On Mon, Feb  9, 2026 at 16:31 (+0100), Arash Esbati wrote:

> Hi Keita,

> Ikumi Keita <[email protected]> writes:

>> Thank you for your report. Though I haven't understand the culprit yet,
>> I noticed a difference between `TeX-LaTeX-sentinel' and
>> `TeX-ConTeXt-sentinel'. Could you try the following patch if you know
>> how to use it?

>> diff --git a/context.el b/context.el
>> index 25cbeb34..d248859f 100644
>> --- a/context.el
>> +++ b/context.el
>> @@ -606,7 +606,7 @@ for a label to be inserted after the sectioning command."
>>             (message (concat name ": problems after "
>>                              (TeX-current-pages)))
>>             (setq TeX-command-next TeX-command-default)))))
>> -  (unless TeX-error-list
>> +  (unless (TeX-error-report-has-errors-p)
>>      (run-hook-with-args 'TeX-after-compilation-finished-functions
>>                          (with-current-buffer TeX-command-buffer
>>                            (expand-file-name

> Thanks for looking at this issue.  I applied your suggestion and tried
> it, but no avail.  Maybe Jim has different results.

Unfortunately, not.

> I only had a brief look, and I think we have to touch
> `TeX-TeX-sentinel-check' again.  First, I think we should do this to
> context.el:

> --8<---------------cut here---------------start------------->8---
> diff --git a/context.el b/context.el
> index 25cbeb34..001d765f 100644
> --- a/context.el
> +++ b/context.el
> @@ -573,7 +573,7 @@ for a label to be inserted after the sectioning command."
>     ;; Mark IV
>     ((with-current-buffer TeX-command-buffer
>        (string= ConTeXt-Mark-version "IV"))
> -    (cond ((TeX-TeX-sentinel-check process name))
> +    (cond ((not (TeX-TeX-sentinel-check process name)))
>            ((re-search-forward "fatal error: " nil t)
>             (message (concat name ": problems after "
>                              (TeX-current-pages)))
> --8<---------------cut here---------------end--------------->8---

Did that change do anything for you?  In my case, it didn't even give me
the "... errors ... Use C-c ` to display" message.


Seeing "Mark IV" in that code did give me one concern...
I suspect most (if not virtually all) ConTeXt users at now using LMTX.
I wonder if any differences between MkIV and LMTX are significant for
AUCTeX.

> ,----[ C-h f TeX-TeX-sentinel-check RET ]
>> TeX-TeX-sentinel-check is a native-comp-function in ‘tex.el’.

>> (TeX-TeX-sentinel-check PROCESS NAME)

>> Cleanup TeX output buffer after running TeX.
>> Return nil only if no errors were found.

> `----

> In the case above, the first cond-clause wins because
> `TeX-TeX-sentinel-check' returns non-nil (it finds an error), so the
> search for "fatal error" never happens.

FWIW, I don't follow you there.  Isn't there a search for "fatal error"
exactly when TeX-TeX-sentinel-check returns non-nil?

Further, ConTeXt (apparently) never outputs "fatal error", so searching for
that doesn't seem fruitful.

> Next, ConTeXt emits a "fancy" error message.  Consider this file:

> --8<---------------cut here---------------start------------->8---
> \starttext
> \xyzzy
> \stoptext

> %%% Local Variables:
> %%% mode: ConTeXt
> %%% TeX-master: t
> %%% ConTeXt-Mark-version: "IV"
> %%% End:
> --8<---------------cut here---------------end--------------->8---

> Running context on it says:

> --8<---------------cut here---------------start------------->8---
> ! Undefined control sequence
> tex error       > tex error on line 2 in file ./context-test.tex: 

> <line 3.2>
> --8<---------------cut here---------------end--------------->8---

> Now take this plain-TeX file:

> --8<---------------cut here---------------start------------->8---
> % \starttext
> \xyzzy
> % \stoptext

> %%% Local Variables:
> %%% mode: plain-TeX
> %%% TeX-master: t
> %%% End:
> --8<---------------cut here---------------end--------------->8---

> Running tex on it gives:

> --8<---------------cut here---------------start------------->8---
> (./context-test.tex
> ./context-test.tex:4: Undefined control sequence.
> l.4 \xyzzy

> )
> --8<---------------cut here---------------end--------------->8---

> which AUCTeX parses correctly.  Again, I only had a very brief look, so
> apologies if the above is off.

Cheers.
                                Jim



_______________________________________________
bug-auctex mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/bug-auctex

Reply via email to