Improvment of and looping in a regular expression pattern

Question

My implemented regex pattern contains two repeating symbols: \d{2}\. and <p>(.*)</p>. I want to get rid of this repetition and asked myself if there is a way to loop in Python's regular expression implementation.

Note: I do not ask for help to parse a XML file. There are many great tutorials, howtos and libraries. I am looking for means to implement repetition in regex patterns.

My code:

import re
pattern = '''
<menu>
<day>\w{2} (\d{2}\.\d{2})\.</day>
<description>
<p>(.*)</p>
<p>(.*)</p>
<p>(.*)</p>
</description>
'''
my_example_string = '''
<menu>
<day>Mi 03.04.</day>
<description>
<p>Knoblauchcremesuppe</p>
<p>Rindsbraten "Esterhazy" (Gem&uuml;serahmsauce)</p>
<p>mit H&ouml;rnchen und Salat</p>
</description>
</menu>
'''
re.findall(pattern, my_example_string, re.MULTILINE)

• konijn: Parsing XML with regex is usually wrong, what are you really trying to accomplish?
• Philipp: The XML is malformed what prevents a usage of LXML and Xpath. I easily can retrieve the deserved data, but I want to find a way to avoid these repetitions in any regex patterns.

Answer 1

Firstly, just for anyone who might read this: DO NOT take this as an excuse to parse your XML with regular expressions. It generally a really really bad idea! In this case the XML is malformed, so its the best we can do.

The regular expressions looping constructs are * and {4} which you already using. But this is python, so you can construct your regular expression using python:

expression = """
<menu>
<day>\w{2} (\d{2}\.\d{2})\.</day>
<description>
"""
for x in xrange(3):
 expression += "<p>(.*)</p>"
expression += """
</description>
</menu>
"""

• Gareth Rees: What about expression += "<p>(.*)</p>\n" * 3 ?