Re: mod_isapi

[Greg Stein <gs...@lyra.org>]

On Wed, Jun 28, 2000 at 02:17:57PM -0500, William A. Rowe, Jr. wrote:
>...
> > > And finally, does mod_isaspi have a maintainer?
> > 
> > you? ;)
> 
> Actually, the ISAPI support is several revisions out of spec, and never
> supported filters (interesting convergence of issues, no :?)
> 
> I may take a crack at this, but the priority is far lower than any of
> the cgi or security issues under Win32, including Bill's point that we
> need to provide somesuch support of chmod() or abstract the security
> and privilage model (if that's even possible).  I'm really suggesting that
> if we have no maintainer, that the module is pulled for a few months until
> someone (including me, perhaps, but much later) steps up and runs with it.

I concur with pulling the module. An alternative is to move it to
modules/experimental/ until somebody picks it back up.

Cheers,
-g

-- 
Greg Stein, http://www.lyra.org/

RE: mod_isapi

["William A. Rowe, Jr." <wr...@lnd.com>]

> From: Greg Stein [mailto:gstein@lyra.org]
> Sent: Wednesday, June 28, 2000 6:56 PM
> 
> On Wed, Jun 28, 2000 at 02:17:57PM -0500, William A. Rowe, Jr. wrote:
> >...
> > > > And finally, does mod_isaspi have a maintainer?
> > > 
> > > you? ;)
> > 
> > Actually, the ISAPI support is several revisions out of 
> spec, and never
> > supported filters (interesting convergence of issues, no :?)
> > 
> > I may take a crack at this, but the priority is far lower 
> than any of
> > the cgi or security issues under Win32, including Bill's 
> point that we
> > need to provide somesuch support of chmod() or abstract the security
> > and privilage model (if that's even possible).  I'm really 
> suggesting that
> > if we have no maintainer, that the module is pulled for a 
> few months until
> > someone (including me, perhaps, but much later) steps up 
> and runs with it.
> 
> I concur with pulling the module. An alternative is to move it to
> modules/experimental/ until somebody picks it back up.

++1 on modules/experimental, and turning it into a Loadable module
(not linked to the core.)  I will proceed on that if the rest of the
list agrees.

This will allow anyone to jump in and contribute the fixes necessary
to bring it up to rev 4 or 5 of the spec, add filters, and fix the
fact that ERROR that was returned in the past had a zero value (it
was a blinkin gdi declaration!!!)  I will spend a few minutes on it
looking at what filtering semantics it supports and how those would
be cross-implementable under Apache.

Bill