2018-01-05 GnuCash IRC logs
00:10:28 *** Aussie_matt has joined #gnucash
00:13:12 *** dougransom has quit IRC
00:58:10 *** Aussie_matt has quit IRC
01:18:37 *** storyjesse has quit IRC
01:21:19 *** Mechtilde has joined #gnucash
01:22:19 *** fell has quit IRC
01:23:31 *** fell has joined #gnucash
01:47:04 *** frakturfreak has quit IRC
02:01:51 *** frakturfreak has joined #gnucash
02:11:55 *** Mechtilde has quit IRC
03:31:36 *** Aussie_matt has joined #gnucash
03:50:04 *** fekepp has joined #gnucash
03:55:21 *** Aussie_matt has quit IRC
03:58:39 *** hoijui has joined #gnucash
04:10:34 *** hoijui has quit IRC
04:13:11 *** fabior has joined #gnucash
04:35:02 *** Cuare has quit IRC
04:56:11 *** Cuare has joined #gnucash
05:11:02 *** fekepp has quit IRC
05:14:09 *** nbk has joined #gnucash
05:14:29 *** nbk has left #gnucash
05:14:45 *** flips has quit IRC
05:28:40 *** flips has joined #gnucash
05:46:58 *** fabior has quit IRC
05:53:45 *** storyjesse has joined #gnucash
06:11:16 *** Almon has joined #gnucash
06:34:38 *** bastianilso has joined #gnucash
06:37:43 *** bastianilso has quit IRC
06:38:02 <chris> jralls: yes!! i managed to get guile to parse html output for easier unit testing!
06:43:08 <chris> no more regex
06:46:38 *** User_ has joined #gnucash
06:47:11 <chris> https://github.com/christopherlam/gnucash/commit/85ff07ffb3a036a8a278d4493100803da34ab394
06:52:22 *** Jimraehl1 has joined #gnucash
06:52:45 *** Jimraehl1 has quit IRC
06:57:04 *** Aussie_matt has joined #gnucash
06:57:09 <chris> jralls - new report unit test ouptut: http://sprunge.us/iKDV
06:57:27 <chris> LVL +1
06:58:25 *** Aussie_matt has quit IRC
07:04:39 *** User_ has quit IRC
07:19:53 *** oozer has joined #gnucash
07:58:52 *** kus has joined #gnucash
08:22:28 *** weasel has quit IRC
08:24:53 *** kus has quit IRC
08:34:42 *** weasel has joined #gnucash
08:43:49 *** Robert has joined #gnucash
08:45:12 *** Robert has left #gnucash
09:08:49 *** fekepp has joined #gnucash
09:11:22 *** fekepp has quit IRC
09:13:02 *** Mechtilde has joined #gnucash
09:16:42 *** fekepp has joined #gnucash
09:21:10 *** fabior has joined #gnucash
09:32:45 *** Aussie_matt has joined #gnucash
09:36:35 *** storyjesse has quit IRC
09:55:49 *** Aussie_matt has quit IRC
09:57:53 *** kus has joined #gnucash
10:02:18 *** kael has joined #gnucash
10:58:47 <lmat> chris is levelling up; great! Is the HTML parsing this: (xml->sxml table-string) ? I wouldn't call this HTML parsing: (substring result table-start (+ 8 table-end))
11:05:29 *** fabior has quit IRC
11:17:28 *** Mechtilde has quit IRC
11:24:08 *** gncbot sets mode: +o fell
11:30:56 *** ArtGravity has joined #gnucash
11:31:04 *** kael has quit IRC
11:33:48 *** kus has quit IRC
11:36:44 *** Almon has quit IRC
11:37:23 *** Almon has joined #gnucash
11:45:08 *** redarrow has quit IRC
12:02:13 <jralls> Cuare: Are you the OP on 661093?
12:07:27 <jralls> chris: The sprunge URI returns an "internal server error". It also looks kind of short,
12:08:21 <jralls> chris: But it looks like the "tests" are hard-wired to pass. Are they intended just as a demo?
12:12:50 <jralls> chris: An aside: All of the scheme tests have a problem: They fail silently. All test conditions should do something like (let ((result (test foo)) (or (= result expected) (begin (format #t "Test foo failed, got ~a expected ~a~%
12:13:22 <jralls> " result expected) #f)))
12:17:51 *** Almon has quit IRC
12:18:19 *** kus has joined #gnucash
12:20:14 <jralls> chris: OTOH it's not really a unit test if it runs the whole report out to html and then parses the html. A unit test tests each function in the code for a particular execution path. Complicated functions with lots of execution paths require lots of unit tests, a good reason to decompose such functions into snaller easier-to-test functions with a controller function taking care of the branching.
12:21:00 <jralls> chris: Mind, functional tests that test the whole report are also useful.
12:26:14 <Cuare> yes im here again
12:28:22 <jralls> Cuare: Sorry to learn that you've downgraded to Windows.
12:28:23 <Cuare> op no just have the same problem as he did on: Bugz 661093
12:29:29 <jralls> Cuare: Do you have a stack trace to be sure that you really have the same bug? Why did you copy-and-paste the bug description instead of describing your own experience?
12:30:59 <Cuare> Well its pretty much identicle to what was going on with that post.
12:32:57 <Cuare> Thinking im just giong to roll back to 2.6.17 had no prblems with it.
12:34:23 <jralls> Cuare: The fact that you dind't have that problem in 2.6.17 strongly suggests that it's *not* the same bug, which was reported in 2.4.something and which has to do with a systemic problem in GnuCash's engine architecture.
12:36:36 <jralls> So please open a new bug describing your crash, explaining that it's a regression from 2.6.17 (you might check 2.6.18 to pin down when it was introduced). At least attach a trace file (https://wiki.gnucash.org/wiki/Tracefile) from a crashed session.
12:38:58 <Cuare> i will open a new post for is and povide a stack trace for it. I will be doing that soon. thanks for the help @jralls
12:39:26 *** kus has quit IRC
12:39:35 *** kus has joined #gnucash
12:41:06 *** Mechtilde has joined #gnucash
12:44:13 *** fabior has joined #gnucash
12:50:11 *** tuxd00d has quit IRC
13:08:33 *** Jeff has joined #gnucash
13:11:06 *** kael has joined #gnucash
13:12:16 <Jeff> I'm trying to direct connect to Wells Fargo using OFX. So far my efforts are a dismal failure. Has someone here gotten this to work? How 'bout tossing me a bone? Please?
13:13:10 <warlord> Jeff: try asking on the gnucash-user mailing list?
13:13:56 <Jeff> not yet, should I?
13:16:20 <warlord> You'll get a better response than here.
13:16:21 <jralls> Jeff: According to https://www.wellsfargo.com/online-banking/software/web-connect Wells Fargo supports only OFX Web Connect. GnuCash requires OFX Direct Connect for direct access. You'll have use download and import instead.
13:16:33 *** fekepp has quit IRC
13:17:42 <Jeff> thanks for that. i'll check out that link.
13:23:37 *** fekepp has joined #gnucash
13:30:34 *** Jeff has quit IRC
13:35:32 *** kael has quit IRC
13:51:43 *** fabior has quit IRC
14:02:15 *** Cuare has quit IRC
14:18:25 <chris> jralls- yes ofcourse they're designed as functional tests for now hardwired to pass; was aiming to test various options and observe output. the pure unit tests will be rather difficult to write with this blob and with other old reports... I supose test-cash-flow is a good example to follow
14:19:41 <chris> lmat no it;s not html parsing, it's xml decoding into internal scheme recursive lists; definitely much better than regex as used by test-generic-net-barchart for example
14:23:11 <chris> https://gist.githubusercontent.com/christopherlam/d18ea7e82ad87e9d57ba66527d6d256b/raw/88e3e686b26b2b51083c48d8d4e490f07c7bc9de/xml-%253Esxml
14:29:39 *** Mechtilde has quit IRC
14:51:47 <jralls> chris: Did you have a look at Yawar's testing framework? https://github.com/yawaramin/ggspec
14:54:15 <chris> will check later... for today I'm meeting rgmerk
14:56:29 <jralls> chris. OK.
14:57:50 <jralls> chris: WRT the critical error from gnc_get_default_report_font_family, the only thing there that would make a critical error is that there are no top-level windows. You'd of course get that when running tests with no GUI.
15:01:23 <jralls> That doesn't look like the best way to get the default font anyway, but it would be easy enough to have it return "Arial" instead of NULL when it fails.
15:01:29 <chris> there's still a non-logged default font issue - see the header for every report output https://gist.github.com/christopherlam/1a8fbba0f49a8d3e4d274eb5d69d8f8a - font-family is reading garbage
15:03:08 <jralls> It's quite possible that the styling has changed for Gtk3. Like I said, not the best way to get the default. /me looking at Gtk3 styling docs.
15:03:08 <chris> anyway this is minor
15:11:31 <jralls> Ah, the problem is that the char* returned by pango_font_description_get_font_family is a pointer into font_desc which gets freed immediately after getting it. It needs to be strduped and then freed. Oh, even worse, the swigfile declares it %newobject, so it should crash.
15:12:40 <jralls> Easy to fix, though.
15:13:05 <chris> thank you wizard :)
15:15:48 <jralls> Ah, it doesn't crash because the invalid ptr gets strduped at the end. Just need to move the strdup to the retrieval and change the no-topwindow condition to return the default instead of NULL. No point in throwing an error there, either, it just means the report was run with no GUI.
15:17:59 <chris> yeah i noticed that I could isolate the reports to run without GUI -- it would be an interesting tool to allow another executable to run reports from the command line and use the official tool
15:18:09 <chris> use the offical API
15:18:12 <chris> *
15:19:38 <chris> well, I wasnot exactly running the reports without GUI; rather running the scm tests which can access API
15:21:33 *** CDB-Away_ has joined #gnucash
15:21:42 <jralls> Except that most of the tests do run the reports. It might be nice to have a command-line report runner that doesn't start the GUI: `gnucash --report=balance_sheet -o balance_sheet.html`
15:22:55 *** CDB-Man has quit IRC
15:23:07 *** CDB-Man_ has quit IRC
15:23:31 *** User_ has joined #gnucash
15:23:52 *** CDB-Man has joined #gnucash
15:24:43 <chris> exactly, they've been asking for cmdline access to data for years
15:26:41 *** User_ has quit IRC
15:27:16 <chris> is #250 still confusing?
15:33:01 <jralls> chris: No. For some reason the date on the commit didn't change even though you force-pushed it with the DecDate change so I thought that I'd mis-remembered the content instead of that you'd found and fixed the problem.
15:33:54 <jralls> chris: I just pushed a fix to the default-font-family issues.
15:34:08 <chris> nice!
16:01:34 *** frakturfreak has quit IRC
17:35:18 *** frakturfreak has joined #gnucash
17:44:52 *** CDB-Man is now known as CDB-Man_
17:53:40 *** Almon has joined #gnucash
18:10:03 *** meb has joined #gnucash
18:11:28 *** kael has joined #gnucash
18:13:08 *** meb has quit IRC
18:20:04 *** kael has quit IRC
18:25:26 *** shakes808 has joined #gnucash
19:07:34 *** xmaka has quit IRC
19:08:26 *** xmaka has joined #gnucash
19:14:25 *** frakturfreak has quit IRC
19:36:48 *** xmaka has quit IRC
19:40:03 *** xmaka has joined #gnucash
19:43:11 *** xmaka has quit IRC
19:59:02 *** chris has quit IRC
20:09:46 *** oozer has quit IRC
20:18:33 *** meb has joined #gnucash
20:22:26 *** Almon has quit IRC
20:38:18 *** Cuare has joined #gnucash
20:39:15 *** gnomey has quit IRC
20:43:54 *** gnomey has joined #gnucash
20:44:42 *** xmaka has joined #gnucash
20:51:32 *** CDB-Man_ has quit IRC
20:52:02 *** CDB-Man has joined #gnucash
20:54:33 *** CDB-Man has quit IRC
20:54:56 *** CDB-Man has joined #gnucash
20:59:28 *** Cuare has quit IRC
21:12:10 *** shakes808 has quit IRC
21:42:56 *** ArtGravity has quit IRC
21:55:01 *** marusich has quit IRC
22:00:37 *** marusich has joined #gnucash
22:18:51 *** Cuare has joined #gnucash
22:28:24 *** kael has joined #gnucash
22:39:28 *** kael has quit IRC
22:42:16 *** kael has joined #gnucash
22:47:29 *** kael has quit IRC
23:10:57 *** bertbob has quit IRC
23:13:48 *** kus has quit IRC
23:20:44 *** bertbob has joined #gnucash
23:26:00 *** marusich has quit IRC
23:26:18 *** marusich has joined #gnucash
23:34:05 *** marusich has quit IRC
23:34:23 *** marusich has joined #gnucash
23:40:23 *** bertbob has quit IRC
23:42:02 *** kael has joined #gnucash
23:45:07 *** kael has quit IRC
23:58:14 *** bertbob has joined #gnucash