Ramprasad A Padmanabhan wrote: > > James Taylor wrote: > > > > On Jan 2, 2004, at 5:38 PM, James Taylor wrote: > >> > >> I'm trying to parse a bit out of an HTML file, where the formatting > >> could change daily as far as spaces/newlines go. Say for example I > >> have something like this: > >> > >> $str=<<EOF; > >> <html> > >> <body> > >> <p>Hello this is juts an example</p> > >> <p><!---begin---><a href="nowhere.com">blahahahaha</a> > >> </p><a href=" > >> http://www.somewhere.com";> > >> HELLO</a> > >> </p><!---end--->Hello world</body> > >> </html> > >> EOF > >> > >> $repl="Replacement Text"; > >> > >> $str =~ s/\<!---begin---\>.+?\<!---end---\>/$repl/im; > >> > >> > >> Well, that doesn't work. > >> > >> Any suggestions on this one? I thought /m was supposed to make > >> regex's span multiple lines, which seems to be the problem here, as : > >> > >> print "good\n" if $str =~ /<!---begin--->.+?<!---end--->/mi > >> > >> comes up with nothing, though I am able to match them on their own. > >> Thanks > > ok, well I ended up dropping the regex and splitting the strings and > > re-building them after being inspired by another thread. > > > > so, > > > > my $str="text <!---begin---> > > text to > > be replac > > ed<!---end--->"; > > > > my $repl="replacement text"; > > my ($a,$b)=split('<!---begin--->',$str); > > my ($c,$d)=split('<!---end--->',$b); > > $str=$a.$repl.$d; > > > > I'd still like to know if you can do this via regex however. > > > > put an 's' switch for mulitple line > s/TEXT/REPLACE/is > > do a "perldoc perlre" for better help
The /m modifier says that the string conatains 'M'ultiple records, so that ^ and $ will match at newlines within the string as well as just at the ends of the whole string. The /s modifier says that the string contains a 'S'ingle record (i.e. newlines are insignificant) so that . will match newline characters as well as any other character. Using /m will have no effect in your case as there are no ^ or $ in your regex. Making this change to your program fixes the problem as below. HTH, Rob use strict; use warnings; my $str=<<EOF; <html> <body> <p>Hello this is juts an example</p> <p><!---begin---><a href="nowhere.com">blahahahaha</a> </p><a href=" http://www.somewhere.com";> HELLO</a> </p><!---end--->Hello world</body> </html> EOF my $repl="Replacement Text"; $str =~ s/\<!---begin---\>.+?\<!---end---\>/$repl/is; print $str, "\n"; **OUTPUT <html> <body> <p>Hello this is juts an example</p> <p>Replacement TextHello world</body> </html> -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED] <http://learn.perl.org/> <http://learn.perl.org/first-response>
