2007-02-19 GnuCash IRC logs

00:32:21 *** hampton is now known as hampton|away
00:55:10 *** Demitar has quit IRC
01:06:49 *** dnw has joined #gnucash
01:08:35 <dnw> Hi, I have just compiled 2.0.5 and have a run time error " /usr/lib/gnucash/libgncmod-engine.so.0: undefined symbol: qof_string_number_compare_func " system is ubuntu 6.1
01:09:20 <dnw> There wa a report of this in oct 2006 in the dev list.
01:11:57 <Wilddev> hmm really?
01:12:18 <Wilddev> let me fire up my ubuntu system
01:14:35 <dnw> I have just installed ubuntu and run the packaged Gnucash version 2.0.2 and installed the dev packages and complied 2.0.5 so this is the first time I have compiled on this system
01:16:39 <Wilddev> yeah that's pretty much where my system is at here, except I havent done the installs
01:16:51 <Wilddev> but I do remember that error from before
01:18:10 <Wilddev> what versions of guile, slib and g-wrap do you have?
01:19:58 <Wilddev> oh I see what your problem is
01:20:18 <Wilddev> dnw: you need to uninstall the current gnucash before you compile
01:22:42 <Wilddev> dnw: and I'd suggest to upgrade to guile 1.8.0, g-wrap 1.9.6 and slib 3.1.1
01:22:53 <dnw> no ! I will do that and have another go at installing
01:23:03 <Wilddev> dont go newer on the last 2 as it will break stuff
01:23:25 <Wilddev> huh?
01:24:22 <Wilddev> dnw: I think what happened is that when you were compiling it linked against the old version so you go the error bc of that
01:52:58 <dnw> Thanks - that seems to be the problem it now runs. I have to remove the old .gnucash settings as it crashed with report option errors-
01:54:10 <dnw> Are you interested in the back trace ?
02:00:38 <Wilddev> hmm, no, I'm not a dev :)
02:00:43 <Wilddev> glad u got it working
02:01:45 <dnw> Thanks for the help - good night.
02:01:49 *** dnw has quit IRC
02:42:21 *** Wilddev has quit IRC
05:15:28 *** conrad has joined #gnucash
05:15:28 *** slicslak has joined #gnucash
05:15:28 *** wizkid238 has joined #gnucash
05:15:28 *** gncbot sets mode: +o conrad
05:34:58 *** minDscrm has quit IRC
06:02:18 *** Demitar has joined #gnucash
06:28:05 *** dgtlmoon has quit IRC
06:51:35 *** dgtlmoon has joined #gnucash
06:55:31 *** egsavage has joined #gnucash
07:13:44 *** dgtlmoon has quit IRC
07:20:05 *** writerz_ has joined #gnucash
07:25:32 *** dgtlmoon has joined #gnucash
07:27:32 *** exper_ has quit IRC
07:38:12 *** twunder has joined #gnucash
07:54:24 *** twunder has quit IRC
07:59:14 *** egsavage has quit IRC
08:04:25 *** prock_ has joined #gnucash
08:13:09 *** prock has quit IRC
08:21:56 *** jpeach has quit IRC
08:25:46 *** twunder has joined #gnucash
08:30:46 *** jpeach has joined #gnucash
08:36:05 <Linuxbeak> Hello, I'm trying to emerge gnucash (again) but it's giving me an error that grep: /usr/lib/libguile-ltdl.la doesn't exist. I've already reemerged guile several times over because gnucash complained that it wasn't built correctly, and now it this happens. Any ideas?
08:36:11 *** IanL has joined #gnucash
08:51:48 *** jpeach has left #gnucash
08:59:32 <Linuxbeak> anyone around?
09:24:55 *** IanL has quit IRC
09:31:03 *** rlaager has quit IRC
09:39:46 *** giant has joined #gnucash
09:40:04 *** warlord-afk is now known as warlord
09:40:21 <warlord> Linuxbeak: find /user -name \*.la | xargs grep /usr/lib/libguile-ltdl.la
09:40:40 <warlord> something is broken in your base system; something expects a .la file that doesn't exist.
09:41:45 <warlord> dgtlmoon: You can't use the QIF importer to pay an invoice. You need to "process payment", which will create a transaction from "Bank -> A/R" and then in the importer you need to match the payment to A/R and call it a duplicate of the existing transaction.
10:08:39 *** IanL has joined #gnucash
10:11:07 <Linuxbeak> warlord: errrr. there are a TON of lines of code
10:11:51 *** minDscrm has joined #gnucash
10:16:54 <warlord> Linuxbeak: so you're saying that lots of stuff (NOT gnucash-related) references that file? But the file doesn't exist? That would imply that your whole system needs to be upgraded.
10:21:38 *** IanL has quit IRC
10:22:52 *** hampton|away is now known as hampton
10:29:19 <jsled> Linuxbeak: re-emerge g-wrap.
11:00:23 <Linuxbeak> jsled: okay
11:00:31 <Linuxbeak> (sorry, I was out delivering paychecks)
11:01:53 <Linuxbeak> jsled: I've followed several different people's advices at this point. I'm using the ~x86 versions of gnucash, g-wrap and guile
11:04:04 <hampton> what? No paychecks for us? :-)
11:07:48 <Linuxbeak> free as in beer ;-)
11:08:38 <Linuxbeak> okay, I remerged gnucash
11:08:39 <Linuxbeak> er
11:08:42 <Linuxbeak> g-wrap
11:08:52 <Linuxbeak> now I'm reemerging... gnucash... which just broke.
11:09:23 <Linuxbeak> okay. I masked the newest version of guile
11:09:29 <Linuxbeak> a la advice of someone else
11:09:46 <Linuxbeak> just removed it from package.mask
11:14:48 <hampton> I'm willing to be paid in beer. ;-)
11:16:00 <warlord> me, too! :)
11:17:27 *** ErKa has joined #gnucash
11:18:02 <Linuxbeak> heh. Sorry, I'm only 20.. can't help you yet ;-)
11:18:17 <Linuxbeak> unless of course you're willing to move me to Europe!
11:18:17 <warlord> hehe
11:18:43 <hampton> lol
11:20:06 <Linuxbeak> okay... gnucash is compiling... no problems *yet8
11:20:07 <Linuxbeak> *
11:20:20 <giant> your only 20?... since when did that stop anybody?
11:22:28 <warlord> yay
11:22:51 <Linuxbeak> it failed!
11:22:52 <Linuxbeak> again
11:22:59 <Linuxbeak> grep: /usr/lib/libguile-ltdl.la: No such file or directory
11:23:09 <Linuxbeak> I thought reemerging g-wrap would fix that?
11:25:40 <Linuxbeak> here's the error:
11:25:40 <Linuxbeak> grep: /usr/lib/libguile-ltdl.la: No such file or directory
11:25:40 <Linuxbeak> /bin/sed: can't read /usr/lib/libguile-ltdl.la: No such file or directory
11:25:40 <Linuxbeak> libtool: link: `/usr/lib/libguile-ltdl.la' is not a valid libtool archive
11:25:41 <Linuxbeak> make[4]: *** [libgw-core-utils.la] Error 1
11:26:14 <warlord> Linuxbeak: run that 'find' command again..
11:26:40 <warlord> or, better yet, find /usr/lib -name \*.la | xargs grep libguile-ltdl.la
11:27:02 <Linuxbeak> done. There are a lot of lines, and way too many for me to post here.
11:27:44 <Linuxbeak> although it doesn't all look like they're libguile-ltdl.la related
11:28:31 <Linuxbeak> as a matter of fact, it looks like there are no instances of it anywhere in the results
11:29:59 <warlord> Linuxbeak: post it to pastebin.ca
11:31:08 <Linuxbeak> http://pastebin.ca/363673
11:32:17 <warlord> g-wrap still has that dependency..
11:32:18 <warlord> /usr/lib/libgwrap-guile-runtime.la:dependency_libs=' /usr/lib/libgwrap-core-runtime.la /usr/lib/libguile.la /usr/lib/libguile-ltdl.la /usr/lib/libqthreads.la -ldl -lpthread -lcrypt -lm /usr/lib/libffi.la'
11:33:13 <Linuxbeak> it doesn't exist on my system, though.
11:34:01 <warlord> well, g-wrap seems to have found it!
11:34:31 <warlord> Here's two more g-wrap libs that pull it in:
11:34:33 <warlord> #
11:34:33 <warlord> /usr/lib/libgw-guile-gw-glib.la:dependency_libs=' /usr/lib/libgwrap-guile-runtime.la /usr/lib/libgwrap-core-runtime.la /usr/lib/libguile.la /usr/lib/libguile-ltdl.la /usr/lib/libqthreads.la -ldl -lpthread -lcrypt -lm /usr/lib/libffi.la /usr/lib/libglib-2.0.la -lrt'
11:34:33 <warlord> #
11:34:33 <warlord> /usr/lib/libgw-guile-standard.la:dependency_libs=' /usr/lib/libgwrap-guile-runtime.la /usr/lib/libgwrap-core-runtime.la /usr/lib/libguile.la /usr/lib/libguile-ltdl.la /usr/lib/libqthreads.la -ldl -lpthread -lcrypt -lm /usr/lib/libffi.la'
11:34:52 <warlord> This tells me that g-wrap is pulling it in, so your g-wrap build is broken.
11:35:17 <warlord> maybe you should remove your existing gnucash and g-wrap installs, then rebuild/reinstall g-wrap.. then rebuild/reinstall gnucash.
11:35:33 <warlord> feel free to run this find test again after you remove gnucash and g-wrap.
11:35:45 <Linuxbeak> okay, so emerge -C gnucash and g-wrap, right?
11:35:53 <Linuxbeak> what about my financial data?
11:36:10 <warlord> I dont know the emerge commands, I dont do Gentoo.
11:36:14 <Linuxbeak> stupid question: is it necessary to back it up
11:36:17 <warlord> But your financial data will be fine.
11:36:24 <warlord> No. It wont touch that.
11:36:41 <warlord> It wont touch your configuration, either. Once you reinstall it'll just pick up agian.
11:38:13 *** andi5 has joined #gnucash
11:38:13 *** gncbot sets mode: +o andi5
11:38:36 <Linuxbeak> okay. I'm uninstalling gwrap and gnucash
11:38:40 <warlord> 'k
11:39:33 <Linuxbeak> okay. rebuilding g-wrap now
11:41:00 <warlord> did you try the 'find' after you removed them both to make sure there wasn't anything else left?
11:42:39 <Linuxbeak> the same command as last time? no.
11:42:39 <warlord> biab..
11:42:46 <Linuxbeak> should've.
11:48:32 <Linuxbeak> It's not there,a gain
11:48:33 <Linuxbeak> *again
11:53:05 <andi5> real distro.... ;-)
12:04:07 *** twunder has quit IRC
12:04:25 <warlord> good.
12:04:36 *** twunder has joined #gnucash
12:06:56 <jsled> Linuxbeak: which versions of guile, g-wrap and gnucash are you using?
12:07:27 <Linuxbeak> jsled: I'm not certain, but the new emerge of gnucash hasn't failed yet :-)
12:07:35 <Linuxbeak> gnucash 2.0.4-r1, I know that
12:11:56 <warlord> I have no idea why the re-emerge of g-wrap didn't force a rebuild..
12:15:36 <warlord> anyways, gotta run. have a good day
12:15:38 *** warlord is now known as warlord-afk
12:22:45 <jsled> yeah ... a revdep-rebuild after upgrading guile should have flagged g-wrap as having a broken dep.
12:24:36 <Linuxbeak> still is compiling
12:27:35 <chris> Should we blacklist the broken slib and g-wrap versions in our configure.in?
12:28:15 <andi5> what is the current problem with slib, btw
12:29:07 <jsled> I'm not sure. gentoo had specified all the (in)compatible configurations before we (would have been able to) released an updated configure.in.
12:29:21 *** benoitg has quit IRC
12:29:29 <jsled> And in some cases, the compatability might be in the packaging. I think the slib-3.1.4 issue is just that.
12:29:56 <jsled> But, I'm not sure what the issue is, actually. :(
12:35:38 *** nomeata has joined #gnucash
12:43:48 <Linuxbeak> wow. This thing is still compiling
13:11:50 <Linuxbeak> IT'S ALIVE!!!
13:11:52 <Linuxbeak> bwahaaha
13:27:30 *** ErKa has quit IRC
13:27:39 *** nomeata has quit IRC
13:37:44 *** ErKa has joined #gnucash
13:52:19 *** dgtlmoon has quit IRC
13:53:06 *** MrN has joined #gnucash
13:53:29 <MrN> hi
14:12:10 <hampton> Gratuitous whitespace changes suck.
14:12:40 <jsled> so do tab characters. :p
14:13:07 <hampton> Fortunately they only caused two conflicts in my tree.
14:15:35 <hampton> If we're going to enforce a coding style, we should first agree on it.. Then we need a set of options to the indent program to create that style. Then we should have a flag day where all files are converted to the proper style.
14:16:30 <hampton> I'm not saying that we shouldn't, but that we should be methodical about it.
14:16:37 <jsled> True. I would like a more consistent style.
14:16:50 <jsled> FWIW, I felt justified in that change, as they've all been re-(mis-)formatted by me recently.
14:17:14 <jsled> I've made other whitespace-only changes in the last few weeks, but only along side other valid changes.
14:19:34 <jsled> And on one, I brought my recent changes in-line with the existing style of the file ... that's gotta count for something. :)
14:20:14 <hampton> lol. I wasn't keeping score, just griping about gratuitous changes causing conflicts.
14:21:38 <jsled> :) I am sorry about that. I did consider the possibility, and frankly didn't think it would be so... where was it?
14:22:07 <andi5> maybe we need some input whether we can agree at all :)
14:22:53 <hampton> app-utils/gnc-sx-instance-model.c. I have changes to make the template_group a template_root
14:23:02 <hampton> no big deal.
14:24:19 <andi5> i only noticed that plongstaff uses a style of bracing that i have not seen in newly written code for a while :)
14:25:19 <hampton> I'm willing to propose a set of options to indent, but I doubt everyone would agree with them.
14:26:29 <jsled> Always a fun discussion. :)
14:26:58 <andi5> well, i would be fine if we agreed on something that differs significantly from my own style...
14:27:00 <hampton> Or should we just use 'indent -gnu' (a style I don't like). I know jsled will insist on the --no-tabs option. :-)
14:27:42 <andi5> +1 for no tabs
14:27:45 <hampton> I have my preference, but I would really just like something that's consistent and the I can configure emacs to enforce for me.
14:28:14 <jsled> There's a couple of areas I care about, and a couple of things I don't really.
14:28:25 *** ErKa has quit IRC
14:29:52 <andi5> hampton: you could make a proposal and count your plusses.... if another proposal gets more then.... well, you will not win ;-)
14:30:33 <jsled> If we've been talking about having The Great Renaming as well, then this naturally goes along side of it.
14:31:05 <jsled> And I think warlord's right re: timing: at the beginning of a test/release series would be good...
14:31:21 <andi5> whereas indentation changes seem more invasive to me than renamings
14:31:23 <hampton> Seems to me that The Great Renaming would go well with the GObject conversion.
14:31:34 <hampton> agreed on timing.
14:31:38 <jsled> Yeah.
14:32:08 <jsled> I'm assuming the Account Group removal branch will fold in soon. And the modules-cleanup won't be too too long lived.
14:32:25 <jsled> The GObject branch will (hopefully) progress pretty quickly in time,
14:32:32 <jsled> That leaves the GDA-dev branch.
14:33:23 <andi5> you seem to be more optimistic than me
14:33:28 <jsled> heh. :)
14:36:28 <jsled> An advantage of the modules-cleanup is that it can fold in kinda whenever. If half the modules are gone, then that's half better than now.
14:37:33 <hampton> I'm hoping to commit the remove-group branch will commit this week. I've moved all the changes to trunk (the reason I kept bugging jsled about trunk yesterday). I need to 1) retest, 2) add support for the "allow_file_incompatability" key, and 3) inspect the diffs again.
14:40:40 *** andi5 has quit IRC
15:04:19 *** egsavage has joined #gnucash
15:16:06 *** Linuxbeak has quit IRC
16:24:28 *** magpie has quit IRC
16:26:20 *** egsavage has quit IRC
17:04:43 *** giant has quit IRC
17:13:30 *** egsavage has joined #gnucash
17:25:55 *** andi5 has joined #gnucash
17:25:55 *** gncbot sets mode: +o andi5
18:01:50 *** benoitg has joined #gnucash
18:02:43 <MrN> n8
18:02:58 *** MrN has left #gnucash
18:03:27 <jsled> nate? neight? naight?
18:03:53 <hampton> Night? I was going to ask too, but he already left.
18:04:21 <andi5> maybe nacht :)
18:04:55 <hampton> ah. makes sense.
18:11:54 <andi5> hampton: r15635 looks good
18:12:03 <hampton> ty
18:38:14 <jsled> hampton: hey. I'm "fixing" the (gnc:{warn,error,msg,debug}) procs to use the g_log stuff, and a side-effect is that nothing goes to stdout by default anymore, including the "found finance::quote ${version}" message...
18:39:19 <jsled> I figure there's two options: either just explicitly (display ...) that message to stdout, or assume that 2.2 will have an improved handler that will write >=message to std{out,err} as well...
18:39:20 <hampton> I kinda liked having that on stdout, but ok.
18:40:20 <jsled> I guess it makes sense to quickly change it to explicitly (display), just in case the other change doesn't get made...
18:41:51 <hampton> yeah, that's fine. I stuck it in because yahoo changed their website and broke currency quotes for F::Q < 1.09, there were lots of user questions, and no easy way for them to find the F::Q version.
18:42:02 <jsled> Yeah, it's a useful message.
18:42:31 <hampton> As long as the output is somewhere consistent that we can point users to I'm ok with the change.
18:43:23 <jsled> Dig. I've got it double-outputting at the moment, but I do intend to make the other change at some point...
18:49:41 *** mnoir has joined #gnucash
18:58:26 *** andi5 has quit IRC
19:01:28 *** warlord-afk is now known as warlord
19:14:14 *** ural0 has joined #gnucash
19:42:28 *** warlord has quit IRC
20:07:13 *** mnoir has quit IRC
20:22:47 *** warlord has joined #gnucash
20:22:48 *** gncbot sets mode: +o warlord
20:27:42 *** motin has joined #gnucash
21:00:12 *** ural0 has quit IRC
21:56:11 *** twunder has quit IRC
23:06:09 *** motin has quit IRC
23:23:52 *** egsavage has quit IRC