DTD documentation issues, collapsing commits etc

[Steven Noels <st...@outerthought.org>]

We are currently experiencing strange problems w.r.t. the generated DTD 
docs: on the Linux platform (where forrestbot runs), 
http://www.krysalis.org/forrest/document-v11.dtdx.html and 
http://www.krysalis.org/forrest/howto-v10.dtdx.html are the same 
document, whereas on my W2K/CygWin system, a local ./build docs yields 
different docs, as it is supposed to do. I don't know since when this is 
happening, I'm suspecting the jar update operation to be the main cause 
since it has caused quite some depreciation errors on libre and nekodtd, 
too.

1) we'll see what we can do to solve this (and the depreciation errors)
2) prior to updating jars on which we depend, we should validate whether 
doing so doesn't cause too much problems and inform the original author 
of upcoming patch work :-<

the crash course CVS updates of the last few days could have been done 
in a more coordinated way, IMO - can't we at least 'claim' on the 
mailing list some work area and inform others that work is in progress? 
Forrest has a small CVS code base, which means there's good chances that 
we step on each other toes working on the same files for different 
purposes. A quick mail to the list could prevent this.

</Steven>
-- 
Steven Noels                            http://outerthought.org/
Outerthought - Open Source, Java & XML Competence Support Center
stevenn@outerthought.org                      stevenn@apache.org

Re: DTD documentation issues, collapsing commits etc

[Steven Noels <st...@outerthought.org>]

Steven Noels wrote:

> We are currently experiencing strange problems w.r.t. the generated DTD 
> docs: on the Linux platform (where forrestbot runs), 
> http://www.krysalis.org/forrest/document-v11.dtdx.html and 
> http://www.krysalis.org/forrest/howto-v10.dtdx.html are the same 
> document, whereas on my W2K/CygWin system, a local ./build docs yields 
> different docs, as it is supposed to do. I don't know since when this is 
> happening, I'm suspecting the jar update operation to be the main cause 
> since it has caused quite some depreciation errors on libre and nekodtd, 
> too.
> 
> 1) we'll see what we can do to solve this (and the depreciation errors)

OK - Marc submitted a patch to me, and as far as I can see, these 
strange problems are solved, now. No depreciation warnings anymore, and 
the DTD documentation issue is solved. See commit mail for the details. 
 From what Marc mentioned to me, we were lucky it ever ran on W2K 
anyhow. We should be fine now.

> 2) prior to updating jars on which we depend, we should validate whether 
> doing so doesn't cause too much problems and inform the original author 
> of upcoming patch work :-<
> 
> the crash course CVS updates of the last few days could have been done 
> in a more coordinated way, IMO - can't we at least 'claim' on the 
> mailing list some work area and inform others that work is in progress? 
> Forrest has a small CVS code base, which means there's good chances that 
> we step on each other toes working on the same files for different 
> purposes. A quick mail to the list could prevent this.

This remark still stands, however :-|

</Steven>
-- 
Steven Noels                            http://outerthought.org/
Outerthought - Open Source, Java & XML Competence Support Center
stevenn@outerthought.org                      stevenn@apache.org