Re: [Emacs-diffs] emacs-25 db436e9: Don't call debug on failed cl-assert

["Clément Pit-Claudel" <cpitclaudel@gmail.com>]

On 2017-01-29 10:37, Noam Postavsky wrote:
> On Fri, Jan 27, 2017 at 10:46 PM, Noam Postavsky
> <npostavs@users.sourceforge.net> wrote:
>> On Fri, Jan 27, 2017 at 10:30 PM, Clément Pit--Claudel
>> <clement.pit@gmail.com> wrote:
>>> On 2017-01-27 20:59, Noam Postavsky wrote:
>>>> I guess let-binding `debugger' to a function which performs the
>>>> logging should do the trick?
>>>
>>> I already do that, in fact, and I do re-throw the exception from there.  Looks like things didn't work because of the way `debugger' is called in `cl--assertion-failed'.  Is that call correct?  The argument in (funcall debugger `(cl-assertion-failed ,form ,string ,@sargs)) doesn't seem to match the docs:
>>>
>>>     If due to error, args are ‘error’ and a list of the args to ‘signal’.
>>>
>>> Is the call just missing an 'error argument?
>>>
>>
>> Oh yeah, I guess it is.
> 
> Um, should this be
> 
> (funcall debugger 'error `(cl-assertion-failed ,form ,string ,@sargs))
> 
> or
> 
> (funcall debugger 'error `(cl-assertion-failed (,form ,string ,@sargs)))
> 
> I'm getting confused with all the levels of nesting and funcall/apply.


I think the second one.  `debugger' gets two arguments, 'error and a list of arguments to `signal'.  That list should have two entries: `cl-assertion-failed' and `(,form ,string ,@sargs).

Thanks!