正则表达式+ Python从制表符分隔文件的值中删除特定的尾随和结束字符

Question

It's been years (and years) since I've done any regex, so turning to experts on here since it's likely a trivial exercise :)

I have a tab delimited file and on each line I have a certain fields that have values such as:

• foo
• bar
• b"foo's bar"
• b'bar foo'
• b'carbar'

(A complete line in the file might be something like:

123\\t b'bar foo' \\tabc\\t123\\r\\n

I want to get rid of all the leading b', b" and trailing ", ' from that field on every line. So given the example line above, after running the regex, I'd get:

123\\t bar foo \\tabc\\t123\\r\\n

Bonus points if you can give me the python blurb to run this over the file.

Answer 1

(^|\\t)b[\\"'] should match the leadings, and for the trailing:

\\"' should do it

In Python, you do:

import re
r1 = re.compile("(^|\t)b[\"']")
r2 = re.compile("[\"'](\t|$)")

then just use

r1.sub("\\1", yourString)
r2.sub("\\1", yourString)

Answer 2

for each line you can use

re.sub(r'''(?<![^\t\n])\W*b(["'])(.*)\1\W*(?![^\t\n])''', r'\2', line)

and for bonus points:

import re

pattern = re.compile(r'''(?<![^\t\n])\W*b(["'])(.*?)\1\W*?(?![^\t\n])''')
with open('outfile', 'w') as outfile:
    for line in open('infile'):
        outfile.write(pattern.sub(r'\2', line))

Answer 3

>>> "b\"foo's bar\"".replace('b"',"").replace("b'","").rstrip("\"'")
"foo's bar"
>>> "b'bar foo'".replace('b"',"").replace("b'","").rstrip("\"'")
'bar foo'
>>>