These -1s cause us a lot of grief.  When is a signed int not
	a signed int?  When it's an unsigned int or long!  {Sigh}


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83249 13f79535-47bb-0310-9956-ffa450edef68

	Make sure that redirect processing looks for the "Location"
	field in both headers_out and err_headers_out.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83247 13f79535-47bb-0310-9956-ffa450edef68

to the correct places so they only get called for the pipes that have
actually created.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83246 13f79535-47bb-0310-9956-ffa450edef68

library. regerror() is needed by PHP3.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83245 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83244 13f79535-47bb-0310-9956-ffa450edef68

work here (causes a new warning) because the stuff isn't used anywhere, of
course.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83243 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83242 13f79535-47bb-0310-9956-ffa450edef68

PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83241 13f79535-47bb-0310-9956-ffa450edef68

PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83240 13f79535-47bb-0310-9956-ffa450edef68

so just avoid any warnings in snprintf.c


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83239 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83238 13f79535-47bb-0310-9956-ffa450edef68

speed "tuneups" but also a renaming of some internal functions.
After some sleep, I thought that making the 'quad' specific functions
as having the "noted" names made better sense, although the
code itself doesn't care.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83236 13f79535-47bb-0310-9956-ffa450edef68

conversion functions when the number value doesn't warrant use of
quads


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83235 13f79535-47bb-0310-9956-ffa450edef68

that var_type isn't defined. To work around possible buggy compilers, we
create an enum type and then the var and then assign it


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83234 13f79535-47bb-0310-9956-ffa450edef68

stupid vars


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83233 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83232 13f79535-47bb-0310-9956-ffa450edef68

ap_snprintf() can now handle quads, assuming the compiler knows
about them, and we now determine which type is the off_t
typedef


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83231 13f79535-47bb-0310-9956-ffa450edef68

quads (long longs). Note that we depend on AP_LONGEST_LONG being
set correctly. This change does not effect the current operation
of ap_snprintf, so it's OK to add right now while we work on
the rest of it


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83230 13f79535-47bb-0310-9956-ffa450edef68

PR: 3815


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83229 13f79535-47bb-0310-9956-ffa450edef68

1. Eliminate unnecessary (I hope :-) call to wait_for_multiple_objects
2. Clean up parent process code that handles shutdown and restart signals
3. Honor shutdown and restart events immediately, even it it means loosing connections in the TCP stack's listen queue. Restarts triggered by MaxRequestPerChild will caue the child process to handle all connections in the listen queue before exiting.
PR: 3815


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83228 13f79535-47bb-0310-9956-ffa450edef68

strong opinion on this stuff because I'm still part of the guys who thing the
major/minor distinction is mostly useless.  A single value is enough IMHO...


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83227 13f79535-47bb-0310-9956-ffa450edef68

This solves problems with DSO modules which use the regex library.

Submitted by: Jens-Uwe Mager <jum@helios.de>, Ralf S. Engelschall
Reviewed by: Ralf S. Engelschall


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83226 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83225 13f79535-47bb-0310-9956-ffa450edef68

PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83224 13f79535-47bb-0310-9956-ffa450edef68

it dereferenced a NULL pointer (well disguised as '\0' in place of NULL)


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83223 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83222 13f79535-47bb-0310-9956-ffa450edef68

	Allow SetEnvIf* to set things according to the protocol of
	the request.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83221 13f79535-47bb-0310-9956-ffa450edef68

	Add a patch and clean out some stalities.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83219 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83218 13f79535-47bb-0310-9956-ffa450edef68

Submitted by: Rex Dieter <rdieter@math.unl.edu>
Ported to 1.3.6: Ralf S. Engelschall
PR: 3997


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83216 13f79535-47bb-0310-9956-ffa450edef68

Submitted by: Jan Gallo <gallo@pvt.sk>
Reviewed by: Ralf S. Engelschall
PR: 3690, 4373


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83215 13f79535-47bb-0310-9956-ffa450edef68

their brain-dead /bin/sh.

PR: 4372


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83214 13f79535-47bb-0310-9956-ffa450edef68

Submitted by: Todd Vierling <tv@pobox.com>
Reviewed by: Ralf S. Engelschall
PR: 4310


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83213 13f79535-47bb-0310-9956-ffa450edef68

PR:		4366, 679
Submitted by:	Petr Lampa <lampa@fee.vutbr.cz>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83212 13f79535-47bb-0310-9956-ffa450edef68

comments made it clear to me that what I'd written wasn't so clear ;)


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83205 13f79535-47bb-0310-9956-ffa450edef68

an evil hack, it causes memory corruption.  You cannot destroy a #included
subrequest of mod_include... look for all uses of "request_config" in
mod_include, in particular look at the code around the ap_join_pool call.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83204 13f79535-47bb-0310-9956-ffa450edef68

	Someone finally stood up and made the ReadmeName and HeaderName
	features use subrequests.  Not only that, but they can be
	parsed for SSIs too!

PR:		1574, 3026, 3529, 3569, 4256
Submitted by:	Raymond S Brand <rsbx@rsbx.net>
Reviewed by:	Ken Coar


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83203 13f79535-47bb-0310-9956-ffa450edef68

	".def" isn't a universally-recognised suffix; versions of
	make that don't know about it will think ".def.a" is a target
	rather than a rule.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83200 13f79535-47bb-0310-9956-ffa450edef68

stat fails

Submitted by:	Ed Korthof <ed@bitmechanic.com>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83199 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@83197 13f79535-47bb-0310-9956-ffa450edef68