On 26/06/2005, at 1:28 AM, Jim Gallacher wrote:
Changed import mechanism.
(I wasn't paying too much attention to this whole discussion, so
maybe someone else could comment on how this changed and the possible
impact on user code?)
New version attribute.
Added publisher.get_page() which allows the import of a published page.
I know I am not part of the development team, so can only make
suggestions,
but I personally don't believe that the new import mechanism which has
been
developed just for publisher should be included.
There are already problems which can arise due to the mixing of the
"import"
statement and "apache.import_module()", and now a third system is being
brought into the picture which behaves different enough that in itself
may
cause confusion, but when wrongly mixed with the others could result in
a
mess. I know I don't want to have to be answering questions about it and
if it did come down to it, I probably wouldn't.
Thus, I would be suggesting leaving this feature out. You shouldn't
however
let this delay 3.2. People understand that module importing is broken
and
how to work around it, leaving it that way a bit longer is not going to
cause any new harm. At least, less harm than if this new scheme has to
be
ripped out at a later date.
Instead, fixing of the main module importer should be carried out. I
know
that others disagree with some of my ideas on that, but I do believe it
can
be fixed and the only features you would loose would be the ability to
import
parts of packages when using "apache.import_module()" explicitly and
this
only because to support packages properly when autoreloading is
supported
overly complicates the design to the point that it isn't practical. This
ability of the current importer to deal with packages is actually broken
in various ways anyway and I don't think dropping it would cause any
real
problem. Note that this wouldn't prevent use of packages or parts of
packages
from standard Python locations from any of the Handler directives.
Anyway, the point of this mail is simply to register my viewpoint that
the
new publisher specific import mechanism should not be included. I am not
at this time going to be drawn into further discussions about how to fix
the current import mechanism though. I am working on a document which
describes the basics of how the module importer works and all its
problems
and issues. When I am finished that, then we might start discussing it
again. I am about half done on this document and hopefully in another
week it will be ready to be used as the basis of some discussions on how
to progress this issue.
Graham
