emacs-orgmode
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: bug#47937: 27.1; Invisible text property updated only for a portion


From: Arthur Miller
Subject: Re: bug#47937: 27.1; Invisible text property updated only for a portion of buffer
Date: Thu, 22 Apr 2021 14:36:03 +0200
User-agent: Gnus/5.13 (Gnus v5.13) Emacs/27.1 (windows-nt)

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Arthur Miller <arthur.miller@live.com>
>> Cc: 47937@debbugs.gnu.org
>> Date: Wed, 21 Apr 2021 22:10:45 +0200
>> 
>> >> Because you have font-lock-mode turned on.  And JIT font-lock begins
>> >> by wiping out all the text properties.
>> >
>> > I should clarify this, I guess: this is specific to Org buffers, see
>> > org-unfontify-region.  And "all text properties" is an exaggeration:
>> > it removes many properties, but not all of them.
>> 
>> Allright, thank you for the excellent clarification. I guess I should go
>> for custom visibility spec instead of plain invisible property, so that
>> my text property survive font lock.
>> 
>> Please close the bug and thanks.
>
> I'm closing the bug, but I suggest to take this up with Org
> developers, because I think org-unfontify-region removes too many
> properties in a way that is too indiscriminate.  They should ideally
> only remove the properties they themselves add.

I just today looked at the function, and I see docs says it should
remove fontification from links, which can partly be set to invisible
since org mode hide markup, so it is probably legit what they do. I am
not sure if that function runs on entire buffer. I'll CC this to
org-mail list so we can see what they say.

I am not really at home with font-lock and text properties, but as it
seems to me from this experience, is that proper way would be to use
visibility spec, and create custom property, since so many different
modes vill fight for default 'visibility property. However I was lazy,
so I just made sure to fontify entire buffer before I do my thing, but
it is not very robust approach :).



reply via email to

[Prev in Thread] Current Thread [Next in Thread]