Home/ Questions/Q 7864883
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-06-02T23:55:13+00:00

I have a regex to match middle names that looks like this : first_name

  • 0

I have a regex to match middle names that looks like this : 

first_name = 'Matthew'
last_name = 'Walsh'
for char in first_name:
    new_first_name+='(' + char.lower() + '|' + char.upper() + ')'
for char in last_name:
    new_last_name+='(' + char.lower() + '|' + char.upper() + ')'


middle_name_regex_str = "\b?((" + new_first_name + " (?P<middle_name1>[A-Z][^ ]?[a-z]* )?" + new_last_name + ")|(" + new_last_name + " (?P<middle_name2>[A-Z][^ ]?[a-z]* )?" + new_first_name + "))"

here is a pattern it works for :

MATTHEW B. WALSH, D.M.D.\nBorn:\nAkron, Ohio\nCollege:\nBachelor of
Arts, Kenyon College, 1998

for this pattern it works fine, and matches the middle name ‘B.’

however I want to be on the safe side and escape the first and last names, but when I add re.escape it fails : 

middle_name_regex_str = "\b?((" + re.escape(new_first_name) + " (?P<middle_name1>[A-Z][^ ]?[a-z]* )?" + re.escape(new_last_name) + ")|(" + re.escape(new_last_name) + " (?P<middle_name2>[A-Z][^ ]?[a-z]* )?" + re.escape(new_first_name) + "))"

and now the regex does not match properly : 

regex = re.compile(middle_name_regex_str)
regex.search('MATTHEW B. WALSH, D.M.D.\nBorn:\nAkron, Ohio\nCollege:\nBachelor of Arts, Kenyon College, 1998')

this returns nothing.

Shouldn’t re.escape be safe to use in the sense that it does not alter the behavior of the expression? What in adding backslashes before non alphanumeric chars could cause it to not match?

Any help would be appreciated!

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    using re.escape on something that already contains regex special chars will look for those literal chars.

    Two suggestions here:

    1. if possible, why not use re.IGNORECASE to test the regex regardless of case?

    2. if not, you can do something like this

    first_name= 'Matthew'

    last_name = 'Walsh'

first_name_re = "".join('(%s|%s)' % (re.escape(c.upper()),re.escape(c.lower())) for c in first_name)
last_name_re = "".join('(%s|%s)' % (re.escape(c.upper()),re.escape(c.lower())) for c in last_name)


# now that they are safe -we can simply put them in the middle of the regex
middle_name_regex_str = "\b?((%s (?P<middle_name1>[A-Z][^ ]?[a-z]* )?%s)|(%s (?P<middle_name2>[A-Z][^ ]?[a-z]* )?%s))" % (first_name_re, last_name_re, first_name_re, last_name_re)

    not sure about the order of formatting args here,but you get the point

    • 0