use URI;
use HTML::Parser;

my $parse = new HTML::Parser( default_h => [ sub {print shift}, 'text'
+ ],
  start_h => [
  sub {
    my ($tag, $attr, $origtext) = @_;
    if (($tag eq 'img') and ($attr->{src} =~ m#^(?:http://)?myserver\.
+#)) {
      $attr->{img} = URI->new($attr->{img})->abs('http://myserver.com'
+);
      print "<$tag " . join (' ', map {$_.'="' . $attr->{$_} . '"'} ke
+ys %$attr) . '>';
    }
    else {
      print $origtext;
    }
  }, "tagname, attr, text"]
);

$parse->parse(join('',<>));
[download]

s|(img src=")myserver\.com/|$1|g;

before: img src="blahblah.com/images/pic.gif"
after: img src="images/pic.gif"
[download]

I use BBEdit on a Mac for coding, and it is as simple as finding:

src="myserver.com/images/

and replacing with:

src="../images/

or whatever is the correct path. You can perfoem this across multiple files/directories with specific suffixes so that you can zero in on the exact files that need the change.

I don't code on a PC, but I am sure that there must be as powerful a search and replace function on one of the text/code editors.

A text editor substitution works well when you have 1-5 files. Suppose that the monk in question is dealing with a webserver that serves 500 documents with this problem. Perl /is/ your friend. (Of course, this brings about the template argument, but we'll save that for another day.. =])

-marius

I am speaking as a relative newcomer to Perl. So with all respect I can say that I have used BBEdit to do a "Find and Replace" on over 100 files in multiple directories. It is fast (~5 files a second), accurate on changes to a small number of characters (15-20 charaters at a time), and will provide a summary of changes as well as automatically save or allow you to manually save.

By the way, I have no conection to BBEdit in any shape or form.

Very soon I need to do the same sort of processing on 1000+ HTML files, and I felt the need for a utility to help both you and I do this.

So I wrote one, and put it in the Code Catacombs under Change Absolute to Relative links in HTML files.

It was a fun experiment. I got to play with alot of CPAN modules I haven't used much before (I mostly work with CGI and html-based database interfaces), like: HTML::TreeBuilder, File::Find, Cwd, File::Copy and a new module to add to my "favorite's list", Getopt::Declare.

That was the exact wheel I was looking for. Thank you. If I was anywhere near you, I'd buy you a beer.

Thanks for the wheel.