You are viewing a plain text version of this content. The canonical link for it is here.
Posted to j-dev@xerces.apache.org by "George T. Joseph" <gt...@peakin.com> on 2000/07/20 03:48:38 UTC

Just what is the status of Xerces-J 1.1.x?

It's bad enough that...

...a re-architecture of the internal datatype validation process was tied to
Schema 4/7 support,

...that Schema support STILL doesn't include features that 1.0.3 does,

...many bug fixes and minor enhancements (including ones I submitted) were
somehow "lost" between 1.0.3 and 1.1.2.,

...Bugzilla is gone,

...the spiffy new validation architecture failed to account for post-parse
validation steps like IDREF,

BUT after spending 8 hours re-enabling Schema ID/IDREF checking under the NEW
validation architecture, I check my mail and find that Eric Ye just checked in a
patch that re-enables Schema ID/IDREF checking USING THE OLD DTD VALIDATION
ARCHITECTURE.

What gives?  Is there a plan somewhere that outlines a development strategy for
Xerces (besides the recent flurry of redesign suggestions)?  Before a redesign
gets going, how about a mechanism that will allow contributors (regardless of
"committer" status) to advertise the modules they are currently working on, and
potential contributors to quickly identify areas they can help in.  The STATUS
file and this list just don't do it.

george



Re: Just what is the status of Xerces-J 1.1.x?

Posted by Eric Ye <er...@locus.apache.org>.


> ...that Schema support STILL doesn't include features that 1.0.3 does,
>

Would you mind giving some more information? as far as I know, 1.1.2
provides far more Schema support than 1.0.3, but it is possible some of
features get lost somewhere.

> ...many bug fixes and minor enhancements (including ones I submitted) were
> somehow "lost" between 1.0.3 and 1.1.2.,
>
That is very possible, between 1.0.3 and 1.1.2 there had been 2 major
rearchitecture on the parser and 1 major refactoring on DOM impl for the
sake of full Schema support and DOM impl performance. Please post the bug
fixes and minor enhancements that you think are "lost".

> ...the spiffy new validation architecture failed to account for post-parse
> validation steps like IDREF,
>
> BUT after spending 8 hours re-enabling Schema ID/IDREF checking under the
NEW
> validation architecture, I check my mail and find that Eric Ye just
checked in a
> patch that re-enables Schema ID/IDREF checking USING THE OLD DTD
VALIDATION
> ARCHITECTURE.
>

Sorry if it had wasted you time. Actually some of the built in datatypes,
such as "ID/IDREF", are for the compatibility with DTD attribute validation,
so it is natural to just use the DVD validation architecture. This was fixed
yesterday because somebody reported a bug on this.

> What gives?  Is there a plan somewhere that outlines a development
strategy for
> Xerces (besides the recent flurry of redesign suggestions)?  Before a
redesign
> gets going, how about a mechanism that will allow contributors (regardless
of
> "committer" status) to advertise the modules they are currently working
on, and
> potential contributors to quickly identify areas they can help in.  The
STATUS
> file and this list just don't do it.
>

Point well taken. Currently I am working mostly on adding Schema Structures
support, mainly org.apache.xerces.validators.common and
org.apache.xerces.validators.schema package.

_____


Eric Ye * IBM, JTC - Silicon Valley * ericye@locus.apache.org





Re: Just what is the status of Xerces-J 1.1.x?

Posted by James Duncan Davidson <du...@x180.com>.
on 7/21/00 11:57 AM, Andy Clark at andyc@apache.org wrote:

> James Duncan Davidson wrote:
>> Take a look at the Tomcat STATUS page.
>> 
>> http://xml.apache.org/websrc/cvsweb.cgi/jakarta-tomcat/STATUS.html
> 
> That's a really nice page. We can do something similar and I don't
> mind being in charge of updating the status page. I'd like to get
> a list of action items, though. The obvious ones are:
> 
> 1) Xerces Redesign
> 2) Schema Work

I'd actually break STATUS up into two docs in that particular case.. :)
Until we get further with XRI, then we shouldn't confuse the current parser
STATUS with it's stuff. :)

.duncan


Re: Just what is the status of Xerces-J 1.1.x?

Posted by Andy Clark <an...@apache.org>.
James Duncan Davidson wrote:
> Take a look at the Tomcat STATUS page.
> 
> http://xml.apache.org/websrc/cvsweb.cgi/jakarta-tomcat/STATUS.html

That's a really nice page. We can do something similar and I don't
mind being in charge of updating the status page. I'd like to get
a list of action items, though. The obvious ones are:

  1) Xerces Redesign
  2) Schema Work

And there's a lot of action items already under "Schema Work".

-- 
Andy Clark * IBM, JTC - Silicon Valley * andyc@apache.org

Re: Just what is the status of Xerces-J 1.1.x?

Posted by James Duncan Davidson <ja...@eng.sun.com>.
on 7/19/00 6:48 PM, George T. Joseph at gtj@peakin.com wrote:
 
> What gives?  Is there a plan somewhere that outlines a development strategy
> for Xerces (besides the recent flurry of redesign suggestions)?  Before a
> redesign gets going, how about a mechanism that will allow contributors
> (regardless of "committer" status) to advertise the modules they are currently
> working on, and potential contributors to quickly identify areas they can help
> in.  The STATUS file and this list just don't do it.

The STATUS file and lists do it for a lot of OSS projects -- and a lot of
Apache projects -- just fine. It's a matter of using them and keeping files
up to date. The version that is in there is two months old and really just
talks about things that could be done -- not work in progress.

Take a look at the Tomcat STATUS page.

http://xml.apache.org/websrc/cvsweb.cgi/jakarta-tomcat/STATUS.html

It gives a pretty detailed list of current work, priorities, and person
working it it (if applicable). We should follow similar practice. Quite
honestly, I'm going to be only tracking what goes on with the redesign, and
the current redesign requirements document could grow quite nicely into
something with responsible folks assigned, etc.

Andy, et al -- it would be useful for folks if something like this happend
for Xerces.now :) 

However, in p