geiser-edit-symbol-at-point (M-.) does not work on fork+exec-command

  • Open
  • quality assurance status badge
Details
2 participants
  • Arun Isaac
  • Ludovic Courtès
Owner
unassigned
Submitted by
Arun Isaac
Severity
normal
A
A
Arun Isaac wrote on 14 May 2019 22:29
(address . guix-patches@gnu.org)(name . Ludovic Courtès)(address . ludo@gnu.org)
cu7a7fo93vf.fsf@systemreboot.net
geiser-edit-symbol-at-point (M-.) does not work on
fork+exec-command. geiser wrongly tries to lookup fork instead of
looking up fork+exec-command. This is because (thing-at-point 'symbol)
does not correctly identify the whole symbol. And, that, in turn, is
because ?+ has been reclassified as an expression prefix character (')
in .dir-locals.el. It should be a symbol constituent character (_) which
is the default for scheme mode.

(modify-syntax-entry ?+ "'")

This is a rather trivial bug. But, I thought I'll just document it
here. I'm not sure if we should fix it. I can't think of any easy way
out.
-----BEGIN PGP SIGNATURE-----

iQEzBAEBCAAdFiEEf3MDQ/Lwnzx3v3nTLiXui2GAK7MFAlzbJSUACgkQLiXui2GA
K7Ohuwf/SVeW6Ml2ctmmCVyBiSJ7Kk6G7+ivch8NWzV+HpSkHxZ2xCqKwCvZI53p
C0hI8801fd7p/55w65C8MgKhiuQl0TboOsh3djmlGCgZ5O7WsVGB27xpurwxwDWz
l1jWDA4vW9IVGeHXDMHwVnVatYTWny7WkbznyfVOjGwK+3VHaBt6Q1piMpjsK8pi
5NFVx8T+OQI01Esc7a+iHiR67aAaPgKpHnnGsBKGj6+AXiEsOzMDU175uXDagpwF
rdAeaydM9ExNE3TpLMre8Uo8946RabbiU4OJjiprT/YYFpRBNxQwQfwcKxKNH4LO
Cnj+iXW6NKnd8TPe1gxeshb/Ay88RA==
=bn5T
-----END PGP SIGNATURE-----

L
L
Ludovic Courtès wrote on 15 May 2019 14:40
(name . Arun Isaac)(address . arunisaac@systemreboot.net)(address . 35736@debbugs.gnu.org)
87y337vql5.fsf@gnu.org
Hi,

Arun Isaac <arunisaac@systemreboot.net> skribis:

Toggle quote (10 lines)
> geiser-edit-symbol-at-point (M-.) does not work on
> fork+exec-command. geiser wrongly tries to lookup fork instead of
> looking up fork+exec-command. This is because (thing-at-point 'symbol)
> does not correctly identify the whole symbol. And, that, in turn, is
> because ?+ has been reclassified as an expression prefix character (')
> in .dir-locals.el. It should be a symbol constituent character (_) which
> is the default for scheme mode.
>
> (modify-syntax-entry ?+ "'")

What about proposing this change to ‘scheme-mode’ to Emacs upstream?

Alternately, or in the meantime, we could carry that fix in Geiser
maybe?

Thoughts?

Ludo’.
A
A
Arun Isaac wrote on 16 May 2019 09:06
(name . Ludovic Courtès)(address . ludo@gnu.org)(address . 35736@debbugs.gnu.org)
cu74l5u98v7.fsf@systemreboot.net
Toggle quote (5 lines)
> What about proposing this change to ‘scheme-mode’ to Emacs upstream?
>
> Alternately, or in the meantime, we could carry that fix in Geiser
> maybe?

I don't think reporting this to scheme-mode or geiser would help. They
are justified in classifying ?+ as a symbol constituent character
(_). It is we (the Guix project) who have reclassified ?+ as an
expression prefix character (') in a "non-standard" way for our use in
gexps. WDYT?
-----BEGIN PGP SIGNATURE-----

iQEzBAEBCAAdFiEEf3MDQ/Lwnzx3v3nTLiXui2GAK7MFAlzdC9wACgkQLiXui2GA
K7M9fQf+LEM/nPo7ryeMktDeg3WGbMj0gKjjJFjC7NH9mL7bOMwjXzFBjfSyYdKW
ADUPrara+HYiiyH4jDf1tuyF0UxQHq36FC/T3xH6X2E1mtxzKT/POP3yssYXDpk2
Ip8yQGslTLynze1PjbN8kgHxWJBCviBAEC2bteCWW/DU9ZD+iHdcEjxRapskVv9e
6fwlRzi0Iy7Y4LXpUMr1N/pocAeU7yeLfRH7IVAtId/tncC+jsrvqki7IZ5p7e4g
B6yxbRlN/d5tDqDprjojgd1Oncvxm08CKkDhQwpFhP6Oq7FV4RHnsfOFGalyLELJ
ILSxQqgK3CNug/GEKvOEczWndxstJw==
=oGKr
-----END PGP SIGNATURE-----

L
L
Ludovic Courtès wrote on 16 May 2019 11:53
(name . Arun Isaac)(address . arunisaac@systemreboot.net)(address . 35736@debbugs.gnu.org)
87a7fmiv2z.fsf@gnu.org
Arun Isaac <arunisaac@systemreboot.net> skribis:

Toggle quote (11 lines)
>> What about proposing this change to ‘scheme-mode’ to Emacs upstream?
>>
>> Alternately, or in the meantime, we could carry that fix in Geiser
>> maybe?
>
> I don't think reporting this to scheme-mode or geiser would help. They
> are justified in classifying ?+ as a symbol constituent character
> (_). It is we (the Guix project) who have reclassified ?+ as an
> expression prefix character (') in a "non-standard" way for our use in
> gexps. WDYT?

Oh true, sorry!

Well I don’t know, I’m confused now. I vaguely remember prior
discussions about this with Alex Kost, but I forgot what the conclusions
were. :-/

Ludo’.
?