Hi Darsin,
If we try it this way
<([^>]+)>;([^<]+)</$1>
this will only work for tags without parameters, like <ul>something</ul> and will not work for tags with parameters like <a href=something>anything</a>
The idea is to repeat as $1 what was matched as an opening tag.
So let's enhance it so it matches tags with parameters:
<(w+)(s+[^>]+)*>(
[^<]+)</$1>
_/ ______/ ___/ ; |
| ; | ; | `-- first
$1 ; $2 ; $3 ; match
base params text ; repeated
part of of the ; between
opening tag ; tags
tag
I have not tested it live, so there might be some flaws/typos.
But I hope you catch the idea.
On 12/23/05, Darsin <gmail.com" target="_blank">darsin gmail.com> wrote:
>;
> Hi all
> I am trying to build a regex which can get an html tag from given html
> text for eg if given:
>;
> <P>this is a test html</P><P
> class=left>saumitra</P><P>class=center>chaturvedi</P>;<P
> class=center>darpan sinha</P>
>
> then i can get above 4 tags as four seperate matches as shown below:
>; <P>this is a test html</P>
> <P class=left>saumitra</P>
> <P class=center>chaturvedi</P>
> <P class=center>darpan sinha</P>
>
> I have built the below string to be used in Visual Basic:
>; <[a-z][a-z]*d?s*(class=[a-z]*)*>[^<;]+</[a-z][a-z]*d?>
> Here d is used to test one or more occurences of a digit (for eg in
> case an H1, H2, etc is used)
> Pattern works fine if the first closing tag after an opening tag is
> same ie. <p>some text</p>. But fails if the tag is soemthing like:
> <ul> <li>saumitra<;/li><;li>chaturvedi</li></ul>
> The result i get are two matches <li>saumitra<;/li> and
> <li>chaturvedi</li> where as i want one match
> i would like to match the starting tag with the ending tag and pickup
> all the text (html or non-html) in between, including the tags as well.
> Thus i should get all the text between <ul> and </ul>; as one match. In
> case i get all its children as the second and third match then i wont
> mind at all.
> Any help in this regard would be appreciated.
>
>
--
best regards, ; Eugeny
|