Matthieu Moy <[EMAIL PROTECTED]> writes:
> Mark Triggs <[EMAIL PROTECTED]> writes:
>
>> I'm not all that familiar with tla's new file escaping stuff, but could
>> we have our own elisp implementation instead?
>
> Merge from the mainline, there's something better: escaping is done
> only when the file contains characters other than "[^a-zA-Z._+,{}-]"
> (actually, we don't use /escaping/ in xtla, but it's implemented just
> in case), and unescaping is done only if the escaped string contains
> '\'.
Cool, that'll work too. Which patch is that in? *tla-missing* isn't
showing any missing patches at the moment.
>> [ lisp code ]
>>
>> seems to work for me.
>
> For the cases you have tested. But escaping is done for many more
> characters than this :
>
> $ touch été
> $ tla add été
> $ tla inventory
> [...]
> U \(U+E9)t\(U+E9)
> $ _
>
> Implementing all this in pure elisp would be probably even slower ...
Ah, damn. I overlooked the "else" block in the hackerlab code :o) I
agree that it would probably be too slow.
Anyway, I'll test my updated replay code with your tla-unescape changes
and commit a patch.
Cheers,
Mark
--
Mark Triggs
<[EMAIL PROTECTED]>
