bug#35476: font-lock-{append, prepend}-text-property and anonymous faces

[Kévin Le Gouguec]

Noam Postavsky <npostavs@gmail.com> writes:

> Kévin Le Gouguec <kevin.legouguec@gmail.com> writes:
>
>> Now that Stefan committed a fix to Emacs 27, and Nicolas a workaround
>> to org-mode, this bug report can probably be closed; I just have a few
>> questions left:
>>
>> 1. Shouldn't Stefan's fix also be applied to
>>    font-lock-append-text-property?
>>
>> 2. Is it worth adding the test suite I posted in bug#35476#8 to the
>>    Emacs repository?
>
> I'd say the answers are yes and yes.  So would you mind adding the fix
> for font-lock-append-text-property to your patch?

Fix and test suite:

0001-Stop-splicing-anonymous-faces-in-font-lock-append-te.patch
Description: Text Data

Further refactoring (not really necessary; feel free to close the report
without applying):

0002-Extract-common-code-for-adding-text-properties.patch
Description: Text Data

Both patches were only "lightly" tested, i.e. by C-x C-e'ing every
function and test, then calling ert-run-tests-interactively; AFAICT the
tests pass[1].  Haven't run a full 'make check' yet (M-: insert-excuse
'battery); I don't think anything broke when Stefan committed the fix to
font-lock-prepend-text-property though.

Still pretty new to this, so let me know if I messed up anything
(e.g. commit message format, conventions when adding files, functions or
tests).

Thank you for your time.



[1] Although running 'make lisp/font-lock-tests' in the test/ folder
    fails on the append test.  Could it be that the Makefile runs the
    tests against the old font-lock.el that is installed on my system,
    rather than the new one in my repository?  I took a look at the
    test_template in test/Makefile, but I could not understand what the
    machinery loads at a glance.