Thanks for doing this.I think my confusions stemmed from thinking that
axkit: requests were used for very different tasks than http:, though
after the discussion I think the two schemes are conceptually quite
similar, though obviously implemented quite differently.
With that in mind, do you think the first bit of the page would read
better as the POD below.
Cheers,
--
M.
=head1 Getting other documents
During an AxKit request it's possible to request other resources. This
is helpful to build Y-shaped rendering pipelines amongst other things.
Of course you can do this with a normal C<http:> request, but if the
other resources lie on the same (virtual) server and are served by
AxKit, it's often better to make a special C<axkit:> request instead:
=over
=item * This is necessary wherever AxKit runs single threaded, for
example under Windows.
=item * C<axkit:> requests are more efficient.
=item * C<axkit:> requests imply a hostname and port rather than
specifying one directly. This is particularly helpful in moving
between proxy and no-proxy setups.
=back
=head2 Examples
An example in XSLT might look something like:
<xsl:apply-templates select="document('axkit:foo.xml')"/>
Where the file F<foo.xml> is relative to the current URI.
In XSP you might want to do something similar using the Util taglib:
<util:include-uri href="axkit:/includes/bar.xml"
xmlns:util="http://apache.org/xsp/util/v1"/>
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
