Re: SourceFactory considered harmful (Bug #15279)
Carsten Ziegeler wrote:
Sylvain Wallez wrote:
Hi team,
Bug #15279 (WriteableSource doesn't work with "context:" URLs) reveals
what seems to me a design flaw in the SourceResolver.
The SourceResolver uses two kinds of classes can to do its job :
SourceFactories and URLFactories. When an URL has to be resolved, it
first checks if any SourceFactory handles the URL's protocol, and if
there's none, it falls back to searching an URLFactory then creates an
URLSource.
This is true for 2.0.x, but not for 2.1 where the new source resolver
package from excalibur is used.
URLFactories are deprecated (starting with 2.1)
With this source resolving, the same implementation, URLSource I think,
is used. I guess the problem is that the current implementation of
the context: protocol does not use the WriteableSource implementation.
I checked ContextSourceFactory, which calls resolver.resolveURI() with
the translated URL. If that URL is a "file:", then we should get a
(writeable) FileSource.
It's been a long time since I wanted to dig in the source resolution in
2.1. Time to do it ;-)
Sylvain
--
Sylvain Wallez Anyware Technologies
http://www.apache.org/~sylvain http://www.anyware-tech.com
{ XML, Java, Cocoon, OpenSource }*{ Training, Consulting, Projects }
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
RE: SourceFactory considered harmful (Bug #15279)
Sylvain Wallez wrote: > Hi team, > > Bug #15279 (WriteableSource doesn't work with "context:" URLs) reveals > what seems to me a design flaw in the SourceResolver. > > The SourceResolver uses two kinds of classes can to do its job : > SourceFactories and URLFactories. When an URL has to be resolved, it > first checks if any SourceFactory handles the URL's protocol, and if > there's none, it falls back to searching an URLFactory then creates an > URLSource. This is true for 2.0.x, but not for 2.1 where the new source resolver package from excalibur is used. URLFactories are deprecated (starting with 2.1) With this source resolving, the same implementation, URLSource I think, is used. I guess the problem is that the current implementation of the context: protocol does not use the WriteableSource implementation. Carsten Carsten Ziegeler Open Source Group, S&N AG - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
Re: SourceFactory considered harmful (Bug #15279)
Jeremy Quinn wrote:
On Saturday, Dec 14, 2002, at 22:44 Europe/London, Sylvain Wallez wrote:
So I propose to strongly deprecate URLFactories (means @deprecate the
source and issue a warning in the logs if some are declared), and
rewrite existing implementations as SourceFactories.
Thoughts ?
You could well be right Sylvain but please keep in mind, when I
re-factored SourceWritingTransformer (in 2.1) after Carsten's
additions, this particular aspect (resolving the different Sources)
was the bit I really did not understand very well! Are we sure that is
is implemented correctly in SWT in the first place? (!!)
Well, this is not SWT-specific, since SWT just uses the SourceResolver
to get a Source. That's what happens inside the SourceResolver that
causes problems !
Sylvain
--
Sylvain Wallez Anyware Technologies
http://www.apache.org/~sylvain http://www.anyware-tech.com
{ XML, Java, Cocoon, OpenSource }*{ Training, Consulting, Projects }
-
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
Re: SourceFactory considered harmful (Bug #15279)
On Saturday, Dec 14, 2002, at 22:44 Europe/London, Sylvain Wallez wrote: So I propose to strongly deprecate URLFactories (means @deprecate the source and issue a warning in the logs if some are declared), and rewrite existing implementations as SourceFactories. Thoughts ? You could well be right Sylvain but please keep in mind, when I re-factored SourceWritingTransformer (in 2.1) after Carsten's additions, this particular aspect (resolving the different Sources) was the bit I really did not understand very well! Are we sure that is is implemented correctly in SWT in the first place? (!!) regards Jeremy - To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]
