Download Regular Expressions: Finding and Manipulating Patterns in Strings and more Study notes Programming Languages in PDF only on Docsity!
Programming Language Paradigms
Regular Expressions
Chris Brooks Department of Computer Science University of San Francisco ?? Department of Computer Science — University of San Francisco – p.1/
Regular Expressions 6-0:
Often, when you’re working with strings or files. you need to find, replace or ignore strings or lines that match some sort of pattern. Find lines containing ’img src’ Find strings that begin with a number Find strings that don’t contain a number Replace all instances of strings beginning with CS with
’USF-CS’
How can we do this in an automated way? ?? Department of Computer Science — University of San Francisco – p.2/
String methods 6-1:
String module has find, count, replace, and split. this works when you know the exact string you’re looking for. Awkward when you have multiple occurrences, or many different matches. Doesn’t deal well with case ?? Department of Computer Science — University of San Francisco – p.3/
Example 6-2:
Replace all instances of ’ROAD’ with ’RD’ s.replace(’ROAD’, ’RD’) What about ’111 BROAD ROAD’? s[:-4] + s[-4:].replace(’ROAD’, ’RD’) Better, but brittle and hard to read Need to know that ’ROAD’ is the last four chars re.sub(’ROAD$’, ’RD’, s) ?? Department of Computer Science — University of San Francisco – p.4/
Example 6-3:
of a line. end $ says to only match occurrences at the What about cases where someone leaves off the street type? Or cases where there’s an apartment number after
’ROAD’
’111 BROAD’
Still matches. ’111 BROAD ROAD Apt 3’ Doesn’t match RD,s ) re.sub(’\bROAD\b’, ’\b’ indicates word boundary; needs to be escaped. ?? Department of Computer Science — University of San Francisco – p.5/
Escaping in Strings 6-4:
character needs to be escaped by putting another ’ín The ’´ front of it. Otherwise, Python thinks it’s a control character for the following character. t’ \ e.g. ’ When you need to work with lots of backslashes, you can use a raw string. r’\bROAD\b’ s2 = No metacharacters are evaluated. ?? Department of Computer Science — University of San Francisco – p.6/
Finding Sequences of Characters 6-5:
Find addresses that begin with 1-3 4s. 4 Main St, 44 Mulberry Lane, 423 First St., etc = ’ˆ44?4?’ pat ’ indicates beginning of the line. ’ˆ ’?’ indicates 0 or 1 occurrence of the previous character. re.match() returns a Match object if there is a match, None otherwise ?? Department of Computer Science — University of San Francisco – p.7/
More complex sequences 6-6:
Now, let’s suppose we need to find sequences that begin with 1-3 5s, and can end with ’01’, ’02’, or 2 followed by 1-3 zeros. 44?4?’ Start with our previous pattern: ’ ˆ ending with ’01’ is ’(01)?’, with ’02’ is ’(02)?’ 44?4?(01|02)$’ will match 401, 44402, 4402, etc ’ ˆ Add 200?0?0? to parens
?? Department of Computer Science — University of San Francisco – p.8/
An easier way to express 6-7:
repetition
We can also use the {n,m} syntax to match repetition. ˆ4{1,3} matches 1-3 4s at the beginning of the string. We can rewrite the previous expression as: 4{1,3}(0(1|2) |20{1,3})$’ p = ’ ˆ ?? Department of Computer Science — University of San Francisco – p.9/
Things to match on 6-8:
’.’ - matches any character x-y - match a range or set of characters. abc will match a or b or c A-Z will match A,B,C, ..., Z Can also tag the complement of a set - [âbc] will match any character except a,b,or c. ?? Department of Computer Science — University of San Francisco – p.10/
Things to match on 6-9:
b ’ - word boundary \ ’ d’ - any digit\ ’ 0-9]) D’ - anything that’s not a digit (same as [ ˆ\ ’ w’ - equivalent to ’[A-Za-z0-9_]’\ ’ s’ - whitespace\ ’ S’ - non-whitespace\ ’ ?? Department of Computer Science — University of San Francisco – p.11/
Matching multiple occurrences 6-10:
’?’ - 0 or 1 occurrences ’+’ one or more occurrence ’’ 0 or more occurrences ’{m,n}’ - m to n occurrences ’{m}’ - exactly m occurrences ’?’ after another multiple-occurrence matcher - non-greedy. e.g. ’<.?>’ ?? Department of Computer Science — University of San Francisco – p.12/
