[squeak-dev] XMLTokenizer problem with ampersand

Levente Uzonyi leves at elte.hu
Tue Jun 2 02:10:17 UTC 2015

XMLTokenizer is not suitable to parse HTML documents. XML and HTML may 
look similar, but are very different.
We used to use Soup[1] to parse HTML pages.


[1] http://squeaksource.com/Soup.html (watch out for versions which may 
not be Squeak-compatible)

On Tue, 2 Jun 2015, karl ramberg wrote:

> Hi,thanks for the info.
> I guess I need a HTMLTokenizer for what I'm doing. I had issues with &nbsp as well, with the current XMLTokenizer
> Karl
> On Mon, Jun 1, 2015 at 11:01 PM, Jakob Reschke <jakob.reschke at student.hpi.de> wrote:
>       I guess this will not help you, but a standalone ampersand is not
>       valid XML (it is the leader for entities, if you want to have a
>       literal ampersand in the text, the markup must be &amp;), hence I
>       would not expect any XML tokenizer or parser implementation to accept
>       it.
>       HTML is more relaxed about this, so a standalone amapersand is valid,
>       but you would need some kind of HTMLTokenizer and I do not know
>       whether there is such thing for Squeak. Anyone else knows one?
>       Best regards
>       Jakob
>       2015-06-01 20:05 GMT+02:00 karl ramberg <karlramberg at gmail.com>:
>       > Hi,
>       > I'm parsing some html docs but the XMLTokenizer chockes on a '&' followed by
>       > a space in a string.
>       > I guess '&' is used for other stuff than a 'and' in html and it causes error
>       > when used in plain text.
>       >
>       > Does anybody have fix for this?
>       >
>       > Karl

More information about the Squeak-dev mailing list