On Jul 16, 9:38 am, Peng Yu <[EMAIL PROTECTED]> wrote: > On Jul 15, 10:29 pm, Gary Herron <[EMAIL PROTECTED]> wrote: > > > > > Peng Yu wrote: > > > Hi, > > > > The following code snippet is from /usr/bin/rpl. I would like the it > > > to match a word, for example, "abc" in ":abc:". But the current one > > > would not match "abc" in ":abc:". I tried to modify it myself. Would > > > you please let me know what is the corrected way to do it? > > > > Thanks, > > > Peng > > > > if opts.whole_words: > > > regex = re.compile(r"(?:(?<=\s)|^)" + re.escape(old_str) + > > > r"(?=\s| > > > $)", > > > opts.ignore_case and re.I > > > or 0) > > > -- > > >http://mail.python.org/mailman/listinfo/python-list > > > The regular expression "\w+" will match (what might be your definition > > of) a word, and in particular will match abc in :abc:. Regular > > expressions have lots of other special \-sequences that might be worth > > your while to read about: http://docs.python.org/lib/re-syntax.html > > > Gary Herron > > I didn't read the docs and tried the following code. > > regex = re.compile(r"\A" + re.escape(old_str) + r"\Z", > opts.ignore_case and re.I or 0) > > But I'm not sure why it is not working. > > Thanks, > Peng
Not sure why you picked \A and \Z -- they are only useful if you are using the re.M flag. What you want is \b -- match word boundary, on either side of your word: regex = re.compile(r"\b" + re.escape(old_str) + r"\b",re.I) re.I is the same as re.IGNORECASE. More than one option may be OR'ed together. There's no such thing as "re.O" in Python. I can understand where you get the idea, as there is an 'o' modifier for REs in Perl. To summarize, \A and \Z match the beginning and end of a STRING, while \b matches the beginning or end of a WORD. -- john -- http://mail.python.org/mailman/listinfo/python-list
