2018-04-21 GnuCash IRC logs
00:04:16 *** gour has joined #gnucash
00:24:53 *** Cuare has quit IRC
01:04:49 *** boldstripe has quit IRC
01:18:35 *** Mechtilde has joined #gnucash
01:30:33 *** fell has quit IRC
01:31:46 *** fell has joined #gnucash
01:57:32 *** gncbot sets mode: +o fell
02:17:52 *** gjanssens has joined #gnucash
02:17:52 *** ChanServ sets mode: +o gjanssens
02:18:33 <gjanssens> .
03:08:10 <chris> unit testing strings-encoding is a no-go in linux... can't reproduce. seems windows is the difficulty here.
03:37:23 *** chris has quit IRC
03:40:00 *** pilotauto has quit IRC
03:47:54 <fell> @tell chris Yep, while "modern" Win versions use utf-16 instead of utf-8 they tend to fall back to the pre-ISO ANSI CP1252.
03:47:54 <gncbot> fell: The operation succeeded.
04:05:04 *** harshitaneja has joined #gnucash
04:13:15 *** fekepp has joined #gnucash
04:21:43 *** fabior has joined #gnucash
04:32:15 *** chris has joined #gnucash
04:46:27 *** gour_ has joined #gnucash
04:48:41 *** gour has quit IRC
04:51:30 *** pilotauto has joined #gnucash
05:00:03 *** vxp has joined #gnucash
05:00:07 <vxp> Hi all
05:00:36 <vxp> I've just compiled gnucash 3 from git and all work well on an Ubuntu 17 box
05:00:58 <vxp> but I need to understand how to get logs from getting update quotes from stock market
05:01:03 *** Mechtilde has quit IRC
05:01:05 <vxp> I just get a generic message
05:01:33 <vxp> "There was an unknow error while retrieving the price quotes"
05:01:48 <vxp> I need help to investigate what is going on
05:02:29 <vxp> ALready did "sudo gnc-fq-update" and it finishes without errors
05:07:28 <vxp> Ok. Just found out how to edit the source for the quotes. Managed to change to Yahoo json and it is working
05:08:04 <vxp> Is there a way to change all sources of the securities in one shot, or should I do manually to each one of them?
05:35:08 *** vxp has quit IRC
05:35:36 *** vxp has joined #gnucash
05:58:06 *** oozer has joined #gnucash
07:00:16 *** fabior has quit IRC
07:00:52 *** fabior has joined #gnucash
07:02:20 *** fabior has quit IRC
07:25:14 *** pilotauto has quit IRC
07:38:56 *** GabrieleV has quit IRC
07:41:05 *** GabrieleV has joined #gnucash
07:44:11 *** Aussie_matt has joined #gnucash
07:51:43 *** boldstripe has joined #gnucash
08:04:53 *** oozer has quit IRC
08:17:22 *** Jimraehl1 has joined #gnucash
08:20:44 *** Jimraehl1 has left #gnucash
08:27:59 *** gour_ is now known as gour
09:06:01 <fell> Hm, while working on htdocs, git pull: Already up-to-date. But git push: To git@code.gnucash.org:gnucash-htdocs! [rejected] master -> master (fetch first)
09:10:18 <fell> There was a commit on code, but not on github
09:12:52 *** Moondhum has joined #gnucash
09:14:09 *** Moondhum has quit IRC
09:54:58 *** fabior has joined #gnucash
10:10:02 *** ArtGravity has joined #gnucash
10:11:37 *** boldstripe has quit IRC
10:12:36 *** boldstripe has joined #gnucash
11:22:38 *** fabior has quit IRC
11:43:40 <gjanssens> fell: That sometimes happens if the link between code and github fails
11:43:56 <gjanssens> Pushing a followup commit usually fixes it.
11:44:41 <gjanssens> (something else you still want to alter, or a dummy change like a whitespace change in a comment somewhere)
11:48:20 *** Aussie_matt has quit IRC
12:12:42 *** oozer has joined #gnucash
12:29:39 *** Agfarmer18 has joined #gnucash
12:43:22 <jralls> fell: And in order to fix your local so that you can actually do the push you have to pull from code.
12:48:39 <jralls> gjanssens: I think I understand where the weird long path like C:\Users\RENROM~1\AppData\Local\Temp\René Romijn\GnuCash\expressions-2.0 is coming from. g_get_tmpdir() DOSifies the path and some matching code somewhere doesn't recognize that it's in the user's direcotry.
12:50:22 *** boldstripe has quit IRC
12:50:53 *** boldstripe has joined #gnucash
12:52:09 <gjanssens> jralls: ok. You do understand by now the tempdir thing is a last resort attempt, right ?
12:52:14 <gjanssens> Which of course should also work
12:52:36 <jralls> vxp: If you're brave you can open your data file in a text editor (make sure to save back plain text and unix line endings) and do a global search-and-replace). Work on a copy.
12:53:39 <gjanssens> jralls: and that's also why I suggested to replace g_get_tmp_dir with the variant provided by boost::filesystem.
12:53:46 <gjanssens> I would expect it to be more compatible.
12:54:16 <gjanssens> And all that only relates to the paths managed by our filepath-utils code.
12:54:39 <gjanssens> I don't think it's used to determine the trace file code.
12:55:07 <gjanssens> Perhaps we should at some point.
12:55:20 <jralls> Hmm, yes. The clean way would be to provide gnc_get_tmp_dir so that we can keep bfs encapsulated in gnc-filepath-utils.
12:56:26 <gjanssens> Right.
12:56:32 <jralls> It *is* used to determine where the trace file goes, and it blows up with a message box when you specify a path with non-ascii characters to logto.
12:56:57 <gjanssens> g_get_tmp_dir is you mean ?
12:57:36 <jralls> Yes.
12:58:11 <jralls> Also apparently an "expressions-2.0", something to do with the expression parser. Haven't started on that one yet.
12:59:25 <gjanssens> The --logto path is passed to qof_log_init_filename_special without any treatment.
12:59:38 <jralls> I was a bit premature declaring victory yesterday. My fix got the utf8 into bfs correctly but bfs::path.string() is still returning latin1.
12:59:48 <gjanssens> We probably have to do a locale to utf-8 or the other way around on that
13:00:13 <gjanssens> expressions-2.0 comes from gnucash-bin.c as well.
13:00:57 <gjanssens> line 64.
13:01:10 <gjanssens> It uses gnc_build_userdata_path so it folds back to filepath-utils
13:01:49 <jralls> Locales to utf-8 don't work very well in Windows. The utf8 codepage (60001?) is broken in many locales. There's another way, with bfs::path.imbue(). I'm working that angle ATM.
13:02:05 <gjanssens> Ok
13:03:02 <gjanssens> If expressions-2.0 is searched for in tempdir, that means filepath-utils initialization has failed.
13:03:45 <gjanssens> It should be searched in GNC_DATA_HOME
13:04:30 <gjanssens> That normally points at %APPDATA%\GnuCash on Windows
13:05:27 <gjanssens> And it falls back to %TMPDIR%\GnuCash if that fails
13:05:49 <gjanssens> I'm not sure how that second username gets in that path.
13:06:22 <jralls> OK, imbuing worked and the state file is getting written correctly now.
13:06:30 <gjanssens> So for some reason %APPDATA%\GnuCash is still not recognized as a valid GNC_DATA_HOME
13:06:51 <gjanssens> Or is that also due to the missing bfs::path.imbue() ?
13:11:07 <gjanssens> Oh and I'm wrong about expressions-2.0
13:11:24 <gjanssens> It's set in gnc-exp-parser.c
13:11:42 <jralls> There are a couple of possiblities. One is that reading %APPDATA% must be done with care. It's in the environment coded in UTF16 and naively narrowing it will make a latin-1 string in the case of René but will fail spectacularly for fa
13:12:05 <gjanssens> Which is initialized when loading gncmod-app-utils.
13:12:43 <jralls> Sigh. For 惹。
13:13:21 <gjanssens> Well we're really using SHGetSpecialFolderLocation not the environment, but that's probably only a detail
13:13:28 <gjanssens> I have to leave now...
13:13:31 <gjanssens> Good luck further
13:16:00 <jralls> The other possibility is that it's getting the DOSified version, e.g. RENROM~1 and the equivalence test fails. It looks like that's done by string-comparing the paths.
13:16:28 *** fabior has joined #gnucash
13:16:50 <jralls> OK, bye.
13:21:39 *** kus has joined #gnucash
13:27:10 *** kus has quit IRC
13:43:19 <jralls> gjanssens: Ah, that's actually in win32_get_userdata_home and I've already fixed it.
13:46:49 *** fabior has quit IRC
14:01:01 *** Agfarmer18 has quit IRC
14:04:17 *** frakturfreak has joined #gnucash
14:09:20 <vxp> jralls: I tried that but the file seemed garbage (encrypted?).
14:09:36 <vxp> jralls: In the end I did one by one.
14:09:59 <vxp> jralls: thanks the same. :-)
14:10:06 <jralls> vxp: It's gzipped.
14:10:30 <vxp> jralls: Oh Man!!!!! How could I have missed it!
14:11:38 <vxp> jralls: Now I am trying to compile latest version from git on debian, but facing some dependencies issues.
14:11:57 <jralls> Which debian?
14:13:21 <vxp> Stretch
14:15:35 <vxp> I had to add backports to get the right libgwengui-gtk3-dev package
14:16:00 <vxp> Now it seems it doesn't like my Boost version
14:17:14 <vxp> jralls: this morning I managed to compile in Ubuntu without much effort. On Debian it is taking me longer than I wanted to.
14:18:01 <vxp> jralls: Is there any reason why Debian and Ubuntu maintainers had not released version 3 on their repos that you're aware?
14:18:41 *** frakturfreak has quit IRC
14:19:52 *** kus has joined #gnucash
14:20:06 <jralls> The Debian packager has it working in sid. I think the rules for debian preclude it going into the official stretch. Ubuntu went to final release for 18.04 just as we released 3.0. Since it's a -.0 and 18.04 is LTS I'm glad they didn't put it in.
14:20:13 <jralls> Have you checked getdeb?
14:21:53 <vxp> jralls: I'll check the sid package
14:22:21 <vxp> jralls: I am running at work Ubuntu 17.10 - and with some effort compiled it ok
14:23:06 <vxp> jralls: I always had small problems with getdeb packages - so normally I avoid
14:23:08 <vxp> i
14:23:09 <vxp> t
14:23:17 <vxp> jralls: thanks again
14:23:24 <jralls> OK. You're welcome.
14:33:41 <vxp> jralls: Just for info the gc3 is available only on experimental - noe even sif - for debian. Did not manage to make it work with a simple apt-get - now I will try to download the deb and check deps.
14:33:53 *** oozer has quit IRC
14:34:53 <jralls> vxp: OK, good luck. It may not link on stretch because the lib versions are all higher on sid.
14:35:45 <vxp> jralls: Ok. Worst case scenario I will buy a Mac. :-)
14:36:06 <jralls> Sounds more like a best-case scenario to me! ;-)
14:38:08 <vxp> jralls: That's not what my wallet tells me! But it's a great machine!
14:38:56 <jralls> Get a used one. Anything from the last 5-6 years will still run the latest MacOS.
14:41:18 <boldstripe> CVS importer on v3.0.1 in Debian Buster failed to import an (AMEX) CVS with a '\' (backslash) character in it. OK when character deleted. Should I report that as a bug?
14:41:37 <boldstripe> \ was in the Description column
14:41:38 <vxp> jralls: good advice
14:44:54 <jralls> boldstripe: According to https://www.csvreader.com/csv_format.php '\' is an escape character, if you want one in the file you need to make it "\\".
14:45:38 <jralls> boldstripe: However, the rules are different on Windows. Is that what you're using?
14:51:08 <fell> vxp: Your data file might be compressed. So you could either save uncompressed or use zcat.
14:52:07 <boldstripe> The '\' was put in my file by Amex. I was happy to delete it. I had some difficulty discovering it as the cause and identifying it. I am on Debian Testing (Buster).
15:02:25 <jralls> boldstripe: What was the whole field with the '\' before you changed it?
15:06:35 <jralls> GTG/BBL
15:06:43 *** jralls is now known as jralls_afk
15:43:25 *** ncv has joined #gnucash
15:47:10 *** boldstripe has quit IRC
15:47:42 *** boldstripe has joined #gnucash
15:50:39 <boldstripe> jralls: the entire description field as I found it in the CSV is (in my single quotations): 'WINNERS GARAGE WINNERS GARAGE\ - WOODSIDE NY'
15:51:37 <boldstripe> It was the only one in three years of Amex CSVs. I don't recall ever having my cursor in that field, so I think it unlikely I accidentally typed it myself.
15:51:51 <boldstripe> (the only backslash)
15:52:47 <boldstripe> LibreOffice Calc (on Linux) and Excel (Mac) opened it without a problem.
15:54:01 <boldstripe> The error message appeared when I chose it in the CSV Importer panel. A pop up window said it could not parse the file.
15:55:00 *** ncv_ has joined #gnucash
15:55:43 *** oozer has joined #gnucash
15:56:14 *** ncv has quit IRC
15:57:58 *** Mechtilde has joined #gnucash
16:01:49 *** kus has quit IRC
16:14:40 *** lastm has joined #gnucash
16:19:21 *** lastm has quit IRC
16:26:16 *** frakturfreak has joined #gnucash
16:26:40 *** gour has quit IRC
16:27:49 *** Mechtilde has quit IRC
16:37:54 *** jralls_afk is now known as jralls
16:41:50 <jralls> boldstripe: OK. Not surprised about Excel, they use a different escape mechanism. LibreOffice probably does too for compatibility. We could do the same, maybe with a preference. We'll see what gjanssens thinks when he returns.
17:08:46 *** User_ has joined #gnucash
17:17:27 *** User_ has quit IRC
17:20:55 <boldstripe> OK, thanks.
17:27:46 *** harshitaneja has quit IRC
17:31:12 *** marusich has joined #gnucash
17:32:15 *** boldstripe has quit IRC
17:32:48 *** boldstripe has joined #gnucash
17:37:01 *** marusich has quit IRC
18:11:30 *** ncv__ has joined #gnucash
18:12:45 *** ncv_ has quit IRC
18:27:18 <fell> I fixed all validator.w3.org issues on www.gnucash.org with exception of the google+ button, which still throws 7 errors.
18:29:21 <jralls> fell: Nice. Is there a new version of the G+ button out there? OTOH, does the G+ account get any traffic?
18:30:17 *** Agfarmer18 has joined #gnucash
18:30:43 <fell> It was written for html5, but we use still XHTML + RDFa
18:35:05 <fell> jralls: I see the 3.0 and 2.6.20 announcement, but no 2.6.21 announcement there.
18:36:42 <fell> 1731 Follower
18:37:56 <jralls> fell: https://github.com/Gnucash/gnucash-htdocs/blob/master/news/180410-2.6.21.news?
18:38:39 <fell> but not on https://plus.google.com/u/0/+gnucash
18:39:23 <jralls> Oh, on G+. I have no idea how that's supposed to get updated. cmarchi set it up and AFAIK never shared.
18:40:39 <fell> Ngewi Fet seems to have some access, too.
18:42:12 <jralls> Looks like somebody's tending it, there are responses from "GnuCash" to user questions.
18:44:39 <fell> About shows only gnucash-user@gnucash.org as contact.
18:46:00 <fell> and source code links to http://svn.gnucash.org/trac/browser/
18:47:31 <jralls> Oh, that's no good. I think warlord finally took that down when he set up the new server last year. It 404'd on me.
18:55:54 <fell> I reported the dead link in feedback and will now wait for an answer.
18:57:44 *** ncv__ has quit IRC
19:14:50 *** oozer has quit IRC
20:01:55 *** oozer has joined #gnucash
20:04:19 <warlord> Sorry, jralls, fell, I'm heading to the airport now so can't look at anything. Is there an email thread about this? If not, please start one? Gotta run.
20:04:22 *** warlord has quit IRC
20:05:05 <jralls> Wish he'd said what he's referring to.
20:24:49 *** shakes808 has joined #gnucash
20:25:43 <shakes808> Hello everyone. I have been looking on how to set up a loan to my small business. Can anyone please point me in the right direction?
20:30:49 *** shakes808 has quit IRC
20:45:20 *** shakes808 has joined #gnucash
20:45:50 *** gjanssens has quit IRC
20:46:38 <shakes808> Hello everyone, I am trying to figure out how to create a loan for my small business. Can anyone point me it the right direction? I haven't been able to find anything. Unless it is the same as a personal one.
20:58:41 <shakes808> When I make an account under Liabilities > loan > loan-1, I put the loan amount in the Increase column and then it immediately creates the same thing but zeroes out the amount
21:00:29 *** frakturfreak has quit IRC
21:09:57 <luc14n0> shakes808: Have you found the guide? http://www.gnucash.org/viewdoc.phtml?doc=guide
21:11:56 <shakes808> luc14n0: Yes, I did find that and I think I just figured out my missing step. When creating the loan, I needed to go into the "Opening Balance" tab to finish the setup. I was making the account and then going into the account and trying to add it after the fact.
21:12:27 <shakes808> id10t error ;) I think I am good on this now.
21:12:44 <shakes808> Thank you for the link though
21:13:37 <luc14n0> There you go! No problem.
22:03:31 *** oozer has quit IRC
22:34:08 *** Agfarmer18 has quit IRC
22:54:16 *** boldstripe has quit IRC
23:06:26 *** wget has quit IRC
23:09:46 *** wget has joined #gnucash
23:13:27 *** boldstripe has joined #gnucash