#linuxcnc-devel Logs

Mar 30 2022

01:51 AM pere: andypugh: <URL: https://github.com/LinuxCNC/linuxcnc/pull/1687 > - small improvement lifted from my po4a branch.
02:37 AM pere: I see several warnings when building linuxcnc. is there any plan to get rid of the compiler warnings?
03:14 AM hansu: Is 'LinuxCNC:petterreinholdtsen/translate-po4a' now our feature branch for po4a? I am a bit confused because all po4a related PRs are still targeting to master.
03:21 AM pere: hansu: I renamed my branch to translate-po4a-close and continue on this one, as the old translate-po4a PR was closed when pulled into 'LinuxCNC:petterreinholdtsen/translate-po4a', and the old translate-po4a is unfit for merge into master.
03:22 AM pere: need to fix the github and buildbot build failures somehow.
03:24 AM pere: hansu: hope my comment in <URL: https://github.com/LinuxCNC/linuxcnc/pull/1689 > cleared up the confusion.
03:55 AM hansu: yes, thank you
03:55 AM hansu: I am talking about this branch: https://github.com/LinuxCNC/linuxcnc/tree/petterreinholdtsen/translate-po4a
03:56 AM hansu: NOT https://github.com/petterreinholdtsen/linuxcnc/tree/translate-po4a ;-)
03:58 AM pere: hansu: I know. It was the branch andypugh created to test my changes with the buildbots.
03:58 AM pere: As far as I know, I can not update it.
04:01 AM hansu: But am I wrong that we wanted to create a branch on the official repo where all po4 changes go in before merging them to master?
04:02 AM pere: I got the same impression. do not believe 'LinuxCNC:petterreinholdtsen/translate-po4a' is the one.
04:03 AM hansu: Ok. Should I create one? "translate-po4a"? Or other suggestions for the name?
04:08 AM pere: hm, perhaps doc-and-translations-staging or something like that?
04:15 AM hansu: hmm too long :D "po4a-doc-translation"?
04:16 AM hansu: i created a label to po4a related stuff. I labeled two PRs. Did I miss one?
04:20 AM pere: depends, do you want to tag the closed ones too?
04:21 AM hansu: not really, only want to see the "open todos"
04:22 AM pere: then I guess you have the lot
04:35 AM pere: what is the story with docs/help/rtfaults.adoc having two languages?
04:44 AM hansu: Moin!
04:44 AM hansu: Do you see the history?
04:47 AM smoe: No, sorry, I just went online.
04:47 AM hansu: http://tom-itx.no-ip.biz:81/~tom-itx/irc/logs/%23linuxcnc-devel/2022-03-30.html
04:48 AM hansu: there a) a question for you and b) we are trying to find a name for the po4 development branch
04:51 AM smoe: There are two kinds of developments for the docs. Maybe that is why it is all so confusing. Let me try to recap.
04:54 AM smoe: Development 1) The documentation is just a tiny bit inconsistent in how Asciidoc is supposed to be used to show tables, or to decide what there should be a table, vs a description vs an itemization vs just some short paragraphs. There is the idea spearheaded by @silopolis to give this all a bit of a guidance via a style-doc. So, any reformatting should go somewhere quickly to avoid conflicts.
04:58 AM smoe: Development 2) The preparation of .po files to seed the weblate needs to have _es and _fr in sync with _en. Whenever something in _fr/_es is found superior to _en, that change should make it into _en to avoid effort in _fr/_es to be lost. So we need something to synchronize. That is master today, which basically means that we do not synchronize. But this could be solved in any fork of ours.
05:00 AM smoe: Development 3) The po4a build system of @pere needs something to run an test with the Ubuntu 20.04 setup (there is nothing newer available) of the Github CI. This is the branch we had discussed how I understood it. Just, @pere's work eliminates the _fr and _es files, so we cannot merge our .po-prep with it.
05:02 AM smoe: Weblate) Once we have it up, please take it for a fact that because of the missing cross-references will break the regular build. Not once - but many times. Basically with every commit this will happen :) It may be reasonable to have a branch that does not check cross-references or at least does not fail because of it but just reports them and provides docs to look at.
05:02 AM smoe: That is what I see.
05:05 AM hansu: 1) leads me to the question: How are tables handled by po4a? Is every cell a separate translation string?
05:05 AM smoe: I do not think I want to know. But cell-wise is what I expect.
05:07 AM smoe: @siloplis (working on _fr) and me (working on _es - still not believing that I am saying this, but I only need to read in those files or fix syntax) have not started the syncing of our work. This will be another 4 to 6 patches, I think, to reach some fix point from which to generate the .po for Petter. I admit I would want this to happen in master, just because it is less overhead - for us.
05:11 AM hansu: Hmm this depends a bit how much these changes are visible for the devel docs. Don't want to confuse the readers much more.
05:15 AM smoe: These should not be visible at all - just introduce/remove white space and move between languages what is already in the documentation today.
05:16 AM smoe: But, I agree, once this is done, the documentation should receive a good proofread by everyone prior to the release of 2.9.
05:18 AM smoe: I see this on our todo list independently from the _es and _fr backfeeds now. It would be preferable if we had a style-guide for the documentation. There would then be little discussion about something needing a change or not.
05:18 AM hansu: Okay then it's no problem for them to go in master. But If there need to be inserted some paragraphs again, then this should better go in a dev branch I supoose.
05:19 AM smoe: Great. I agree.
05:20 AM smoe: https://github.com/LinuxCNC/linuxcnc/pull/1675 ?
05:25 AM hansu: Yes that's next on my todo list ;-)
05:28 AM smoe: Ok. I need to reach out to @silopolis and his work on the _fr files.
09:55 AM pere: smoe: did you try to build my branch?
09:55 AM smoe: @pere, have not tried this, yet. Would like to sync with @siloplis first.
11:03 AM hansu: petter, how do you generate the po files from the existing adoc files taking the current translations into account? Is this something po4a is shipped with?
11:04 AM hansu: smoe What is the purpose of the doc-base files?
11:06 AM smoe: It points users of Debian to documentation that is availble. There should not be any changes except for me having adapted for the _fr files that got unified to the _en schema. The _ to - change I had reverted.
11:07 AM hansu: I only wonder why you removed the /usr/share/doc/linuxcnc/LinuxCNC_HAL_fr.pdf
11:13 AM pere: hansu: I generally use 'make -C src translateddocs'
11:14 AM pere: hansu: it was removed because it was a unique french collection of documents, not matching the english documentation.
11:15 AM hansu: okay
11:16 AM hansu: And the target 'translateddocs' is part of your makefile on your po4a branch, right?
11:18 AM pere: hansu: yes.
11:19 AM pere: there is also the pofiles target, which only do part of the job, updating the PO files. the translateddocs target will generate the translated docs too.
11:21 AM smoe: @pere already answered for me.
11:22 AM hansu: okay I will have a look how that magic works later :)
11:22 AM smoe: A "main achievement" of that branch was to a) remove English files with an _es or _fr ending and b) to have the same files contributing to each document - if existing.
11:23 AM smoe: po4a will generate all files, so all files do exist once we are in @pere's build env
11:24 AM hansu: But why did you remove touchy_es.adoc?
11:24 AM hansu: Yes I understand
11:26 AM smoe: Did I make a mistake and that was truly Spanish?
11:31 AM hansu: It seems so
11:33 AM hansu: But that is the only one
11:35 AM smoe: I just got the diff from master and squashed this with the commit the removed the file. Thank you for spotting this.
11:40 AM hansu: Hmm when I look on the diff the file is still deleted ...
12:12 PM JT-Shop: https://hosted.weblate.org/projects/mesa-configuration-tools/
12:12 PM JT-Shop: does that look like it's configured correctly?
12:24 PM hansu: Looks good from the user point of view (y)
12:32 PM JT-Shop: thanks
12:46 PM andypugh: pere smoe hansu silopolis[m] I looked for po4 branches, and found that some work on this was done back in 2015….https://github.com/LinuxCNC/linuxcnc/tree/seb/master/po4a
12:53 PM hansu: interesting. Seems it never made it into master
01:33 PM pere: andypugh: interesting. lets hope we manage to get it into master this time.
01:36 PM linuxcnc-build: build #1868 of 1660.rip-buster-python3 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1660.rip-buster-python3/builds/1868 blamelist: Hans <hansunzner@gmail.com>, Steffen Moeller <moeller@debian.org>
01:36 PM andypugh: i wonder if seb_kuzminsky remembers why it stalled?
02:08 PM linuxcnc-build: build #2269 of 1640.rip-buster-rtpreempt-amd64 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1640.rip-buster-rtpreempt-amd64/builds/2269 blamelist: Hans <hansunzner@gmail.com>, Steffen Moeller <moeller@debian.org>
02:41 PM linuxcnc-build: build #2027 of 1650.rip-buster-rtpreempt-rpi4 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1650.rip-buster-rtpreempt-rpi4/builds/2027 blamelist: Hans <hansunzner@gmail.com>, Steffen Moeller <moeller@debian.org>
02:41 PM linuxcnc-build: build #8668 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/8668 blamelist: Hans <hansunzner@gmail.com>, Steffen Moeller <moeller@debian.org>
03:22 PM -!- #linuxcnc-devel mode set to +v by ChanServ
05:58 PM -!- #linuxcnc-devel mode set to +v by ChanServ
06:00 PM linuxcnc-build: build #1869 of 1660.rip-buster-python3 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1660.rip-buster-python3/builds/1869 blamelist: Steffen Moeller <moeller@debian.org>, andypugh <andy@bodgesoc.org>
06:00 PM linuxcnc-build: build #2270 of 1640.rip-buster-rtpreempt-amd64 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1640.rip-buster-rtpreempt-amd64/builds/2270 blamelist: Steffen Moeller <moeller@debian.org>, andypugh <andy@bodgesoc.org>
06:10 PM linuxcnc-build: build #2028 of 1650.rip-buster-rtpreempt-rpi4 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1650.rip-buster-rtpreempt-rpi4/builds/2028 blamelist: Steffen Moeller <moeller@debian.org>, andypugh <andy@bodgesoc.org>
06:10 PM linuxcnc-build: build #8669 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/8669 blamelist: Steffen Moeller <moeller@debian.org>, andypugh <andy@bodgesoc.org>
06:27 PM linuxcnc-build: build #1870 of 1660.rip-buster-python3 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1660.rip-buster-python3/builds/1870 blamelist: elovalvo <ernesto.lovalvo@unipa.it>, Steffen Moeller <moeller@debian.org>, Hans Unzner <hansunzner@gmail.com>, andypugh <andy@bodgesoc.org>, Jeff Epler
06:27 PM linuxcnc-build: <jepler@gmail.com>
06:28 PM linuxcnc-build: build #2271 of 1640.rip-buster-rtpreempt-amd64 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1640.rip-buster-rtpreempt-amd64/builds/2271 blamelist: elovalvo <ernesto.lovalvo@unipa.it>, Steffen Moeller <moeller@debian.org>, Hans Unzner <hansunzner@gmail.com>, andypugh <andy@bodgesoc.org>,
06:28 PM linuxcnc-build: Jeff Epler <jepler@gmail.com>
06:29 PM andypugh: Expect a lot of this, chaps, I am trying to clear out a pull request backlog.
06:37 PM linuxcnc-build: build #2029 of 1650.rip-buster-rtpreempt-rpi4 is complete: Failure [4failed compile] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1650.rip-buster-rtpreempt-rpi4/builds/2029 blamelist: elovalvo <ernesto.lovalvo@unipa.it>, Steffen Moeller <moeller@debian.org>, Hans Unzner <hansunzner@gmail.com>, andypugh <andy@bodgesoc.org>, Jeff
06:37 PM linuxcnc-build: Epler <jepler@gmail.com>
06:37 PM linuxcnc-build: build #8670 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/8670 blamelist: elovalvo <ernesto.lovalvo@unipa.it>, Steffen Moeller <moeller@debian.org>, Hans Unzner <hansunzner@gmail.com>, andypugh <andy@bodgesoc.org>, Jeff Epler <jepler@gmail.com>
07:48 PM roguish[m]: andypugh: You're doing a hell of a job............ let's go 2.9
09:09 PM -!- #linuxcnc-devel mode set to +v by ChanServ