code errors and making sure it is disabled when DBM support is not available
on the used platform.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80397 13f79535-47bb-0310-9956-ffa450edef68

- Add usage information
- Add Linux hint
- Correct regex recently added by Ken


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80396 13f79535-47bb-0310-9956-ffa450edef68

Same as the voted take 3 but I've replaced the *(cp-N) syntax with cp[-N]
because most of us like this one more.

Submitted by: Ralf S. Engelschall
Reviewed by:  Jim Jagielski, Dean Gaudet, Martin Kraemer,
              Dirk-Willem van Gulik, Ralf S. Engelschall


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80395 13f79535-47bb-0310-9956-ffa450edef68

	collision insurance.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80394 13f79535-47bb-0310-9956-ffa450edef68

already truncated in that case.

Joel's patch worked for subrequests (i.e. via #include virtual)... I fixed
it to deal with all requests... and I couldn't help but clean up some crufty
code in proxy_http_canon().

Submitted by:	Joel Truher <truher@wired.com>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80393 13f79535-47bb-0310-9956-ffa450edef68

the comparisons are done against the url as stored in r->filename... it's
kind of lame, but this is just how the proxy works.  So for example,
if you're using ProxyPass the url stored in r->filename won't match what
was passed to parse_uri() ... so r->parsed_uri isn't useful.

I reverted to what was before, plus a few performance tweaks I couldn't
resist.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80392 13f79535-47bb-0310-9956-ffa450edef68

when I looked at old vs. new logic in this area I don't see the problem
that I thought was there.  So I think this is closed.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80391 13f79535-47bb-0310-9956-ffa450edef68

that would be nice to hide.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80390 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80389 13f79535-47bb-0310-9956-ffa450edef68

Support for Linux... which incidentally has an R symbol type for read-only
data (this is a gnu feature maybe, it may even be an optional posix thing).


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80388 13f79535-47bb-0310-9956-ffa450edef68

And if we're going to do that it seems silly not to export day_snames
and make use of that.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80387 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80386 13f79535-47bb-0310-9956-ffa450edef68

Submitted by:	Mark J Cox <mark@awe.com>


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80385 13f79535-47bb-0310-9956-ffa450edef68

there some reason this one is still around?  I'm ditching it anyhow unless
someone explains why it's necessary.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80384 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80382 13f79535-47bb-0310-9956-ffa450edef68

	API routines - and fix a problem with os/unix/os.h and -DHIDE.
	A similar fix may be needed for the other os/* directories,
	and a new 'make depend' needs to be run once a default value for
	the HIDE rule is determined.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80381 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@80380 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@80379 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@80378 13f79535-47bb-0310-9956-ffa450edef68

all of its usages with ap_cpystrn() variants in the past.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80377 13f79535-47bb-0310-9956-ffa450edef68

platforms don't define it although they provide the rand() function :-(


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80376 13f79535-47bb-0310-9956-ffa450edef68

and commit an updated version of this header file.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80375 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@80374 13f79535-47bb-0310-9956-ffa450edef68

PR:		1908


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80373 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80371 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80370 13f79535-47bb-0310-9956-ffa450edef68

Aha!  Now I understand why Martin modified the uri regex the way he did...
and I documented it.  Fix the "GET //foo/bar" bug.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80368 13f79535-47bb-0310-9956-ffa450edef68

and I documented it.  Fix the "GET //foo/bar" bug.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80367 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80366 13f79535-47bb-0310-9956-ffa450edef68

wasting resources trying to solve what may end up being bugs in solaris.

PR:		1779, 1854, 1904


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80365 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80363 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80362 13f79535-47bb-0310-9956-ffa450edef68

fill the `Committed Code Changes' to make Ken happy ;-)
(and update the available votes)


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80361 13f79535-47bb-0310-9956-ffa450edef68

	mailed regularly whereas CHANGES is not, and STATUS is typically
	*slightly* more accurate (people often forget to update CHANGES).


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80360 13f79535-47bb-0310-9956-ffa450edef68

in 1.3b5 (because users would assume that a known bug listed under 1.3b3
but not 1.3b5 is fixed).


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80359 13f79535-47bb-0310-9956-ffa450edef68

an interpreter for a script. PR#1712 will be reopened.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80358 13f79535-47bb-0310-9956-ffa450edef68

- remove 'Committed Code Changes' because it's still too late to update it


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80357 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80356 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80355 13f79535-47bb-0310-9956-ffa450edef68

fragments to new-httpd already though, and I've run it through a lot
of tests.  There's still a few TODOs but what's here should work for
almost everyone.

general:
- remove check_fulluri, replace parse_uri with a much simplified version
    based on util_uri functions
- remove r->hostlen, add r->unparsed_uri (mmn bump coming up shortly)
- move r->proxyreq calculation to the translate_names phase

vhosting:
- fix absoluteURI problems with vhosting
- update vhtest suite to test absoluteURI vhosting
- serveralias support tweaked to consume less memory/run faster (submitted
    by Chia-liang Kao <clkao@cirx.org>)
- add matches_request_vhost() function which can be used to ask http_vhost
    if a hostname:port pair would match r->server

util_uri:
- remove WITH_UTIL_URI, make WITH_UTIL_URI the default
- remove the has_foo fields since they're inferred by testing the strings for
    NULL vs. non-NULL
- #if 0 Martin's hand-coded parse_uri_components because it doesn't pass my
    test suite (see below)
- change parse_uri_components_regex to parse_uri_components -- so the
    implementation is hidden from everyone's view
- write mod_test_util_uri which does various exhaustive tests on
    parse_uri_components


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@80354 13f79535-47bb-0310-9956-ffa450edef68