regular expressions 1 day 6 - 9/08/14 ling 3820 & 6820 natural language processing harry howard...
TRANSCRIPT
Regular expressions 1Day 6 - 9/08/14LING 3820 & 6820
Natural Language Processing
Harry Howard
Tulane University
Course organization
08-Sept-2014NLP, Prof. Howard, Tulane University
2
http://www.tulane.edu/~howard/LING3820/
The syllabus is under construction. http://www.tulane.edu/~howard/CompCu
ltEN/
The quiz was the review.
Review
08-Sept-2014
3
NLP, Prof. Howard, Tulane University
Open Spyder
08-Sept-2014
4
NLP, Prof. Howard, Tulane University
§4. Regular expressions
08-Sept-2014
5
NLP, Prof. Howard, Tulane University
Regular expressions, or regex >>> import re re.findall(pattern, target string)
08-Sept-2014NLP, Prof. Howard, Tulane University
6
4.2. Fixed-length matching
08-Sept-2014
7
NLP, Prof. Howard, Tulane University
The test string
>>> S = '''This above all: to thine own self be true,
... And it must follow, as the night the day,
... Thou canst not then be false to any man.'''
08-Sept-2014NLP, Prof. Howard, Tulane University
8
Strings as regular expressions>>> re.findall(' be ', S)
[' be ', ' be ']
08-Sept-2014NLP, Prof. Howard, Tulane University
9
Match one character of a disjunction with |>>> re.findall(' to | be | it | as ', S)
[' to ', ' be ', ' it ', ' as ', ' be ', ' to ']
>>> set(re.findall(' to | be | it | as ', S))
set([' it ', ' as ', ' to ', ' be '])
08-Sept-2014NLP, Prof. Howard, Tulane University
10
Match a group of characters with capturing or non-capturing parentheses, ()>>> re.findall(' (to|be|it|as) ', S)
['to', 'be', 'it', 'as', 'be', 'to']
R>>> re.findall(' (?:to|be|it|as) ', S)
[' to ', ' be ', ' it ', ' as ', ' be ', ' to ']
The default behavior of parentheses is to capture the string inside them in the output. The ?: prefix turns capturing off. For the rest of this discussion, we prefer to exclude the spaces from the output.
08-Sept-2014NLP, Prof. Howard, Tulane University
11
Match one character of a range with [] and its negation with [^]>>> re.findall(' ([a-z][a-z]) ', S)
['to', 'be', 'it', 'as', 'be', 'to']
>>> re.findall(' ([^0-9][^0-9]) ', S)
['to', 'be', 'it', 'as', 'be', 'to']
>>> re.findall(' ([a-e][a-e]) ', S)
['be', 'be']
>>> re.findall(' ([^a-e][^a-e]) ', S)
['to', 'it', 'to']
08-Sept-2014NLP, Prof. Howard, Tulane University
12
Match a number of repetitions of a character with {}
>>> re.findall(' ([a-z]{2}) ', S)['to', 'be', 'it', 'as', 'be', 'to']
08-Sept-2014NLP, Prof. Howard, Tulane University
13
Match any character with .
>>> re.findall(' (..) ', S)['to', 'be', 'it', 'as', 'be', 'to']>>> re.findall(' (.{2}) ', S)['to', 'be', 'it', 'as', 'be', 'to']
08-Sept-2014NLP, Prof. Howard, Tulane University
14
4.2.7. and following
Next time
08-Sept-2014NLP, Prof. Howard, Tulane University
15