Regular expression to find all words possible

[arindam31]

Hi guys,

My intention is to find all possible words starting with any letter of my choice example 'p' from a paragraph.I want all the possible results . How can i do that .

Take the example :

'This rule says that any match that begins earlier in the string is always preferred over any plausible match that begins later. This rule doesn\'t say anything about how long the winning match might be (we\'ll get into that shortly), merely that among all the matches possible anywhere in the string, the one that begins the leftmost in the string is chosen. Actually, since more than one plausible match can start at the same earliest point, perhaps the rule should read "a match...\'\' instead of "the match...,\'\' but that sounds odd'

The results should be
preferred
plausible
..
..
..

Also i want to know how can i get multiple search results if we already know that there will be multiple matches .

Thanks

[JoshuaBurleson]

Show us the code you have so far, you need to show effort around here.

[arindam31]

Show us the code you have so far, you need to show effort around here.

Ofcourse...

These are the things i was trying.

>>> str1

'This rule says that any match that begins earlier in the string is always preferred over any plausible match that begins later. This rule doesn\'t say anything about how long the winning match might be (we\'ll get into that shortly), merely that among all the matches possible anywhere in the string, the one that begins the leftmost in the string is chosen. Actually, since more than one plausible match can start at the same earliest point, perhaps the rule should read "a match...\'\' instead of "the match...,\'\' but that sounds odd'

>>> var=re.compile("p+")
>>> var2=var.search(str1)
>>> var2.group()

Traceback (most recent call last):
File "<pyshell#82>", line 1, in <module>
var2.group()
AttributeError: 'NoneType' object has no attribute 'group'

This leads me to doubt that whether , the '+' sign can be used with letters.
I was expecting to get the first match atleast.
Next i have no idea how to get multiple search results.

By the way ,Thank you for replying.

[arindam31]

This is what happens when i use this code.

>>> var=re.compile("per+")
>>> var2=var.search(str1)
>>> var2.group()
'per'

I cant understand why the plus sign is not working here.

[TrustyTony]

The regex differs from globbing in that * and + modify previous symbol and any chatacter is . not ?. So p+ matches any number of p's at least one ie p, pp, ppp, .... Look for special symbols starting with \ to put suitable one between p and +

Read re documentation from Internernet or from IDLE help file.

[arindam31]

The regex differs from globbing in that * and + modify previous symbol and any chatacter is . not ?. So p+ matches any number of p's at least one ie p, pp, ppp, .... Look for special symbols starting with \ to put suitable one between p and +

Read re documentation from Internernet or from IDLE help file.

Thank you boss.That was indeed the problem.The below code worked .

>>> var=re.compile("p\w+")
>>> var2=var.search(str1)
>>> var2.group()
'preferred'

Now, for the second part, how can i get all the results matching the pattern .?

Will findall() work?

[arindam31]

Hurrah i found the solution. Myself......Eh.sorry for that.

>>> var=re.compile(r"p\w+")
>>> var2=[]
>>> var2=var.findall(str1)
>>> var2
['preferred', 'plausible', 'possible', 'plausible', 'point', 'perhaps']

Findall does work..

[TrustyTony]

Yes, I checked it also with Python without re:

print(','.join(w for w in (''.join(c for c in word if c.isalpha()) for word in 'This rule says that any match that begins earlier in the string is always preferred over any plausible match that begins later. This rule doesn\'t say anything about how long the winning match might be (we\'ll get into that shortly), merely that among all the matches possible anywhere in the string, the one that begins the leftmost in the string is chosen. Actually, since more than one plausible match can start at the same earliest point, perhaps the rule should read "a match...\'\' instead of "the match...,\'\' but that sounds odd'.split()) if w.startswith('p')))
preferred,plausible,possible,plausible,point,perhaps

[snippsat]

Just a note var2=[] is not needed as findall() is returing a list. finditer() can also be a good solution for iterating over larger files.
Always use raw string with regex r' '

import re

data = '''\
This rule says that any match that begins earlier
in the string is always preferred over any plausible match that begins later.'''

pattern = re.compile(r'p\w+')
for match in pattern.finditer(data):
    print match.group()

'''Out-->
preferred
plausible
'''

[arindam31]

Just a note var2=[] is not needed as findall() is returing a list. finditer() can also be a good solution for iterating over larger files.
Always use raw string with regex r' '

import re

data = '''\
This rule says that any match that begins earlier
in the string is always preferred over any plausible match that begins later.'''

pattern = re.compile(r'p\w+')
for match in pattern.finditer(data):
    print match.group()

'''Out-->
preferred
plausible
'''

Nice tip dude..I thought that we have to create a list variable ourselves first......Thanks for the correction....