Getting content without tags

Watt

Hello,

How (using preg match all) can I get all content form my page and don't get HTML tags, I wrote:

	preg_match_all("/[>](.*?)[<-]/ims", $contents, $getall, PREG_SET_ORDER);

But it does not work for all texts on my page :

<div>text</div>

- text will be matched

<div>
text
</div>

- it wont.

Can you help me?

bradgrafelman

I would suggest that you use [man]DOM[/man] instead. Much more versatile than regular expressions, since it actually parses and understands the HTML.

halojoy

you should look into what bradgrafelman adviced.
it is probably much more convenient than preg_match

One trick I use sometimes
is to remove all newlines in HTML before the preg_match search.
This makes it one long string.

This is what I use

$thehtm = str_replace(array("\r\n","\n"), '', $html);

Watt

Hi, thanks for answering, method with str replace unfortunately didnt work.

I haven't used DOM method - I forund:

$doc = new DOMDocument();
$doc->loadHTML($myhtml);
echo $doc->saveHTML();

But it doesn't change anything. Can you show/guide me how should it look like ?

cahva

Try PHP Simple HTML DOM Parser. I like it and its easy to use.

Watt

I found that, but it's a class, but I need 'basic' code. Is there any possibility to do it in easier way? (Without including unnessesary classes and functions)

halojoy

If you just want al but not HTML tags
then you can use [man]strip_tags[/man]
I would think you have considered this.
So, maybe you can not use it for some reason.

$string = strip_tags ( $str [, $allowable_tags ] )

NogDog

How about:

$stripped = strip_tags($contents);

Watt

No, no, no I want to translate my page and send contents to google translate on get translated content. Bud google 'damages' tags so I have to split it - leave markers in my page and send to google only texts, than I have to replace markers with translated text, so I realy need DOM function 🙂

NogDog

In your original code, why do you have the hyphen ("-") in the 2nd character class? I would think all you'd need is:

'/>(.*?)</s'

...or even:

'/(?<=>).*?(?=<)/s'

However, using the DOM would be more robust and allow you to be more selective/specific as to what gets selected.

Weedpacket

Watt wrote:
No, no, no I want to translate my page and send contents to google translate on get translated content.

Isn't the translation done by Google and then you hand-tune it?

Or are you just wanting to add Google's translation script on your pages?

What sort of "damage" does the translator do? I'd be surprised if Google doesn't parse HTML properly, since it's a big chunk of what it does all day.