«

I am using regular expressions to search a string (always full
sentences, maybe more than one sentence) for common abbreviations and
remove the periods. I need to break the string into different
sentences but split('.') doesn't solve the whole problem because of
possible periods in the middle of a sentence.

So I have...

----------------

import re

middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')

# this will find abbreviations like e.g. or i.e. in the middle of a
sentence.
# then I want to remove the periods.

----------------

I want to keep the ie or eg but just take out the periods. Any
ideas? Of course newString = middle_abbr.sub('',txt) where txt is the
string will take out the entire abbreviation with the alphanumeric
characters included.

On Jul 31, 3:07*pm, chrispoliq...@gmail.com wrote:
> I am using regular expressions to search a string (always full
> sentences, maybe more than one sentence) for common abbreviations and
> remove the periods. *I need to break the string into different
> sentences but split('.') doesn't solve the whole problem because of
> possible periods in the middle of a sentence.
>
> So I have...
>
> ----------------
>
> import re
>
> middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
>
> # this will find abbreviations like e.g. or i.e. in the middle of a
> sentence.
> # then I want to remove the periods.
>
> ----------------
>
> I want to keep the ie or eg but just take out the periods. *Any
> ideas? *Of course newString = middle_abbr.sub('',txt) where txt is the
> string will take out the entire abbreviation with the alphanumeric
> characters included.

>>> middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
>>> s = 'A test, i.e., an example.'
>>> a = middle_abbr.search(s) # find the abbreviation
>>> b = re.compile('\.') # period pattern
>>> c = b.sub('',a.group(0)) # remove periods from abbreviation
>>> d = middle_abbr.sub(c,s) # substitute new abbr for old
>>> d
'A test, ie, an example.'

On Jul 31, 3:56*pm, Mensanator <mensana...@aol.com> wrote:
> On Jul 31, 3:07*pm, chrispoliq...@gmail.com wrote:
>
>
>
>
>
> > I am using regular expressions to search a string (always full
> > sentences, maybe more than one sentence) for common abbreviations and
> > remove the periods. *I need to break the string into different
> > sentences but split('.') doesn't solve the whole problem because of
> > possible periods in the middle of a sentence.
>
> > So I have...
>
> > ----------------
>
> > import re
>
> > middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
>
> > # this will find abbreviations like e.g. or i.e. in the middle of a
> > sentence.
> > # then I want to remove the periods.
>
> > ----------------
>
> > I want to keep the ie or eg but just take out the periods. *Any
> > ideas? *Of course newString = middle_abbr.sub('',txt) where txt is the
> > string will take out the entire abbreviation with the alphanumeric
> > characters included.
> >>> middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
> >>> s = 'A test, i.e., an example.'
> >>> a = middle_abbr.search(s) * * *# find the abbreviation
> >>> b = re.compile('\.') * * * * * # period pattern
> >>> c = b.sub('',a.group(0)) * * * # remove periods from abbreviation
> >>> d = middle_abbr.sub(c,s) * * * # substitute new abbr for old
> >>> d
>
> 'A test, ie, an example.'


A more versatile version:

import re

middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
s = 'A test, i.e., an example.'
a = middle_abbr.search(s) # find the abbreviation
b = re.compile('\.') # period pattern
c = b.sub('',a.group(0)) # remove periods from abbreviation
d = middle_abbr.sub(c,s) # substitute new abbr for old

print d
print
print

s = """A test, i.e., an example.
Yet another test, i.e., example with 2 abbr."""

a = middle_abbr.search(s) # find the abbreviation
c = b.sub('',a.group(0)) # remove periods from abbreviation
d = middle_abbr.sub(c,s) # substitute new abbr for old

print d
print
print

s = """A test, i.e., an example.
Yet another test, i.e., example with 2 abbr.
A multi-test, e.g., one with different abbr."""

done = False

while not done:
a = middle_abbr.search(s) # find the abbreviation
if a:
c = b.sub('',a.group(0)) # remove periods from abbreviation
s = middle_abbr.sub(c,s,1) # substitute new abbr for old ONCE
else: # repeat until all removed
done = True

print s

## A test, ie, an example.
##
##
## A test, ie, an example.
## Yet another test, ie, example with 2 abbr.'
##
##
## A test, ie, an example.
## Yet another test, ie, example with 2 abbr.
## A multi-test, eg, one with different abbr.

On Jul 31, 3:07*pm, chrispoliq...@gmail.com wrote:
>
> middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
>

When defining re's with string literals, it is good practice to use
the raw string literal format (precede with an 'r'):
middle_abbr = re.compile(r'[A-Za-z0-9]\.[A-Za-z0-9]\.')

What abbreviations have numeric digits in them?

I hope your input string doesn't include something like this:
For a good approximation of pi, use 3.1.

-- Paul

On Jul 31, 9:07*pm, chrispoliq...@gmail.com wrote:
> I am using regular expressions to search a string (always full
> sentences, maybe more than one sentence) for common abbreviations and
> remove the periods. *I need to break the string into different
> sentences but split('.') doesn't solve the whole problem because of
> possible periods in the middle of a sentence.
>
> So I have...
>
> ----------------
>
> import re
>
> middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
>
> # this will find abbreviations like e.g. or i.e. in the middle of a
> sentence.
> # then I want to remove the periods.
>
> ----------------
>
> I want to keep the ie or eg but just take out the periods. *Any
> ideas? *Of course newString = middle_abbr.sub('',txt) where txt is the
> string will take out the entire abbreviation with the alphanumeric
> characters included.

It's recommended that you should use a raw strings for regular
expressions.

Capture the letters using parentheses:

middle_abbr = re.compile(r'([A-Za-z0-9])\.([A-Za-z0-9])\.')

and replace what was found with what was captured:

newString = middle_abbr.sub(r'\1\2', txt)

HTH

On Jul 31, 10:07*pm, chrispoliq...@gmail.com wrote:
> I am using regular expressions to search a string (always full
> sentences, maybe more than one sentence) for common abbreviations and
> remove the periods. *I need to break the string into different
> sentences but split('.') doesn't solve the whole problem because of
> possible periods in the middle of a sentence.
>
> So I have...
>
> ----------------
>
> import re
>
> middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
>
> # this will find abbreviations like e.g. or i.e. in the middle of a
> sentence.
> # then I want to remove the periods.
>
> ----------------
>
> I want to keep the ie or eg but just take out the periods. *Any
> ideas? *Of course newString = middle_abbr.sub('',txt) where txt is the
> string will take out the entire abbreviation with the alphanumeric
> characters included.

Its impossible with regex. U could try it with a statistical analysis;
and even this would give u a good split.

On Aug 1, 12:53*pm, dusans <dusan.smit...@gmail.com> wrote:
> On Jul 31, 10:07*pm, chrispoliq...@gmail.com wrote:
>
>
>
>
>
> > I am using regular expressions to search a string (always full
> > sentences, maybe more than one sentence) for common abbreviations and
> > remove the periods. *I need to break the string into different
> > sentences but split('.') doesn't solve the whole problem because of
> > possible periods in the middle of a sentence.
>
> > So I have...
>
> > ----------------
>
> > import re
>
> > middle_abbr = re.compile('[A-Za-z0-9]\.[A-Za-z0-9]\.')
>
> > # this will find abbreviations like e.g. or i.e. in the middle of a
> > sentence.
> > # then I want to remove the periods.
>
> > ----------------
>
> > I want to keep the ie or eg but just take out the periods. *Any
> > ideas? *Of course newString = middle_abbr.sub('',txt) where txt is the
> > string will take out the entire abbreviation with the alphanumeric
> > characters included.
>
> Its impossible with regex. U could try it with a statistical analysis;
> and even this would give u a good split.

"and even this wont* give u a good split."