#linuxcnc-devel Logs
Nov 20 2021
#linuxcnc-devel Calendar
12:00 AM pere: silopolis: are you working on the linuxcnc translation?
12:06 AM silopolis: pere yes... well, at least I want to try !
12:06 AM silopolis: I'm french btw
12:06 AM pere: silopolis: right. I would recommend waiting with the translation, and work on the translation migration first.
12:19 AM silopolis: pere followed this on the ML and my first ToDo entry was to assert french docs can be dumped
12:19 AM silopolis: let's read this first then !
12:21 AM Tom_L: yes it is bridged with matrix
12:25 AM silopolis: Tom_L Hi, thanks... Must be something I don't get in Element to join here
12:27 AM pere: silopolis: one minor problem is the use of roff macros in some manual pages, which po4a do not understand. I disabled translating these manual pages in 086fe01ac0a22c9e389a42cb8538843ba3c72a95, but hope to find a solution for them later.
12:27 AM pere: s/roff macros/roff macro definitions/
12:32 AM silopolis: pere another, for Crowdin, could be the mixing of the translations in the same tree vs separate en/, es/, fr/ subtrees
12:33 AM silopolis: but have to refresh myself about this and make a test
12:33 AM pere: silopolis: nah, that is fairly trivial to handle in the po4a.conf file.
12:33 AM silopolis: oh ? nice !
12:35 AM silopolis: pere will renew Crowdin setup asap pulling from the translate-po4a branch
12:49 AM pere: silopolis: if you look at po4a.conf there, you will notice the asciidoc translations are alongside the original, while the man translations are in separate directories.
12:58 AM silopolis: Another thing that catches eyes are the image files at the root of the tree :/ Would find it cleaner to have them in an "images" or "pictures" subdirectory, or even a "media/images" subtree
01:00 AM pere: silopolis: I did not try to adjust the file structure in the po4a patch, both to reduce its scope and because I assume the current structurure is the way the developers want it...
02:11 AM silopolis: pere aaaall around :'(
02:12 AM silopolis: pere do you want me to fix this in main/en doc too ?
04:46 AM silopolis: pere ^
04:47 AM pere: silopolis: did not see anything between 08:34 and 11:32 CET.
04:48 AM silopolis: pere oh yes, gotta get used to IRC again sorry
04:49 AM pere: silopolis: any progress on the french migration?
04:49 AM silopolis: pere about fixed line width... Was telling you that there was all around and offering to fix this
04:50 AM pere: I suspect the restructuring should go into master if it consist of translation improvements, and the po4a branch if it is adding 'dummy' strings.
04:52 AM silopolis: as fixed line width removal is reaaly linked/needed for gettext magic to really happen, I'd vote to include this in structural refactoring going with po4a... But you're the boss ;-)
04:56 AM pere: silopolis: do you have an example of what kind of change you talk about?
05:01 AM pere: For example my docs/src/Master_Integrator.pdf look good with linear right and left margins. not sure what kind of problem you are talking about.
05:01 AM silopolis: @pere afk, will talk later
05:02 AM pere: see you! :)
06:10 AM silopolis: re
06:22 AM pere: :)
07:11 AM silopolis: pere see this screencap please
07:16 AM silopolis: pere lines squared in both languages are sentences cut, in both en and fr by fixed line width, but not the same number of lines. Besides breaking the semantics of paragraphs in both languages, it also forbids po4a/gettext to manage corresponding blocks
07:20 AM silopolis: pere for semantically sound pot/po files, main documentation has to have cleanly formated blocks
07:25 AM silopolis: pere screencap link should help ! https://i.imgur.com/z0MOKOH.png
07:29 AM pere: silopolis: I assumed those were intended to be separate paragraphs.
07:33 AM silopolis: no they're not sadly
07:35 AM silopolis: pere and see here how this looks like in Crowdin https://i.imgur.com/GnY3H7s.png
07:44 AM pere: silopolis: right. then I understand, and I agree. gettext work best with one paragraph per string.
08:03 AM silopolis: pere well, further investigations tend to show po4a is smarter than Crowdin, most probably because it understands asciidoc syntax (and so ignores single linebreaks), whereas ATM Crowdin treats it like raw text files
08:08 AM silopolis: pere see this cap https://i.imgur.com/meXNY7u.png where the same sentence as in previous picture is actually recognised as a single block, whereas it is split on several lines in source
08:08 AM Tom_L: or
08:10 AM silopolis: Tom_L oh nice , thank you ! :+1:
08:10 AM Tom_L: it's not the most reliable but i try to keep it going
08:11 AM Tom_L: CaptHindsight[m], how do you join devel with libera?
08:11 AM Tom_L: seems my account is still on freenode
08:12 AM silopolis: men, that french documentation is in such state... '=(
08:33 AM Tom_L: silopolis, are you running riot client?
08:34 AM silopolis: Tom_L element on laptop and mobile
08:34 AM Tom_L: i thought i figured out how to join it but i'm not sure yet
08:36 AM Tom_L: add the server to your server list
08:36 AM Tom_L: libera.chat
08:44 AM silopolis: Tom_L can't find server list anywhere
08:44 AM Tom_L: i'm not that familiar with it
08:45 AM Tom_L: still looking myself
08:46 AM Tom_L: looks like i made it with riot
08:47 AM Tom_L: silopolis, join #linuxcnc-devel:libera.chat
08:47 AM Tom_L: try that
09:14 AM silopolis[m]: Tom_L: u there ?
09:17 AM * silopolis there too '=D
09:19 AM pere: silopolis: good to hear. :)
09:21 AM silopolis: pere for po4a smartness I presume lol
09:21 AM silopolis: pere
09:22 AM silopolis: pere (sorry) for french translation, that's another story...
09:41 AM Tom_L: am now
10:15 AM silopolis[m]: Tom_L: works like a charm !
10:17 AM Tom_L: great
10:17 AM silopolis[m]: Tom_L: thanks a lot for your help 🙏
12:59 PM CaptHindsight[m]: https://matrix.to/#/#linuxcnc-devel:libera.chat here is the room link if you get bumped and the bread crumb trail is gone
03:25 PM linuxcnc-build: build #1554 of 1660.rip-buster-python3 is complete: Failure [4failed compile runtests] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1660.rip-buster-python3/builds/1554 blamelist: Rene Hopf <renehopf@mac.com>, ddotldot <94646173+ddotldot@users.noreply.github.com>
03:27 PM linuxcnc-build: build #1957 of 1640.rip-buster-rtpreempt-amd64 is complete: Failure [4failed compile runtests] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/1640.rip-buster-rtpreempt-amd64/builds/1957 blamelist: Rene Hopf <renehopf@mac.com>, ddotldot <94646173+ddotldot@users.noreply.github.com>
05:20 PM linuxcnc-build: build #8355 of 0000.checkin is complete: Failure [4failed] Build details are at http://buildbot.linuxcnc.org/buildbot/builders/0000.checkin/builds/8355 blamelist: Rene Hopf <renehopf@mac.com>, ddotldot <94646173+ddotldot@users.noreply.github.com>