Re: [AUCTeX] Patch to tex-info.el to support node name completion in @..ref commands, and solve a few other pbs

[Vincent Belaïche]

Dear Mosé,

No there was no strong reason for assoc-string rather than assoc, the
point was just that when you know that the key is a string assoc-string
is supposed to be faster.

My opinion --- with no regard to this specific patch --- is that it
would better in the case of XEmacs to define assoc-string as an alias of
assoc if the function is unbound.

Furthermore, I must say that the way I wrote this patch is quite generic
because I have anticipated that there are other specials characters
(hashsign, backslash, etc...) that we may want some day to
de-escape. However currently the map contains only the comma character,
so using assoc[-string] to search a map of one element might be a little
over-engineered. The code could be simplified without use of a map if 
everybody is content with not anticipating any other escapable characters.

VBR,
    Vincent

----------------------------------------
> From: address@hidden
> Date: Sun, 8 Nov 2015 01:06:26 +0100
> Subject: Re: [AUCTeX] Patch to tex-info.el to support node name completion in 
> @..ref commands, and solve a few other pbs
> To: address@hidden
> CC: address@hidden
>
> Hi Vincent,
>
> 2015-10-25 21:15 GMT+01:00 Vincent Belaïche <address@hidden>:
>> Dear AUCTeX experts,
>>
>> I would like to submit to you the attached patch.
>>
>> [...]
>
> XEmacs doesn't have `assoc-string', is it possible to use `assoc' or
> there was a strong reason for using `assoc-string' instead?
>
> Bye,
> Mosè