«

I m working on my perl regex code, where I have to parse a html line like :

<a href="/question?id=15422849"><p>MY text here 1</p><p>MY text here 2</p><p>MY text here 3</p></a>

I am doing something like:
$string =~ m/(.*)href(.*)/;

But this is not helping me in what I want. I want something closer to following text:

"MY text here 1 MY text here 2 MY text here 3"

Can some give some ideas ?

vivek_12315 <> wrote:
>I m working on my perl regex code, where I have to parse a html line like :
>
> <a href="/question?id=15422849"><p>MY text here 1</p><p>MY text here 2</p><p>MY text here 3</p></a>
>
>I am doing something like:
>$string =~ m/(.*)href(.*)/;
>
>But this is not helping me in what I want. I want something closer to following text:
>"MY text here 1 MY text here 2 MY text here 3"
>
>Can some give some ideas ?

Your Question used to be Asked Frequently. Please see

perldoc -q "remove html"

jue

In article <678ed33b-3479-46bb-b6d3->,
vivek_12315 <> wrote:

> I m working on my perl regex code, where I have to parse a html line like :
>
> <a href="/question?id=15422849"><p>MY text here 1</p><p>MY text here
> 2</p><p>MY text here 3</p></a>
>
> I am doing something like:
> $string =~ m/(.*)href(.*)/;
>
> But this is not helping me in what I want. I want something closer to
> following text:
>
> "MY text here 1 MY text here 2 MY text here 3"


http://search.cpan.org/dist/HTML-Strip/Strip.pm

Henry Law <> wrote:
>On 13/02/13 00:16, vivek_12315 wrote:
>> I m working on my perl regex code, where I have to parse a html line like :
>>
>> <a href="/question?id=15422849"><p>MY text here 1</p><p>MY text here 2</p><p>MY text here 3</p></a>
>
>I appreciate that you call yourself a newbie, and to you what I'm about
>to suggest may seem complicated and difficult; but that's the way we all
>learn ...
>
>Have you thought of parsing the HTML properly, using a module like
>HTML::Tree or HTML::TreeBuilder? The hardest part is choosing the
>module; after that you should find it moderately easy to use it do what
>you want, since it's pretty simple. And once you've done it it will
>probably be a lot better than hand-cranked parsing code.
>
>Note to all concerned: I'm not joining in the "you can't parse HTML with
>regexes" thread. In this case, at least, I'm sure that's perfectly
>possible.

Actually for this particular example it is almost trivial(*):
s/<.*?>//g;
Of course this is going to fail as soon as the HTML code becomes a tiny
bit more complex.

*: almost because it doesn't add the space characters between the
individual paragraph elements.

jue