textcatcher is a small library to help you parse and filter multiline text.

the Catcher class is the matcher API.  Some concrete classes of Catcher are
* REMatch - match a regular expression
* LineCatcher - match a full text line
* TextCather - match a partial text line

Implementing your own Catcher class is easy.  Here's a catcher that watches
the output of mysqldump and looks for multi-line table defintions.

class SQLTable(textcatch.Catcher):
    """ e.g.
        CREATE TABLE `example` (
        `id` int(11) NOT NULL AUTO_INCREMENT,
         PRIMARY KEY (`id`),
        ) ENGINE=InnoDB DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
    """
    start = re.compile('^CREATE TABLE ')
    end = re.compile('\) ENGINE=')
    def parse(self):
        print "here's a full table defition!"
	print "\n".join(self.lines)

If you want to watch for multiple matches in a stream use a CatchQueue,
a catcher-alike that dispatches to more than one catcher.  You can also
use a queue to get fancier behavior by chaining objects together.  The
above SQLTable catcher will print all the tables so if we follow that
with a TextCatcher that muffles all text we get a unix filter program
that extracts table definitions from dumps!

#!/bin/env python
import textcatcher
import sys
outstream = textcatcher.CatchQueue()
outstream.add(SQLTable(listen=True))
outstream.add(textcatcher.TextCatcher('', muffle=True))
for text in sys.stdin:
    text = outstream.line(line)
    if text:
        sys.stdout.write(text)

Code origianlly from Leanlyn http://bit.ly/leanlyn