«

I frequently have to clean up lines with:

s/^\s+//;
s/\s+$//;

Is there ONE regex (as in ONE, SINGLE s///, not one compound
statement) that can do both of these ?

Thanks & happy Gnu Year.

On Jan 19, 1:24*pm, Spydo <spy...@gmail.com> wrote:
> I frequently have to clean up lines with:
>
> s/^\s+//;
> s/\s+$//;
>
> Is there ONE regex (as in ONE, SINGLE s///, not one compound
> statement) that can do both of these ?
>
> Thanks & happy Gnu Year.

DOH Nevermind I found it.. I didnt know ^ and $ could be in parens..

s/(^\s+)|(\s+$)//g;

>>>>> "Spydo" == Spydo <> writes:

Spydo> s/^\s+//;
Spydo> s/\s+$//;

Spydo> Is there ONE regex (as in ONE, SINGLE s///, not one compound
Spydo> statement) that can do both of these ?

Only if you sacrifice runtime efficiency.

--
Randal L. Schwartz - Stonehenge Consulting Services, Inc. - +1 503 777 0095
<> <URL:http://www.stonehenge.com/merlyn/>
Smalltalk/Perl/Unix consulting, Technical writing, Comedy, etc. etc.
See http://methodsandmessages.posterous.com/ for Smalltalk discussion

Spydo <> wrote:
>I frequently have to clean up lines with:
>
>s/^\s+//;
>s/\s+$//;
>
>Is there ONE regex (as in ONE, SINGLE s///, not one compound
>statement) that can do both of these ?

Yes, there is but it doesn't make much sense to use it.
See the relevent FAQ
perldoc -q "strip blank"
for details.

jue

Spydo wrote:
> On Jan 19, 1:24 pm, Spydo<spy...@gmail.com> wrote:
>> I frequently have to clean up lines with:
>>
>> s/^\s+//;
>> s/\s+$//;
>>
>> Is there ONE regex (as in ONE, SINGLE s///, not one compound
>> statement) that can do both of these ?
>>
>> Thanks& happy Gnu Year.
>
> DOH Nevermind I found it.. I didnt know ^ and $ could be in parens..
>
> s/(^\s+)|(\s+$)//g;

There is no need for the parentheses

s/^\s+|\s+$//g;

Does the same without having to save data to $1 and $2.



John
--
Any intelligent fool can make things bigger and
more complex... It takes a touch of genius -
and a lot of courage to move in the opposite
direction. -- Albert Einstein

Spydo <> writes:
>
> s/(^\s+)|(\s+$)//g;

Or Regexp::Common::whitespace for a big way of getting a little
expression. Text::Trim is effective too, in a scary context-sensitive
do-what-i-mean way.

(Could the faq cross reference those? If it doesn't make a long entry
even longer ...


--
Some people say cricket is like watching grass grow.
That's not true, it's watching grass die from not watering it.

Στις 19/1/2011 8:28 μμ, ο/η Spydo Î*γÏ�αψε:
> On Jan 19, 1:24 pm, Spydo<spy...@gmail.com> wrote:
>> I frequently have to clean up lines with:
>>
>> s/^\s+//;
>> s/\s+$//;
>>
>> Is there ONE regex (as in ONE, SINGLE s///, not one compound
>> statement) that can do both of these ?
>>
>> Thanks& happy Gnu Year.
>
> DOH Nevermind I found it.. I didnt know ^ and $ could be in parens..
>
> s/(^\s+)|(\s+$)//g;
>


here is a pretty one

s/^\s*(.*?)\s*$/$1/

In article <iipnm2$1l35$>,
George Mpouras <> wrote:
>Στις 19/1/2011 8:28 μμ, ο/η Spydo Î*γÏ�αψε:
>> On Jan 19, 1:24 pm, Spydo<spy...@gmail.com> wrote:
>>> I frequently have to clean up lines with:
>>>
>>> s/^\s+//;
>>> s/\s+$//;
>>>
>>> Is there ONE regex (as in ONE, SINGLE s///, not one compound
>>> statement) that can do both of these ?
>>>
>>> Thanks& happy Gnu Year.
>>
>> DOH Nevermind I found it.. I didnt know ^ and $ could be in parens..
>>
>> s/(^\s+)|(\s+$)//g;
>>
>
>
>here is a pretty one
>
>s/^\s*(.*?)\s*$/$1/

That's really neat.

--
David Canzi | Life is too short to point out every mistake. |

On 2011-02-09 01:34, David Canzi wrote:
> In article<iipnm2$1l35$>,
> George Mpouras<> wrote:
>> Στις 19/1/2011 8:28 μμ, ο/η Spydo Î*γÏ�αψε:
>>> On Jan 19, 1:24 pm, Spydo<spy...@gmail.com> wrote:

>>>> I frequently have to clean up lines with:
>>>>
>>>> s/^\s+//;
>>>> s/\s+$//;
>>>>
>>>> Is there ONE regex (as in ONE, SINGLE s///, not one compound
>>>> statement) that can do both of these ?
>>>>
>>>> Thanks& happy Gnu Year.
>>>
>>> DOH Nevermind I found it.. I didnt know ^ and $ could be in parens..
>>>
>>> s/(^\s+)|(\s+$)//g;
>>>
>>
>>
>> here is a pretty one
>>
>> s/^\s*(.*?)\s*$/$1/
>
> That's really neat.

I don't think so, because it also replaces when there is no heading or
tailing whitespace at all.

I normally use:

s/\s+\z//, # remove trailing whitespace
s/\A\s+//, # remove leading whitespace
s/[^\S ]/ /g, # replace non-space whitespace by space
s/ {2,}/ /g, # unify spaces
for $input;

See also tr/ in perlop, think about tr/\t\n / /s.

--
Ruud

Στις 9/2/2011 3:07 πμ, ο/η Dr.Ruud Î*γÏ�αψε:
> On 2011-02-09 01:34, David Canzi wrote:
>> In article<iipnm2$1l35$>,
>> George Mpouras<> wrote:
>>> ΣÃ�„ιÃ�‚ 19/1/2011 8:28 μμ, ο/η Spydo ÃŽÂ*γÃ�Â�αÃ�ˆÎµ:
>>>> On Jan 19, 1:24 pm, Spydo<spy...@gmail.com> wrote:
>
>>>>> I frequently have to clean up lines with:
>>>>>
>>>>> s/^\s+//;
>>>>> s/\s+$//;
>>>>>
>>>>> Is there ONE regex (as in ONE, SINGLE s///, not one compound
>>>>> statement) that can do both of these ?
>>>>>
>>>>> Thanks& happy Gnu Year.
>>>>
>>>> DOH Nevermind I found it.. I didnt know ^ and $ could be in parens..
>>>>
>>>> s/(^\s+)|(\s+$)//g;
>>>>
>>>
>>>
>>> here is a pretty one
>>>
>>> s/^\s*(.*?)\s*$/$1/
>>
>> That's really neat.
>
> I don't think so, because it also replaces when there is no heading or
> tailing whitespace at all.
>
> I normally use:
>
> s/\s+\z//, # remove trailing whitespace
> s/\A\s+//, # remove leading whitespace
> s/[^\S ]/ /g, # replace non-space whitespace by space
> s/ {2,}/ /g, # unify spaces
> for $input;
>
> See also tr/ in perlop, think about tr/\t\n / /s.
>


too much noise for something simple