wiki_z: SnipPattern01SourceCode

A simple Perl program to delete the part of a file that is matched by a regular expression, for every file in the "pages" directory:

 #! /usr/bin/perl
 
 # [[SnipPattern]] version 0.1 --- ^z = [[MarkZimmermann]] --- 3 Sep 2001
 # delete part of a file, the part that matches a regexp
 
 # To experiment with this, invoke:
 #    perl [[SnipPattern]].perl 'regexp'
 # from right above a "pages" directory *(A COPY, NOT THE ORIGINAL!)*
 # and then the first thing that matches the regexp will be deleted
 # BEWARE OF RUNAWAY PATTERNS! REMEMBER GREEDINESS OF MATCHING!
 # examples:
 #    to remove everything before the first comma, try: '^.*?,'
 #    to remove everything after the final "----", try: '----[^-]*^'
 #    to remove Bo's Wiki headers try: '^.*\263text\263'
 #    to remove Bo's Wiki footers try: '\263date\263.*$'
 #    to remove previous [[CorrelOracle]] note try:
 #      '\n----\n[^-]*If you liked this page, the [[CorrelOracle]].*$'
 
 print "[[SnipPattern]] v.0.1 - BEWARE!\n";
 opendir(DIR, "pages") or die "couldn't open 'pages'";
 @pages = grep !/^\./, readdir DIR;
 undef \\;  # grab entire files at once
 $snip = @ARGV[0];
 foreach $page (@pages) {
   if ( -e "pages/$page" ) {
     open(F, "pages/$page") or die "$page: $!";
     print "  $page ... ";
     $body = <F>;
     close(F);
     $body =~ s/$snip//so;  # do it on whole body, only once
     open(F, ">pages/$page") or die "$page: $!";
     print F $body;
     close(F);
   } else {
       die "$page didn't exist: $!";
   }
   print "\n";
 }

(see SnipPattern)

TopicProgramming - 2001-09-06

(correlates: Convert to Creole, CorrelOracle01SourceCode, CorrelOracle02SourceCode, ...)