On Wednesday, May 26, 2021 10:00 AM, pelzflorian (Florian Pelz) <pelzflorian@pelzflorian.de> wrote:

> On Tue, May 25, 2021 at 10:03:01PM -0400, Julien Lepiller wrote:
>
> > Le 25 mai 2021 16:16:42 GMT-04:00, Luis Felipe luis.felipe.la@protonmail.com a écrit :
> >
> > > #:authors (G_ "Jane Roe, Eva Lua and Luis Felipe")
> >
> > I think so, but I think we'll have to test, because I don't really know the i18n process either for the website ^^'
>
> Yes, it is correct and the best way, but testing by running the
> sexp-xgettext script is good too.
>
> > > That's what I thought, and I was about to ask if it was possible to
> > > have <date> and <authors> instead of <1> and <2>. I guess not...
> > > I need a course on the i18n system :)
>
> I wanted the sexp-xgettext i18n system so developers would not have to
> add a tag name to every element … Well, we can use regexp helper
> function find-empty-element like website/sexp-xgettext.scm uses at the
> beginning of the deconstruct procedure:
>
> (define (find-empty-element msgstr name)
> "Return the regex match structure for the empty tag for XML
> element of type NAME inside MSGSTR. If the element does not exist or
> is more than the empty tag, #f is returned."
> (string-match (string-append "<" (regexp-quote name) "/>") msgstr))
>
> and call it on a translation with G_, but I would prefer not to
> increase code complexity in the website.

Ok, here is the new patch with the suggested changes.

I updated the POT to check the translation strings and I think they are well formed. For example:

#. TRANSLATORS: <1/> is a date, and <2/> a list of authors.
#: apps/media/templates/components.scm:50
#, xml-text
msgid "Published <1/> by <2/>"
msgstr ""

I didn't include the POT update in the patch though. Maybe for future patches I'll do the whole i18n process.