2018-05-18 GnuCash IRC logs
00:49:22 *** Skeletron has joined #gnucash
01:00:15 *** Skeletron has quit IRC
01:22:58 *** Mechtilde has joined #gnucash
01:25:46 *** fell has quit IRC
01:26:56 *** fell has joined #gnucash
01:27:26 *** gncbot sets mode: +o fell
01:55:55 *** Cuare has joined #gnucash
02:02:21 *** Mechtilde has quit IRC
02:17:24 *** gour has joined #gnucash
02:19:04 *** storyjesse has quit IRC
02:45:21 *** Mechtilde has joined #gnucash
02:56:18 *** gjanssens has joined #gnucash
02:56:18 *** ChanServ sets mode: +o gjanssens
02:57:01 *** fabior has joined #gnucash
02:57:52 *** Mechtilde has quit IRC
03:34:08 *** Mechtilde has joined #gnucash
03:45:32 *** fekepp has joined #gnucash
03:51:09 *** fekepp has quit IRC
03:51:17 *** GabrieleV has quit IRC
04:09:09 *** harshitaneja has joined #gnucash
04:24:02 <gjanssens> .
04:28:17 *** fekepp has joined #gnucash
04:29:03 *** fekepp has joined #gnucash
04:50:06 *** fabior has quit IRC
05:04:00 *** fabior has joined #gnucash
05:39:09 *** fabior has quit IRC
05:46:58 *** jotrago has quit IRC
05:47:21 *** jotrago has joined #gnucash
05:53:25 *** fekepp has quit IRC
05:56:04 *** pilotauto has quit IRC
05:57:21 *** fekepp has joined #gnucash
06:00:21 *** fekepp has quit IRC
06:02:29 *** fekepp has joined #gnucash
06:04:29 *** fabior has joined #gnucash
06:38:11 *** fabior has quit IRC
06:52:49 *** Jimraehl1 has joined #gnucash
06:54:18 *** Jimraehl1 has left #gnucash
07:07:44 *** Mechtilde has quit IRC
07:25:27 *** oozer has joined #gnucash
07:48:30 *** boldstripe has joined #gnucash
07:54:07 *** GabrieleV has joined #gnucash
07:55:39 *** Mechtilde has joined #gnucash
08:10:16 *** chris has joined #gnucash
08:12:58 <warlord> Finally pulled down all the bugs. But it failed pulling the user list.
08:13:12 <warlord> Retrying to see if it was a one-off or if my request is too bug.
08:13:13 <warlord> big
08:30:54 *** Agfarmer18 has joined #gnucash
08:32:49 *** fekepp has quit IRC
08:33:33 *** fekepp has joined #gnucash
08:34:56 *** Agfarmer18 has quit IRC
08:44:57 *** Agfarmer18 has joined #gnucash
08:49:05 <warlord> Looks like I need to chunk up my data. :(
08:49:46 <warlord> The user request is too long. And I can't seem to get POST data to work, so I'm limited by the GET URL limit. :(
09:03:40 <chris> gour I've cleaned up maint-TR-average. enjoy :)
09:03:44 *** kael has joined #gnucash
09:16:43 <gour> chris: is it already pushed? is it going to be applied upstream?
09:32:45 *** chris has quit IRC
10:13:41 *** storyjesse has joined #gnucash
10:22:52 *** Agfarmer18 has quit IRC
10:30:00 *** fekepp has quit IRC
10:37:30 <warlord> We have 5047 users attached to our bugs.
10:37:44 <warlord> 8598 bug
10:37:45 <warlord> s
10:39:44 <warlord> Trying to find the sweet spot for the # of users I can request at a time.
10:40:01 <warlord> The number is somewhere between 100 and 1000. 100 definitely works. 1000 fails.
10:40:28 *** Mechtilde has quit IRC
10:46:45 <warlord> OKay, 500 didn't work. Let's drop back down to 100
10:48:47 *** storyjesse has quit IRC
10:50:30 *** ArtGravity has joined #gnucash
11:04:15 *** chris has joined #gnucash
11:08:59 <chris> (just putting it out there for gjanssens if he can spare some time to debug test-invoice.scm for me on https://github.com/christopherlam/gnucash/tree/scheme-progress-business-tests -- this test tries to put numbers to create an invoice from scratch...... qty, unit price, taxtable used, discount%, etc. - see the output at https://imgur.com/IOUpmzy - this shows all calculations are ignoring discount% and TaxTable amount)
11:09:41 <chris> I also found a bug in subtotal-grid - this cannot handle subtotals with multiple commodities... I'll work on it when I can.
11:10:11 <chris> otherwise night all, of to sleep now
11:10:21 <chris> and good luck warlord
11:11:11 *** chris has quit IRC
11:24:04 *** fekepp has joined #gnucash
11:34:00 *** O01eg has quit IRC
11:46:35 *** harshitaneja has quit IRC
11:47:06 *** harshitaneja has joined #gnucash
12:15:21 <gjanssens> @tell chris I quickly looked at test-invoice.scm yesterday and I see your issue. I haven't had time to figure it out yet...
12:15:21 <gncbot> gjanssens: The operation succeeded.
12:26:35 *** linuturk has quit IRC
12:35:58 *** User has joined #gnucash
12:37:51 <warlord> w00t. Fixed the user download issue. Apparently there's something that has a null user, so trying to download a null-string user throws an error. Fixed that, so now I've got everything. Yay.
12:39:28 <jralls> Yay indeed.
12:40:59 <warlord> Well, "everything". Still don't have access to watchers or user permissions; still need to manually pull the list of gnucash developers.
12:42:20 <warlord> Still need to determine how to massage the data..
12:42:24 <jralls> I think that we probably want to manually set up the GnuCash developers. There's no point in creating accounts for the long-departed.
12:43:59 <jralls> I imagine users will have to set up their own watch lists once they've created accounts.
12:44:04 <warlord> Perhaps. But it's really not that log a list. The more important issue is that we will need to manually set up the additional list of people who can edit and/or confirm bugs
12:44:25 <warlord> Yeah, I think they will have to do that.
12:44:47 <warlord> We also need to think about what we want to do about the different status/resolved states.
12:44:50 <jralls> Do we have anyone who isn't a current developer who's editing and/or confirming bugs?
12:45:01 <warlord> I don't know.
12:45:12 <warlord> You can see the list of users at https://bugzilla.gnome.org/page.cgi?id=browse.html&product=GnuCash
12:45:24 <jralls> Aside from André Klapper, that is, and he will presumably not be joining us. ;-)
12:45:34 <warlord> LOL
12:45:52 <warlord> Well, he will have an account..
12:46:08 <warlord> Whether or not he sets it up is up to him, of course.
12:46:28 <warlord> But he's one of the 5047 users
12:54:04 *** User has quit IRC
12:54:40 *** gour_ has joined #gnucash
12:55:24 <jralls> In the Gnome instance I think anyone who's a developer on any project gets the confirm and edit privs. I'm a developer only on GnuCash and gtk-mac-integration but I sometimes edit bugs on Pango, Gtk, and GLib.
12:56:35 <jralls> And FWIW since Gnome doesn't use "confirmed" I don't think having the confirm priv means anything.
12:56:47 *** gour has quit IRC
13:11:47 *** fabior has joined #gnucash
13:17:18 <warlord> I wonder if bugzilla needs special code to handle NEEDINFO. I don't see anything in the code about it.
13:18:39 <jralls> https://www.bugzilla.org/docs/4.4/en/html/bug_status_workflow.html
13:19:19 <warlord> FWIW, we are running 5.0.3
13:20:50 <warlord> Ahh. I suspect that might need to be set up manually -- I don't know if I can script it. And I don't know if it gets destroyed if I drop the database. Someting to check I guess.
13:20:55 <warlord> Gotta run. BIAB
13:21:23 <jralls> I'm sure it gets destroyed if we drop the database.
13:21:44 <jralls> https://bugzilla.readthedocs.io/en/5.0/administering/workflow.html
13:23:07 *** harshitaneja has quit IRC
13:23:15 <jralls> And yes, I expect there's a lot we'll need to set up manually.
13:55:15 *** fabior has quit IRC
14:01:04 *** boldstripe has quit IRC
14:06:14 *** eagles0513875 has quit IRC
14:13:05 *** gour_ is now known as gour
14:46:05 *** frakturfreak has joined #gnucash
14:47:57 *** harshitaneja has joined #gnucash
14:49:51 <warlord> Perhaps we should think about what we want as our workflow before we import all the data?
14:51:21 *** eagles0513875 has joined #gnucash
14:56:55 *** Caretaker13 has joined #gnucash
15:01:36 <Caretaker13> rant: what is the point of android gnucash if you can't sync with pc gnucash? if all you can really do is "jot down" a few transactions in android gnucash, to later import (hard!) into pc gnucash, why not forget android and just write down transaction on paper, and then hand-enter into pc gnucash?
15:03:19 <jralls> warlord: I don't think we have time. If Gnome is on schedule (and I guess I should pop over there and ask) we've got only two weeks to get up and running.
15:04:14 <jralls> Caretaker13: This is the desktop side. Your rant will be more effective somewhere the Ngewi is listening.
15:04:45 *** User has joined #gnucash
15:04:51 <Caretaker13> is ngewi the android gnucash developer?
15:06:57 <Caretaker13> and where can ngewi be found?
15:07:56 <jralls> Caretaker13: Yes. At the GfA Google Group, the CodingUser github repo, and on UserVoice.
15:08:15 <Caretaker13> okay, thanks.
15:10:36 <Caretaker13> btw, does what i said make sense? may i ass-ume that no one here uses android gnucash at all, only desktop gnucash?
15:11:12 <jralls> Lurkers aside I believe that's correct.
15:11:31 <Caretaker13> okay, got it.
15:12:07 <warlord> Caretaker13: Moreover, apparently you CAN "share" your (XML) datatfile between desktop and android.
15:12:13 <warlord> At least from 2.6. Dont know about 3.x
15:12:51 <warlord> jralls: I think we can take the time to duplicate the gnome work flow, and maybe modify it a bit..
15:14:24 <Caretaker13> warlord: tried that a while back, didn't seem to work very well, but can't remember why. maybe i'll try again.
15:14:45 <jralls> warlord: As long as our workflow includes the status values that Gnome uses I don't think that there's any problem with modifying it as we please. I don't think there's any obstacle to that after we go "live".
15:17:56 <warlord> jralls: Adding new states is always easy. I dont know how hard it is to remove them. We can change NOTGNOME to NOTGNUCASH.. Things like that.
15:18:15 <warlord> I just need to know how to translate as I import -- and make sure the right values are in the DB to start with.
15:18:41 <warlord> FWIW, the bugs (in json form) take up 444MB !!
15:19:22 <jralls> We should absolutely change NOTGNOME to NOTGNUCASH. You don't have to do it as you import; in fact it's probably easier to do it with SQL afterwards.
15:20:24 <warlord> It's honestly very easy to do in my import. I have a translate_bug() function where I can do anything I want, so changing status tags is easy.
15:20:42 *** kael has quit IRC
15:20:57 <warlord> Let's verify their date is still June 1?
15:21:45 <jralls> Actually NOTGNOME is a resolution, not a status. The corresponding status is FIXED and is one of three required ones (see the 5.0 workflow link from 13:21.
15:21:54 <jralls> )
15:22:58 <jralls> I asked csoriano on #gnome. Neither ovitters nor andre are there ATM.
15:28:14 <warlord> Sorry, you're right. I'm conflating resolution and status flags.
15:28:42 <warlord> Is it "FIXED" or just "CLOSED"?
15:31:51 *** Caretaker13 has quit IRC
15:32:01 <jralls> FIXED.
15:35:51 <jralls> Actually, Gnome uses "RESOLVED", but they're using 4.4 which doesn't require FIXED and DUPLICATE. That implies that we can't exactly duplicate the Gnome workflow...
15:40:44 <warlord> hmm
15:47:54 <jralls> The documentation isn't very helpful. I imagine that RESOLVED:FIXED should be transformed to FIXED and RESOLVED:DUPLICATE to DUPLICATE, but https://bugzilla.readthedocs.io/en/5.0/administering/field-values.html implies that resolutions are associated with FIXED. That would lead to FIXED:WONTFIX and FIXED:NOTABUG that seem a bit nonsensical.
15:50:29 <jralls> Maybe that part is configurable, so we can have terminal statuses of FIXED, DUPLICATE, and RESOLVED; in that case only RESOLVED would need a resolution field.
15:52:27 <jralls> I'm also in favor of a CLOSED status that prevents reopening except by a dev or admin to prevent users from reopening "WONTFIX", "NOTABUG", or "INVALID".
15:53:03 *** gour has quit IRC
16:03:52 <warlord> jralls: Any chance you could write all this down, either in the wiki or in an email? I'm kinda heads-down in $DAYJOB right now.
16:05:08 <jralls> warlord: OK. Probably a good time to start a wiki page for all of this, we'll need instructions for users and admins at some point.
16:08:16 <warlord> Indeed.
16:09:16 <warlord> I have all the (current) data, so we're in a good shape to play more. Hopefully I'll have time this weekend to try an initial import of all data -- I suspect I'll need to fix some additional history mappings.
16:09:35 <warlord> BZ is really annoying in that they translate a bunch of fields, but then I need to translate them back again!
16:21:04 *** oozer has quit IRC
16:51:18 *** gjanssens has quit IRC
16:58:16 <jralls> warlord: https://wiki.gnucash.org/wiki/QA/BugzillaAdministration#New_Installation
17:03:08 *** fekepp has quit IRC
17:06:39 <warlord> jralls: Thanks. Made a minor update to the users section.
17:07:02 <jralls> OK, that's what a wiki is for.
17:10:16 <warlord> :)
17:10:39 <warlord> Where did you get that "9-digit id" from?
17:10:43 *** kael has joined #gnucash
17:15:04 <jralls> warlord: https://www.bugzilla.org/docs/2.16/html/dbschema.html
17:15:35 <jralls> A bit obsolete, but probably has only grown.
17:19:58 *** gnomey has quit IRC
17:23:49 <warlord> Ah. So yeah, I can pull down the ID, but I can't save it. And it doesn't matter, because there's no external API that leverages a user's internal ID#. So it's okay if those change.
17:27:27 <jralls> The BZ import code must take care of remapping, then. The other tables use the userid as the foreign key, but I suppose the JSON is returning one of the other fields so that it doesn't depend on the userid.
17:40:16 *** boldstripe has joined #gnucash
17:41:31 *** boldstripe has quit IRC
17:41:31 <warlord> it uses the string email
17:41:52 <warlord> internally the DB uses the ID, but the JSON/Perl APIs use the string
17:42:35 *** boldstripe has joined #gnucash
17:43:15 <jralls> So someone could really mess things up by changing their email on one system but not the other.
17:47:54 <warlord> Huh?
17:48:00 <warlord> I dont understand.
17:50:55 <jralls> You have a couple thousand bugs cached with "jralls at ceridwen dot fremont dot ca dot us" as a commenter. If I go to bugzilla.gnome.org and change my email address to "jralls at ceridwen dot us" then the import will create a new user on the newer bugs.
17:53:17 <jralls> That's not quite right: When you drop the user table and reimport it with the new userid then none of the cached bugs will find a match.
17:54:07 <jralls> If you don't drop the user table then when you pull new bugs those won't find a match.
17:55:30 <warlord> Ah, yes.. if someone goes and changes their email address right now then my cached bug data will be incorrect.
17:55:47 *** Mechtilde has joined #gnucash
17:56:05 <warlord> but I'll see that when it refreshes the users list (which is does every time) -- and I can go and manually refresh those bugs.
17:56:20 <warlord> Hopefully nobody will do that between now and when we do the final import.
17:56:46 <jralls> Seems unlikely that anyone would.
17:57:14 *** gnomey has joined #gnucash
17:57:47 <jralls> csoriano hasn't responded. :-(
18:08:28 *** Mechtilde has quit IRC
18:16:20 *** kael has quit IRC
18:31:17 *** boldstripe_ has joined #gnucash
18:31:20 *** ArtGravity has quit IRC
18:31:35 *** boldstripe has quit IRC
18:31:37 *** boldstripe_ is now known as boldstripe
18:32:41 *** kael has joined #gnucash
18:38:44 *** frakturfreak has quit IRC
19:04:19 *** kael has quit IRC
20:03:41 *** oozer has joined #gnucash
20:15:07 *** bertbob has quit IRC
20:17:39 *** User has quit IRC
20:24:36 *** harshitaneja has quit IRC
20:30:27 *** boldstripe has quit IRC
20:31:38 *** bertbob has joined #gnucash
20:46:25 *** pilotauto has joined #gnucash
20:53:58 *** boldstripe has joined #gnucash
21:38:49 *** oozer has quit IRC
21:43:57 *** frank_laptop_ has joined #gnucash
21:58:29 *** marusich has joined #gnucash
23:10:00 *** marusich has quit IRC
23:15:58 *** wget has quit IRC
23:20:09 *** wget has joined #gnucash