Checking if string inside quotes?

[castironpi at gmail.com]

On May 9, 4:31 pm, "Michael Yanowitz" <m.yanow... at kearfott.com> wrote:
> Thanks, but it is a little more complicated than that,
>   the string could be deep in quotes.
>
>    The problem is in string substitution.
> Suppose I have a dictionary with MY_IP : "172.18.51.33"
>
>   I need to replace all instances of MY_IP with "172.18.51.33"
> in the file.
>   It is easy in cases such as:
>   if (MY_IP == "127.0.0.1"):
>
>   But suppose I encounter:"
>  ("(size==23) and (MY_IP==127.0.0.1)")
>
>    In this case I do not want:
>  ("(size==23) and ("172.18.51.33"==127.0.0.1)")
>     but:
>  ("(size==23) and (172.18.51.33==127.0.0.1)")
>     without the internal quotes.
>  How can I do this?
>   I presumed that I would have to check to see if the string
> was already in quotes and if so remove the quotes. But not
> sure how to do that?
>   Or is there an easier way?
>
> Thanks in advance:
> Michael Yanowitz
>
> -----Original Message-----
> From: python-list-bounces+m.yanowitz=kearfott.... at python.org
>
> [mailto:python-list-bounces+m.yanowitz=kearfott.... at python.org]On Behalf
> Of half.ital... at gmail.com
> Sent: Wednesday, May 09, 2007 5:12 PM
> To: python-l... at python.org
> Subject: Re: Checking if string inside quotes?
>
> On May 9, 1:39 pm, "Michael Yanowitz" <m.yanow... at kearfott.com> wrote:
> > Hello:
>
> >    If I have a long string (such as a Python file).
> > I search for a sub-string in that string and find it.
> > Is there a way to determine if that found sub-string is
> > inside single-quotes or double-quotes or not inside any quotes?
> > If so how?
>
> > Thanks in advance:
> > Michael Yanowitz
>
> I think the .find() method returns the index of the found string.  You
> could check one char before and then one char after the length of the
> string to see.  I don't use regular expressions much, but I'm sure
> that's a more elegant approach.
>
> This will work. You'll get in index error if you find the string at
> the very end of the file.
>
> s = """
> foo
> "bar"
> """
> findme = "foo"
> index = s.find(findme)
>
> if s[index-1] == "'" and s[index+len(findme)] == "'":
>         print "single quoted"
> elif s[index-1] == "\"" and s[index+len(findme)] == "\"":
>    print "double quoted"
> else:
>    print "unquoted"
>
> ~Sean
>
> --http://mail.python.org/mailman/listinfo/python-list

In "nearby" quotes or in quotes at all?
import re
a='abc"def"ghijk'
b=re.sub( r'([\'"])[^\1]*\1', '', a )
b.replace( 'ghi', 'the string' )
#fb: 'abcthe stringjk'
edit()

Here, you get the entire file -in b-, strings omitted entirely, so you
can't write it back.

I've used `tokenize' to parse a file, but you don't get precisely your
original back.  Untokenize rearrages your spacings.  Equivalent
semantically, so if you want to compile immedately afterwords, you're
alright with that.  Short example:
from tokenize import *
import token
from StringIO import StringIO
a= StringIO( 'abc "defghi" ghi jk' )
from collections import deque
b= deque()
for g in generate_tokens( a.readline ):
	if g[0]== token.NAME and g[1]== 'ghi':
		b.append( ( token.STRING, '"uchoose"' ) )
	else:
		b.append( g )

untokenize( b )
#fb: 'abc "defghi""uchoose"jk '
edit()
acb