I had a look at this quite a few weeks ago when I was looking to do some
aggregation where the <map:part>'s were different web sites. I
trawled the archive and found the patch referred to in bug 10208.
In the http case it is more difficult to solve because the exception
was further down. You can't easily determine that you'll get a
problem until you try to access the URL and it fails (because the
machine is down). I can't remember the details
but I had a good fiddle with it. I'm still working with 2.0.1
(because we had to put a stake in the ground a while back) so I
figured I'll investigate it when I upgrade and actually need to do the
work requiring the aggregation. (The code from CVS HEAD used some
different pieces and some other things which might have made it
easier to get the exception).
WebServiceProxyGenerator has been announced since then. Would it
help in trying to solve the problem of aggregating parts from remote
sites and being able to easily trap the error. I say that without
having investigated - I just made a mental note of it.
One thing that would also be nice is if such a thing would keep track
of sites which are down so you don't get the long network delay. That
is a whole can of worms, though...
So while it's on topic any hints? I don't mind doing some work on it.
...Peter
Vadim Gritsenko writes:
> > From: icewind [mailto:[EMAIL PROTECTED]]
> >
> > I have two separate questions relating to
> > <map:aggregate>
> >
> > 1. I have a bunch of <map:part> tags within a
> > <map:aggregate> tag. Each xml file that is referenced
> > in the map:part tags contains lists of the full path
> > names to other xml files. (The whole purpose of this
> > thing is to make a list of links the user can click
> > on) The problem is, if for instance, one of the
> > filenames in one of the <map:part> files is
> > mis-spelled, the whole things will stop, and none of
> > the following map:part files will be processed. (A
> > mis-spelling causes this because I extract the
> > contents of a tag out of each xml file listed in each
> > of the <map:part> files. If it cant find a filename,
> > it just stops.) So, the question is how can I have
> > cocoon just skip over a file that it cant find and go
> > on to the next map:part gracefully?
>
> See http://nagoya.apache.org/bugzilla/show_bug.cgi?id=10208
---------------------------------------------------------------------
Please check that your question has not already been answered in the
FAQ before posting. <http://xml.apache.org/cocoon/faq/index.html>
To unsubscribe, e-mail: <[EMAIL PROTECTED]>
For additional commands, e-mail: <[EMAIL PROTECTED]>
