[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Entity resolution vs. URI resolution

• To: Edwin Goei <edwingo@s...>
• Subject: Re: Entity resolution vs. URI resolution
• From: David Brownell <david-b@p...>
• Date: Fri, 25 Jan 2002 11:05:32 -0800
• Cc: xml-dev@l...
• References: <3C4DE133.21C45105@s...><027f01c1a5cb$c99342e0$6800000a@b...> <3C51A69E.9FF1AC3E@s...>

> I think you are saying that resolving 1) an external entity containing a
> systemID and 2) a URI in general should go through the same resolver: a
> SAX EntityResolver.  To do otherwise, would cause confusion.  Is this
> correct?

Pretty much, though the API (EntityResolver2 or whatever) is a
different issue from the implementation underlying it.  The core
issue that URIs should be interpreted consistently.

As for the API issue, I fail to see a point for adding another API,
and certainly not inside SAX, duplicating existing functionality.
Defining another API encourages use of secondary/tertiary/...
implementations, creating inconsistency/confusion.  But I'm also
aware that large applications are rife with such stuff, maintainance
problems and all.  :)


> By resolve, I mean mapping either 1) an external entity to a SAX
> InputSource or 2) mapping some kind of URI in general to a SAX
> InputSource.  The URIs in #2 may come from sources other than those
> mentioned in the XML REC itself, such as in XInclude or XML Schema
> processing.

I think it's perfectly reasonable to pass such stuff through an object
that happens to be a SAX entity resolver.  Architecturally, the fact
that some of the URIs happen to come from entity declarations is
one that should be irrelevant to how they are "resolved".

- Dave

• References:
  • Entity resolution vs. URI resolution
    • From: Edwin Goei <edwingo@s...>
  • Re: Entity resolution vs. URI resolution
    • From: David Brownell <david-b@p...>
  • Re: Entity resolution vs. URI resolution
    • From: Edwin Goei <edwingo@s...>

• Prev by Date: RE: Generality of HTTP
• Next by Date: Re: Non-validating parser, non-existing DTD,undefined entity and error reporting
• Previous by thread: Re: Entity resolution vs. URI resolution
• Next by thread: an XML data Editor
• Index(es):
  • Date
  • Thread