2014-09-11 GnuCash IRC logs
00:28:22 *** webmanoffesto has quit IRC
01:10:41 *** GabrieleV_ has joined #gnucash
01:10:43 *** GabrieleV has quit IRC
01:10:43 *** GabrieleV_ is now known as GabrieleV
01:24:59 *** MechtiIde has joined #gnucash
02:07:45 *** wol has joined #gnucash
02:08:02 *** MechtiIde has quit IRC
02:23:01 *** O01eg has quit IRC
02:46:08 *** floh1111 has joined #gnucash
03:23:36 *** webmanoffesto has joined #gnucash
03:38:13 *** GabrieleV_ has joined #gnucash
03:38:16 *** GabrieleV has quit IRC
03:38:16 *** GabrieleV_ is now known as GabrieleV
04:31:03 *** webmanoffesto has quit IRC
04:37:09 *** webmanoffesto has joined #gnucash
05:04:50 *** gjanssens has joined #gnucash
05:04:50 *** gncbot sets mode: +o gjanssens
05:54:06 *** david has joined #gnucash
06:33:05 *** Jimraehl1 has left #gnucash
06:36:40 *** Jimraehl1 has joined #gnucash
07:42:50 *** wol has quit IRC
07:51:51 *** wol has joined #gnucash
08:05:17 *** floh1111 has quit IRC
08:31:21 *** david has quit IRC
08:38:42 *** benjamin-agaric has quit IRC
08:39:26 *** benjamin-agaric has joined #gnucash
08:51:54 *** floh1111 has joined #gnucash
09:16:52 *** himaxx has joined #gnucash
09:18:13 *** GabrieleV_ has joined #gnucash
09:18:15 *** GabrieleV has quit IRC
09:18:15 *** GabrieleV_ is now known as GabrieleV
09:19:10 *** himaxx has quit IRC
09:21:40 *** Gbarr1 has quit IRC
09:22:12 *** Gbarr has joined #gnucash
09:34:04 *** lmat_ has joined #gnucash
09:53:13 *** GabrieleV_ has joined #gnucash
09:53:17 *** GabrieleV has quit IRC
09:53:17 *** GabrieleV_ is now known as GabrieleV
10:32:38 *** bhardwajs has joined #gnucash
10:37:14 *** GabrieleV_ has joined #gnucash
10:37:23 *** GabrieleV has quit IRC
10:37:23 *** GabrieleV_ is now known as GabrieleV
10:50:45 *** O01eg has joined #gnucash
11:05:24 *** floh1111 has quit IRC
11:18:11 *** bhardwajs has quit IRC
11:22:55 *** wol has quit IRC
11:44:45 *** floh1111 has joined #gnucash
11:52:18 *** wol has joined #gnucash
11:55:58 *** bhardwajs has joined #gnucash
12:03:09 *** MechtiIde has joined #gnucash
12:09:25 *** benjamin-agaric has quit IRC
12:12:29 *** jralls has quit IRC
12:12:51 *** jralls has joined #gnucash
12:12:51 *** gncbot sets mode: +o jralls
12:16:00 *** benjamin-agaric has joined #gnucash
12:20:25 *** floh1111_ has joined #gnucash
12:20:54 *** floh1111 has quit IRC
12:25:36 *** wol has quit IRC
12:26:39 *** fell has joined #gnucash
12:32:18 *** gncbot sets mode: +o fell
12:38:32 *** wol has joined #gnucash
12:51:31 <jralls> lmat: Didn't see your message until today. I'm still slogging away at numeric, just starting on the boost::rational experiment, and I'm planning to do gnc-date->boost::calendar::gregorian after that.
12:59:24 *** bhardwajs1 has joined #gnucash
13:01:16 <jralls> KVP might be feasible, but it might be better to figure out the best way to replace GObject properties first. You might look at http://www.boost.org/doc/libs/release/libs/property_map/ and http://www.boost.org/doc/libs/release/libs/property_tree/ as a way to handle both. Mind, I have no idea if they're suitable, I haven't even looked at the overviews.
13:04:57 <jralls> QofSession is a stand-alone that begs for refactoring, as is QofBackend. The latter defines the interface for the backends as well as the shim between QofInstance and the backends, so it requires some deep thought about how best to implement it.
13:07:43 *** bhardwajs has quit IRC
13:08:10 * warlord would love to see a QofQueriable abtract Interface class...
13:08:51 <warlord> OTOH, it would also be nice if QOF could have some knowledge of Accounts in order to speed up searches when looking in particular accounts.
13:09:20 <jralls> QofCollection (in qofid.(h|cpp)) holds the pointers for all open engine objects in linked lists. It's the major performance problem in GnuCash right now for large data sets.
13:09:37 * warlord isn't sure how to do that generically... Perhaps by making an Account contain a "Collection" and then you search by collection, somehow
13:10:40 <warlord> Prior to QOF, the Query engine knew that when you limited searching to an Account (or set of accounts), it would only search the SplitLists from those specific Accounts
13:11:01 <warlord> That "feature" was lost during the QOFification of the Query engine.
13:11:22 <jralls> warlord: I think we want to move searching into SQL so that we don't have to load the whole dataset into memory at startup. We need to move that way for multiuser.
13:11:54 <warlord> that would be nice, too.
13:12:46 <warlord> It was actually PART of the original plan -- the Qof Query does have a callback to the backend to perform the query..
13:14:55 <jralls> I'd like to just write queries in SQL. I think that would make it much easier for users to write custom reports, since SQL is pretty easy to learn.
13:16:41 <warlord> Is it stable enough across all our different backend solutions?
13:22:58 <jralls> No. The XML backend would have load an in-memory SQL database. There is a query language for XML, XQuery, but we don't want to go there. It's severely arcane.
13:25:21 <jralls> Sorry, "would have to load...". In real practice, SQLite3 would become the default backend and XML would become an import/export format.
13:34:03 *** aqua___ has joined #gnucash
13:36:51 <lmat> Just got back from a self defense class at work :)
13:36:56 <lmat> jralls: Thanks for the responses.
13:37:44 <lmat> jralls: KVP winds up being a boost::variant and it may be straightforward from there. I'll certainly take a look at GObject and backends, too!
13:38:11 <lmat> jralls: Thanks for the pointer that QofCollection is a performance problem. This gives me a lot of research to do in the QOF.
13:38:30 <warlord> jralls: yeah, loading xml into an in-memory sqlite db sounds pretty cool :)
13:39:12 *** bhardwajs1 has quit IRC
13:40:11 <jralls> lmat: re boost::variant: Good idea. I was focussing more on passing values in and out without leaking the existence of KVP.
13:57:34 *** bhardwajs has joined #gnucash
14:00:10 <lmat> hey, just got the response to our inspection response :D
14:00:39 <lmat> It looks good, my wife and her mom and my mom are for it, so it looks like we'll be new home owners.
14:01:34 <warlord> lmat: congrats
14:01:38 <jralls> Congrats!
14:01:51 <lmat> Thanks!
14:03:05 *** floh1111_ has quit IRC
14:13:25 *** wol has quit IRC
14:19:55 *** benjamin-agaric has quit IRC
14:20:50 *** aqua___ has quit IRC
14:28:31 *** bhardwajs has quit IRC
14:29:42 *** bhardwajs has joined #gnucash
14:35:51 *** bhardwajs1 has joined #gnucash
14:36:11 <gjanssens> lmat: yes, congrats!
14:36:25 <lmat> thanks!
14:41:02 *** bhardwajs has quit IRC
14:44:51 *** benjamin-agaric has joined #gnucash
14:50:35 <bhardwajs1> jralls & lmat: have you been able to edit C++ next steps page? I will try to do some work tonight.
14:53:21 <jralls> bhardwajs1: What "C++ next steps" page?
14:53:23 *** fell has quit IRC
14:54:33 *** MechtiIde has quit IRC
14:55:31 <jralls> bhardwajs1: Anyway, look at the log from around a couple of hours ago for some things to look at if you want to take on a module.
14:58:46 <bhardwajs1> jralls: let me take a look.
15:07:30 <lmat> bhardwajs1: Oh, you're a dev? great! :D
15:10:08 *** aqua___ has joined #gnucash
15:10:18 *** david has joined #gnucash
15:55:08 *** Gbarr has quit IRC
16:04:08 <lmat> sheesh... Looks like the KvpValueBinaryData's size parameter, datasize, is a 'int' (not unsigned or anything!). and the kvp_value_new_binary given a guint64.
16:04:11 <lmat> ugh
16:04:48 <lmat> I like guint64 more, but does that have a negative effect on either struct _KvpValue or struct KvpValueBinaryData?
16:23:37 *** bhardwajs has joined #gnucash
16:24:42 *** jralls has quit IRC
16:25:54 *** jralls has joined #gnucash
16:25:55 *** gncbot sets mode: +o jralls
16:26:25 *** kenyon has joined #gnucash
16:27:31 *** kenyon has left #gnucash
16:32:32 *** bhardwajs1 has quit IRC
16:38:06 *** aqua___ has quit IRC
16:57:19 *** GabrieleV_ has joined #gnucash
16:57:24 *** GabrieleV has quit IRC
16:57:24 *** GabrieleV_ is now known as GabrieleV
17:09:54 *** david has quit IRC
17:14:21 *** david has joined #gnucash
17:15:19 *** Jimraehl1 has left #gnucash
17:16:13 *** bhardwajs1 has joined #gnucash
17:16:14 *** Jimraehl1 has joined #gnucash
17:23:02 *** bhardwajs has quit IRC
17:30:34 *** gjanssens has quit IRC
17:47:20 *** lmat has quit IRC
18:01:51 *** fell has joined #gnucash
18:03:20 *** bhardwajs1 has quit IRC
18:04:54 *** bhardwajs has joined #gnucash
18:06:26 <bhardwajs> lmat: I try to code. Having said that, last 6 months were rough. A very stressful role that prompted me to change employers took way too much time. Hopefully, I can now contribute. Sorry for being AWOL. :(
18:06:26 <gncbot> bhardwajs: Sent 8 hours and 13 minutes ago: <lmat> Hi, there! I'm ready to talk, but you weren't here when I was. Feel free to ping me again.
18:07:37 *** bhardwajs has quit IRC
18:21:02 *** david has quit IRC
18:36:50 *** GabrieleV_ has joined #gnucash
18:37:14 *** GabrieleV has quit IRC
18:37:14 *** GabrieleV_ is now known as GabrieleV
18:37:18 *** bhardwajs has joined #gnucash
18:38:39 *** jralls has quit IRC
18:39:00 *** jralls has joined #gnucash
18:39:01 *** gncbot sets mode: +o jralls
18:51:14 *** benjamin-agaric has quit IRC
18:56:17 <jralls> @tell lmat It doesn't look to me that kvp_frame_.*_binary is actually used anywhere. It doesn't even seem to be supported by the SQL backend. I'd try deleting all references to it and its handlers in backends/xml and see if GC still compiles. No sense in worrying about an overly complicated, yet unused detail.
18:56:17 <gncbot> jralls: The operation succeeded.
18:59:39 *** benjamin-agaric has joined #gnucash
19:29:37 *** jralls has quit IRC
19:30:45 *** jralls has joined #gnucash
19:30:46 *** gncbot sets mode: +o jralls
19:32:19 *** GabrieleV_ has joined #gnucash
19:32:20 *** GabrieleV has quit IRC
19:32:21 *** GabrieleV_ is now known as GabrieleV
19:36:16 *** jralls has quit IRC
19:36:38 *** jralls has joined #gnucash
19:36:38 *** gncbot sets mode: +o jralls
19:51:49 *** GabrieleV_ has joined #gnucash
19:52:21 *** GabrieleV has quit IRC
19:52:22 *** GabrieleV_ is now known as GabrieleV
20:00:17 *** benjamin-agaric has quit IRC
20:18:00 *** jralls has quit IRC
20:18:53 *** jralls has joined #gnucash
20:18:53 *** gncbot sets mode: +o jralls
20:26:35 *** bhardwajs1 has joined #gnucash
20:31:01 *** kpreid has quit IRC
20:31:16 *** kpreid has joined #gnucash
20:34:50 *** bhardwajs has quit IRC
20:46:48 *** bhardwajs1 has quit IRC
20:47:51 *** bhardwajs has joined #gnucash
20:49:38 *** benjamin-agaric has joined #gnucash
22:59:11 *** fell_ has joined #gnucash
23:06:47 *** fell has quit IRC
23:21:31 *** cigarshark has joined #gnucash
23:34:29 *** fell_ has quit IRC
23:39:42 *** fell has joined #gnucash
23:39:43 *** gncbot sets mode: +o fell