scripts/nlp/re-test.py

#! /usr/bin/env python

## This script tests the raw version of given regular expression on the given input string. 
##
## Version: 1.1												(25-Sep-2012)
##


import sys, optparse, re


##======================================================================
## main
def main(argv=None):
	if argv is None:
		arv = sys.argv

	parser = optparse.OptionParser(usage="%prog <REGEX>" +
                                             "\nTests the raw version of given regular expression on the given input string.", version="%prog 1.1")
	
	parser.add_option("-i", help="ignore case", dest="ignoreCase", action="store_true")
	parser.add_option("--ure", help="decode REGEX to unicode", dest="unicodeRE", action="store_true")
	parser.add_option("--uin", help="decode input to unicode", dest="unicodeIn", action="store_true")
	parser.add_option("--umatch", help="use unicode match", dest="unicodeMatch", action="store_true")
	
	(opts, posArgs) = parser.parse_args()
	
	if len(posArgs) < 1:
		parser.error("At least 1 arguments is required")
	
	if opts.unicodeRE:
		vRE = posArgs[0].decode("UTF-8")
	else:
		vRE = posArgs[0]
	
	# processing input      
	vLineCntr = 0
	vMatchCntr = 0
	while True:
		vInput = sys.stdin.readline()
		
		if not vInput:
			break
		
		if opts.unicodeIn:
			vInput = vInput.decode("utf-8")
		
		if opts.ignoreCase:
			if opts.unicodeMatch:
				vPattern = re.compile(vRE, re.IGNORECASE | re.UNICODE)
			else:
				vPattern = re.compile(vRE, re.IGNORECASE)
		else:
			if opts.unicodeMatch:
				vPattern = re.compile(vRE, re.UNICODE)
			else:
				vPattern = re.compile(vRE)
		
		vOutput = re.findall(vPattern, vInput)
		vMatchCntr += len(vOutput)
		if vOutput:
			if opts.unicodeIn:
				vInput = vInput.encode("utf-8")
			
			print vInput,
			print vOutput
		vLineCntr += 1
		
	print "%d matches in %d lines" % (vMatchCntr, vLineCntr)
	

##======================================================================
## calling main
if __name__ == "__main__":
	sys.exit(main())