bug#67117: [PATCH] Tree-sitter: fix an issue when searching subtree backward

[Yuan Fu]

> On Nov 18, 2023, at 10:40 PM, Eli Zaretskii <eliz@gnu.org> wrote:
> 
>> From: Yuan Fu <casouri@gmail.com>
>> Date: Sat, 18 Nov 2023 22:13:49 -0800
>> Cc: dvzubarev@yandex.ru,
>> 67117@debbugs.gnu.org
>> 
>> 
>> 
>>> On Nov 18, 2023, at 9:47 PM, Eli Zaretskii <eliz@gnu.org> wrote:
>>> 
>>>> From: Yuan Fu <casouri@gmail.com>
>>>> Date: Sat, 18 Nov 2023 10:47:47 -0800
>>>> Cc: Denis Zubarev <dvzubarev@yandex.ru>,
>>>> 67117@debbugs.gnu.org
>>>> 
>>>> 
>>>>>> Steps to reproduce the issue:
>>>>>> 1. emacs -Q
>>>>>> 2. M-x find-file /tmp/t.py
>>>>>> 3. paste to the buffer
>>>>>> 
>>>>>> Temp(1, 2) 
>>>>>> 
>>>>>> 4. M-x python-ts-mode
>>>>>> 5. Call search-subtree with backward flag
>>>>>> M-x eval-expression (treesit-search-subtree
>>>>>>             (treesit--thing-at (point) "call")
>>>>>>             (lambda (n) (equal (treesit-node-type n ) "integer"))
>>>>>>             t)
>>>>>> It should return the second int node (#<treesit-node integer in 9-10>), 
>>>>>> but it returns nil.
>>>>> 
>>>>> Yuan, any comments to the patch and the issue in general?
>>>> 
>>>> LGTM. Thanks Denis, and Eli :-)
>>> 
>>> Should we install this on the emacs-29 branch or on master?
>> 
>> Emacs-29, I’d say, since it’s a bug fix. Why do you ask?
> 
> Because I want to install it, obviously.

Of course ;-) I mean you are usually the people who answer this type of 
questions, and our convention is to install bug fixes on emacs-29, so I was 
wondering if there’s some other considerations that I don’t know about.

Yuan