httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "William A. Rowe, Jr." <>
Subject RE: cvs commit: apache-1.3/src/os/win32 service.c
Date Tue, 13 Jun 2000 00:25:08 GMT
> From: Greg Stein []
> Sent: Monday, June 12, 2000 6:56 PM
> Actually, I have little concern about the stability of the Win32 stuff. I
> can't imagine you putting in all that effort to make it worse :-). My
> hesitance was pretty simple: 1) have these Win32 changes had enough testing?
> (i.e. we knew the behavior and problems before, but we're starting over
> now)

Actually, not a far over as the 2.0 port already is.  But yes, 9x is a 
radical change, Andrew and I have both been pounding the he\\ out of it,
but you never can be too sure.  I invited those following another
"when will 1.3.13 get here" post on to grab the
next tarball, try it, read the revised docs, and comment.  Our problem,
not yours :-)  So no, everyone *please* test away.

> and 2) did something outside of Win32 get tweaked accidentally.

I'm nearly certain, no.

> I'll just hit up ViewCVS and do a diff between the head and the 1.3.12
> release for http_main.c and see what the total diffs are. My 
> problem, not yours :-)

Speaking of which, has anyone told you today how cool ViewCVS really 
is???  If I wouldn't gag with a spoon, I'd tell you it's bitchin'

One suggestion, a one-click to compare a version with the absolute 
current version (no selecting two files for diffs, etc).  Your own
review might even been an incentive to do it :-)

> And have no fear: I think the work that you've done for the 
> Win32 stuff (on both branches) is great.

Actually, I have more hope and joy for 2.0 - I refused to backport the
argument passing via StartService (that will allow us to bring up the
Services panel and key -D NO_THAT_SERVER in the params :-)  (Or do the
same with the -k start command line.)  And I'm looking at the list for
2.0, let's see here...

ScriptInterpreterSource registry

Invokes the script specified by the registered "ExecCGI" handler, or if
absent, then the "Open" command handler.

ScriptInterpreterSource registry-strict

Invokes the script specified by the registered "ExecCGI" handler.
If absent, then the cgi will -not- be executed.

I'm just full of ideas :-)  Not for 1.3.x though... just want to see
it work, and some user requests and contributions be fulfilled.

View raw message