Re: [Caml-list] new emacs tuareg mode release

["Stéphane Glondu" <steph@glondu.net>]

Sam Steingold a écrit :
>>  I have the same problem with 22.3.1
>>
>>  playing around, this construction seems to be causing some invalid modifier
>> errors:
>>  (skip-syntax-backward "\s-")
>>
>>  in tuareg-find-colon-typespec and other places
> 
> what kind of error?
> I.e., start emacs like this:
> $ emacs -q --no-site-file
> and type in the *scratch* buffer and tell me what you type and cut and
> paste the error message from the *Messages* buffer.

On emacs 22.3.1,

(skip-syntax-backward "\s-")

gives the backtrace:

Debugger entered--Lisp error: (error "Invalid modifier in string")
  read(#<buffer *scratch*>)
  preceding-sexp()
  eval-last-sexp-1(nil)
  eval-last-sexp(nil)
  call-interactively(eval-last-sexp)
  recursive-edit()
  byte-code("\306^P
@\307=\203!^@\310\311\312\"\210\313\311!\211^ZA@)\242\314=\203!^@\310\315\312\"\210\316^K!\210\317
\210\320     !\210\f\203c^@\321ed\"^MV\203W^@eb\210\322^$
  debug(error (error "Invalid modifier in string"))
  read(#<buffer *scratch*>)
  preceding-sexp()
  eval-last-sexp-1(nil)
  eval-last-sexp(nil)
  call-interactively(eval-last-sexp)

(and nothing in *Messages*)

By the way, I wasn't able to find documentation for this construct. The
"\s" is responsible, but I have no idea what is its semantics. I have
not been able to observe a behaviour different than with just
(skip-syntax-backward "-").


Cheers,

-- 
Stéphane