W3C home > Mailing lists > Public > www-rdf-interest@w3.org > August 2002

CWM patches?

From: David Saff <saff@MIT.EDU>
Date: 16 Aug 2002 10:53:40 -0400
To: RDF-Interest <www-rdf-interest@w3.org>
Message-Id: <1029509620.5754.35.camel@jane>

All,

I'm doing some hacking on a module that uses pieces of cwm, and have a
few suggested patches.  The CWM page at
http://www.w3.org/2000/10/swap/doc/cwm.html
suggests this mailing list as a place for discussion on such matters,
but if that is wrong, please ignore and forgive the rest of this message
and point me in the correct direction.

Anyway, I was looking at some possible optimization of CWM (especially
in endFormula), and decided that I should run against the latest CVS
source for best results.  My code in some places creates an RDFSink with
a null genPrefix, expecting that the usingRunNamespace functionality
will auto-generate.  It appeared, however, that in the latest checked-in
version, usingRunNamespace was broken.  This patch in RDFSink (against
the CVS version current as of noon 2002/08/16 GMT) seemed to fix the
problem:

>>>>>>>>>>
diff /home/saff/cwm_playspace/2000/10/swap/RDFSink.py ./RDFSink.py
112c112,113
< 	    self._genPrefix = ns + "#_g"
---
>             # dsaff note: # had been added twice
> 	    self._genPrefix = ns + "_g"
113a115
> 
196c198
< 	    ns =  self._genPrefix[1]
---
> 	    ns =  self._genPrefix # not sure why this was indexed before: [1]
198c200
< 	    self.bind("run", ns[:hash+1])
---
> 	    self.bind("run", (SYMBOL, ns[:hash+1]))
<<<<<<<<<<

In addition, I didn't think it made sense to force clients of diag.py to
set the global variable chatty_flag:

>>>>>>>>>>
diff /home/saff/cwm_playspace/2000/10/swap/diag.py ./diag.py
19a20
> chatty_flag = 0
<<<<<<<<<<

Finally, Symbol.representation seemed broken:
>>>>>>>>>>
diff /home/saff/cwm_playspace/2000/10/swap/thing.py ./thing.py
161c161
<         return "<" + self.uriref(base) + ">"
---
>         return "<" + self.uriref2(base) + ">"
195a196,197
>         if base == None:
>             return self.uriref()
<<<<<<<<<<

Corrections and gentle reproof appreciated.  Thanks,

	David Saff
Received on Friday, 16 August 2002 11:10:52 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:51:55 GMT