As Athanasius's examples here demonstrate, the \s class requires a whitespace character (if * ? quantifiers are not present), and that's not what you have if you chomp off the newline. You may be interested to investigate the $ \z \Z line end anchors. In your particular case, \Z (big-Z) might be useful to "Match only at end of string, or before newline at the end" (see perlre); a \Z would match match the end of either a chomp-ed or un-chomp-ed line (assuming you're chomp-ing newlines).
Update: Fixed ^ vice $ in "... the ^ \z \Z line end anchors." per Lotus1's /msg. Thanks!
Give a man a fish: <%-{-{-{-<
| [reply]
[d/l]
[select] |
Since I'm matching chomped single lines, not multiple lines, is there something which "\Z" can do which "$" does not?? (Did not know about "\z" anchor - that may have been put in after I learned perl in 1996.)
| [reply] |
My personal regex best practices (solidly in line with TheDamian's as given in his PBP book) include reserving ^ $ for matching around embedded newlines (which means that the /m regex modifier is always asserted) and using \A \z \Z as "absolute" string start/end anchors. To the best of my recollection, \A \z \Z were introduced with Perl 5, sometime prior to 1996 (update: but see hippo's correction).
... is there something which "\Z" can do which "$" does not??
It's the other way around: ^ $ match also around embedded newlines (with the ever-present /m modifier), while \A \z \Z match only at string start/end; hence, in a sense, they do less. The advantage of this practice is that you never have to think about what ^ $ do; also, \A \z \Z cannot be modified.
Give a man a fish: <%-{-{-{-<
| [reply]
[d/l]
[select] |