Re: Re: [Zope-dev] WebDav Bug? -- id != name?

2001-09-20 Thread Casey Duncan

On Thursday 20 September 2001 11:03 am, Kevin Dangoor allegedly wrote:
[snip]

 Less magic is good.

 Longer term, it seems like this can be addressed by the Component
 architecture, can't it? Ideally the WebDAV Presentation Component would be
 given some clues about how to represent the object in WebDAV. Or maybe
 there will be a selection of WebDAV components that provide a variety of
 ways to deal with it.

  soapbox
 
  IMHO, I'd rather see the things that make using file extensions
  a pain in the first place go away. The death to index_html part
  of my proposal is a first step toward that. If web-oriented
  tools expect to be able to use file extensions, then Zope needs
  to do what people expect.

 ...while still balancing the desire to easily access objects and methods
 via Python. Tricky.

 Kevin

Possibly just giving the user the ability to specify a name for an object 
separate from it's id so that an extension could be used could be a 
short-term solution. 

Or even just a property DAV_extension that lets you specify the file 
extension when using webDAV. When files are uploaded via WebDAV, the ext is 
stripped from the id and stored in this property. It is reappended by Zope 
when accessing the object through WebDAV.

At least that way it is magic is under your direct control.

-Casey

___
Zope-Dev maillist  -  [EMAIL PROTECTED]
http://lists.zope.org/mailman/listinfo/zope-dev
**  No cross posts or HTML encoding!  **
(Related lists - 
 http://lists.zope.org/mailman/listinfo/zope-announce
 http://lists.zope.org/mailman/listinfo/zope )



RE: Re: [Zope-dev] WebDav Bug? -- id != name?

2001-09-20 Thread Brian Lloyd

 Or even just a property DAV_extension that lets you specify the file 
 extension when using webDAV. When files are uploaded via WebDAV, 
 the ext is 
 stripped from the id and stored in this property. It is 
 reappended by Zope 
 when accessing the object through WebDAV.
 
 At least that way it is magic is under your direct control.

The problem (a problem) is that accessing the object through
WebDAV is not a tightly bounded thing. DAV is built over HTTP,
and (intentionally) GET and PUT requests by DAV clients are not 
distinguishable from regular HTTP intent (thus the source vs.
rendered problem).

An example: a DAV client does a PROPFIND on a folder, and the 
folder lies and says mymethod.html is there (when its real 
id is just mymethod). The client will likely then do a HEAD 
or a GET on it - which doesn't go through any DAV code at all 
and the lookup will fail with a 404 Not Found.

That means that if we were to try to implement this, it is not a 
simple matter of changing the DAV code, as this would affect 
publishing, security, potentially all name lookup code and who 
knows what else :(

Again, that does not mean it could not happen, but someone will 
need to invest a non-trivial amount of time to come up with a 
proposal that comprehensively assesses (and provides recommendations
for addressing) all of the side effects so that the community can 
make a reasonable decision on whether it is worth the effort.

Brian Lloyd[EMAIL PROTECTED]  
Zope Corporation   www.zope.com



___
Zope-Dev maillist  -  [EMAIL PROTECTED]
http://lists.zope.org/mailman/listinfo/zope-dev
**  No cross posts or HTML encoding!  **
(Related lists - 
 http://lists.zope.org/mailman/listinfo/zope-announce
 http://lists.zope.org/mailman/listinfo/zope )