Hi Christian,

Thanks for your feedback on this (and there is absolutely no urgency
here, so please don't increase the time you spend on Worg.)

> 1. What does the CLI do?

The index.html file at https://bzg.fr/org-faq/index.html is generated in
two steps. In a directory with org-faq.org (like the worg/ directory),
you can run

~$ org-parse -r html org-faq.org > org-faq.json
~$ topics org-faq.json

org-parse is a parser that takes an org file as input, parses it, output
it as json. Here we use the "-r html" which renders the contents of the
sections as HTML within the json file.

topics is a script that takes a json file and exports it as a small html
page. The json file is expected to contain "topics" (here FAQ entries).

So the cron job would run both scripts (or more likely a shell script
calling them both) every day to generate the standalone index.html file.

I acknowledge that this adds some complexity: I commit to maintaining
the cron job and the scripts, which should not be a burden for anyone
other than me.

> 2. Switch back to standard?

The proposal is to have both https://orgmode.org/worg/faq and the
current page.  We can add a link between both, of course.

> 3. Numbering / referencing
>
> With regard to the questions raised by Gregor about numbering
> (https://list.orgmode.org/[email protected]) and the further
> discussion about date-added numbering
> (https://list.orgmode.org/[email protected]):
>
> I agree with you that numbering should be kept out of your views and
> that numbering by date is confusing (without conveying all that much
> helpful information).
>
> Since numbering changes, references to the FAQ should preferably use
> permalinks (custom IDs). Your solution makes those helpfully available
> at a click, so no further action is needed.

Note that if CUSTOM_ID is present in an entry, topics will use it as the
permalink.

Thanks!

-- 
 Bastien

Reply via email to