2010-05-07 GnuCash IRC logs
00:39:51 *** gour has joined #gnucash
00:53:38 *** blizgerg has quit IRC
00:58:42 *** blizgerg has joined #gnucash
02:09:16 *** tim_abell has joined #gnucash
03:13:34 *** gour1 has joined #gnucash
03:20:00 *** gour has quit IRC
04:11:22 *** gjanssens has joined #gnucash
04:11:23 *** gncbot sets mode: +o gjanssens
04:22:15 *** tim_abell has quit IRC
04:40:54 *** ErKa has joined #gnucash
04:57:23 *** daedeloth has joined #gnucash
05:00:42 *** fDavid has joined #gnucash
05:03:44 *** fDavid has quit IRC
05:44:37 *** CaMason has quit IRC
06:41:26 *** fDavid has joined #gnucash
07:36:56 *** Jimraehl has left #gnucash
07:48:39 *** Jimraehl has joined #gnucash
08:39:40 *** fbond has quit IRC
08:45:55 *** fbond has joined #gnucash
08:47:36 *** warlord-afk is now known as warlord
08:53:51 <warlord> gjanssens: THANK YOU!
08:54:11 <gjanssens> warlord: you're welcome :)
08:55:18 <warlord> I figured it had to be a bug in the qofquery code somewhere.. Because I knew the code worked the first time I wrote it, but then major changes happened to QOF and it stopped working. But it's not a feature I use so I didn't notice the failure until months (or years) later when the bug was filed.
08:55:29 <warlord> That will definitely make people happy!
08:57:50 <gjanssens> Yes, it was a nasty bug, that only surfaced via very indirect ways. Glad we got rid of it.
09:05:14 *** fbond has quit IRC
09:05:40 <warlord> Indeed! Thank you.
09:06:43 *** fbond has joined #gnucash
09:08:03 <gour1> warlord: would you agree with the following mark for gnucash: "GNUcash is a corner case but seems to
09:08:06 <gour1> lack an API"
09:08:22 <gour1> see http://gnumed.blogspot.com/2010/05/billing-invoicing-applications-for.html
09:08:23 *** fbond has quit IRC
09:08:33 * gour1 suggested gnucash
09:08:35 *** gour1 is now known as gour
09:08:56 <warlord> GnuCash has an API. It's just not easy to use for third-party applications.
09:09:27 *** fbond has joined #gnucash
09:09:37 <gour> warlord: i see...where can one learn more about it?
09:10:14 <gour> gnumed is wxpython server/client app looking got billing app (it uses postgresql)
09:10:48 <warlord> http://code.gnucash.org/docs/HEAD/
09:11:38 <warlord> I would say PG/MySQL/SQLite/XML
09:11:51 <warlord> And there IS an API (but it's in C)
09:11:56 <warlord> .. And Scheme
09:12:02 <warlord> .. And now python
09:13:41 <gour> pytghon api is via those python-gnucash bindings?
09:13:51 <warlord> I'm not sure the difference between i18n and i18n capable
09:14:52 <gour> me too
09:14:52 <warlord> yeah, which have been partially integrated into the mainline sources.
09:17:52 <gour> you said 'partially', what about 'the rest' ?
09:18:46 <warlord> We've been waiting for them to provide it.
09:19:19 <gour> so, intention is to have everything integrated?
09:23:05 <warlord> That's the hope, but who knows if/when it'll happen. I think they ran out of funding to maintain it.
09:23:48 <gour> i see...how useful is the present 'part' to access gnucash in python?
09:26:29 * gour forwarded the above comments to gnumed list
09:28:40 <warlord> I have no idea. I dont use python. and I never build the python bindings myself.
09:29:41 <gour> ok. so 'recommended' way is to use guile?
09:31:57 <warlord> recommended way to do what?
09:32:27 <gour> to access gnucash remotely?
09:33:23 <warlord> define "remotely"
09:33:33 <warlord> client-server?
09:34:07 <warlord> It's hard to answer generic questions without knowing the background of what you're really trying to do.
09:34:42 <gour> well, gnumed wants to use 3rd party application for billing
09:35:06 <gour> the question how easy/hard gnumed can do it
09:35:55 <warlord> I dont know. I have no idea what gnumed is trying to do or how they want to 'do billing'
09:36:00 <gour> s/gnumed/gnucash
09:36:49 <gour> yeah..you're right. if they're interested, let'am ask
09:37:50 <warlord> While I appreciate your putting the good word in for gnucash, when it comes to technical integration it's hard to be a middleman.
09:38:39 <gour> i know...i just consider that gnucash would be better solution than php-based invoicing app meant to be use with mysql
09:39:11 <warlord> Maybe...
09:39:22 <warlord> Or maybe not. Maybe sql-ledger would be better for their use.
09:39:27 <warlord> I dont know.
09:40:39 <gour> heh..let's see...
10:19:20 <roe_> I recently came from sql-ledger to gnucash for accounting, and have been toying with the idea of integrating simple invoices with gnucash
10:19:52 <warlord> What is "simple invoices"?
10:20:09 <roe_> it is a web based invoiceing system
10:20:50 <roe_> not that robust yet, but I hope to allow clients to login to see their past invoices and such eventually.
10:21:15 <roe_> I'm waiting for 2.4 before I give it a good effort
10:22:18 <warlord> ok
10:37:27 <mishehu_> I rolled my own php invoicing system, but use gnucash for the other financial tasks.
10:37:32 *** mishehu_ is now known as mishehu
10:37:46 <mishehu> so yes, I do have to enter payments into two separate systems.
10:38:04 <warlord> I just use GnuCash's fancy-invoice
10:38:24 <mishehu> but my php script runs by cron, so most of the invoice production is automated.
10:39:15 <mishehu> some of my clients have monthly recurring services, it's much easier to just let it automatically get made and e-mailed.
10:44:38 <warlord> ok
11:23:43 *** fDavid has quit IRC
11:31:44 *** daedeloth has quit IRC
11:36:28 <roe_> mishehu, did you ever try to integrate the two systems?
12:52:52 <gjanssens> warlord: the list search issue keeps bugging me
12:52:59 <gjanssens> would you run a test for me ?
12:53:21 <warlord> Sure.
12:53:27 <gjanssens> Simply go to the list search pages, and select "Nederlands" as language
12:53:39 <gjanssens> Does it come properly formatted or not for you ?
12:54:14 <warlord> Looks right to me
12:54:43 <gjanssens> Ok, thanks, but that kills my theory once more...
12:55:03 <warlord> I'm looking at http://code.gnucash.org/search/nl/-
12:55:06 <warlord> Looks right to me.
12:55:17 <gjanssens> You see, nl is the only language that is not explicitly defined inside nmz that does work properly
12:55:36 <gjanssens> It works fine for me as well, but it's my native locale language.
12:55:44 <warlord> HOWEVER, https://lists.gnucash.org/search/en/ does NOT look correct.
12:55:47 <mishehu> roe_: negative on that
12:55:54 <gjanssens> Yes, I know
12:56:09 <mishehu> roe_: it might be more possible now with db backing support now in gnucash unstable.
12:56:10 <warlord> Neither does https://lists.gnucash.org/search/en/nl/-
12:56:11 <gjanssens> Firstly: why is that https suddenly ?
12:56:30 <gjanssens> Then that's our clue
12:56:31 <warlord> because I clicked off the 'search' link off www.gnucash.org
12:56:41 <roe_> mishehu, right I'm waiting for the release of unstable before I take a serious look at it
12:58:16 <warlord> Another bug: I keep switching languages and now my URL is: http://code.gnucash.org/search/zh_CN/ja/de/en/-
12:58:20 <gjanssens> Your last link is a secondary effect of the problem
12:58:30 <gjanssens> It's all the same bug
12:58:37 <warlord> Oh, I missed that..
12:58:45 <gjanssens> There are a limited set of languages that work out of the box
12:59:03 <gjanssens> But you have to get to the proper page via the proper link
12:59:42 <warlord> See, it should work from any link... IMHO.
13:00:05 <gjanssens> The good languages are nl, de, fr, es, ja, and nb
13:00:22 <gjanssens> If you choose one of the other languages, the thing gets messed up
13:00:53 <gjanssens> And the problem is mainly that for some reason the relative path is set to .. for the good languages and . for the bad ones
13:01:11 <gjanssens> I haven't figured out why yet
13:01:56 <gjanssens> But I suspect it is related to the use of $top_dir
13:02:19 <gjanssens> At least that explains why the links get messed up
13:02:31 <gjanssens> and continue to mess up more recursively
13:05:13 <gjanssens> I also see that for all languages that don't work, the locale is set to en_UTF-8
13:05:17 <warlord> Hmmm..
13:08:59 <gjanssens> And even that top_dir is set to .
13:09:44 <gjanssens> I'm pretty sure this is not getting set from one of the files in svn
13:09:50 <gjanssens> Another wild guess...
13:10:19 <gjanssens> Can you check somewhere in the nmz installation if there is a po directory
13:10:46 <gjanssens> And if there is something in it ?
13:11:21 <gjanssens> I wonder if the nmz language information is not setup via softlinks to the actual po files on the website
13:11:40 <warlord> gjanssens: what about the local.php in the template dir?
13:12:29 <gjanssens> you have a local.php in the template dir ? That's not in svn for sure
13:12:42 <gjanssens> What is in it ?
13:13:18 <gjanssens> Does your template dir also have language directories ?
13:14:25 <warlord> yes
13:14:40 <gjanssens> For all the supported languages ?
13:14:43 <warlord> It's from file:///home/svn/repo/htdocs/trunk
13:14:49 <warlord> Yes.
13:15:25 <warlord> The /home/nmz/nmz/local.php is the same as htdocs/local.php
13:15:30 <warlord> It *is* in svn
13:15:47 <gjanssens> Yes, but not in the *templates* directory
13:17:03 <gjanssens> So just to be clear: /home/nmz/nmz/ has got a local.php
13:17:20 <gjanssens> Does /home/nmz/nmz/ also have the language directories ?
13:17:35 <warlord> Yes. /home/nmz/nmz is a checkout of htdocs
13:17:47 <warlord> htdocs/trunk to be exact
13:17:52 <gjanssens> Ok, thanks
13:17:55 <warlord> It has everything that htdocs/trunk has
13:18:27 <gjanssens> And it is updated automatically whith each commit to htdocs/trunk ?
13:20:05 <warlord> yes
13:21:04 <gjanssens> I'm just trying to figure out what information nmz uses to find its files...
13:21:48 <gjanssens> If you look in /home/nmz/nmz/en/local.php, is top_dir set to . or .. ?
13:22:17 <warlord> $top_dir = "..";
13:22:52 <gjanssens> Thanks, that definitely confirms it has been updated properly.
13:23:45 <warlord> I could have told you it was being updated.. you own a bunch of stuff in there ;)
13:24:13 <gjanssens> Heh, indeed :)
13:25:11 <gjanssens> What's in /home/nmz/index ? That defined as "the default directory" according to namazurc
13:26:39 <warlord> directories for each list.
13:27:12 <warlord> http://pastebin.com/sS8mFYLz
13:30:17 <gjanssens> The search indexes, no doubt...
13:30:48 <gjanssens> Did you compile nmz from source, or did you install it from rpm ?
13:30:57 <warlord> RPM
13:31:04 <gjanssens> Ok
13:31:11 <warlord> namazu-2.0.18-6.fc10.i386
13:31:27 <gjanssens> I'll go and search the rpm myself and see what gets installed where
13:31:44 <gjanssens> The source is quite a mess to sift through
13:34:54 <warlord> Yeah.
13:35:27 <warlord> However, I just found http://wpkg.org/Integrating_Mailman_with_a_Swish-e_search_engine
13:37:20 <warlord> Just another option...
13:38:08 <gjanssens> I had seen that one yesterday as well
13:38:43 <gjanssens> I didn't suggest it because at first sight it doesn't work with individual list messages
13:39:25 <gjanssens> I had the impression it indexed the full index files. So if you search for gnucash, the result will show the september digest for example
13:39:42 <gjanssens> But perhaps that demo was just poorly configured.
13:39:57 <gjanssens> I didn't dig much deeper then.
13:41:34 <gjanssens> Perhaps we should look deeper
13:41:50 <gjanssens> (in Swish-e's capabilities, I mean)
13:42:45 <gjanssens> There's search/namazu-cgi.sh in svn
13:42:56 <gjanssens> Is this file acually called ?
13:44:42 <warlord> yes, but it needs to be copied into place manually
13:45:30 <gjanssens> Ok, so that means that eventually /var/www/cgi-bin/namazu.cgi gets executed
13:45:44 <gjanssens> That file is not part of the rpm, so what is it ?
13:46:29 <gjanssens> A custom script or just a softlink ?
13:48:00 <warlord> A copy of /usr/lib/namazu/namazu.cgi
14:05:01 *** ErKa has quit IRC
14:08:44 *** andyt has joined #gnucash
14:09:28 *** fbond has left #gnucash
14:11:04 *** fbond has joined #gnucash
14:16:55 <gjanssens> Ok, I didn't look in the right rpm, there's also namazu-cgi....rpm and namazu-lib....rpm
14:25:03 <gjanssens> Hmm, just to rule out the obvious: are the locales for the missing languages installed ? it, zh_CN ?
14:30:49 <warlord> it_IT exists.
14:31:05 <warlord> as does zh_CN.
14:31:08 <warlord> (according to locale -a)
14:32:19 <gjanssens> Ok, that should confirm they are there.
14:34:45 <gjanssens> Back to /home/nmz/nmz/search/templates
14:34:56 <gjanssens> You said there was a local.php file there, right ?
14:36:26 <gjanssens> Or did you mean /home/nmz/nmz to have the local.php ?
14:40:42 <warlord> /home/nmz/nmz
14:41:13 <gjanssens> Yes, I was expecting that.
14:42:09 <warlord> http://pastebin.com/BWPBHCvr
14:42:20 <warlord> that's the ls from /home/nmz/nmz/search/templates
14:43:18 <gjanssens> It's plain svn checkout as far as I can see
14:43:21 <gjanssens> The source shows that environment variable NAMAZULOCALEDIR can be used to override the locale directory for nmz.
14:43:40 <gjanssens> Do you know if this parameter is set ? I think it should be.
14:43:49 <gjanssens> Question is where.
14:44:22 <gjanssens> Does the apache config set up some environment for the cgi scripts ?
14:45:47 <warlord> I dont think it's set.
14:45:55 <gjanssens> Hmm yet another wild guess
14:46:28 <gjanssens> I have added some .htaccess files. Do you think the webserver config has to be reloaded before they are detected ?
14:46:39 <warlord> I suppose we can test it in the namazu-cgi.sh
14:46:48 <warlord> Shouldn;t.
14:47:57 <gjanssens> I didn't think so either. It is weird though that the .htaccess files were missing for the languages that have issues
14:49:12 <warlord> I can restart httpd if you want me to?
14:49:31 <gjanssens> Well, it certainly won't hurt...
14:49:35 <warlord> ok...
14:49:49 <warlord> Done.
14:50:46 <gjanssens> Now we're sure: it didn't help at all ... :(
14:55:21 <warlord> k
14:55:37 <warlord> It was Neil that did all this.....
15:08:05 <kimmo> hrm
15:37:46 *** gour has quit IRC
15:49:23 *** cortana has joined #gnucash
16:00:02 *** fDavid has joined #gnucash
16:24:07 *** fDavid has quit IRC
16:32:29 *** fbond has joined #gnucash
16:38:25 *** fbond has joined #gnucash
16:38:29 *** fbond has left #gnucash
16:52:08 <gjanssens> Almost got to go for tonight.
16:52:37 <gjanssens> One final thing.
16:53:14 <gjanssens> I see there's a user "nmz" on the system. I guess its homedir is /home/nmz, but you can doublecheck in /etc/password
16:53:28 <gjanssens> Is there a .namazurc in this user's homedir ?
16:53:41 <gjanssens> If so, can you drop it in pastebin ?
16:54:08 <gjanssens> You can leave your feedback on irc. I'll read the logs tomorrow.
16:54:17 *** gjanssens has quit IRC
16:55:20 <warlord> Yes:
16:55:21 <warlord> nmz:x:485:480::/home/nmz:/sbin/nologin
16:55:29 <warlord> [root@code ~]# ls -la /home/nmz/.nmz*
16:55:29 <warlord> ls: cannot access /home/nmz/.nmz*: No such file or directory
17:05:12 *** warlord is now known as warlord-afk
17:28:50 *** thomas_ has quit IRC
17:47:03 *** fbond has joined #gnucash
18:05:26 *** aindilis has joined #gnucash
18:05:28 *** aindilis` has joined #gnucash
18:46:10 *** mib_wsrkzd has joined #gnucash
19:00:18 <Simon> does anyone have per-country (scotland/england/wales/ni) party vote counts?
19:03:39 <Simon> erm. this is not window 62
19:03:41 * Simon sighs
19:07:36 <jsled> heh
19:19:12 *** fbond has quit IRC
19:41:33 <Simon> I had to go through all the windows from 1 to find where I actually sent that line too
19:42:08 <Simon> I stayed up far too late last night to watch the person I didn't vote for win
19:47:01 *** cortana has quit IRC
20:22:04 *** ErKa has joined #gnucash
20:27:57 *** message144 has joined #gnucash
21:13:51 *** linas_ has quit IRC
21:14:43 *** linas_ has joined #gnucash
21:18:07 *** fbond has joined #gnucash
21:30:43 *** ErKa has quit IRC
21:34:09 *** mib_wsrkzd has quit IRC
22:03:45 *** aindilis has quit IRC
22:03:45 *** aindilis` has quit IRC
22:29:50 *** Linuturk has quit IRC
22:35:13 *** fbond has quit IRC
22:53:25 *** Linuturk has joined #gnucash
23:31:46 *** message144 has quit IRC