re.match

[Bob Cannard]

"Michael P. Soulier" wrote:
> 
>     Hey people. Could someone explain the purpose of the match function in the
> re module? If it's only difference from search is to match at the beginning of
> the line, why does it exist? That's what the ^ in the regexp is for.

I'd like to add my own recent experience as a Python newbie getting to
grips with its implementation of regexps. Once upon a time there was
a regular expression which matched a - well, let's call it a Thing.
There were 3 questions that needed to be asked:

   - Does this string contain any Things? (re.search, obviously)

   - Find all the Things in this string. (re.findall)

   - Is this string a Thing? (re.match - er, no, that doesn't work,
     it only asks if this string begins with a Thing.)

Life would have been perfect if re.match, or some similar function,
required a match of the entire string against the regular expression.
But it doesn't, and there doesn't seem to be a function that does,
so the regular expression spawned another with a dollar sign on the
end. (Curious assymetry.) Alternatively the regexp could have been
left uncompiled so it could be manipulated, but that would have
caused a loss of speed and generality in my lower-level functions.
Another possibility would have been to combine compiled regular
expressions, but there doesn't seem to be a way to do this. Does
anyone know of a function or piece of magic that I've missed?

Anyway, my point is that there does seem to be a case, from my
experience, for having different functions that do different things
with the same regular expression. If anything I'd like to see more
of them.

Cheers,

       Bob.