git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84218 13f79535-47bb-0310-9956-ffa450edef68

(and WILL, if he had apache-1.3.[0-9] before) have a file called
index.html in his .../htdocs/
Accept the presence of either index.html or index.html.en as
a clue for an existing htdocs tree.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84216 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84212 13f79535-47bb-0310-9956-ffa450edef68

	Add the long-awaited IndexOptions DescriptionWidth keyword.
	It was waiting for a way to avoid breaking HTML tags, but
	that's already been done as much as possible -- we can't
	do anything about breaking HTML elements without putting
	a full parser in.

PR:		2324


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84211 13f79535-47bb-0310-9956-ffa450edef68

for the presence of index.html, while it's been renamed to index.html.en
a couple of weeks ago.

Also, didn't we want to move /usr/local/apache/htdocs/manual/ out of
the user's htdocs/ tree to /usr/local/apache/manual/
and Alias /manual/ to /usr/local/apache/manual/ ?
And perhaps note the presence of the i18n versions of the entry page
in the CHANGES file?  [none of these are done in this commit]


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84209 13f79535-47bb-0310-9956-ffa450edef68

representation of the special characters, not the latin1
representation. This would result in invalid URI references
for, e.g., filenames generated by mod_autoindex.c
(when they had special chars in them)


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84208 13f79535-47bb-0310-9956-ffa450edef68

between implied conversion for protocol parts and configured
conversion for document data was missing. The effect of this was that
PDF files could not be read by Acrobat Reader (which sends long
lists of byte ranges in each request) when the server was apache
on ebcdic machines.

Submitted by:	Oliver Reh <Oliver.Reh@FAEDV-N.Bayern.de>
Reviewed by:	Martin Kraemer


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84207 13f79535-47bb-0310-9956-ffa450edef68

	Add a FoldersFirst keyword to the IndexOptions directive, which
	causes subdirectories to always appear first in FancyIndexed
	listings.

Reviewed by:	Ryan Bloom, Martin Kraemer, Jim Jagielski


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84206 13f79535-47bb-0310-9956-ffa450edef68

than write(), and it doesn't have a maximum transfer size of 16kB.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84204 13f79535-47bb-0310-9956-ffa450edef68

the check fails


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84195 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84184 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84183 13f79535-47bb-0310-9956-ffa450edef68

to allow one to override the corresponding variables from config.layout.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84182 13f79535-47bb-0310-9956-ffa450edef68

for supporting Berkeley-DB/3.x....


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84181 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84180 13f79535-47bb-0310-9956-ffa450edef68

More details were described in my original posting two weeks ago:

-----------------------------------------------------------------------
Today (on my birthday ;) I've hacked on and cleaned up snprintf
implementations for a forthcoming string library and discovered the following
bug in ap_snprintf.  If one tries to print a "long long" value which is just a
little bit lower than the maximimum "unsigned long" (not "unsigned long
long"!) value, ap_snprintf treats the value as a (signed) "long" instead of a
"long long":

| $ cat test.c
| #include <stdio.h>
| #include "ap_config.h"
| #include "ap.h"
|
| int main(int argc, char *argv[])
| {
|     char buf[1024];
|
|     ap_snprintf(buf, sizeof(buf), "%qd %qd", 4294967290ULL, 4294967297ULL);
|     printf("ap_snprintf: \"%s\"\n", buf);
|
|     sprintf(buf, "%qd %qd", 4294967290ULL, 4294967297ULL);
|     printf("sprintf:     \"%s\"\n", buf);
| }
| $ cc -I../os/unix -I../include -o test test.c libap.a && ./test
| ap_snprintf: "-6 4294967297"
| sprintf:     "4294967290 4294967297"

The reason is a wrong optimization test in ap_snprintf.c which
can be fixed this way:

Index: ap_snprintf.c
===================================================================
RCS file: /e/apache/REPOS/apache-1.3/src/ap/ap_snprintf.c,v
retrieving revision 1.37
diff -u -r1.37 ap_snprintf.c
--- ap_snprintf.c   1999/10/21 20:44:10 1.37
+++ ap_snprintf.c   1999/11/17 12:22:11
@@ -414,7 +413,7 @@
      * If the value is less than the maximum unsigned long value,
      * then we know we aren't using quads, so use the faster function
      */
-    if (num <= ULONG_MAX)
+    if (num <= ULONG_MAX && is_unsigned)
        return(conv_10( (wide_int)num, is_unsigned, is_negative,
           buf_end, len));

With this patch ap_snprintf and the (FreeBSD) vendor sprintf print the same
values:

| $ cc -I../os/unix -I../include -o test test.c libap.a && ./test
| ap_snprintf: "4294967290 4294967297"
| sprintf:     "4294967290 4294967297"

I think the patch is correct, although I have to admit that because of time
I've not checked the whole quad-fiddling in ap_snprintf.c now. So I do not
commit this myself. Instead someone else should review this first.
-----------------------------------------------------------------------


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84179 13f79535-47bb-0310-9956-ffa450edef68

Submitted by: Andrew McRae <andrew@liquid.com>
Reviewed by: Ralf S. Engelschall
PR: 4973


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84178 13f79535-47bb-0310-9956-ffa450edef68

rlim_t to ap_config.h.

Submitted by: Mark Miller <markm@swoon.net>
PR: 4906


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84177 13f79535-47bb-0310-9956-ffa450edef68

between 1.3.7 and 1.3.9) which changed the DSO extension from `sl' to
`so'. This worked only for modules (where we load the DSO manually),
but horribly fails under HPUX for DSO-based/shared libraries (where our
$SHLIB_SUFFIX_NAME is used, too).

Submitted by: Gary Silverman <gary.silverman@abnamro.com>
Reviewed by: Ralf S. Engelschall
PR: 4974


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84176 13f79535-47bb-0310-9956-ffa450edef68

Submitted by: Steve Atkins <steve@blighty.com>
Cleaned up and reviewed by: Ralf S. Engelschall
PR: 5382


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84175 13f79535-47bb-0310-9956-ffa450edef68

checked against the wrong failure value.

Submitted by: Rick Ohnemus <rjohnemus@systemware-inc.com>
Reviewed by: Ralf S. Engelschall
PR: 5292


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84174 13f79535-47bb-0310-9956-ffa450edef68

in ap_config.h to make C++ compiler happy, too.

Submitted by: SAKAI Kiyotaka <ksakai@netwk.ntt-at.co.jp>
Reviewed by: Ralf S. Engelschall
PR: 5363


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84173 13f79535-47bb-0310-9956-ffa450edef68

for AllowCONNECT and ProxyDomain.

Submitted by: Patrik Grip-Jansson <patrikj@gnulix.org>, Ralf S. Engelschall
Reviewed by: Ralf S. Engelschall
PR: 5319


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84172 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84170 13f79535-47bb-0310-9956-ffa450edef68

Submitted by: Chris Pepper <pepper@tgg.com>
PR: 5360


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84169 13f79535-47bb-0310-9956-ffa450edef68

replaced the wrong "$AWK" with a plain "awk" in the last releases. So we now
both fix this and move the comment which already tried to explain it more
closer to the location to which it applies.

Submitted by: Paul Gilmartin <pg@sweng.stortek.com>
Extended and reviewed by: Ralf S. Engelschall
PR: 5304


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84167 13f79535-47bb-0310-9956-ffa450edef68

explicitly recognized platforms) to avoid side-effects with APACI's --shadow
feature and similar uses where GuessOS' output is used directly on the
filesystem (where pipes are meta-characters!).

Submitted by: Paul Gilmartin <pg@sweng.stortek.com>
Adjusted and reviewed by: Ralf S. Engelschall
PR: 5303


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84166 13f79535-47bb-0310-9956-ffa450edef68

configure script more robust and this way support also a plain `/' as
the argument without resulting in an empty name.

PR: 5291


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84165 13f79535-47bb-0310-9956-ffa450edef68

by always using square brackets consistently.

Submitted by: Masashi Kizaki <kizaki@cpo.dnp.co.jp>
Reviewed by: Ralf S. Engelschall
PR: 5230


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84164 13f79535-47bb-0310-9956-ffa450edef68

integer'' stuff ``#ifndef+#undef+#endif'' pairs were generated instead of
``#ifdef+#undef+#endif'' pairs.

Submitted by: Greg Siebers <gsiebers@vignette.com>
Reviewed by: Ralf S. Engelschall
PR: 5231


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84163 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84162 13f79535-47bb-0310-9956-ffa450edef68

fails to load, the name of the dll that caused the problem. This can be
different from the module dll if it depends on another dll and that
other dll can't be found.


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84161 13f79535-47bb-0310-9956-ffa450edef68

scripts by default.

PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84160 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@84159 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84158 13f79535-47bb-0310-9956-ffa450edef68

be uniformly random. using floating point will use all the bits.

PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84157 13f79535-47bb-0310-9956-ffa450edef68

off the distribution of the random value. The much simpler forms works
just as well, and provides a normal distribution.

PR:
Obtained from:
Submitted by:
Reviewed by:


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84156 13f79535-47bb-0310-9956-ffa450edef68

git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84155 13f79535-47bb-0310-9956-ffa450edef68

on EBCDIC based machines: a character range [a-z] would match much more
than only the islower() characters because in the EBCDIC charset there
are "holes in the contiguity" between a-i, j-r and s-z.

This patch fixes [<lowercase>-<lowercase>] and [<uppercase>-<uppercase>]
ranges by only regarding alphabetic characters between the lower and
upper bound. (Any other range definition remains unchanged.)


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84154 13f79535-47bb-0310-9956-ffa450edef68

want a debugging tool losing information for me for the sake of "looking
good".


git-svn-id: https://svn.apache.org/repos/asf/httpd/httpd/branches/1.3.x@84153 13f79535-47bb-0310-9956-ffa450edef68