in reply to matching paragraphs

Could you explain what you mean with a paragraph of text?

You mean lines of text seperated by an empty line, you could do something like this to read a file by paragraph:

{ # create an extra scope local $/ = ""; # set record seperator to empty lines while (<INPUT>) { # $_ contains a paragraph of text from <INPUT> } } # $/ is restored to its original value

Note that this code will tread more than one blank line as a single separator. See also the section $INPUT_RECORD_SEPARATOR in perlvar.