From: Kenneth Pronovici (pronovic@skyjammer.com)
Date: Wed Nov 08 2000 - 17:50:48 EST
> I think this is completely bogus and the reason of your problem.
> You just assume that in all and any URI-Reference computation leading
> to an external fetching this sequence should be added. You don't have
> the right to do this.
First, I'm open to suggestions on a better way to do this... Maybe
I've misunderstood the point of the entity loader?
I don't have a corporate DTD server, but I use XML on multiple machines.
I'm generating XML on machine A, which has its copies of a DTD in
/dvl/fds/dtd. I'm using TIBCO to send messages from machine A to a process
on machine B, which has its own copies of the DTDs in a (possibly different)
directory. The idea is that each document references a DTD, and then each
individual program knows where to find that file. Am I mis-using my entity
loaders?
> However I must admit (I discovered it today) that the entity loader
> is called and the URI location of the resource effectively used has
> been changed, then this must me made available back to the parser. I
> think it is through the filename field of the returned
> xmlParserInputPtr but I must double check that it's done the right way
> (especially if it uses HTTP and a redirection).
I'm also confused on this. Regardless, the issue still is that the entity
loader is being called twice for the same entity. I don't see how there is
any way I could implement an entity loader and make this work under
those circumstances... if it's already been called once, and libxml
saves the result, and then calls it again with that saved result,
how could my external entity loader ever be right, even disregarding
whether you think I should be using it the way I am?
KEN
---- Message from the list xml@rpmfind.net Archived at : http://xmlsoft.org/messages/ to unsubscribe: echo "unsubscribe xml" | mail majordomo@rpmfind.net
This archive was generated by hypermail 2b29 : Wed Nov 08 2000 - 18:43:30 EST