Jean-Francois Arcand wrote:
>
>
> Costin Manolache wrote:
>
>>Jean-Francois Arcand wrote:
>>
>>
>>
>>>>At least in tomcat5 I never use xerces - only the bundled parser, and so
>>>>far I've seen no problems.
>>>>
>>>>
>>>>
>>>Crimson is a very good parser, and it still faster that Xerces in some
>>>case. The only reason I see for bundling xerces is for when we turn on
>>>validation & the web.xml used a schema for validation.
>>>
>>>
>>
>>So why not making the "web.xml validator" a separate download ? If
>>people want to turn on validation, they'll need to get the validator -
>>which will include xerces and some simple script to do the validation.
>>
>>
> That means removing DTD validation from the current implementation? For
> me, seems a feature regression. If we stop validating, we should have a
> very good documentation in other to avoid confusion and hours or
> debugging.
No, I'm not sugesting to remove DTD validation.
Just that it hurts people - and without too much gain.
But it's actually the reverse - the DTD validation is the one that causes
confusion and hours of debugging ( plus slow startup time ). Good software
should be tolerant with the user input.
A tool that generates the web.xml should be strict and make sure the output
is valid. A user who wants to be sure the webapp it ships is correct should
validate the web.xml ( against DTD or schema ). A deploy tool could also
validate web.xml and the tlds.
It is too late to change the behavior for DTD ( for backward compat, etc ),
but at least schema validation can be avoided.
BTW - in the web.xml files that I write I usually remove the declaration on
the top ( to make sure the validation doesn't happen ).
Costin
---------------------------------------------------------------------
To unsubscribe, e-mail: tomcat-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: tomcat-dev-help@jakarta.apache.org
|