2017-09-08 GnuCash IRC logs
01:14:40 *** Mechtilde has joined #gnucash
01:35:16 *** Mechtilde has quit IRC
02:26:59 *** CDB-Away_ has joined #gnucash
02:27:57 *** CDB-Man_ has joined #gnucash
02:29:22 *** CDB-Away has quit IRC
02:29:39 *** CDB-Man has quit IRC
02:31:21 *** hoijui has joined #gnucash
02:46:51 *** mrklintscher111 has joined #gnucash
02:52:56 *** fabior has joined #gnucash
03:13:05 *** gjanssens has joined #gnucash
03:13:05 *** ChanServ sets mode: +o gjanssens
03:16:39 <gjanssens> .
04:24:57 *** fabior has quit IRC
04:25:14 *** Mechtilde has joined #gnucash
04:41:20 *** mrklintscher112 has joined #gnucash
04:41:20 *** mrklintscher111 has quit IRC
04:46:10 *** fabior has joined #gnucash
04:50:18 *** Mechtilde has quit IRC
05:02:47 *** User has joined #gnucash
06:06:25 <lmat> As I'm testing the build, I'm getting unreproducible failures :
06:06:28 <lmat> :-/
06:07:13 <lmat> In order to test the after failure script, I modified a test to fail: https://github.com/limitedAtonement/gnucash/commit/743a17dee467c456bd8de452360c44d3e401b47a
06:07:34 <lmat> The build before this commit succeeded: https://travis-ci.org/limitedAtonement/gnucash/builds/273119768
06:09:37 <lmat> After this, the build failed with a segmentation fault while building business-reports: https://travis-ci.org/limitedAtonement/gnucash/jobs/273126550#L3599 I don't think this is related to my change.
06:11:28 <lmat> That is the autotools build in arch linux. I thought the ubuntu autotools build failed in an unexpected way, too, but investigation reveals that it didn't.
06:12:38 <lmat> I do see quite a bit of shared object errors in the arch linux builds; perhaps I should be surprised that they ever finish successfully.
06:13:39 <lmat> (libdbi "No such file or directory" shared object errors that is)
06:18:22 <lmat> Oh, it was the ubuntu ninja build that failed compiling scheme: https://travis-ci.org/limitedAtonement/gnucash/jobs/273126552
06:25:02 *** fekepp has joined #gnucash
06:25:53 <gjanssens> lmat: the scheme failure is again due to the changes I did for moving .gnucash
06:26:12 <gjanssens> compiling report-collectors.scm at some point runs this code
06:26:31 <gjanssens> I your environment doesn't have a writable temporary directory this code will fail
06:30:57 <lmat> gjanssens: hmm, I don't see why it wouldn't, but I haven't explicitly checked.
06:37:36 <gjanssens> lmat: as for your other build failure are you sure the dbi drivers for mysql and postgresql are installed ? The build seems to fail because they are missing.
06:42:51 *** Jimraehl1 has joined #gnucash
06:44:19 *** Jimraehl1 has quit IRC
06:50:45 <lmat> gjanssens: The build succeeds, too, sometimes on the same system :-)
06:52:29 <gjanssens> lmat: a timing issue then...
06:52:35 <gjanssens> Those are ugly :(
06:54:36 <lmat> /usr/lib/dbd/libdbdmysql.so /usr/lib/dbd/libdbdpgsql.so /usr/lib/dbd/libdbdsqlite3.so are all present
06:55:20 <lmat> But the error says "libpq.so.5: cannot open shared object file: No such file or directory" Why *.5?
06:57:59 <lmat> Ahh...
06:58:22 <lmat> https://www.archlinux.org/packages/extra/i686/postgresql-libs/files/ And I'm guessing similar for mysql.
06:58:41 <lmat> It's not enough to install the libdbi drivers, I'll need to install the individual packages, too.
07:00:51 <lmat> I'll add those to the build, and see what changes.
07:15:37 *** fabior has quit IRC
07:16:34 *** fekepp has quit IRC
07:17:21 *** Mechtilde has joined #gnucash
07:35:40 *** pilotauto has quit IRC
08:05:55 *** rickoehn has joined #gnucash
08:06:38 *** hoijui has quit IRC
08:33:04 *** Aussie_matt has joined #gnucash
09:05:21 *** librarian-bob has joined #gnucash
09:11:44 *** mrklintscher112 has quit IRC
09:12:08 *** librarian-bob has quit IRC
09:17:01 *** mrklintscher112 has joined #gnucash
09:20:07 <warlord> .
09:29:06 *** Chris has joined #gnucash
09:52:20 *** Mechtilde has quit IRC
09:55:33 *** codesmythe has joined #gnucash
10:05:35 *** mikee has quit IRC
10:09:18 *** mrklintscher112 has quit IRC
10:13:57 <Chris> @tell warlord Well I have to announce that I finally created a report that I wanted to create, approximately 10 years after my first encounter with this software. The impenetrable wall of scheme has finally been smashed to smithereens, and I can see the endless ( and ) in falling green characters.
10:13:57 <gncbot> Chris: The operation succeeded.
10:14:00 <Chris> @tell gjanssens Well I have to announce that I finally created a report that I wanted to create, approximately 10 years after my first encounter with this software. The impenetrable wall of scheme has finally been smashed to smithereens, and I can see the endless ( and ) in falling green characters.
10:14:00 <gncbot> Chris: The operation succeeded.
10:14:33 *** Aussie_matt has quit IRC
10:15:29 *** codesmythe has left #gnucash
10:16:15 <gjanssens> Chris: when the people you want to talk to are online there's no need to add a @tell. Just start with the nick and your message :)
10:16:15 <gncbot> gjanssens: Sent 2 minutes ago: <Chris> Well I have to announce that I finally created a report that I wanted to create, approximately 10 years after my first encounter with this software. The impenetrable wall of scheme has finally been smashed to smithereens, and I can see the endless ( and ) in falling green characters.
10:16:56 <gjanssens> Congratulations on your breakthrough :D
10:17:16 <warlord> .
10:17:16 <gncbot> warlord: Sent 3 minutes ago: <Chris> Well I have to announce that I finally created a report that I wanted to create, approximately 10 years after my first encounter with this software. The impenetrable wall of scheme has finally been smashed to smithereens, and I can see the endless ( and ) in falling green characters.
10:17:27 <warlord> Chris: great job! What rport?
10:17:38 <warlord> report
10:18:11 <Chris> warlord: https://github.com/Gnucash/gnucash/pull/177 - always found it annoying that I couldn't get calculated cells in multicolumns
10:23:01 <warlord> Ah, the business tax report.
10:24:00 *** kael has joined #gnucash
10:24:46 <warlord> Still, welcome to the world of scheme. It's not as bad as you thought, eh?
10:25:50 *** codesmythe has joined #gnucash
10:26:00 <Chris> Thank you :D it's got (turtles) all the way down
10:26:11 <warlord> haha
10:26:47 *** mikee has joined #gnucash
10:26:51 *** ChanServ sets mode: +o mikee
10:27:09 <warlord> Drowning in a sea of parentheses?
10:28:30 <Chris> Not anymore
10:29:15 <Chris> Still would be keen to submit for 2.8 and not sure which menu-item, nor a preferred name for this... "Business Tax Report" isn't quite right
10:31:05 <Chris> (and this is a definite upgrade from my first PR https://github.com/Gnucash/gnucash/pull/153/commits/385ca0c3240bfc2ab6cc7eb329052e3acc2c0e45 :-P)
10:37:27 <gjanssens> Chris: you're still in time to get it in gnucash 2.8
10:39:57 <gjanssens> jralls and myself are currently working to get the windows build running to completion so we can release our first 2.7 release candidate.
10:40:22 <gjanssens> After that I'll get back to all outstanding PR's (unless jralls wants them in before 2.7 still)
10:41:31 *** kael has quit IRC
10:42:58 <Chris> Sure and I'll continue bughunting meanwhile
10:43:28 <warlord> Cool
10:49:22 <gjanssens> warlord: fyi I'm currently testing the new scripts to build the package and create the installer on our buildserver on oVirt
10:49:38 <gjanssens> It's currently building the dependencies
11:03:22 <codesmythe> gjanssens: I've submitted a PR to add the missing python bindings to CMake. But we don't currently enable the python bindings in the travis script. Should we?
11:03:22 <gncbot> codesmythe: Sent 6 days, 7 hours, and 2 minutes ago: <gjanssens> That's even better!
11:03:36 <codesmythe> bindings test, that is
11:04:36 <warlord> gjanssens: great. hopefully it wont hang! if it does, it would be useful to know where... and if it's consistent.
11:04:47 *** Mechtilde has joined #gnucash
11:05:41 <gjanssens> warlord: indeed but so far so good
11:05:59 <warlord> great.
11:08:00 <gjanssens> warlord: what's your opinion on this: https://bugzilla.gnome.org/show_bug.cgi?id=787226#c3
11:08:27 <gjanssens> For me as foreigner I can't estimate whether the proposals are more clear or not
11:08:53 *** Groan has joined #gnucash
11:09:50 <warlord> gjanssens: maybe "My Reference" and "Their Reference"
11:32:17 *** fekepp has joined #gnucash
11:37:03 *** Chris has quit IRC
11:38:58 *** Derperperd has quit IRC
11:39:39 *** Derperperd has joined #gnucash
11:59:20 *** Derperperd has quit IRC
12:00:27 *** Derperperd has joined #gnucash
12:08:42 <gjanssens> warlord: can you check whether I have accidentally blacklisted the build server over ssh ? I think the keypair was not in place yet for the Mingw64 environment.
12:09:28 <gjanssens> warlord: oh no. It looks like the uploads went fine
12:10:14 <lmat> @tell Chris Congratulations on decoding the ( ) matrix! It sounds like you're a bit ahead of me in this regard!
12:10:14 <gncbot> lmat: The operation succeeded.
12:10:48 <lmat> codesmythe: I think the only reason we don't cmake test the python bindings is because they were missing.
12:11:25 <lmat> codesmythe: I incorporated both your suggestions on my travis/docker pull request: use the correct branch (not some default), and refactor common code to a separate script.
12:11:46 <gjanssens> yes! the first Windows installer for master got just uploaded. I haven't had time to test whether it runs yet though :D
12:12:01 <lmat> gjanssens: It compiles? Ship it!
12:12:03 <lmat> ;-)
12:12:24 <warlord> gjanssens: the blocker system ignores my internal addresses, so code will never block the builder.
12:12:32 <codesmythe> lmat: Great. I am looking forward to the docker support. I'd eventually like to add support for Debian Jessie and some Fedoras
12:12:37 <warlord> gjanssens: GREAT NEWS!
12:12:58 <codesmythe> gjanssens: Excellent.
12:13:59 <lmat> codesmythe: Should be quite straightforward. My latest branch just succeeded with all the features we want (I think!), so I'll create a new pull request
12:17:22 <lmat> codesmythe: I would think an Ubuntu 16.04 (LTS) would make sense, too.
12:20:15 <codesmythe> lmat: Yep. Re: the python bindings. The tests were indeed missing. But since we don't use WITH_PYTHON=ON for CMake, they still won't run after my PR is merged.
12:22:23 *** fell has joined #gnucash
12:22:47 *** gncbot sets mode: +o fell
12:23:27 <fell> I am back for a few hours.
12:25:11 *** fekepp has quit IRC
12:25:22 <fell> Gjanssens, PR 182 was part of https://bugzilla.gnome.org/show_bug.cgi?id=787226
12:26:37 <fell> And on the german mailing list rised the question, if something should be changed in the english text.
12:27:03 *** fekepp has joined #gnucash
12:27:54 <fell> https://lists.gnucash.org/pipermail/gnucash-de/2017-September/010032.html ff.
12:29:55 <fell> See the images there
12:34:16 <gjanssens> fell: I know, but I don't have an answer to that so the bug is still open. warlord added a suggestion on my request.
12:35:14 <gjanssens> I considered the current state of PR 182 to be fixing a bug so I pushed that already. The question whether the English text should change is much less critical IMO.
12:35:35 *** fell is now known as fell_
12:35:41 *** fell_ is now known as fell
12:47:40 <fell> If the german translators are confusing their users, in other languages, it might happen, too.
12:50:14 *** kael has joined #gnucash
12:51:21 <fell> And I would like to hear other opinions, e.g from Mechtilde.
12:51:59 *** Groan has quit IRC
12:52:23 <Mechtilde> fell, opinions of what?
12:53:16 <Mechtilde> I was creating some transactions with chiptan in GnuCash
12:53:36 <fell> gnucash-de Lieferantenrechnungen / IDs
12:53:37 <Mechtilde> it works without "flickr code"
12:54:09 <Mechtilde> I didn't have my printed translation byhand
12:55:13 <Mechtilde> where are the differences
12:55:14 <fell> Oh, interesting, how did you confirm without flickr?
12:55:59 <Mechtilde> you get a dialog describing which values you have to input
12:59:21 <fell> Can you document it somewhere? E.g. similar to https://wiki.gnucash.org/wiki/De/Onlinebanking-Einrichtungsassistent
13:00:37 *** kael has quit IRC
13:01:07 <gjanssens> fell: to be clear - I'm in favour of clarifying the strings. I just didn't know better alternatives. The ones in proposed on the bug report didn't strike me as less confusing.
13:01:32 <gjanssens> Once we agree on better alternatives we can proceed.
13:02:00 <Mechtilde> I have to start a fake transaction without executing it
13:02:24 *** Groan has joined #gnucash
13:02:42 <fell> gjanssens, I like warlords suggestion, but we need to verify, it wors similar in the invoice dialog.
13:04:52 <Mechtilde> At Vendor the translation is more specific than at customer
13:08:35 <gjanssens> warlord: I'd have configure an e-mail address for git on the build server. Which one do you propose ?
13:08:48 <gjanssens> I don't think it will ever go outside of the build server so it can be a fake one also
13:08:55 <warlord> fell: I do believe that Customer Invoices also have two fields, one for the local ID (Invoice ID) and one for the customer reference #
13:09:25 *** kael has joined #gnucash
13:10:09 <warlord> gjanssens: What do you mean "configure an e-mail address for git on the build server"?
13:10:52 <gjanssens> warlord: the "git stash" command fails because git can't detect an e-mail address https://wiki.gnucash.org/builds/win32/build-logs/gnucash-build-log-2017-09-08-12-04-10.log
13:11:26 <warlord> why are we stashing?
13:11:28 <gjanssens> I was mostly wondering whether we had some common e-mail address that gets used on the build server.
13:11:42 <warlord> No.
13:12:07 <gjanssens> The stash is something done automatically by the jhbuild when it detects the source tree isn't clean.
13:12:26 <warlord> gjanssens: Oh. We can't just run git clean ... ?
13:12:33 <Mechtilde> fell, I compared the two pictures
13:13:13 <gjanssens> warlord: I don't know. It's something jhbuild does and I don't know whether we can intercept this or not. That's jralls' area of expertise.
13:13:15 <Mechtilde> It is poible that I swape the two terms
13:13:31 <gjanssens> But for now I'll just use a fake address. It's for internal use only anyway
13:13:36 <Mechtilde> that I didn't understand it in the right way
13:13:53 <Mechtilde> s/poible/possible
13:17:28 <warlord> Then put in "build@builder" for the email. it shouldn't matter.
13:18:25 <fell> gjanssens: cccbec6 is definitive wrong. Billing ID is used in several dialogs for both types of invoices.
13:19:36 <Mechtilde> I'm not sure what is wrong or right
13:21:16 <fell> Oops, I read it the wrong direction.
13:23:39 <fell> we had for "Bill ID" before "Rechnungsnummer des Lieferanten", now "Abrechnungs-ID", IMHO better : "Nr. der Lieferantenrechnung"
13:26:30 <fell> But we had for "Billing ID", which is used in invoices, too, "Abrechnungs-ID", now "Rechnungsnummer des Lieferanten", which is in 50% wrong.
13:27:12 <gjanssens> fell: yes you're right about that :(
13:27:32 <gjanssens> This is an untranslatable situation
13:31:36 <warlord> Yeah, I think part of the issue is that English doesn't have truly different terms for Customer Invoice and Vendor Bill. The words "Invoice" and "Bill" are used interchangibly in both cases.
13:32:39 <gjanssens> warlord: I'm getting permission denied when trying to overwrite a file via scp. Is that expected ?
13:33:27 <gjanssens> This happens because an (almost) empty log file is uploaded at the beginning of the packaging script
13:33:48 <gjanssens> At the end the complete log file should replace this initial one.
13:34:22 <gjanssens> ISTR this works fine on the maint branch (different scripts though) so I wonder why it fails on master?
13:40:38 <warlord> Not sure. It SHOULD work. What path is working, and what path is not?
13:43:10 *** fell has quit IRC
13:51:17 <gjanssens> warlord: the exact same path is working and isn't.
13:51:44 <gjanssens> The first upload works, the second doesn't. So I can't overwrite an existing file on the server
13:51:52 <gjanssens> But I can add new files
13:52:30 *** jralls has joined #gnucash
13:52:30 *** ChanServ sets mode: +o jralls
13:53:01 <gjanssens> jralls: right in time. I was just starting to write you a message as I have to leave soon
13:53:26 <jralls> gjanssens: OK. I have a couple of hours.
13:53:34 <gjanssens> jralls: I have played a bit with the build_package.ps1 script and fixed a few minor issues
13:53:51 *** fell has joined #gnucash
13:53:57 <gjanssens> I managed to upload our first master built Windows installer
13:54:18 <gjanssens> There are a few issues still
13:54:27 *** gncbot sets mode: +o fell
13:54:58 <gjanssens> 1. I can't overwrite a file on the upload server. So I can upload the start log, but uploading the end log fails (permission denied)
13:55:10 *** Mechtilde has quit IRC
13:55:37 <gjanssens> Same happens if we try to re-upload an installer file that was previously uploaded (or at least has the same name as a previously uploaded one)
13:56:12 <jralls> Are you uploading from your own system or did you manage a build on the build VM?
13:56:17 <gjanssens> 2. The build fails from time to time, complaining a certain dll can't be opened for writing.
13:56:27 <gjanssens> I'm working on the build vm currently so that's where the upload happened
13:56:52 <gjanssens> The only way to get past this second error is to manually remove the offending dll.
13:57:16 <jralls> OK, cool.
13:57:19 <gjanssens> It has happened on my system locally once and again today on the build server. The build result is probably still open.
13:57:31 <gjanssens> That is, if you scroll back a bit in the logs
13:58:29 <jralls> Only one of us can connect to the VM at a time unless Ovirt has changed something in the last 6 months since warlord set it up.
13:58:38 <jralls> warlord, are you present?
13:58:45 <gjanssens> Well I just logged out for you :)
13:58:53 <warlord> I am here. And no, ovirt only allows one person to connect to the console.
13:58:58 <warlord> I'm not sure how to change that.
13:59:16 <gjanssens> Or at least I closed my virt-viewer window. I'm not sure that's sufficient. I think so.
13:59:25 <warlord> That should be sufficent.
14:00:01 <gjanssens> 3. A minor issue: the log file has some garbage in there. I think they are shell escape codes to print in color or bold.
14:00:55 <jralls> I saw the noise in the log and I agree they look like ansi escapes.
14:01:29 <jralls> That should be fixable with a "no color" option to the shell.
14:01:37 <gjanssens> 4. The package_build.ps1 script seems to be missing the logic to avoid building if nothing changed. build_package.sh checked commit hashes against a list of already built ones. If found it just stopped.
14:01:57 <gjanssens> I think that logic was flawed as well to be honest
14:02:34 <gjanssens> So I'm not sure what our future policy should be. Do we want to rebuild by default every night ? Or only of either gnucash-on-windows.git or gnucash.git has changed ?
14:03:39 <gjanssens> Oh and fyi, I have altered the script to allow the caller to set the hostname for uploading from the commandline rather than hardcoding it. If no hostname is set, not upload is attempted.
14:04:31 <gjanssens> Of course if we want to trigger a rebuild manually we'd have to think of adding the parameter or we shoud write a local wrapper script the only lives on the build server
14:04:47 <gjanssens> It think that's mostly it for now.
14:04:57 <gjanssens> I have to leave in a couple of minutes...
14:05:17 <gjanssens> You can take it from here again. I won't have much time for the weekend.
14:08:24 <gjanssens> Oh yes, one last thing we should fix before 2.8 gets released: the executable upload is hardcoded to directory "master"
14:08:46 <gjanssens> That's still correct now, but won't be any more once 2.8 is the new stable
14:08:55 *** kael has quit IRC
14:09:45 <gjanssens> We should probably re-add some branch dectection code so executables are uploaded to a directory named after the branch they are built from.
14:09:52 <jralls> buid_package.sh's tag file is only for releases. There's logic somewhere else than notices if there are new commits on maint or master. Jhbuild is supposed to do that too and I had in mind to rely on that.
14:10:32 <jralls> But you're right that there needs to be logic in the script to not fire up inno setup if there's no build.
14:12:16 <jralls> Anyway, the script is a first pass, as is the moduleset. The fact that it builds successfully is excellent; thanks for testing.
14:13:18 <gjanssens> You're welcome.
14:13:29 <gjanssens> I'm leaving now
14:13:34 *** gjanssens is now known as gjanssens_afk
14:13:43 <jralls> Bye. TTyl
14:15:12 *** frakturfreak has joined #gnucash
14:35:30 *** Groan has quit IRC
14:36:20 <jralls> warlord: Well, that didn't take long. As soon as I opened a Mingw32 shell and tried to type something Cortana grabbed focus, task manager launched, and I couldn't communicate with the VM. I just told it to power down and I'm waiting.
14:43:16 <jralls> warlord: It supposedly powered off and back on but it's in the same state.
15:20:05 <warlord> jralls: shutdown != power-down.
15:23:03 <jralls> warlord: OK. The ovirt control panel does indeed say "shutdown". VMWare Fusion's term is "power down" and it nukes the running VM just like removing power. What does Ovirt's shutdown button do?
15:34:54 <warlord> A soft shutdown. Basically it sends a message to the OS to "shut down". But if the OS is wedged then this message doesn't get sent.
15:35:21 <warlord> There is a power-down option in the right-click menu on the admin portal, I don't know if there is one on the user portal.
15:35:42 <warlord> I wonder if there is some way to disable cortana?
15:41:19 *** fekepp has quit IRC
15:44:42 <jralls> warlord: Right click only gets me the regular browser context menu on the user console.
15:45:24 <jralls> warlord: https://www.howtogeek.com/265027/how-to-disable-cortana-in-windows-10/
15:45:26 *** hoijui has joined #gnucash
15:46:37 <jralls> warlord: I've now done both a shutdown and a reset to no effect. I also get a dialog box from user console about the VM not responding.
15:47:44 <warlord> Yeah, you wont. it gets into a state where ovirt thinks the VM is alive -- you can connect to it -- technically the VM is *running*, but it's internally wedged.
15:47:49 <warlord> Let me go power it down.
15:49:02 <warlord> I did get an event that the shutdown failed.
15:49:44 <warlord> Okay, it should be powered down now. You should be able to start it up?
15:56:15 <warlord> jralls: and feel free to disable cortana first thing! :)
16:02:11 <jralls> OK, I'm back in a nd starting regedit.
16:07:26 <jralls> Eh, this is pro, so it's the Group Policy Editor... but Cortana should now be disabled.
16:07:53 <jralls> Let's see if I can freeze it a different way this time...
16:09:22 <warlord> LOL. Lets see if you can freeze it the same way this first ;)
16:09:28 <warlord> Just let me know and I can power cycle again
16:19:06 <jralls> It didn't freeze. I'm going to do the next couple of steps on my local VM, it's faster and I can commit more easily from there.
16:20:25 <warlord> Ok
16:20:32 <warlord> (Yay for not freezing)
16:27:41 *** hoijui has quit IRC
16:34:02 *** fabior has joined #gnucash
16:55:27 <lmat> Do we have a way to do gnucash database conversions?
16:56:05 <lmat> jralls: Like you said earlier, we'll want to convert map data from "/" separators to some non-printable character. But this should only be done once!
16:56:54 <lmat> jralls: Or maybe a one-time bayes map blow-away/start-over?
16:57:08 <lmat> (I don't know how much people depend on this functionality...never used it)
16:57:57 <jralls> lmat: The usual way is to read liberally and write strictly. Forcing users to retrain their import matching will not be popular.
16:59:06 <jralls> One can force a complete rewrite of a SQL backend with "safe-save".
17:00:02 <lmat> jralls: Is there any way to tell if the conversion has already happened? (database versioning is what I'm envisioning, but I guess it doesn't exist in gnucash.)
17:01:06 <jralls> Yes. There's a database backend number that's defined in configure/CMakelists.txt and written into the versions table.
17:06:23 <jralls> Oops, looks like it's not in CMakeLists.txt. Its name is GNUCASH_RESAVE_VERSION. Having a newer one than is in the table triggers a safe-save after everything is loaded; an older one will block loading.
17:08:47 <lmat> jralls: Excellent.
17:10:00 <lmat> jralls: You said there is deep nesting of kvps somewhere; where was that? bayes maps?
17:10:30 <jralls> Oh, it *is* in CMakeLists.txt, just missed it the first time around. The number that's there is from the SVN revision when it was created. Now I think we can just use the release numbers jammed together, e.g. 20700.
17:10:41 <gjanssens_afk> jralls: while working on the VM earlier today I also had short freezes (up to say 20 seconds) because of visual effects. It always recovered though.
17:11:00 <gjanssens_afk> But I'm glad you disabled it. It's not helpful in a vm over the internet.
17:11:37 <gjanssens_afk> And those are my last words for tonight, other than perhaps... good night :D
17:12:05 *** gjanssens_afk has quit IRC
17:14:22 <jralls> Yes, the bayes maps get nested pretty deeply.
17:24:10 *** mrklintscher112 has joined #gnucash
17:25:29 <jralls> Hmm, on looking at XML, maybe not as deeply as I thought. It's 'import-map-bayes'/<tokent>/<account> and 'import-map-bayes'/<tokent>/<count>.
17:26:48 <jralls> But there can be thousands of them.
17:27:03 <jralls> I've got to go, prolly not back until tomorrow morning.
17:27:14 *** jralls has quit IRC
17:35:55 *** frakturfreak has quit IRC
17:53:45 *** fell has quit IRC
18:00:25 *** Groan has joined #gnucash
18:14:43 *** Groan has quit IRC
18:15:33 *** mrklintscher112 has quit IRC
18:31:56 *** Groan has joined #gnucash
19:40:54 *** fabior has quit IRC
19:59:56 *** jethrogb has quit IRC
19:59:57 *** Groan has quit IRC
20:09:50 *** pilotauto has joined #gnucash
20:45:35 *** rickoehn has quit IRC
21:20:26 *** fell has joined #gnucash
22:17:33 *** User has quit IRC
22:41:56 *** bertbob has quit IRC
22:55:54 *** bertbob has joined #gnucash
23:18:38 *** Chris has joined #gnucash