2025-11-13 GnuCash IRC logs
00:01:41 *** Robert847 has joined #gnucash
00:02:09 *** ChanServ sets mode: +v Robert847
00:05:05 <Robert847> Hi. I am about to use gdb to get a stack trace. I found that I already have gdb 8.2.1 installed from a previous project. would that version work ok today?
00:05:57 <Robert847> Windows 10 under extended support
00:07:34 <Robert847> I will try it and see what happens
01:00:31 <Robert847> It seems to be cranking very slowly, just opening threads and closing them with code 0
01:01:57 <Robert847> using about 200MB of RAM according to task manager
01:03:51 *** mechtilde has joined #gnucash
01:04:18 <Robert847> It is trying to open the last file it used which apparently had a lock file present from when I last closed it
01:12:58 <Robert847> OK, Gnucash is running. Now I need to figure out how to set it up to crash on closing. I think viewing a report might work
01:14:37 <Robert847> Once, just trying to switch to a different register did it
01:15:33 *** fell has quit IRC
01:16:52 *** fell has joined #gnucash
01:16:52 *** ChanServ sets mode: +o fell
01:19:04 <Robert847> I am not seeing anything in gdb except opening and closing threads
01:21:20 <Robert847> Taask manager says Gnucash is up to 970 MB of RAM
01:24:13 *** mechtilde has left #gnucash
01:24:45 *** mechtilde has joined #gnucash
01:25:25 *** mechtilde has left #gnucash
01:26:24 *** mechtilde has joined #gnucash
01:26:24 *** ChanServ sets mode: +v mechtilde
01:26:31 <mechtilde> Do you use the xml file or an external databse for the data?
01:26:52 <mechtilde> Robert847:
01:28:13 <Robert847> In this case it is xll on a USB key
01:29:05 <Robert847> Gnucash is finally closed so I will look for a lock file
01:29:30 *** silverballer47 has joined #gnucash
01:33:41 <Robert847> did not crash that time
01:34:22 <Robert847> Still, I think I should see som diagnostics while it is running
01:41:11 <Robert847> The only evidence of a crash is that the lock file is not deleted and changes to the gcm file are not recorded
01:42:15 <Robert847> when it crashes, which is not every time
01:42:40 <Robert847> not every time I close the data file
01:45:36 <Robert847> I thought gdb woould give a lot more than new thread and thread exited messages
01:53:10 <Robert847> So I have two problems. One this is too slow to running to just use gnucash and wait for it to crash. 2 Why am I not getting any diagnostic messages during normal operation.
01:53:55 *** Robert847 has left #gnucash
04:16:25 *** silverballer47 has quit IRC
04:23:02 *** silverballer47 has joined #gnucash
04:53:54 *** gjanssens has joined #gnucash
04:53:54 *** ChanServ sets mode: +o gjanssens
05:59:46 *** dcomp has joined #gnucash
06:03:23 *** ChanServ sets mode: +v dcomp
06:03:53 *** dcomp has left #gnucash
06:04:00 *** dcomp has joined #gnucash
06:04:00 *** ChanServ sets mode: +v dcomp
06:04:56 <dcomp> Is there anyway to use an IPv6 address as a hostname when connecting to postgres. It seems to only accept domains and IPv4 addresses?
06:35:19 <Gry> depends on postgres server setup i think
06:43:36 <dcomp> I just put the address in my hosts file and then connected that way. I think the GUI doesn't understand the IPv6 addresss literal in the bracketed format [IPv6] like browsers and other tools. So if I try to to connect to [fd7a:115c:a1e0:b1a:0:1:a6f:e88a]:5432. It errors saying "Can't connect to postgres://postgres@[fd7a:115/database"
06:56:37 <Gry> um
06:56:39 <Gry> why not
06:56:53 <Gry> oh, because :
07:12:19 <warlord> dcomp, I THINK you can do it inside [ipv6]. So, I believe that pg://[::1]/... /should/ work. but of course, not been tested.
09:30:23 *** Gimp3 has quit IRC
09:32:40 *** Gimp3 has joined #gnucash
09:32:40 *** ChanServ sets mode: +v Gimp3
12:15:40 *** phebus has joined #gnucash
12:15:40 *** ChanServ sets mode: +v phebus
13:45:43 <dcomp> looks like gnc_uri_get_components isn't rfc3986 compliant and doesn't support address literals. Any reason not to use Boost::URL as boost is already a dep?
13:46:46 <warlord> dcomp, that would be a question for jralls. I see no objection, personally. I'm sure the C api was written well before the C++ integration.
13:50:00 <jralls> gnc-uri-utils.c is actually gjanssens's handywork and as the file extension indicates it hasn't been converted to C++ yet. A PR to rewrite it based on boost::url would be welcome.
14:27:28 *** mechtilde has quit IRC
14:40:03 *** silverballer47 has quit IRC
15:24:00 <gjanssens> Saw my name mentioned :)
15:25:46 <gjanssens> Sure gnc-uri-utils.c was a naive attempt to unify datafile/db uris for our different backends. It served its purpose at the time it was written, but I agree it's probably time to replace it with something more appropriate.
15:26:32 <gjanssens> Does Boost::URL also handle file uris ? gnc-uri-utils is in many ways very tailor-made for our use case.
17:08:11 *** gjanssens has quit IRC
17:33:06 <jralls> @tell gjanssens re boost::url and file uris, I suppose you mean plain file paths, not those beginning with file:. It doesn't appear to but it allows adding custom parsing functions that I think would work.
17:33:06 <gncbot> jralls: The operation succeeded.
19:21:59 *** jralls-m1 has joined #gnucash
19:21:59 *** jralls has quit IRC
19:22:11 *** jralls-m1 is now known as jralls
19:22:17 *** ChanServ sets mode: +o jralls
22:02:17 *** chris has joined #gnucash
22:02:17 *** ChanServ sets mode: +v chris
22:02:21 *** gncbot sets mode: +o chris
22:08:17 *** chris has quit IRC
23:10:00 *** jonakeys has quit IRC
23:10:07 *** jonakeys has joined #gnucash
23:29:34 *** chris has joined #gnucash
23:29:34 *** ChanServ sets mode: +v chris
23:29:38 *** gncbot sets mode: +o chris