Shawn O. Pearce wrote:
> Ian Clatworthy <> wrote:
>> Shawn O. Pearce wrote:
>>> If we are going to start parsing the message content we probably
>>> should have some sort of escape that allows the frontend to protect
>>> some of it from the replacement parsing, so literal :n can be passed
>>> through at the same time that a different :n is updated with the
>>> revision identifier.
>> Would it be better to explicitly mark commit messages needing expansion?
>> By using a 'template' command in place of a 'data' command say?
> How does that help escaping ":8080" in a commit message that needs
> this replacement for ":43" but not for ":8080" (as the latter was
> literally in the source)?

Good point - it doesn't.

I was thinking more about stopping accidental interpolation, assuming
the vast majority of messages wouldn't want it, regardless of whatever
templating syntax we selected.

Ian C.

Mailing list:
Post to     :
Unsubscribe :
More help   :

Reply via email to