Re: Introducing face in comments for various modes

[Thibaut Verron <thibaut.verron@gmail.com>]

On 13/12/2022 11:40, Heime wrote:
> ------- Original Message -------
> On Tuesday, December 13th, 2022 at 10:05 AM, Thibaut Verron <thibaut.verron@gmail.com> wrote:
>
>
>> On 13/12/2022 10:46, Heime wrote:
>>
>>> ------- Original Message -------
>>> On Tuesday, December 13th, 2022 at 9:35 AM, Thibaut Verron thibaut.verron@gmail.com wrote:
>>>
>>>> On 13/12/2022 09:30, Heime wrote:
>>>>
>>>>> ------- Original Message -------
>>>>> On Tuesday, December 13th, 2022 at 8:04 AM, Heime heimeborgia@protonmail.com wrote:
>>>>>
>>>>>> ------- Original Message -------
>>>>>> On Tuesday, December 13th, 2022 at 7:52 AM, Jean Louis bugs@gnu.support wrote:
>>>>>>
>>>>>>> Heime, I like the idea of highlighting comments, though I do it this
>>>>>>> way to specify (syntax comment-start) as that works for multiple
>>>>>>> modes.
>>>>>>>
>>>>>>> ; one
>>>>>>> ;; two
>>>>>>> ;;; three
>>>>>>> ;;;; and more
>>>>>>>
>>>>>>> ;;; Highlighting comments
>>>>>>>
>>>>>>> (setq rcd-regexp-comment (rx line-start
>>>>>>> (one-or-more (syntax comment-start))
>>>>>>> (one-or-more space)
>>>>>>> (group (one-or-more not-newline))
>>>>>>> line-end))
>>>>>>>
>>>>>>> ;; (highlight-regexp regexp nil 1)
>>>>>>> ;; (unhighlight-regexp regexp)
>>>>>>>
>>>>>>> Jean
>>>>>>> I would like to introduce (syntax comment-start) in place
>>>>>>> of ";;" in "^;;\s+\\[.+\\].*$".
>>>> It's not something you can "introduce" in your regexp, it only makes
>>>> sense within the context of a regexp built with rx:
>>>> https://www.gnu.org/software/emacs/manual/html_node/elisp/Rx-Notation.html
>>>>
>>>> Rewriting your regexp as an rx regexp would not be too difficult if you
>>>> want to go that route, Jean's example is a good template.
>>>>
>>>> Otherwise, as an approximation of that feature, you could built your
>>>> regex string using the value of the variable comment-start ( ";" in
>>>> emacs-lisp-mode) instead of hardcoding ;; .
>>>> How can one specify two comment characters next to each other?
>>
>> Using the variable, the same way you would normally do for a regexp, by
>> putting \{2\} after the matched regexp.
>>
>> E.g.
>>
>> ELISP> (format "\\(%s\\)\\{2\\}" comment-start)
>>
>> "\\(;\\)\\{2\\}"
>>
>> (don't trust me for the number of escape characters)
>>
>> But I don't think it's really what you want: for example, in a mode
>> where comment-start is "# " (e.g. org or python), this will match "# # "
>> but not "##".
> Let's disregard org and python at this time.
>   

Even then, avoiding the need to carefully count string expansions for 
backslashes would be enough reason to use syntax tables for me.


>> Using the syntax table is better here, as the character # alone has the
>> syntax of a comment-start.
>>
>> I had forgotten about the simplest option, that is the regexp equivalent
>> of the rx syntax: \s< in a regexp will match exactly a character with
>> the syntax of a comment-start.
>>
>> So: "\s<\{2\}" should be exactly what you want (at least in languages
>> where comments start with a single character).