Help with Regex for domain names

I'm trying to figure out how to write efficiently write a regex for
domain names with a particular top level domain. Let's say, I want to
grab all domain names with country codes .us, .au, and .de.

I could create three different regexs that would work:
regex = re.compile(r'[\w\-\.]+\.us)
regex = re.compile(r'[\w\-\.]+\.au)
regex = re.compile(r'[\w\-\.]+\.de)

How would I write one to accommodate all three, or, better yet, to
accommodate a list of them that I can pass into a method call? Thanks!

Feyo

Feyo wrote:
> I'm trying to figure out how to write efficiently write a regex for
> domain names with a particular top level domain. Let's say, I want to
> grab all domain names with country codes .us, .au, and .de.
>
> I could create three different regexs that would work:
> regex = re.compile(r'[\w\-\.]+\.us)
> regex = re.compile(r'[\w\-\.]+\.au)
> regex = re.compile(r'[\w\-\.]+\.de)
>
> How would I write one to accommodate all three, or, better yet, to
> accommodate a list of them that I can pass into a method call? Thanks!

Just a point of interest: A correctly formed domain name may have a
trailing period at the end of the TLD [1]. Example:

foo.bar.com.

Though you do not often see this, it's worth accommodating "just in
case"...


[1] http://homepages.tesco.net/J.deBoyne...main-name.html



--
----------------------------------------------------------------------------
Tim Daneliuk
PGP Key: http://www.tundraware.com/PGP/

Tim Daneliuk

Feyo wrote:
> I'm trying to figure out how to write efficiently write a regex for
> domain names with a particular top level domain. Let's say, I want to
> grab all domain names with country codes .us, .au, and .de.
>
> I could create three different regexs that would work:
> regex = re.compile(r'[\w\-\.]+\.us)
> regex = re.compile(r'[\w\-\.]+\.au)
> regex = re.compile(r'[\w\-\.]+\.de)
>
> How would I write one to accommodate all three, or, better yet, to
> accommodate a list of them that I can pass into a method call? Thanks!
>
regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)')

If you have a list of country codes ["us", "au", "de"] then you can
build the regular expression from it:

regex = re.compile(r'[\w\-\.]+\.(?:%s)' % '|'.join(domains))

MRAB

On Jul 30, 11:56*am, MRAB <pyt...@mrabarnett.plus.com> wrote:
> Feyo wrote:
> > I'm trying to figure out how to write efficiently write a regex for
> > domain names with a particular top level domain. Let's say, I want to
> > grab all domain names with country codes .us, .au, and .de.
>
> > I could create three different regexs that would work:
> > regex = re.compile(r'[\w\-\.]+\.us)
> > regex = re.compile(r'[\w\-\.]+\.au)
> > regex = re.compile(r'[\w\-\.]+\.de)
>
> > How would I write one to accommodate all three, or, better yet, to
> > accommodate a list of them that I can pass into a method call? Thanks!
>
> *>
> regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)')
>
> If you have a list of country codes ["us", "au", "de"] then you can
> build the regular expression from it:
>
> regex = re.compile(r'[\w\-\.]+\.(?:%s)' % '|'.join(domains))

Perfect! Thanks.

Feyo

On Jul 30, 9:56 am, MRAB <pyt...@mrabarnett.plus.com> wrote:
> Feyo wrote:
> > I'm trying to figure out how to write efficiently write a regex for
> > domain names with a particular top level domain. Let's say, I want to
> > grab all domain names with country codes .us, .au, and .de.
>
> > I could create three different regexs that would work:
> > regex = re.compile(r'[\w\-\.]+\.us)
> > regex = re.compile(r'[\w\-\.]+\.au)
> > regex = re.compile(r'[\w\-\.]+\.de)
>
> > How would I write one to accommodate all three, or, better yet, to
> > accommodate a list of them that I can pass into a method call? Thanks!
>
> >
> regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)')

You might also want to consider that some country
codes such as "co" for Columbia might match more than
you want, for example:

re.match(r'[\w\-\.]+\.(?:us|au|de|co)', 'foo.boo.com')

will match.

rurpy@yahoo.com

On Thu, 30 Jul 2009 10:29:09 -0700, rurpy wrote:

>> regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)')
>
> You might also want to consider that some country
> codes such as "co" for Columbia might match more than
> you want, for example:
>
> re.match(r'[\w\-\.]+\.(?:us|au|de|co)', 'foo.boo.com')
>
> will match.

.... so put \b at the end, i.e.:

regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)\b')

Nobody

Nobody wrote:
> On Thu, 30 Jul 2009 10:29:09 -0700, rurpy wrote:
>
>>> regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)')
>> You might also want to consider that some country
>> codes such as "co" for Columbia might match more than
>> you want, for example:
>>
>> re.match(r'[\w\-\.]+\.(?:us|au|de|co)', 'foo.boo.com')
>>
>> will match.
>
> ... so put \b at the end, i.e.:
>
> regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)\b')
>
It would still match "www.bbc.co.uk", so you might need:

regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)\b(?!\.\b)')

MRAB

In article <mailman.3998.1248989346.8015.python->,
MRAB <> wrote:
>Nobody wrote:
>> On Thu, 30 Jul 2009 10:29:09 -0700, rurpy wrote:
>>
>>>> regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)')
>>> You might also want to consider that some country
>>> codes such as "co" for Columbia might match more than
>>> you want, for example:
>>>
>>> re.match(r'[\w\-\.]+\.(?:us|au|de|co)', 'foo.boo.com')
>>>
>>> will match.
>>
>> ... so put \b at the end, i.e.:
>>
>> regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)\b')
>>
>It would still match "www.bbc.co.uk", so you might need:
>
>regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)\b(?!\.\b)')

If it's a string containing just the candidate domain, you can do

regex = re.compile(r'[\w\-\.]+\.(?:us|au|de)$')
--
Aahz () <*> http://www.pythoncraft.com/

"Many customs in this life persist because they ease friction and promote
productivity as a result of universal agreement, and whether they are
precisely the optimal choices is much less important." --Henry Spencer

Aahz