2007-06-12 GnuCash IRC logs
00:12:33 *** warlord is now known as warlord-afk
01:49:10 *** thebanana has joined #gnucash
01:52:29 <thebanana> can i use gnucash in a networked mode?
01:53:00 <thebanana> i'm evaluating my options for small business accounting and will need access to the data from several workstations
01:54:35 *** bonez39 has quit IRC
02:15:05 *** thebanana has quit IRC
03:13:18 *** BlackBsd has quit IRC
03:46:29 *** ErKa has joined #gnucash
03:59:41 *** Rol1 has joined #gnucash
04:32:48 *** jakin has quit IRC
04:44:50 *** chemaja has joined #gnucash
04:47:43 *** motin has quit IRC
05:16:25 *** motin has joined #gnucash
05:30:29 *** andi5 has joined #gnucash
05:30:29 *** gncbot sets mode: +o andi5
06:03:02 *** Demitar has quit IRC
06:09:09 *** nomeata has joined #gnucash
06:34:19 *** ErKa has quit IRC
07:19:15 *** ErKa has joined #gnucash
07:26:13 *** twunder has joined #gnucash
07:36:09 *** twunder has quit IRC
08:18:04 *** deconvolution has joined #gnucash
08:19:37 *** twunder has joined #gnucash
08:21:38 <deconvolution> hello, my broadband ISP set up an account for me with some initial credits. What's the best way to maintain this type of account? Assets, expense, equity or income?
08:26:57 <cortana> asset
08:31:52 <deconvolution> cortana: thanks
08:31:54 *** ErKa has quit IRC
08:46:00 *** warlord-afk is now known as warlord
08:51:01 *** ErKa has joined #gnucash
09:18:37 *** chemaja has quit IRC
09:19:32 *** chemaja has joined #gnucash
09:28:44 *** motin_ has joined #gnucash
09:37:44 *** motin has quit IRC
09:43:45 *** chemaja has quit IRC
09:55:59 *** motin_ has quit IRC
10:02:37 *** andi5 has quit IRC
10:35:00 *** ErKa has quit IRC
10:37:22 *** Ardonik has joined #gnucash
11:34:43 *** ErKa has joined #gnucash
11:44:37 *** Rol1 has quit IRC
11:47:17 *** Rol1 has joined #gnucash
12:38:02 *** andi5 has joined #gnucash
12:38:02 *** gncbot sets mode: +o andi5
13:06:19 *** sjc has joined #gnucash
13:26:35 *** Demitar has joined #gnucash
13:43:40 *** andi5 has quit IRC
14:10:46 *** deconvolution has quit IRC
14:28:37 *** ErKa has quit IRC
15:17:13 *** _gunni_ has joined #gnucash
15:17:52 *** _gunni_ has quit IRC
15:18:04 *** _gunni_ has joined #gnucash
15:46:30 *** dbr has joined #gnucash
16:06:46 <dbr> warlord: are you running a case sensitive file system on your mac?
16:07:41 <dbr> 2.1.3 is crashing for me with a symbol problem between fink's libjpeg and Apple's libJPEG
16:08:40 <dbr> the fink folk last week pointed to DYLD_LIBRARY_PATH as being a bad idea (but the problem in this case won't occur if you're running case sensitive)
16:09:53 <dbr> curiously, I haven't seen the problem until now (fink version 2.0.5 tarball or svn in early May), well after the DYLD_LIBRARY_PATH additions
16:12:03 <warlord> dbr: Um, I honestly dont know.. How do I tell?
16:13:28 <dbr> if you didn't consciously set it, I can't imagine you are...
16:13:37 <dbr> default is not cs
16:13:55 <warlord> well, i'm logged in.. what's a good test?
16:14:46 <dbr> I suppose you could ls something and see if it fails on upper case?
16:15:09 <warlord> Okay, it's case-insensitive.
16:18:50 <dbr> ok. I'm going to test DYLD_FALLBACK_LIBRARY_PATH to see if that solves my problem without messing with make check
16:19:19 <dbr> was make check the primary motivation for adding DYLD_LIBRARY_PATH?
16:19:56 <warlord> Honestly, I dont recall.
16:20:06 <warlord> I thnk that might've been part of it.
16:20:11 <dbr> OK. I'll go test...
16:21:41 <warlord> ok
16:43:21 *** ural0 has joined #gnucash
16:53:11 *** benoitg has left #gnucash
16:55:01 *** ural0 has quit IRC
16:59:49 *** twunder has quit IRC
17:51:06 *** nomeata has quit IRC
18:04:51 *** ErKa has joined #gnucash
18:14:05 *** andi5 has joined #gnucash
18:14:05 *** gncbot sets mode: +o andi5
18:16:42 *** dbr is now known as dbr-afk
18:22:36 <andi5> warlord: you are too fast, why do i even bother...
18:23:26 <warlord> LOL
18:23:37 <warlord> It was good timing on my part.
18:24:04 <andi5> warlord: hiho :-) .... well, i have got a Q: is there any reason why i can find my customer, can show her invoices in the search dialog, but _cannot_ find the invoices themselves as new search?
18:24:33 <warlord> Ummm........
18:24:48 <warlord> Not that I know of..
18:25:03 <warlord> How are you specifying the invoice search?
18:25:09 <andi5> i think i have struggled here before .... must be some newbie mistake :-)
18:26:13 <warlord> could be
18:26:38 <andi5> say, i see the results of the "view invoices"... the first column says "num==ii2", so i search for "Invoice ID contains "i" ... i can search for other things as well, empty results whatever i do
18:28:14 <andi5> i cannot find my job as well :( ... somehow this is unsatisfactory
18:28:26 <warlord> That sounds very... weird.
18:28:30 <warlord> Which version?
18:28:34 <andi5> 213
18:29:39 <warlord> Do your invoice numbers have accented chars?
18:29:49 <andi5> nono
18:30:07 <warlord> Just checking; the qofquery code isn't UTF-8 safe
18:30:14 <andi5> yes, it is a pity
18:30:50 <andi5> there is also a bug about case-insensitive search, which is.... *bad*
18:30:57 <warlord> eh, not too hard to fix that. Just need to move the utf8 compare functions and make them available and use them instead of strstr()
18:32:26 <warlord> unfortunately i can't answer you; and the qofquery debugging messages are... less than useful. Try single-stepping through the qof_query_run() call?
18:32:34 <andi5> well, there is one big question: are those qof strings guaranteed to be valid utf-8 ones? ... or is there some "entry" where a check should be added?
18:32:44 <andi5> ok, will do that
18:32:57 <andi5> s,added,checked,
18:33:05 <warlord> Yes, the strings are all guranteed UTF8 because the gtk input is utf8 and the datafile is utf8.
18:33:09 <andi5> ouch.... that last sentence was weird
18:33:24 * andi5 is tire
18:33:37 <andi5> fine
18:34:48 <warlord> but... i have alphanumeric entries and I can find them just fine.. (although I admit I haven't tried with 2.1.x recently.
18:36:53 * warlord builds current SVN
18:46:13 <warlord> andi5: feel like forwarding me your data file? ;)
18:46:23 <andi5> not yet :-)
18:46:50 <andi5> well, it is just for playing with the business features, but hey....
18:47:06 <warlord> ok....
18:47:49 <andi5> i wanted to know whether i have some major misunderstanding... now i want to solve that problem for me :-)
18:48:40 *** ErKa has quit IRC
18:50:41 *** _gunni_ has quit IRC
18:52:48 <warlord> ok.
18:53:02 <warlord> keep in mind sub-searches -- you can refine existing searches.
18:53:05 <warlord> (or start over)
18:53:29 <andi5> yep
19:17:38 <andi5> warlord: is it intended that only one search criterion survives a search? (all others are removed from the dialog)
19:21:06 <warlord> yes.. the search criterion are now a part of the underlying query / results..
19:21:34 <andi5> even if the results are empty and "new search" is selected?
19:22:07 <warlord> The dialog doesn't differentiate that case from any other case.
19:23:50 <warlord> It builds a new query based on the criterion, runs the query, builds the results, and resets the page.
19:23:52 <andi5> oh... there is a difference... if you find something, "refine current search" is actived, if not then "new search" .... but suppose you have 5 criteria and mistyped crit 4 .... you will have to reconfigure all criteria again then
19:24:06 <andi5> oh
19:24:39 <warlord> Yes, the default action is different based on the result count
19:24:49 <warlord> (new search v. refine search)
19:25:08 <warlord> I suppose that it would be better to not reset the page if there are no results.
19:25:34 <andi5> does that mean that "new search" does not always mean "new search"?
19:25:48 <warlord> no, new search always means new search..
19:26:02 <warlord> but 'refine search' changes its meaning based on the last set of results.
19:26:12 <warlord> (as do all the other options)
19:26:20 <andi5> yes, so the dialog is consistent, but hazardous ;-)
19:26:34 <warlord> i suppose. ;)
19:27:15 <warlord> I guess I tend to use few criteria at once, and use 'refine search'.. until I get a small list.
19:28:01 <andi5> yes, this is fine as long as you do not accidentally intersect too much :)
19:28:49 <andi5> well, does not matter now
19:28:50 <warlord> right, because there's no way to step back one search.
19:31:18 <andi5> well, maybe one more thought: .... do not you think that less magic would be better? i.e. why cannot we just NOT reset the dialog... refining would be like pressing "+ add", whereas you can also use a slightly modified criteria set for additions too
19:33:27 <warlord> Because there's no way to do (OR (AND x1 x2 x3) (AND x4 x5 x6) (AND x7 x8 x9)) in a single dialog.
19:34:49 <andi5> well, i mean: x1 x2 x3 <find> <add results> x4 x5 x6 <find> <add results> x7 x8 x9 <find>
19:35:49 <andi5> and now consider x1==x4==x7 and x3==x6==x9 :-)
19:36:48 <warlord> Well, that's what's going on now, except it stops showing the criteria.
19:37:56 *** Rol1 has quit IRC
19:38:01 <andi5> yes, i just vote for merging "new search" and "refine current search" into "new search" or something like that
19:38:05 <warlord> but keeping them visible between executions would just add to confusion.
19:38:14 <andi5> would it?
19:38:40 <warlord> Yes, because there's no way to visually differentiate AND and OR criteria.
19:40:27 <warlord> Let's assume that the criteria stay visible. You now want to extend the search by adding some "any criteria are met".. Well, what does it do with the already visible criteria? Or worse, what happens if you Remove one of the "previous" criteria that's already a part of the current Query?
19:41:43 <warlord> Keep in mind that the search maintains the last Query, so "refine" search is an AND into that query, Add results to current search is an OR into that query, and I forget what "Delete results from current search" does.. XOR, maybe?
19:41:49 <warlord> (no, can't be xor)
19:42:34 <warlord> Ahh, it's a NAND
19:42:36 <andi5> i would <add results> <remove x1> <remove x2> <remove x3> <search for: any crit met> <add z4> <add z5> <find>
19:42:47 <warlord> EWW!
19:42:48 <warlord> Too much work!
19:42:54 <andi5> is it? i think it is old and (not new)
19:42:55 <warlord> The current method is much easier.
19:43:08 <warlord> Anyways, I gotta run.. Dinnertime.
19:43:10 *** warlord is now known as warlord-afk
19:43:32 <andi5> bye
19:43:47 <andi5> maybe someone else feels like me and speaks up someday ;-)
19:45:41 *** warlord-afk is now known as warlord
19:45:45 <warlord> okay, i have a minute..
19:45:50 <warlord> (but only a minute)
19:45:52 <andi5> hehe
19:46:02 <warlord> you're right about one thing.. If the results are empty it shouldn't reset.
19:46:06 <warlord> okay, gotta run
19:46:08 *** warlord is now known as warlord-afk
19:46:11 <andi5> lol
19:50:33 <andi5> warlord-afk: ah, i know got your point... the problem with "merging" those two choices is that one cannot refine a search by an AND clause later on .... maybe i should not try to think about such complicated issues ;-)
20:09:29 *** Def has quit IRC
20:09:29 *** dacc has quit IRC
20:09:29 *** Hydroxid1 has joined #gnucash
20:09:29 *** Def has joined #gnucash
20:09:29 *** Hydroxide has quit IRC
20:10:07 *** andi5 has quit IRC
20:10:38 *** wizkid238 has quit IRC
20:10:46 *** dacc has joined #gnucash
20:11:05 *** cortana has quit IRC
20:12:05 *** cortana has joined #gnucash
20:14:29 *** wizkid238 has joined #gnucash
20:25:58 *** sjc has quit IRC
20:43:20 *** cortana has quit IRC
20:55:13 *** cortana has joined #gnucash
21:13:02 *** dbr-afk has quit IRC
21:28:59 *** warlord-afk is now known as warlord
21:30:36 <warlord> @tell andi5 Exactly! You can't combine sets of AND and OR clauses in your merge scenario whereas you can in mine. The only problem with NOT clearing the criteria on empty-result is that it WOULD be an "odd" case and potentially confusing.. Why does it clear the criteria always EXCEPT on no-results?
21:30:36 <gncbot> warlord: The operation succeeded.
21:55:37 *** twunder has joined #gnucash
22:13:55 *** dbr-afk_ has joined #gnucash
22:21:11 *** twunder has quit IRC
22:23:42 *** wizkid239 has joined #gnucash
22:30:06 *** jakin has joined #gnucash
22:31:35 *** wizkid238 has quit IRC
22:44:47 *** cparker has joined #gnucash
23:06:19 *** Ardonik has quit IRC
23:13:12 *** Ardonik has joined #gnucash
23:29:42 *** BlackBsd has joined #GnuCash
23:39:23 *** benoitg has joined #gnucash
23:45:44 *** warlord is now known as warlord-afk