[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [uddi-spec] Use of maxRows and listHead in V3
I was wondering if we might use E_invalidValue - I know
it usually relates to keyed references, but it seems more appropriate than
E_fatalError, which I usually take to be an error inside the server (something
like a hardware error, for example). As long as we require the message to be
explicit in calling out the maxRows value as the flaw, there shouldn't be any
confusion.
Tony Rogers
From: Daniel Feygin [mailto:feygin@unitspace.com] Sent: Monday, 16 August 2004 8:39 To: Rogers, Tony; uddi-spec@lists.oasis-open.org Cc: 'Andrew Hately' Subject: RE: [uddi-spec] Use of maxRows and listHead in V3 Ok, maxRows=0 is useful indeed.
For negative numbers the only error code that seems to make
sense is E_fatalError.
Daniel From: Rogers, Tony [mailto:Tony.Rogers@ca.com] Sent: Friday, August 13, 2004 2:57 AM To: Daniel Feygin Cc: uddi-spec@lists.oasis-open.org; Andrew Hately Subject: RE: [uddi-spec] Use of maxRows and listHead in V3 If we return no results on maxrows = 0, but return the number of results,
then the program can use that number to prepare for a second call
to receive the results - that's the usual rationale for including a "return
0 results but tell me how many there are" call. It's less useful for these
results, because they vary in size, but that is what I'd see it being used for.
That, or some statistical analysis application, where they are interested in the
number of results rather than the content.
Anyway, it seemed to me that there was some (albeit small) use for
accepting maxRows=0.
As for negative numbers - you have a point - maybe we should reject the
request. Would you care to suggest a suitable error?
Tony Rogers
|
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]