Since you want all the rest of the string, how about
match = filename.gsub(/-?\d+.(jpg|gif)$/,‘’)
Or to match it with a regex, add a (.?) in front
match = /(.?)-?\d+.(jpg|gif)$/.match(filename)
which will capture your string in match[1]
The .*? is a non-greedy match, so it’ll match only until the rest of the
regex does. \d is the digit character class (saves writing [0-9]). The
gsub’s a lot more efficient, though.
And here’s the sexeger page I referred to earlier:
http://japhy.perlmonk.org/sexeger/sexeger.html
martin
···
the Bare grrr@wild.net wrote:
Hi:
Is there anyway to have a regular expression match from the end of a
string? I would like to match something like the following:
TextThatIWantToMatchThatMayHaveNumbersAndHyphensInIt-0123.EXT
The hyphen may or may not be there. For example, from
“TK-23-ab-355.bnd” I would want to find “TK-23-ab”.
Because, the text I want to match may have numbers in it, it seems to me
that I somehow have to match from the end of the string. I want to I’ve
tried expressions like the following:
match = /^(.)[^0-9][0-9]+.(jpg|gif)$/i.match(filename)
But it didn’t work (of course!). Is there anyway to do what I want
without actually reversing the string?
“sexeger” sounds very interesting for optimization. In my study of
regular expressions, I have never encountered starting from the end of
a string.
Because it eliminates backtracking, Martin DeMello’s first suggestion
may be substantially faster (you would still need to delete any file
name that wasn’t modified by the substitution operation, however). If
every filename examined will have a .jpg or .gif (or .png, etc.) ending
you could also eliminate the alternation part of your regex (just use
.…$).
Other optimization techniques may work for you, but would have to be
benchmarked. Jeffrey Friedl’s “Mastering Regular Expressions” has a
lot of tips for optimization.
···
On Sunday, December 1, 2002, at 02:42 AM, Martin DeMello wrote:
the Bare grrr@wild.net wrote:
Hi:
Is there anyway to have a regular expression match from the end of a
string? I would like to match something like the following:
TextThatIWantToMatchThatMayHaveNumbersAndHyphensInIt-0123.EXT
The hyphen may or may not be there. For example, from
“TK-23-ab-355.bnd” I would want to find “TK-23-ab”.
Because, the text I want to match may have numbers in it, it seems to
me
that I somehow have to match from the end of the string. I want to
I’ve
tried expressions like the following:
match = /^(.)[^0-9][0-9]+.(jpg|gif)$/i.match(filename)
But it didn’t work (of course!). Is there anyway to do what I want
without actually reversing the string?
Since you want all the rest of the string, how about
match = filename.gsub(/-?\d+.(jpg|gif)$/,‘’)
Or to match it with a regex, add a (.?) in front
match = /(.?)-?\d+.(jpg|gif)$/.match(filename)
which will capture your string in match[1]
The .*? is a non-greedy match, so it’ll match only until the rest of
the
regex does. \d is the digit character class (saves writing [0-9]). The
gsub’s a lot more efficient, though.
And here’s the sexeger page I referred to earlier:
http://japhy.perlmonk.org/sexeger/sexeger.html
martin