regular expressions

Large topic – somewhat more advanced than ‘basic programming’, but worth a digression. (reference)

What if we are looking for some number, but we don’t know what number?

import re
bool(re.search('[0-9]', 'Plan 9'))

Pattern	Description
^	Beginning of line
$	End of line
.	Any single character except newline
[...]	Any single character in brackets
[^...]	Any single character not in brackets
re*	0 or more occurrences of preceding expression
re+	1 or more occurrence of preceding expression
re?	0 or 1 occurrence of preceding expression
re1|re2	match re1 or re2
()	grouping

• How would you test whether a string contains a numeric value at the end (e.g. “Plan 99”)?
• What if the string might contain a comma (e.g. “Plan 99,478”)?
• What if you’re looking for the abbreviations of rooms in Hamilton Hall (my office is HH314)?
• … rooms in LSB or HH?
• substitution:

import re
x = "(Smith, John), (Schmo, Joe)"
re_pat = "\(([A-Z][a-z]+), ([A-Z][a-z.]*)\)"
print(re.sub(re_pat,"(\\2 \\1)",x))

## (John Smith), (Joe Schmo)

• in general if you want something to be interpreted literally in a regular expression (e.g. you want to look for + or ( and not have them interpreted in their regular-expression sense), you need to use a backslash \ as an “escape” character: for example, "a\+b" will look for the letter a, followed by a plus sign, followed by the letter b – not one or more as, followed by a b (xkcd)