2019-01-14 GnuCash IRC logs

00:23:48 *** phebus has quit IRC
01:02:07 *** bertbob has quit IRC
01:12:50 *** bertbob has joined #gnucash
01:43:20 *** ordex has quit IRC
01:45:28 *** ordex has joined #gnucash
01:58:51 *** fell has quit IRC
02:00:00 *** fell has joined #gnucash
02:01:54 *** fabior has joined #gnucash
02:47:53 *** fabior has quit IRC
02:58:26 *** fabior has joined #gnucash
02:59:57 *** fabior has quit IRC
03:01:26 *** fabior has joined #gnucash
03:02:53 *** fabior has quit IRC
03:09:39 *** Laurentius has joined #gnucash
03:10:36 *** GabrieleV has quit IRC
03:12:38 *** GabrieleV has joined #gnucash
03:21:49 *** Aussie_matt has quit IRC
03:30:26 *** gour has joined #gnucash
03:41:47 *** Mechtilde has joined #gnucash
03:51:38 *** gjanssens has joined #gnucash
03:51:38 *** ChanServ sets mode: +o gjanssens
03:51:49 <gjanssens> .
04:14:07 *** fabior has joined #gnucash
04:17:30 *** fabior has quit IRC
04:37:09 *** fabior has joined #gnucash
04:47:35 *** kapil___ has joined #gnucash
05:11:37 *** fabior has quit IRC
05:17:07 *** fell has quit IRC
05:17:15 *** fell has joined #gnucash
05:24:14 *** Mechtilde has quit IRC
05:47:27 *** trying2Simplify has joined #gnucash
05:48:46 *** fell has quit IRC
06:27:02 *** trying2Simplify has quit IRC
06:27:15 *** trying2Simplify has joined #gnucash
06:28:28 *** User has joined #gnucash
06:33:01 *** User has quit IRC
06:57:21 *** Jimraehl1 has joined #gnucash
06:57:41 *** Jimraehl1 has left #gnucash
07:00:15 *** Laurentius has quit IRC
07:02:58 *** fell has joined #gnucash
07:07:47 *** kapil___ has quit IRC
07:10:39 *** trying2Simplify has quit IRC
07:28:04 *** simeon has joined #gnucash
07:31:07 <simeon> Hello! Trying to build gnucash 3.4 on Linux Mint 18.3 Cinnamon 64 bit, but get a segmentation fault each time I'm trying to input some data into the leder of the compiled program.
07:43:58 *** Laurentius has joined #gnucash
07:57:44 *** oozer has joined #gnucash
08:01:53 <gjanssens> simeon: can you make a backtrace in gdb for that segmentation fault and post it somewhere, like pastebin ?
08:02:56 *** Mechtilde has joined #gnucash
08:03:46 *** nimish has joined #gnucash
08:04:48 *** boldstripe has joined #gnucash
08:08:24 *** nimish has quit IRC
08:09:05 *** nimish has joined #gnucash
08:23:47 *** nimish has quit IRC
08:24:56 *** nimish has joined #gnucash
08:26:27 *** gour has quit IRC
08:28:31 *** gour has joined #gnucash
08:30:28 *** benoitg has joined #gnucash
08:34:48 *** nimish has quit IRC
08:35:29 *** nimish has joined #gnucash
08:38:46 <warlord> .
08:38:59 <warlord> gjanssens, hi -- did you see the email I sent you?
08:39:42 <gjanssens> warlord: yes, though I haven't had time to look at it yet.
08:41:03 <gjanssens> But the immediate question would be what version of gpg do you have ?
08:41:15 <gjanssens> Mine does provide a --export option
08:41:34 <gjanssens> (1.4.23)
08:42:51 <warlord> I only have gpg2, no gpg
08:43:04 <warlord> gpg2 --version
08:43:04 <warlord> gpg (GnuPG) 2.1.13
08:44:45 <gjanssens> Oh, I had installed gpg, which appears to be available as well on F25
08:45:44 <warlord> Oh. GnuPG2 was already installed
08:45:45 <gjanssens> But gpg2 on my system also has the --export option
08:46:22 <gjanssens> ...as does gpg2 on F25...
08:46:26 * gjanssens is puzzled now
08:47:19 <simeon> gjanssens: gdb report: https://pastebin.com/HYta9L4R
08:48:51 <gjanssens> warlord: have you set gpg_home to point at the directory where the gpg keys are stored ?
08:49:25 <gjanssens> By default it will search for the keys in the gpg directory relative to your build script
08:49:25 <warlord> no. it's the 'default' directory
08:49:36 <gjanssens> If you stored it somewhere else, you should set that environment variable
08:49:39 <warlord> Oh, it wont use the gnupg default of ~/.gnupg?
08:50:12 <gjanssens> No. I probably should change that though
08:50:16 *** nimish has quit IRC
08:50:35 <gjanssens> Can you try with gpg_home set just to test for further issues ?
08:50:42 <warlord> Oh, well in that case... Yes.. give me a secondf..
08:50:49 *** nimish has joined #gnucash
08:51:01 <gjanssens> simeon: It appears you are using the xim input method.
08:51:20 <gjanssens> That is known to fail (I believe that's a gtk3 bug)
08:51:33 <gjanssens> Please change the input method and try again
08:51:34 <simeon> gjanssens: xim? What doest it mean? And what should I do?
08:51:54 <warlord> Still has the problem "gpg: Note: '--export' is not considered an option"
08:54:21 <simeon> gjanssens: Thanks a lot! :)
08:55:13 *** fabior has joined #gnucash
08:57:24 *** fell has quit IRC
08:59:39 <gjanssens> warlord: is the gpg_key parameter properly set ?
08:59:51 <gjanssens> I get that error if that one is unset
08:59:55 <warlord> gjanssens, it's weird, because gpg2 --help | grep -i export seems to imply the argument exists.
09:00:20 <warlord> gpg_key=6EE....
09:00:28 <warlord> which is what I see from "gpg2 -k"
09:01:24 *** fell has joined #gnucash
09:05:33 <gjanssens> ok, can you manually run that command (with parameters properly filled in:
09:05:41 <warlord> gjanssens, note, that I DID include the full fingerprint in gpg_key
09:06:03 *** Mechtilde has quit IRC
09:06:41 <gjanssens> gpg2 --homedir=<path to .gnugpg> --export <key fingerprint>
09:07:34 <gjanssens> what do you mean with "the full fingerprint" ? I have a single string with all uppercase hexadecimal characters, but nothing else
09:08:11 <warlord> It's like 32 characters long
09:08:15 <warlord> (vs 8)
09:08:54 <warlord> When I run that command manually it works,
09:08:56 <gjanssens> Yeah same here.
09:10:15 *** Mechtilde has joined #gnucash
09:10:58 <gjanssens> Ok, and what do you see if you add a 'set -x' at the start of the "prepare_gpg" function ? You may want to complement it with a 'set +x' at the end of that function to limit debug output.
09:11:17 <gjanssens> Is it properly detecting the value of gpg_key there ?
09:13:17 <warlord> Aha, it is missing "--homedir="
09:14:46 <warlord> Oh, I set gpg_home, not gpg_dir
09:15:26 <warlord> Okay, NOW it seems to be working better...
09:15:35 *** nimish has quit IRC
09:16:13 *** nimish has joined #gnucash
09:16:23 <warlord> gjanssens, I'll ping you when the build is done and we can talk about where to move/put it -- and how to best handle multiple branches
09:23:55 <gjanssens> Good
09:26:45 <simeon> gjanssens: It seems, you are right.
09:27:14 <simeon> gjanssens: "GTK_IM_MODULE="ibus" gnucash" solved the problem.
09:30:58 *** nimish has quit IRC
09:31:40 *** nimish has joined #gnucash
09:32:17 <gjanssens> simeon: I'm glad that worked
09:33:54 <warlord> gjanssens, got this:
09:33:59 <warlord> -- Installing: /app/share/glib-2.0/schemas/org.gnucash.window.pages.gschema.xml
09:33:59 <warlord> Compiling gschema files in /app/share/glib-2.0/schemas
09:33:59 <warlord> sed: can't read /app/share/appdata/*.xml: No such file or directory
09:33:59 <warlord> Error: module gnucash: Child process exited with code 2
09:37:22 *** kael has joined #gnucash
09:38:21 *** User has joined #gnucash
09:40:03 <gjanssens> Ah, yes, right before the gnucash 3.4 release, appdata has been moved to a new directory
09:40:18 <gjanssens> It should be /apps/share/metainfo/*.xml
09:40:24 <gjanssens> I'll update the build script
09:40:34 <warlord> gjanssens, okay. :)
09:41:25 *** User has quit IRC
09:42:52 <gjanssens> Pushed
09:42:57 <gjanssens> Can you retry ?
09:43:05 <gjanssens> You may have to remove the json file
09:43:27 <gjanssens> org.gnucash.GnuCash.json
09:43:52 <gjanssens> I'm not sure it will be regenerated, though not removing it may be an interesting test
09:46:25 *** nimish has quit IRC
09:47:19 *** nimish has joined #gnucash
09:49:51 *** tienne has joined #gnucash
09:53:08 *** nimish_ has joined #gnucash
09:54:29 *** nimish has quit IRC
09:54:29 *** nimish_ is now known as nimish
09:57:25 <warlord> gjanssens, rebuilding
09:57:42 <warlord> I did not remove anything.
09:58:07 <warlord> If something needs to be removed then it will need to be removed every time by the script, or the script will someone need to know when to remove it if not every time.
10:04:44 *** gour1 has joined #gnucash
10:05:20 <warlord> gjanssens, success
10:05:30 <warlord> So, next question: where to copy it to?
10:06:32 *** gour has quit IRC
10:07:55 *** nimish has quit IRC
10:08:23 *** nimish has joined #gnucash
10:08:54 <gjanssens> Yay!
10:09:32 <gjanssens> I was thinking of making it available under https://code.gnucash.org/builds/flatpak
10:09:55 <gjanssens> The build script can do the uploading for you if you specify an upload path
10:10:55 <gjanssens> But for flatpak refs to work you need to set both a path to be used by scp/rsync and one to the public side of it
10:11:08 <gjanssens> There are two parameters in custom.sh for this.
10:11:44 <warlord> Right.. But how do we properly deal with multiple branches?
10:12:04 <gjanssens> flatpak does that for you
10:12:22 <gjanssens> It will store all branches in one repo
10:12:42 <gjanssens> And users can specify the branch to install when running flatpak
10:12:51 <warlord> So even if I have multiple different flatpak trees it will figure it out during export?
10:13:05 *** nimish has quit IRC
10:13:10 <gjanssens> No, I don't so
10:13:26 <gjanssens> But I don't think you need multiple flatpak trees ?
10:13:33 *** Mechtilde has quit IRC
10:14:10 <warlord> right now I have a directory flatpak-maint/ where I build the maint branch -- I was expecting I would duplicate this and create a flatpak-master for master --- not sure how to do releases..
10:14:28 <gjanssens> I think the flatpak build is meant to start all from the same flatpak build environment
10:14:40 *** Mechtilde has joined #gnucash
10:14:49 <gjanssens> It handles branch separation itself
10:15:07 <warlord> So it doesn't keep the separate branch sources available?
10:15:09 <gjanssens> In that it will set up different build roots as needed
10:15:34 <gjanssens> It's been a while so I have to jog my memory
10:16:32 <warlord> So how do I tell it to build multiple branches? Or is that set up in the gnucash-on-flatpak tree itself?
10:17:28 <gjanssens> No, IIRC you simply pass the branch name to the build script as a command line argument
10:18:13 <warlord> To build_package.sh ?
10:18:25 <warlord> Or build_package.sh needs to be modified to build multiple branches?
10:18:26 <gjanssens> Yes: build_package -r <branch>
10:18:45 <gjanssens> or build_package -r <tag>
10:19:27 <warlord> Okay, so it's building branches, now.. Need to add smarts to new tags, I guess, similar to the gnucash-on-windows tree.
10:19:38 <warlord> (although I only have -r maint in there)
10:20:16 <gjanssens> If you do build_package.sh -r <tag> it will perform a release build
10:20:25 *** nimish has joined #gnucash
10:20:35 <gjanssens> so for example build_package.sh -r 3.4
10:21:29 <gjanssens> I think that's the only thing I haven't automated yet: detect when new tags appear in the repo
10:21:31 <warlord> Right -- -but I need to script to detect when there is a new tag.
10:21:52 <gjanssens> Yep
10:22:31 <gjanssens> At the time I wasn't sure we wanted release builds to happen fully autmated or not.
10:22:41 <gjanssens> Do we still do so on the Windows build server ?
10:23:03 <warlord> shall it be https://code... ?
10:23:08 *** nimish has quit IRC
10:23:22 <warlord> I thought we did, unless jralls changed it.
10:23:50 <warlord> The logic for that is all in the gnucash-on-windows ... it has the "do_build" script in there. Right now I'm sort of duplicating that here.
10:24:14 <gjanssens> Ok
10:24:18 <warlord> BUT I feel like windows is a "do_build per branch" -- i.e., I have a do_build maint, do_build master... ?
10:24:23 <warlord> I dont recall offhand.
10:24:26 <warlord> Anyways, ... https://code .. ?
10:24:30 <warlord> Or another URL?
10:25:11 <gjanssens> Isn't https://code what we promote everywhere for the nightly builds ?
10:25:37 <warlord> I think so, yes. So I'll keep that.
10:25:40 <gjanssens> Ok
10:26:05 <warlord> Testing now.
10:26:18 <warlord> Although I need to leave in about 5 minutes.
10:26:27 <gjanssens> On flatpak you can combine maint and master in one do_build_branches
10:26:49 <warlord> Right, just need to run build_package twice
10:26:56 <gjanssens> flatpak will do it's own evaluation of whether a rebuild is required, I didn't have to add anything about it
10:27:19 <gjanssens> detecting new tags is something we do have to add for release builds
10:27:51 <gjanssens> Just to clarify this a bit
10:28:26 <gjanssens> For branch builds you don't have to cache commit hashes and check whether a new commit was pushed
10:28:32 <gjanssens> flatpak does all that
10:28:52 <warlord> Right.
10:28:53 *** puck has quit IRC
10:29:12 <warlord> Although it is apparently rebuilding maint again...
10:29:12 <gjanssens> So for branch builds your wrapper is probably only calling build_package.sh <branch_name>
10:29:28 <warlord> Well, it's calling build_package.sh -r <branch_name>
10:29:35 <gjanssens> Eh right
10:30:56 <gjanssens> Hmm, perhaps existing build detection was added in a more recent version of flatpak-builder ? Or some parameter changed...
10:33:01 <warlord> well, https://code.gnucash.org/builds/flatpak now has data
10:34:20 <warlord> Next step: make this a nightly build.
10:36:05 <gjanssens> Yay :)
10:36:18 *** puck has joined #gnucash
10:36:30 <gjanssens> Note I have pushed an additional fix to the gnucash-on-flatpak repo
10:37:04 <gjanssens> As someone pointed out rightfully, the whole sed command can be removed as we fixed the uppercase issue it was running for.
10:38:01 <warlord> Okay. Well, my (nightly) script does a git pull on that repo, so it will get used the next time it's called.
10:39:31 <warlord> BIAW.
10:44:16 *** oozer has quit IRC
10:48:14 *** nimish has joined #gnucash
10:55:46 *** nimish has quit IRC
11:07:32 <gjanssens> jralls: Thomas' last comment on https://bugs.gnucash.org/show_bug.cgi?id=797039 has me thinking again on whether -Werror is really something we should enforce in our public repo
11:08:04 <gjanssens> His comment is not the only one making me reconsider this.
11:08:39 <gjanssens> But I don't have time to discuss this in more detail right now. Maybe somewhere next week I'll get back to this.
11:17:33 *** Mechtilde has quit IRC
11:23:00 *** Mechtilde has joined #gnucash
11:35:24 *** nimish has joined #gnucash
11:37:29 *** oozer has joined #gnucash
11:50:41 <warlord> gjanssens, I think we should still enforce -Werror because either it's really a bug, or it's documenting some undefined behavior. If it's undefined behavior it's better to actually know that explicitly.
11:57:51 <gjanssens> warlord: the idea is that everyone can decide for themselves to enable it and I think we should recommend it
11:58:07 <gjanssens> And us devs should set it locally anyway and perhaps even on our CI
11:58:23 <gjanssens> That's very different from forcing it on downstream packagers IMO
11:58:43 <gjanssens> But again, I'll have to defer a thorough discussion for later
12:00:13 <warlord> Well, I think we want it set by default when building from git, and we want it set by default when doing "make (dist)check". But I can see, arguably, that we might want to not have it set by default if building from a release tarball.
12:00:53 *** nimish_ has joined #gnucash
12:01:08 *** nimish has quit IRC
12:01:39 *** nimish_ is now known as nimish
12:07:10 <warlord> gjanssens, at what time of day should I fire off the nightly flatpak build?
12:07:22 <warlord> the docs get built at 1am US/ET
12:14:52 <gjanssens> I don't know. Right after the docs are finished perhaps ?
12:19:16 *** jrp_ has joined #gnucash
12:20:02 <warlord> Well, do we want to skip the flatpak build if the docs-build fails?
12:20:40 *** jrp has quit IRC
12:22:44 <gjanssens> No, because the docs build may fail in a part of the build that doesn't affect the flatpak builds
12:24:03 *** Mechtilde has quit IRC
12:27:02 *** Mechtilde has joined #gnucash
12:29:44 <warlord> Yeah, so -- maybe I just fire it off at 2am..
12:30:49 <warlord> Er, hmm.. The reason I chose 1am was to not be a problem at DST.. 2am doesn't help. So I could go at midnight (but the build might go longer than an hour) or I can do it at 4am (which means it'll only be starting at 10am your time)
12:31:05 <warlord> Decisions... decisions...
12:31:16 <warlord> Or I suppose I can move the docs build the midnight and put the flatpak at 1?
12:31:23 <warlord> s/the/to
12:34:04 *** zipppy has quit IRC
12:35:55 <warlord> OKay, decision made.
12:36:22 <gjanssens> Can't you start them sequentially from the same cron script ?
12:36:40 <gjanssens> And have them start at midnight looks like a reasonable compromise
12:36:42 <warlord> No, because if the docs fail it wont get to the next step
12:37:14 <gjanssens> I thought one could string commands separated with semi colons
12:37:32 <gjanssens> Or in yet another wrapper script :)
12:37:42 <gjanssens> But do as you see fit
12:38:02 <gjanssens> The exact timing is not that important I think
12:38:59 <warlord> The docs dont take an hour to build, even in the worst of times.
12:39:09 <warlord> This way we know when the flatpak build will start.
12:39:19 <warlord> instead of having to guess
12:39:34 <gjanssens> A reasonable point indeed
12:40:31 <gjanssens> If the docs build never take more than say half an hour you could advance the flatpak build even further
12:40:51 <gjanssens> Or if the docs build is not that heavy you could even consider starting them in parallel
12:41:08 <gjanssens> That would slow down the initial part of the flatpak build of course
12:41:16 <gjanssens> And only if enough resources are allocated
12:41:37 <gjanssens> (I mean only then parallel start is possible)
12:42:11 <gjanssens> That's just a wild thought of course ;)
12:42:12 <warlord> I probably have enough resources to do a parallel build.
12:42:25 <warlord> Honestly, it doesn't matter to me.
12:42:50 <warlord> Hmm, I only have code set up with 2 vCPUs.
12:45:14 <warlord> Anyways, I'm glad to finally get this off my plate! :)
12:45:23 <gjanssens> Me too! Thank :)
12:45:30 <gjanssens> Thank*s* !
12:45:43 <warlord> My pleasure :)
12:45:47 <warlord> Sorry it took so long.
12:45:49 <warlord> Happy New Year.
12:45:54 *** Laurentius has quit IRC
12:46:02 <gjanssens> You too :)
12:46:23 <gjanssens> warlord: will your nightly build detect the 3.4 tag and build it ?
12:46:34 <warlord> Currently no
12:46:50 <gjanssens> Ok, can you just trigger that one manually then ?
12:46:51 <warlord> it's just building maint every night. It's not doing any other branches or tag-detection.
12:47:08 <gjanssens> Would be nice to have a 3.4 to point users at when they ask for it
12:48:29 *** WebManOfFesto has joined #gnucash
12:49:01 <warlord> Um, that didn't work.
12:49:12 <gjanssens> ?
12:50:11 *** zipppy has joined #gnucash
12:50:18 <warlord> gjanssens, I emailed you the log.
12:51:23 <gjanssens> Ok
12:52:54 <gjanssens> Are you running this as root ?
12:54:32 <warlord> no
12:55:19 <warlord> 'but the last one wasn't run as root, either.
12:55:53 <gjanssens> It shouldn't
12:56:08 <gjanssens> I don't understand why it tries to access root's homedir
12:57:35 <warlord> I don't know, either.. Why would it do this now when it didn't a few moments ago when I built the maint branch?
12:58:29 <warlord> And re-trying maint, works just fine (but didn't notice it was the same commit -- unless something was committed in the last hour)
13:00:05 <gjanssens> I don't think anything was committed
13:00:31 *** nimish has quit IRC
13:00:52 <warlord> Then why would it rebuild maint?
13:01:51 <gjanssens> I don't know. As I replied earlier, perhaps existing build detection was only added in a more recent version of flatpak-builder
13:02:19 <gjanssens> But perhaps the dconf warning is just a red herring
13:02:36 <gjanssens> Right after is a message about a missing Sha256
13:04:09 <warlord> Yeah. it is a red herring... Just re-ran and didn't get the dconf issues.
13:04:18 <warlord> Creating new flatpak [gnucash=3.4, gnucash-docs=3.4]
13:04:18 <warlord> Emptying app dir 'build'
13:04:19 <warlord> Downloading sources
13:04:19 <warlord> Failed to download sources: module gnucash-docs: Sha256 not specified
13:04:19 <warlord> Uploading log file '/home/gnucash-docs/flatpak/logs/build-3.4-2019-01-14-13-03-28.log'
13:04:28 <warlord> so yeah, Sha256 issue
13:04:43 <warlord> I feel like we dealt with this somehow before?
13:05:47 <gjanssens> can you send me the generated org.gnucash.GnuCash.json manifest file ?
13:06:07 <gjanssens> It's building fine on my system with the newer flatpak-builder code
13:06:31 *** fabior has quit IRC
13:06:37 <gjanssens> The gnucash-docs module is a git module and as far as I understand it doesn't make sense to have an sha256 there
13:07:25 <warlord> where would I find that file?
13:08:37 <gjanssens> in your local gnucash-on-flatpak directory
13:09:45 <gjanssens> By the way it looks like jralls did push a commit to maint
13:10:43 <gjanssens> Eh no that was yesterday evening in your timezone
13:11:23 <warlord> gjanssens, mailed to you
13:12:52 <gjanssens> How odd, it's downloading tarballs from sourceforge in your manifest. And the sha256 is indeed missing.
13:13:02 <gjanssens> I'll investigate and get back on this.
13:13:07 <gjanssens> No 3.4 for now
13:13:09 <warlord> okay.
13:14:58 <gjanssens> Ok, at least I can reproduce. I missed the "-r" in my command line :(
13:15:34 <warlord> Ahh
13:15:41 <warlord> Oops. Well, glad you can reproduce.
13:16:06 <warlord> And strangely, my version of builder is supplying new logs even with the same commit-id
13:16:37 <warlord> hopefully I can upgrade code to F29 and then it'll be better. I've started upgrading some of my other infrastructure..
13:16:44 <gjanssens> Ok
13:16:47 <warlord> Then again, my mail server is still sitting at F20!!
13:16:58 <gjanssens> :)
13:17:25 <warlord> :-/
13:17:34 <warlord> Some things are easier to upgrade than others ;)
13:19:50 <gjanssens> True
13:20:26 <gjanssens> For my next upgrade somewhere in the coming months I'm seriously considering a move to a docker/kubernetes based setup
13:21:05 <gjanssens> I have done a first couple of tests with docker containers and once I got my head wrapped around it, I truly think it will help in upgrades
13:21:40 <gjanssens> I'd still want a hypervisor based lower level separation, but on top of that each service would be wrapped in a docker container
13:21:55 <gjanssens> I still have to look at kubernetes though
13:22:29 <gjanssens> Or perhaps I'd rather go podman/buildah instead, which seems to be a similar system from Redhat/Fedora
13:22:59 <gjanssens> Their new Fedora CoreOS is also something I want to learn more about by that time
13:28:01 <warlord> Hmmm
13:29:33 *** Mechtilde has quit IRC
13:29:59 *** Mechtilde has joined #gnucash
13:40:22 *** bertbob has quit IRC
13:43:27 *** bertbob has joined #gnucash
13:56:55 *** jerryq has joined #gnucash
14:05:33 *** Mechtilde has quit IRC
14:07:27 *** frakturfreak has joined #gnucash
14:16:34 <gjanssens> So the issue ith the flatpak tag build is we have a situation I hadn't anticipated: the last gnucash and gnucash-docs release tags are not the same: 3.4 an 3.4-1
14:17:04 <gjanssens> The tag build assumes they are
14:17:15 <gjanssens> So back to the drawing board here...
14:17:21 <gjanssens> I'll get back to that
14:17:40 <gjanssens> At least we have a maint nightly already
14:31:49 <warlord> I could do a master nightly, too, if you want. But it will install a new build log every night, even if there are no changes to the repo. (this is true of maint, too)
14:37:16 <gjanssens> Yeah true. The log is created every night regardless of whether or not there was a new build
14:37:17 *** kael has quit IRC
14:38:36 <gjanssens> Master is hardly changing currently so activating a nightly would only generate noise...
14:41:37 *** bertbob has quit IRC
14:42:39 <warlord> Okay, I wont activate it :)
14:44:24 *** User has joined #gnucash
14:45:09 *** kael has joined #gnucash
14:47:28 *** User has quit IRC
14:55:19 *** Laurentius has joined #gnucash
15:04:03 *** Laurentius has quit IRC
15:12:17 *** Laurentius has joined #gnucash
15:15:51 *** oozer has quit IRC
15:30:51 *** Laurentius has quit IRC
15:43:12 *** Laurentius has joined #gnucash
15:49:16 *** oozer has joined #gnucash
16:02:31 *** jonp` has joined #gnucash
16:03:30 *** Laurentius has quit IRC
16:03:30 *** jerryq has quit IRC
16:03:30 *** jrp_ has quit IRC
16:03:30 *** tienne has quit IRC
16:03:30 *** fell has quit IRC
16:03:30 *** boldstripe has quit IRC
16:03:30 *** GabrieleV has quit IRC
16:03:30 *** O01eg has quit IRC
16:03:30 *** CDB-Away has quit IRC
16:03:30 *** jethrogb has quit IRC
16:03:30 *** Yotson has quit IRC
16:03:30 *** jonp has quit IRC
16:03:30 *** mmkodali[m] has quit IRC
16:03:30 *** yyoshino has quit IRC
16:03:31 *** luwum[m] has quit IRC
16:03:31 *** peter-butler[m] has quit IRC
16:03:31 *** Farelli has quit IRC
16:03:31 *** Unhammer has quit IRC
16:03:31 *** luc14n0 has quit IRC
16:03:31 *** akv has quit IRC
16:03:31 *** jchonig has quit IRC
16:03:31 *** crossan007 has quit IRC
16:03:31 *** chf has quit IRC
16:03:31 *** gnomey has quit IRC
16:03:31 *** g5pw has quit IRC
16:03:31 *** arahael1 has quit IRC
16:03:31 *** MatrixTraveler[m] has quit IRC
16:03:31 *** Simon has quit IRC
16:03:32 *** jrp has joined #gnucash
16:03:35 *** chf has joined #gnucash
16:04:06 *** CDB-Away has joined #gnucash
16:04:25 *** boldstripe has joined #gnucash
16:04:29 *** tienne has joined #gnucash
16:05:06 *** jchonig has joined #gnucash
16:05:25 *** Farelli has joined #gnucash
16:06:18 *** fell has joined #gnucash
16:07:09 *** Simon has joined #gnucash
16:07:17 *** akv has joined #gnucash
16:07:23 *** crossan007 has joined #gnucash
16:10:02 *** GabrieleV has joined #gnucash
16:10:06 *** g5pw has joined #gnucash
16:10:39 *** Yotson has joined #gnucash
16:10:50 *** luc14n0 has joined #gnucash
16:10:51 *** yyoshino has joined #gnucash
16:10:52 *** gnomey has joined #gnucash
16:11:05 *** yyoshino has joined #gnucash
16:11:08 *** jethrogb has joined #gnucash
16:11:37 *** arahael1 has joined #gnucash
16:12:09 *** jerryq has joined #gnucash
16:12:37 *** O01eg has joined #gnucash
16:12:41 *** Unhammer has joined #gnucash
16:12:46 *** Laurentius has joined #gnucash
16:28:40 *** gjanssens has quit IRC
16:41:22 *** Agfarmer18 has joined #gnucash
16:44:52 *** peter-butler[m] has joined #gnucash
16:49:36 *** bertbob has joined #gnucash
16:53:28 *** Laurentius has quit IRC
16:55:07 *** mmkodali[m] has joined #gnucash
17:00:32 *** calvinct has joined #gnucash
17:00:46 *** jralls has quit IRC
17:00:57 *** jralls has joined #gnucash
17:00:57 *** ChanServ sets mode: +o jralls
17:02:52 *** Agfarmer18 has quit IRC
17:03:29 <jralls> warlord: I turned off the automatic tag-building on Windows. It's easier and fits my release workflow better to just log into the VM and fire it up after I've built the tarballs.
17:04:42 *** MoeD has joined #gnucash
17:06:19 <jralls> @tell gjanssens I had to re-tag and re-spin the docs release this time. There was a problem with the new build system that I didn't catch in time. The docs tag is 3.4a, not 3.4-1.
17:06:19 <gncbot> jralls: The operation succeeded.
17:10:31 <warlord> jralls, I think geert's issue was the main tag and docs tag being "different"
17:12:25 <jralls> warlord: Yeah, that will happen occasionally when I find something wrong between the first tag and finalizing the release. If I've already pushed the tag I can't really reuse it.
17:13:27 <jralls> IMO it's best to build releases from the tarballs. I don't know if flatpak build supports that.
17:14:44 <warlord> No clue
17:15:32 <jralls> warlord, fell: How familiar are you all with getting FOP to use fonts? The standard Times, Helvetica, and Courier don't support Cyrillic so the pdfs and probably epub/mobi are full of # where letters belong.
17:16:18 <chf> What is "FOP", jralls?
17:16:36 <jralls> I've installed fonts that are supposed to support all of Unicode but I can't get FOP to use it.
17:17:00 <jralls> chf: The XML Format-Object Processor.
17:17:33 <chf> What you mention ny names are ancient postscript fonts, they've traditionally no cyrillic letters.
17:17:46 <chf> Oh, thanks.
17:18:07 <chf> ny->by
17:18:27 <jralls> Yeah, got that part figured out. What I can't figure out is why FOP insists on using them when I'm telling it to use something better.
17:18:57 <chf> Just a guess: it uses non-cyrillic-aware versions that are installed as well?
17:19:17 <fell> https://xmlgraphics.apache.org/fop/2.1/fonts.html
17:19:32 <jralls> fell: thx
17:19:59 <fell> warlord, which fop version is installed?
17:20:18 <warlord> fop-2.0-3.fc24.noarch
17:21:04 <jralls> That's what I've been trying to use. I'm using version 2.1 on Debian stretch. The doc build is automated only on Windows. For the website versions I build locally and upload.
17:21:08 <warlord> If (when?) I upgrade code to F29, it'll be fop-2.2-2.fc29
17:24:05 <chf> What fell posted also mentions that some applications perform internal replacements by name, so could some libraries do.
17:24:39 <chf> Please note that recent versions of Adobe Acrobat Reader replace "Helvetica" with "Arial" and "Times" with "Times New Roman" internally. GhostScript replaces "Helvetica" with "Nimbus Sans L" and "Times" with "Nimbus Roman No9 L". Other document viewers may do similar font substitutions.
17:24:44 <jralls> The errors I get are similar to FOUserAgent - Glyph "П" (0x41f, Pecyrillic) not available in font "Times-Italic".
17:25:36 *** Aussie_matt has joined #gnucash
17:26:39 <jralls> chf: This is happening before the reader stage. I think the fonts will have to be in the pdf.
17:27:03 * warlord has to run
17:27:05 <warlord> TTFN
17:28:59 *** Wilsona has joined #gnucash
17:29:29 <Wilsona> Hi All,
17:30:23 <chf> Hello, Wilsona.
17:30:59 <Wilsona> Hi Chf!
17:31:21 <Wilsona> one question! Any place or somebody to download a template with the PGC?... (Official Plan General Contable or General Accounting Plan)
17:31:31 <Wilsona> Spanish PGC.... ;-)
17:33:28 <fell> Until now nobody sentus the PGC as template.
17:34:01 <chf> No clue, unfortunately. If you set your language to spanish, however, Gnucash should offer you the already included templates for Spain.
17:34:57 <chf> Chances are that nobody has yet made a Gnucash template of the PGC.
17:35:14 <fell> If you have sometime to create one, https://wiki.gnucash.org/wiki/Translation#How_to_translate_the_files_containing_the_new_account_hierarchies
17:36:10 <fell> and https://wiki.gnucash.org/wiki/Account_Hierarchy_Template
17:36:59 *** calvinct has quit IRC
17:38:30 <chf> I've made a german template and updated another one myself; if you decide to make one for the PGC, Wilsona, I can give you some advice, but consider to invest at least 60 hours of work.
17:38:40 *** badger92 has quit IRC
17:38:44 *** badger92 has joined #gnucash
17:41:12 <fell> it depends on the number of accounts, if you can automatc convert some document from the web in our xml format, ...
17:41:45 <chf> I speak a bit spanish, so don't fear to be left alone with the task, but that's normal spanish, not "accounting jargon".
17:44:21 <fell> https://es.wikipedia.org/wiki/Plan_General_de_Contabilidad seems a good starting point to understand the plan
17:56:27 *** gour1 has quit IRC
17:58:15 *** jerryq has quit IRC
18:02:27 *** boldstripe has quit IRC
18:03:23 *** boldstripe has joined #gnucash
18:06:27 *** luwum[m] has joined #gnucash
18:06:33 *** ordex has quit IRC
18:08:41 *** ordex has joined #gnucash
18:27:28 *** WebManOfFesto has quit IRC
18:49:50 *** jerryq has joined #gnucash
18:50:53 *** trying2Simplify has joined #gnucash
18:56:29 <fell> jralls: in guide/ja we have a fop.xconf
18:58:34 <fell> and related ttfs in fonts
19:00:28 *** MatrixTraveler[m] has joined #gnucash
19:00:48 *** kael has quit IRC
19:27:25 *** WebManOfFesto1 has joined #gnucash
19:35:46 <jralls> fell: Yes, I'd tried copying that but didn't get the value for embed-url or font-triplet name right. Once I figured that out I got the russian docs to build.
19:36:07 <jralls> Now I'm working on integrating it into the build properly.
20:07:08 *** nimish has joined #gnucash
20:10:42 *** nimish has quit IRC
20:14:38 *** nimish has joined #gnucash
20:19:24 *** frakturfreak has quit IRC
20:39:09 *** ordex has quit IRC
20:41:17 *** ordex has joined #gnucash
20:53:20 *** badger92 has quit IRC
20:53:33 *** badger92 has joined #gnucash
21:19:21 *** oozer has quit IRC
21:32:25 *** tienne has quit IRC
21:39:23 *** storyjesse has joined #gnucash
21:51:49 *** trying2Simplify has quit IRC
21:53:54 *** WebManOfFesto1 has quit IRC
22:09:38 *** simeon has quit IRC
22:12:21 *** nimish has quit IRC
23:24:56 *** ordex has quit IRC
23:27:05 *** ordex has joined #gnucash