Re: master fbd2ea1 2/2: Fix two parse-time-months invalid assumptions

Re: master fbd2ea1 2/2: Fix two parse-time-months invalid assumptions

[Michael Albinus]

eggert@cs.ucla.edu (Paul Eggert) writes:

Hi Paul,

> --- a/lisp/net/tramp-smb.el
> +++ b/lisp/net/tramp-smb.el
> @@ -1814,6 +1814,9 @@ Result is the list (LOCALNAME MODE SIZE MTIME)."
>  	    (if (and sec min hour day month year)
>  		(encode-time
>  		 sec min hour day
> +		 ;; FIXME: Won't this fail if parse-time-months is configured
> +		 ;; by the user?  See "The date/time prompt" in the Org manual.
> +		 ;; If the code is OK as-is, perhaps explain why in a comment.
>  		 (cdr (assoc (downcase month) parse-time-months))
>  		 year)
>  	      tramp-time-dont-know))

Thanks for the hint. I understand `parse-time-months' that one could
*add* own entries, w/o removing existing ones. That's why nobody has
reported an error here for the last 15+ years.

But you are right, it is not guaranteed. So I have adapted the code in
order to ensure proper `parse-time-months' value.

Best regards, Michael.

Re: master fbd2ea1 2/2: Fix two parse-time-months invalid assumptions

[Lars Ingebrigtsen]

Michael Albinus <michael.albinus@gmx.de> writes:

> Thanks for the hint. I understand `parse-time-months' that one could
> *add* own entries, w/o removing existing ones. That's why nobody has
> reported an error here for the last 15+ years.

Yes, removing entries from those variables will break stuff.  Perhaps
the doc string should just say so?

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no