The helper function _weblinks_checker_fix_url in weblinks_checker module uses the parse_url php function.

With URLs that have a # character in the path, like:
http://play.cuatro.com/on-line/#/21-dias/ver/21-dias-en-la-mina

the parse_url function assigns to ["fragment"] element of the array returned by the function: /21-dias/ver/21-dias-en-la-mina

and the url saved by weblinks is chopped to: http://play.cuatro.com/on-line losing the rest of the path.

Is there any way to avoid that?

Thanks

Comments

nancydru’s picture

nancydru
she/her/hers
Primary language English
Location Boston
commented

At the moment the only thing that comes to mind is to encode the "#" ().

miguel_angel’s picture

miguel_angel commented

Hi.

I've tried to encode de '#' into the URL:

http://play.cuatro.com/on-line//21-dias/ver/21-dias-en-la-mina

or

http://play.cuatro.com/on-line/%23/21-dias/ver/21-dias-en-la-mina

but it doesn't work.

The browser doesn't decode the %23 and the result is a page not found error.

If I write:

http://play.cuatro.com/on-line/%40/21-dias/ver/21-dias-en-la-mina

The browser decodes the %40 to 'A' correctly. I don't know why it refuses to decode the %23 to '#'.

I think that the first way (encoding with ) will have the same parsing problems due to the existence of the '#' character.