Re: Publishing on Worg succeeds or fails? -- I can't say!

[Fabrice Niessen <fni-news-TA4HMoP+1wHrZ44/DZwexQ@public.gmane.org>]

Nicolas Goaziou wrote:
> Fabrice Niessen writes:
>
>> Could that be stated clearly in the output?
>
> You should ask to whom implemented Worg.

And that is?  Bastien?

>> The thing is that I did not touch those links, as you can see in my
>> commit. So, somehow, Org became more strict. And the fact that that
>> file wasn't modified for some time means it wasn't re-exported for
>> the same period; hence, the problem was never found -- until someone
>> modifies whatever in the file...
>
> Correct.
>
>> IOW, shouldn't all files be published each time, to spot such
>> problems as soon as they become problems (because of syntax changes,
>> for example), instead of being much later, whenever the file is
>> changed (if it ever is)?
>
> So you get 50-100 files to fix instead of the one you modified?

No much fun, that's right.

> IMO, it would be a waste to re-publish unmodified files. You only have
> to consider syntax changes when updating a file.

OTOH, I find it uncomfortable to have a "project" (in the publishing
sense) which is, in fact, not publishable anymore, because many files
won't export anymore -- some could even still rely on Org 7...

But I understand your point. The question comes down to what we want:
a uniform, coherent set of Org files (all up-to-date, and exportable) or
a set of HTML pages which we don't touch anymore if their Org
counterpart is not touched either.

> Note that you need to have a very recent Org (less than 7 hours) to use
> it.

Fixed the problems. Now that page is up-to-date on Worg again. Thanks.

Best regards,
Fabrice

-- 
Fabrice Niessen
Leuven, Belgium
http://www.pirilampo.org/