I think I did overload param names and data types and assume they are similar identifiers.
16:06:08 *** kielein has quit IRC
16:06:55 <esodan> Yes for example the ID "account-guid" refers to the GncAccount's GUID property...
16:08:09 <warlord> esodan: well, yes and no.. It referes to the GUID property in a special way.
16:08:14 <esodan> This will be as is, for now, and I'll use g_type_name and g_type_from_name to convert a GType to a string...
16:09:05 <esodan> well, of course I don't study the code enough to say any thing about this now, but I'll do, in the second step...
16:09:12 <warlord> ok
16:17:59 *** hampton has quit IRC
16:21:53 <BlackBsd> http://wiki.gnucash.org/wiki/SoC2007
16:22:02 <BlackBsd> is this stuff difficult to finish over 3 months?
16:22:20 <BlackBsd> how much time are you guys guessing will be put into the project?
16:22:24 <BlackBsd> full time?
16:22:35 <warlord> BlackBsd: I think each project should be doable in 3 months, full-time.
16:25:25 <jsled> Or less, depending.
16:26:57 <warlord> True
16:27:55 <jsled> I doubt we'll get full-time out of the students.
16:28:09 <jsled> Though ... $4500 /3 = $1500/mo ... that should be full time.
16:28:51 *** thkoch has quit IRC
16:30:22 <warlord> Maybe.. Probably depends on the student. When I was a student I'd have considered that about 100 hours / month..
16:40:59 *** benoitg has left #gnucash
16:45:31 <esodan> This changes compiles well now...
16:45:51 <warlord> yay
16:46:02 <esodan> but I have a problem with libgnomeprint-2-2 library in Ubuntu... the deb package doesn't install the .la file required to link, I'll install this from sources...
16:46:29 <esodan> this is stoping me to continue hacking in the code...
16:46:42 <warlord> You shouldn't need to .la
16:46:50 <warlord> What's referring to it?
16:48:17 <esodan> the linker tells about to need gnomeprint in order to create libgncmod-gnome-utils.la
16:48:43 <warlord> it still shouldn't need the .la, unless something ELSE is referring to the .la.
16:49:11 <warlord> find /usr -name \*.la | xargs grep libgnomeprint<name>.la
16:52:23 <esodan> realy I don't now... Ok I'll try it...
16:54:17 <warlord> esodan: just fill in <name> with the actual "missing" library name
16:55:31 *** andi5 has joined #gnucash
16:55:32 *** gncbot sets mode: +o andi5
16:57:52 <esodan> it founds a lot of .la gnucash related files: libgw-engine, libgncmod-engine, libgncmod-netowr-utils, libgncmod-app-utils, libgncmod-engine, libncmod-gnome-utils and so..
16:58:38 <warlord> sure, but what about NON-GnuCash files?
16:58:55 <warlord> (it's POSSIBLE that your problem is that the installed versions are being used during compile-time)
16:59:52 <esodan> don't worry it is compiling again, if I found a problem with his again I'll back here... I need to go to lunch... see you...
17:00:02 <warlord> happy eating
17:04:49 <esodan> thanks...
17:04:53 *** esodan has left #gnucash
17:07:52 *** twunder has quit IRC
17:09:41 <Adri2000> jsled: around?
17:09:49 <jsled> Adri2000: yes
17:11:27 <Adri2000> jsled: you confirm that 3.14 isn't compatible with 3.8 api?
17:11:39 <warlord> Adri2000: yes.
17:11:58 <warlord> they removed gtk_html_print in 3.13.9x
17:12:20 <jsled> 3.13.92 isn't, at least. I suppose it's possible that 3.14 restored that functionality.
17:12:39 <andi5> Adri2000, jsled: i am going to take a look at that issue, hopefully this evening
17:13:08 <jsled> But, gnucash does't know to look for the package config identifying itself as "libgtkhtml-3.14".
17:13:26 <jsled> (which is awesome if they were able to release that yesterday on 3.14... ;)
17:15:26 <andi5> at least, the tarballs seems to declare GTKHTML_API_VERSION as 3.14, in contrast to 3.13.* with 3.8
17:15:37 <andi5> so i guess jsled will be right
17:16:56 <jsled> I'll be departing for dinner plans in a bit ...
17:18:04 <Adri2000> jsled: I'm talking with the right ubuntu guy about that, he says "ABI version != source version", so libgtkhtml3.8 should work...
17:19:04 <jsled> We're well aware about the difference between abi and source versions.
17:19:27 <jsled> We know that gtkhtml-3.8 is implemented by gtkhtml-3.{8,10,12}.
17:19:30 <Adri2000> sure more aware than me, thay's why I forward his message here :)
17:20:07 <jsled> We also know that libgtkhtml-3.13.92 claims to support the "3.8" api, but doesn't. Probably for the same reasons that they started calling it "3.14".
17:20:26 <warlord> Adri2000: except libgtkhtml3.8-3.13.92 BROKE the ABI (AND API)
17:20:34 <Adri2000> ok ok, I see
17:20:46 <andi5> Adri2000: what exactly is the problem? aptitude shows me packages libgtkhtml3.8, libgtkhtml3.8-dev, libgtkhtml-3.14 and libgtkhtml3.14-dev....
17:21:22 <Adri2000> andi5: libgtkhtml3.8 is 3.13.92 with the API/ABI change that breaks the build of gnucash
17:21:32 <Adri2000> well, he says he will upload a "without the GtkPrint API change"
17:21:44 <Adri2000> a libgtkhtml3.8*
17:21:47 <warlord> andi5: libgtkhtml3.8 is 3.13.92, which.... what Adri2000 said
17:22:22 <andi5> oh... me is slow today.... mr feisty packager did a mistake :)
17:22:36 <warlord> Yes, he did ;)
17:22:54 <jsled> biaw.
17:26:41 *** conrad has quit IRC
17:42:34 *** ecki has joined #gnucash
17:44:43 *** torchte has joined #gnucash
17:47:42 <torchte> hi. My question is how to set accounts tu hbci accounts? Ich möchte gern die Kontoauszüge meines Girokontos im Konto Girokonto verbuchen. wie mach ich das?
17:48:54 <andi5> torchte: did you use the online banking setup druid (maybe hbci setup druid)? .... in one step you can associate each aqbanking account to a gnucash one
17:50:11 <torchte> how is it tu use? I made the hbci setup but there was no opportunity to chose an account.
17:51:25 <andi5> well, the druid spawns another program, namely aqbankings qt3-wizard.... when you exit that, you can proceed with the next druid step which should list you all accounts set up, just (double) click one of them
17:51:38 <andi5> at least that is what i did in my minimal setup :)
17:53:45 *** ecki has quit IRC
17:57:50 *** torchte has quit IRC
17:57:52 *** torchte has joined #gnucash
17:59:30 <torchte> >andi5 sorry and thank you. it works you have to run the hbci setup twice and chose the gnucash account. Thank you
18:00:31 <andi5> no problemo :) ... freut mich, auch mal geholfen haben zu können
18:01:20 <torchte> na und warum quäle ich mich da so mit englisch? :)
18:01:55 <andi5> because this is an english channel (and logged, btw)
18:02:39 <torchte> yes of course
18:03:20 <andi5> torchte: i am sorry, i did not want to sound harsh... at least on the aqbanking mailing list folks typically speak in german =)
18:04:43 <torchte> no no all right. you were not harsh. It is only that my english lessons seems to be far far away a long time ago
18:09:00 *** ErKa has quit IRC
18:28:26 *** warlord2 has joined #gnucash
18:28:26 *** gncbot sets mode: +o warlord2
18:29:10 *** warlord has quit IRC
18:34:15 *** warlord2 is now known as warlord
19:00:21 *** esodan has joined #gnucash
19:00:34 *** nbinont has joined #gnucash
19:01:01 <esodan> I have commited revision 15726 with my lastest changes, hope to finish soo the first step...
19:05:54 <warlord> yay
19:09:17 <esodan> I have the same problem with libgnomeprint-2-2.la...:-/
19:10:33 <esodan> please help me...
19:11:22 <andi5> problems with gtkprint, goffice, now gnomeprint.... /me gets frustrated...
19:11:31 <andi5> s,gtkprint,gtkhtml,.... of course
19:11:47 *** torchte has quit IRC
19:12:16 <warlord> esodan: find /usr -name \*.la | grep -v gnucash | xargs grep libgnomeprint-2-2.la
19:19:09 <nbinont> Hi! Anyone have a clue how I might go about debugging undefined references to gnome-utils stuff?
19:19:15 <esodan> Ok, now I have a lot of la files, then?
19:20:04 <esodan> warlord: what dp I need to search for?
19:20:19 <andi5> nbinont == nathan?
19:20:43 <nbinont> yes
19:21:08 <andi5> great :) ... first of all, the build succeeded for me, so i really hope we only search for oldies in your build tree :)
19:22:36 <andi5> the symbols you mention should be in the app-utils dll, you might run "nm" on it .... i think it there should be a "T" in front of that symbol, at least i think so
19:23:17 <esodan> warlord: please, talk me about the results I'll get with that command...
19:27:02 <andi5> nbinont: i.e. you might run `nm .libs/libgncmod-app-utils.dll | grep gnc_register_gui_component' from c:\soft\gnucash\build\src\app-utils
19:28:08 <nbinont> ok, good to know that it's working for others. I do not get the symbol gnc_ui_to_account from nm on /c/soft/gnucash/dist/lib/gnucash
19:28:19 <warlord> esodan: sorry, i'm on the phone,.
19:28:29 <nbinont> I get others (a few pages of them)
19:28:36 <warlord> um, you're looking for non-gnucash references to that .la
19:28:49 <warlord> you might need to remove the gnucash installed in /usr
19:30:11 <andi5> nbinont: note that there are three, maybe four dlls .... one in build/src/app-utils/.libs, one in inst/lib/gnucash, (one may still be in inst/lib/bin, it will be moved later in install.sh) ... and one in dist/... if you ran dist.sh .... oh, no. 5 would be an setup-installed one :)
19:30:55 <nbinont> I don't have libgncmod-app-utils.dll anywhere else in /c/soft/gnucash
19:30:58 <andi5> so if you have compilation problems, do not leave gnucash/build :)
19:31:03 <warlord> andi5: would the linker find all of those locations?
19:32:16 <andi5> the linker has no chance to know no. 4 and 5 .... no. 1 should be found... no 2. and 3. are installed file locations and the linker _should not_ find them, imho
19:32:29 <nbinont> ando5: I blew away /c/soft/gnucash/inst and build and repos before my last build attempt...could it be trying to link against the one in dist?
19:33:13 <andi5> did you put anything from dist into your paths or environment variables? ... we do not do anything with it
19:33:39 <andi5> it is just a place to gather all the bits and pieces we want to bundle in an installer
19:35:02 <andi5> nbinont: actually, i do not remove repos... just build, inst (and dist)... but i hope that is not the problem :)
19:36:02 <andi5> so you do not have an app-utils dll? ... please step into build/src/app-utils and run "make check all"
19:36:15 <andi5> maybe it prints you some errors and falls back to static libraries... eek
19:36:44 <andi5> argh... may keyboard is ill... it meant "make clean all"
19:37:55 <nbinont> andi5: I don't think it's in my path...I'm doing a quick search to be sure
19:42:02 <andi5> [ot] http://abock.org/2007/03/15/banshee-in-summer-of-code-not/ ... sheds a different light on gnucash's approval :)
19:45:20 <nbinont> I got \n Making clean in test \n ../../../../repos/src/gnc-test-env: exec: guile: not found
19:45:49 <nbinont> and further on \n libtool: link: warning: `/c/soft/guile/lib/libguile.la' seems to be moved \n libtool: link: warning: `/c/soft/guile/lib/libguile-ltdl.la' seems to be moved
19:45:58 <andi5> do not mind those warnings
19:46:07 <andi5> your miss some paths....
19:46:51 <andi5> nbinont: did you install to c:\soft?
19:46:53 <nbinont> ok...which path do I need? /c/soft/guile/lib?
19:46:56 <nbinont> yes
19:47:15 <andi5> quick workaround.... paste this to your shell: export PATH="/c/soft/hh:/c/soft/gnucash/inst/bin:/c/soft/inno:/c/soft/aqbanking/bin:/c/soft/gwenhywfar/bin:/c/soft/libofx/bin:/c/soft/opensp/bin:/c/soft/goffice/bin:/c/soft/libgsf/bin:/c/soft/pcre/bin:/c/soft/swig:/c/soft/gnome/bin:/c/soft/openssl/bin:/c/soft/guile/bin:/c/soft/autotools/bin:/c/soft/readline/bin:/c/soft/regex/bin:/c/soft/tools/bin:$PATH"
19:48:07 <nbinont> ok
19:48:53 <andi5> nbinont: whenever you finish a install.sh run, it shows you what is missing in your environment and what to do about it, i.e. it prints you something to copy and paste into the terminal (you will have to do it yourself)
19:50:02 <nbinont> ok, so that sounds like it was my problem
19:50:51 *** rlaager has quit IRC
19:51:16 <andi5> nbinont: i would suggest: comment out every add_step before "svn_up" in custom.sh... run install.sh .... copy & paste what it gives you .... uncomment again, rerun install.sh (will take some time, of course)
19:52:31 <andi5> nbinont: out of curiosity: did it install the html help workspace correctly for you?
19:55:47 *** warlord is now known as warlord-afk
19:58:28 <andi5> [ot again] .... lol, is http://www.abisource.com/twiki/bin/view/Abiword/GoogleSummerOfCode2007 a way to keep away lazy/resistant_to_learning students? (nice idea)
20:00:15 *** Adri2000 has quit IRC
20:03:35 *** andi5 has quit IRC
20:08:08 <jsled> esodan: why did you drop a bunch of '_'s in names?
20:08:17 <jsled> e.g. [[[
20:08:17 <jsled> - g_list_foreach(entity_list, qof_entity_list_foreach, qecd);
20:08:17 <jsled> + g_list_foreach(entity_list, qof_instancelist_foreach, qecd);
20:08:18 <jsled> ]]]
20:09:01 <jsled> Also, [[[
20:09:02 <jsled> - SchedXactions *sxes = GNC_SXES(ent);
20:09:02 <jsled> - SchedXaction *sx = GNC_SX(evt_data);
20:09:02 <jsled> + SchedXactions *sxes = GNC_SCHEDULE_ACTIONS (ent);
20:09:02 <jsled> + SchedXaction *sx = GNC_SCHEDULE_ACTION (evt_data);
20:09:29 <jsled> ]]] is wrong. They're not "schedule actions", but "scheduled xactions" where "x" is short for "trans".
20:09:55 <jsled> "SX", while perhaps terse, is a very good abbreviation.
20:10:15 <jsled> But I don't know why you'd change that at all.
20:13:02 <jsled> Ugh. There's going to be nasty conflicts between your changes and the Group removal.
20:16:01 <jsled> Oh, you changed all the SchedXaction names too.
20:16:05 <jsled> :(
20:17:27 *** BlackBsd has quit IRC
20:17:48 <jsled> Why'd you add 'FIRST_SIGNAL' to the Split.c GncSplitSignalType enum?
20:18:54 <jsled> Re: [[[
20:18:54 <jsled> - * Transaction.c -- transaction implementation *
20:18:54 <jsled> + * GncTransaction.c -- transaction implementation *
20:19:22 <jsled> ]]] you didn't actually rename the file [which is a good thing[, and please maintain the block formatting.
20:20:35 <jsled> Uh, why are there "commit:beginning" and "begin-edit" signals.
20:20:36 <jsled> ?
20:20:48 <jsled> I mean, why are there signals at all, and why are there two different conventions?
20:21:14 <jsled> g_message isn't a valid replacement for PERR.
20:21:37 <jsled> Oh, and you're already changing xaccFreeTransaction to g_object_unref() ?
20:21:47 <jsled> I thought you'd agreed to hold off on that...
20:24:06 <jsled> Oh, you are renaming GncTransaction and GncAccount.
20:24:06 <jsled> Wow.
20:26:22 <jsled> Moving, e.g., the Budgets fields into a pimpl doesn't seem like something that needs to happen now.
20:29:38 <jsled> something happened @ dialog-search.c:192 [[[
20:29:39 <jsled> - /* Call the first button (usually view/edit) */
20:29:39 <jsled> + /* Call the first button (uSPLIT_ACCOUNT_GUIDsually view/edit) */
20:29:40 <jsled> ]]]
20:36:53 *** esodan has quit IRC
20:37:44 <jsled> Also, I guess I'd like some justification that all the changes from [safe_]strcmp(GNCIdTypeConst, GNCIdTypeConst) to "==" are valid.
20:41:05 <jsled> [ot] hee. http://lkml.org/lkml/2003/12/23/307
20:44:33 *** twunder has joined #gnucash
21:14:48 *** BlackBsd has joined #GnuCash
21:26:28 *** nbinont has quit IRC
21:31:04 *** stodge has joined #gnucash
21:31:15 <stodge> Hi - can I set a starting balance for an existing account?
21:32:37 <stodge> The account already exists....
21:39:35 *** stodge has left #gnucash
21:58:34 <jsled> chris: thanks.
22:07:10 <chris> jsled: np.
22:13:18 *** BlackBsd has quit IRC
22:14:09 *** guitsaru has joined #gnucash
22:32:53 *** sjc has quit IRC
23:05:25 <jsled> http://www.barisione.org/blog.html/p=79
23:05:34 <jsled> ^- [[[
23:05:35 <jsled> This morning I committed GRegex to svn, so glib 2.14 will have regular expression support. GRegex is a wrapper around the PCRE library by Philip Hazel that has several interesting features:
23:05:36 <jsled> ]]]
23:05:47 <jsled> :)
23:33:29 *** guitsaru has quit IRC
23:46:30 *** hampton has joined #gnucash
23:46:31 *** gncbot sets mode: +o hampton