Configuration files with long lines and continuation characters are not read properly. PR 55910. Submitted By: Manuel Mausz <manuel-as mausz.at> Committed By: covener

SECURITY (CVE-2014-0117): Fix a crash in mod_proxy. In a reverse proxy configuration, a remote attacker could send a carefully crafted request which could crash a server process, resulting in denial of service. Thanks to Marek Kroemeke working with HP's Zero Day Initiative for reporting this issue. * server/util.c (ap_parse_token_list_strict): New function. * modules/proxy/proxy_util.c (find_conn_headers): Use it here. * modules/proxy/mod_proxy_http.c (ap_proxy_http_process_response): Send a 400 for a malformed Connection header. Submitted by: Edward Lu, breser, covener

Remove some 'register' in variable declaration. Save a few cycles by calling 'apr_isalnum' instead of 'apr_isalpha' and 'apr_isdigit'.

Follow up to r1384924 . Update comment and allocate one extra byte to be safe, even if not needed in the particular case described in r1384924.

s/apr_pstrndup/apr_pstrmemdup/ when applicable

Use %pm in order to save 8k of stack in 'ap_pcfg_strerror'

Be more clever when allocating memory for log item to be escaped. This should be faster and save about 70-100 bytes in the request pool with the default config.

Revert change in order to apply a better solution.

core: Stop the HTTP_IN filter from attempting to write error buckets to the output filters, which is bogus in the proxy case. Create a clean mapping from APR codes to HTTP status codes, and use it where needed.

core: Add the ability to do explicit matching on weak and strong ETags as per RFC2616 Section 13.3.3.

Turn a int into a apr_size_t for a variabe used to compute a string length

Part two of r1467523

Static var not neccessary here

Be more clever when allocating memory for log item to be escaped. This should save about 70-100 bytes in the request pool with the default config.

Remove useless tests. Turn if (*x && apr_isspace(*x)) into if (apr_isspace(*x))

Can't figure out why we allocate len+2 bytes here. Len+1 should be enough.

Add helper function to execute command w args and get one line of output. Allow AuthLDAPBindPassword to have exec: argument like SSLPassPhraseDialog

Add some caching for password hash validation. Password hash functions must be expensive in order to be secure. But if they have to be re-evaluated for every request, performance suffers. As a minimal remedy, cache the most recent result for every connection. This gives a great performance boost if a web browser does many requests on the same connection with the same user+password. In principle, this may keep the plain text password around longer than before. But in practice, there won't be much difference since user+password can already remain in some unused data bucket for longer than the request duration. A proper solution still needs to be found for connections from proxies which may carry requests for many different users. While it currently only requires the conn_rec, the new ap_password_validate() function takes username and request_rec to allow future extensions, like detection of brute-force attempts.

Add an option to enforce stricter HTTP conformance This is a first stab, the checks will likely have to be revised. For now, we check * if the request line contains control characters * if the request uri has fragment or username/password * that the request method is standard or registered with RegisterHttpMethod * that the request protocol is of the form HTTP/[1-9]+.[0-9]+, or missing for 0.9 * if there is garbage in the request line after the protocol * if any request header contains control characters * if any request header has an empty name * for the host name in the URL or Host header: - if an IPv4 dotted decimal address: Reject octal or hex values, require exactly four parts - if a DNS host name: Reject non-alphanumeric characters besides '.' and '-'. As a side effect, this rejects multiple Host headers. * if any response header contains control characters * if any response header has an empty name * that the Location response header (if present) has a valid scheme and is absolute If we have a host name both from the URL and the Host header, we replace the Host header with the value from the URL to enforce RFC conformance. There is a log-only mode, but the loglevels of the logged messages need some thought/work. Currently, the checks for incoming data log for 'core' and the checks for outgoing data log for 'http'. Maybe we need a way to configure the loglevels separately from the core/http loglevels.

remove unnecessary cast

add new ap_bin2hex() utility function

follow-up to r1389481: find/use getloadavg() prototype on Solaris

follow up to r1390564: we just need scope, not control structure

Extract some more data through the server load API: * include/httpd.h (struct ap_sload_t): Add bytes_served, access_count fields. * server/util.c (ap_get_sload): Fill in those fields. Submitted by: Jan Kaluza <jkaluza redhat.com>

wrap var inits in the macro w/ a do-while.

More accurate counts of busy and ready/idle

Break out loadavg from Apache load.... one is quick, the other isn't so much, and so why load things up when wanting just the server loadavg?

minor name change, but I expect most will just want/need/use the current load average, so simplify the name

Would be nice to have some sort of canonical definition of server loading for Apache. So create a struct that holds some useful data. The hope is that for those platforms that lack getloadavg(), people will write replacements.

make the CT check in ap_parse_form_data case-insensitive

Ditch the !, since that would result in the opposite of what was intended

core: use a cheaper way of matching the content-type when parsing form data

core: Be less strict when checking whether Content-Type is set to "application/x-www-form-urlencoded" when parsing POST data, or we risk losing data with an appended charset. PR 53698 Reported by: Petter Berntsen < sluggr gmail.com >

make varbuf functions treat AP_VARBUF_UNKNOWN consistently, improve docs ap_varbuf_pdup(): copying the whole buffer in case strlen == AP_VARBUF_UNKNOWN does not make sense as the caller can not set the exact buffer size, only a minimum. No API change as previously the behavior with AP_VARBUF_UNKNOWN was undocumented. regsub_core(): Checking for vb->buf is useless, it cannot be NULL unless ap_varbuf_init has not been called. ap_varbuf_cfg_getline(): Initially, allocate enough memory to hold an empty line. If strlen == AP_VARBUF_UNKNOWN, use strlen(buf) instead of undefined behavior.

Make ap_regcomp() return AP_REG_ESPACE if out of memory. Make ap_pregcomp() abort if out of memory. This raises the minimum PCRE requirement to version 6.0, released in 2005.

Fix treatment of regex backreferences. r904765 only made half of the necessary changes to remove the use of '&' as an alias for '$0' and allow to escape any character with a backslash.

Clean up size_t abuse, part 2. ap_malloc/calloc/realloc are explicitly excluded from this cleanup as they must be signature identical to the clib functions, and although the definition of size_t has been flakey, the definition of those functions appears to be generally clean since ANSI C.

Limit length of lines in .htaccess to 8K again, to reduce DoS potential. Make ap_varbuf_cfg_getline() strictly enforce the max_len parameter.

Add lots of unique tags to error log messages

Fix integer overflow in ap_pregsub. This can be triggered e.g. with mod_setenvif via a malicious .htaccess CVE-2011-3607 http://www.halfdog.net/Security/2011/ApacheModSetEnvIfIntegerOverflow/

No need to zero memory that we will overwrite anyway

Remove magic numbers; ensure that an invalid nmatch is corrected tagged

mod_include: Add support for application/x-www-form-urlencoded encoding and decoding.

Improve handling of maxlen = APR_SIZE_MAX, noticed by Jim. Use apr_pregsub_ex() and maxlen = 0 for unlimited in mod_substitute.

Limit ap_pregsub() to 64K, add ap_pregsub_ex() for longer strings and with better error reporting. Modify ap_varbuf_regsub() to be similar to ap_pregsub_ex().

flip operator in comment for r1182887 and add some whitespace to highlight the restriction on nmatch.

Force the honoring of AP_MAX_REG_MATCH

Shut up gcc/glibc warning about ignoring write()'s return value. This may actually fix a real bug in case the error log is directed to a FIFO.

Some varbuf enhancements: - Introduce new ap_varbuf_pdup() and ap_varbuf_regsub() functions. - Fix some bugs in ap_varbuf_strmemcat(). - Make ap_varbuf.buf point to an empty string if no buffer has been allocated, yet.

Cleanup effort in prep for GA push: Trim trailing whitespace... no func change

Add wrappers for malloc, calloc, realloc that check for out of memory situations. Use them in most places where malloc, and friends are used. This results in clean error messages in an out of memory situation instead of segfaulting or silently malfunctioning. In some places, it just allows to remove some logging code. PR 51568, PR 51569, PR 51571.

Give a slightly more direct hint when no global ServerName is set. I still think this buries the lead, but it's hard to rewrite to change the emphasis yet maintain googleability.

Add ap_varbuf API for resizable buffers. Increase length limit of lines in the configuration file to 16MB. Increase length limit of lines in the group file to 16MB. PR: 45888, 50824, 43084 Windows and Netware build changes are untested.

Various code cleanup PR: 51398 Submitted by: Christophe Jaillet <christophe jaillet wanadoo fr>

We already have ap_str_tolower(), so also add ap_str_toupper() function and use it where possible.

Introduce ap_(get|set)_core_module_config() functions/macros and use them everywhere. We know that the core module has module_index 0. Therefore we can save some pointer operations in ap_get_module_config(cv, &core_module) and ap_set_module_config(cv, &core_module, val). As these are called rather often, this may actually have some (small) measurable effect.

Change the ap_cfg_getline() and ap_cfg_getc() to return an error code. Also: - Make ap_cfg_getline() return APR_ENOSPC if a config line is too long. - Add ap_pcfg_strerror() function to convert ap_cfg_getline's return value into a nice message. - Adjust definition of ap_configfile_t accordingly. Not bumping MMN because it has already been bumped today.

core: AllowEncodedSlashes new option NoDecode to allow encoded slashes in request URL path info but not decode them. Change behavior of option "On" to decode the encoded slashes as 2.0 and 2.2 do. PR 35256, PR 46830.

Expose "new" ap_parse_form_data() function instead of requiring mod_request for any module that may want to parse form data...

Make ap_pstr2_alnum accept const char dest

Move ap_resolve_env to core.c, in preparation for adding mod_define-like variable support.

Fix compile error

Make it clear that '_' is a possible char and accepted

Make the balancer shared memory slot more human readable. But we need to make it a "safe" filename, so create 2 funcs which do that.

Fix some Windows build issues

Another ooops: properly return APR_SUCCESS

Add a generic pool cleanup function that sets a pointer to NULL and use it to replace various pool cleanup functions.

Added header include for getpid() prototype on Win32.

c isn't used after being set (Fortify)

Pull out "does request have a body" logic to a central canon function and use that for the 100-Continue OK check. Should likely also start using this in the various other places we do this "have body" check thruout the codebase...

Use the new APLOG_USE_MODULE/AP_DECLARE_MODULE macros everywhere to take advantage of per-module loglevels

Make ap_pregsub(), used by AliasMatch and friends, use the same syntax for regex backreferences as mod_rewrite and mod_include: Remove the use of '&' as an alias for '$0' and allow to escape any character with a backslash. Document the use of $0. PR: 48351

Spelling fix in comment.

Bring back OS/2 support. Reverses r758929 with a little bit of conflict resolution.

remove OS/2 platform support

Add support for escaping all non-ascii chars to ap_escape_html, and use it to fix PR#25202: encoding="entity" doesn't work as advertised in mod_include. For backport, this'll need an ABI-preserving version that'll be a minor MMN bump. But if we do that in /trunk/, it'll never change.

* Change some "apr_palloc / memcpy" to apr_pstrmemdup PR: 39519 Submitted by: Christophe JAILLET <christophe.jaillet wanadoo.fr> * Remove unnecessary casts

* Move ap_timeout_parameter_parse from mod_proxy.c to server/util.c and thus make it part of the public API.

Remove all references to CORE_PRIVATE.

Rename the ap_escape_path_segment_b function as suggested by Ruediger Pluem, and make a corresponding minor bump.

mod_session: Add a generic session interface to unify the different attempts at saving persistent sessions across requests.

* Revert to old behaviour: Even if we detect a bad character: Formally decode string correctly and do not mess up the string. Let the caller decide what it wants to do with the result according to our return code.

Fix r578332 for backslashing filesystems, as noted by rpluem

* Rationalise the two ap_unescape_url versions * Fix ap_unescape_url_keep2f to work as documented (keep %2F encoded, bringing it into line with the documentation as well as common sense). * Introduce suggestion of an extension to the API The most relevant open PR is 41798

* We need to supply a (char *) to ap_str_tolower not a char.

Multiple trivial fixes from Christophe JAILLET PR 38699, 39518, 42005, 42006, 42007, 42008, 42009 The patches are all his, and are sufficiently trivial to review at a glance.

update license header text

Update the copyright year in all .c, .h and .xml files

Bring forward the fix for CVE-2005-3352 already on the branches: * modules/mappers/mod_imagemap.c (imap_url): Escape the referer. * server/util.c (ap_escape_html): Escape the " character. Submitted by: mjc Reviewed by: fielding, jorton

No functional Change: Removing trailing whitespace. This also means that "blank" lines consisting of just spaces or tabs are now really blank lines

Clean up c2x code-path by not double-setting the prefix variable. * server/util.c (c2x): Take a new prefix parameter instead of hard-coding '%' and then conditionally setting 'x' for half of the cases. Reviewed by: nd

Add ap_append_pid(); This is performed enough to warrant a function I think, especially with the fact that the mapping of getpid() to APR_PID_T_FMT isn't consistant in some areas.

Fix trailing backslashes in configuration directives. PR: 34834 Submitted by: Timo Viipuri <viipuri dlc.fi> Reviewed by: Paul Querna

Ensure that we always return a FQDN... This is not guaranteed by either apr_sockaddr_info_get() or apr_getnameinfo(). Also note that under OS X, apr_getnameinfo() always seems to fail (at least on my machines) so if we don't specific ServerName at the global setting, we get a warning.

APRized ap_get_local_host()

Move the POSIX reg* implementations into the ap_* namespace; internalise the ap_reg*<->PCRE wrapper: * configure.in: Add srclib/pcre to the include path. * include/ap_regex.h: Renamed from include/pcreposix.h. Prefix all constants with AP_; prefix all functions and types with ap_. Define AP_DECLARE to nothing if necessary. Remove regcomp error codes. * include/httpd.h: Include ap_regex.h not pcreposix.h. (ap_pregcomp, ap_regexec, ap_regfree): s/regex_t/ap_regex_t/. (ap_regexec, ap_regerror): Prototypes moved to ap_regex.h. * server/util.c (regex_cleanup, ap_pregcomp, ap_pregsub, ap_pregfree): Adjust for ap_ prefixed types. (ap_regexec, ap_regerror): Removed. * server/Makefile.in: Build util_pcre.c. * server/util_pcre.c: Copied from srclib/pcre/pcreposix.c; remove use of PCRE-internals to do error mapping; rename types to add AP_/ap_ prefixes as above. Use APR includes. (ap_regerror): Use apr_snprintf. * srclib/pcre/Makefile.in: Don't build pcreposix.c into libpcre.la. * modules/*: Update to use new type and constant names. PR: 27750 (part one) Submitted by: Andres Salomon <dilinger voxel.net>, Joe Orton

Update copyright year to 2005 and standardize on current copyright owner line.

veto and revert win64 patch: 64bit changes must percolate from the bottom (APR/system) up -- we can't give the client a 64bit API and then cast it to 32bits internally without introducing security holes on other platforms.

WIN64: API changes to clean up Windows 64bit compile warnings

* server/util.c (ap_resolve_env): Rewrite to avoid buffer overflows. Submitted by: Andr�� Malo Reviewed by: jorton, mjc

handling of encoded non-slashes was borked in the AllowEncodedSlashes path thanks to FirstBill for pointing that out!

Fix the handling of URIs containing %2F when AllowEncodedSlashes is enabled. Previously, such urls would still be rejected with 404.

fix name of The Apache Software Foundation

fix copyright dates according to the first check in

apply Apache License, Version 2.0

update license to 2004.

SECURITY [CAN-2003-0020]: escape arbitrary data before writing into the errorlog. Reviewed by: Mark J Cox

switch to APR 1.0 API (which is still in flux) because of the changes to the argument lists of apr_mmap_dup and apr_socket_create, 2.1-dev won't build with apr and apr-util's 0.9 branch anymore

axe a special check for EOF+data from apr_file_gets()... apr_file_gets() doesn't return that combination of conditions any more

Following the previous commit, adjust the actual parameters of ap_strcmp_match() and ap_strcasecmp_match() to use 'expected' rather than 'exp' to avoid shadowing the global exp() function. Submitted by: Justin Erenkrantz <justin@erenkrantz.com>

Don't run regex cleanups before exec Submitted by: Jeff Trawick Reviewed by: Brian Pane

Minor MMN bump: Forward port: Escape special characters (especially control characters) in mod_log_config to make a clear distinction between client-supplied strings (with special characters) and server-side strings. This was already introduced in version 1.3.25. Obtained from: Patch in 1.3.25-dev by Martin

Avoid passing too many characters to getenv. Make sure that only the environment variable string is passed.

finished that boring job: update license to 2003. Happy New Year! ;-))

here we go. add a directive that will keep %2f from being decoded into '/', allowing the *_walk to do their magic and return 404 if it's in the path, and allowing it in the path-info.

2.0 -> 2.1

Use APR_STATUS_IS_EOF rather than a check against APR_EOF

Style police were called back from their vacation *No functional changes*

A couple of observations

Security rollup for 2.0.40 release. Tag and roll baby.

Removed "tolower(++s)" idiom from ap_strcasestr(), to avoid side-effects on any platform with a macro-based tolower() that references its argument multiple times.

stop using APLOG_NOERRNO in calls to ap_log_?error()

Updated ap_getword_white() to use the same coding style (and optimizations) as ap_getword()

Added a check to make sure that h_aliases is not NULL before we try to dereference it in the for(...) loop. Attempting to dereference a NULL pointer was causing a fault if there were no aliases found.

Adapt to the rename of apr_explode_localtime to apr_time_exp_lt in APR. Submitted by: Thom May <thom@planetarytramp.net>

One for 2.0.34 - to fix the problem of utf-8 configuration files saved from MeSsed up editors that believe zero byte joiner lead bytes signify a utf-8 file. Someone point me at an RFC ... gheesh. FWIW; fixes bugs report 10125

Update callers of apr_explode_gmt to apr_time_exp_gmt. The function was renamed in apr. Submitted by: Thom May <thom@planetarytramp.net>

Correct our list of escape characters to include percent and the carriage return on Win32/OS2, and modify the \r \n escaping to account for the fact that Win32/OS2 don't pass these characters through a true argv[] mechansim; replace them with a whitespace since they effectively are for most applications.

Update our copyright for this year.

apr-utils rename apr_ansi_time_to_apr_time and apr_exploded_time_t. PR: Obtained from: Submitted by: Thom May <thom@planetarytramp.net> Reviewed by: Ian Holsman

Whoops - this patch is required by the last request.c change. Fulfills Brian Pane's suggestion that we always overallocate by one to make /foo/bar quickly convertable to /foo/bar/

Optimization for ap_getparents: skip past all the leading characters of the path that aren't '.' rather than copying those bytes onto themselves

Optimization: replace apr_pstrndup() with apr_pstrmemdup() in ap_escape_html() to eliminate a redundant scan through the string.

Reduced the number of strlen operations in ap_getword()

Modified ap_make_full_path to minimize the number of strlen operations

Added NetWare to the Win32 ifdef list

Fix a seg fault in mod_include. When we are generating an internal redirect, we must set r->uri to "", not a bogus string, and not NULL. [Ryan Bloom]

Revert this patch. It indicates some fundamental brokenness with the apr_filepath_root() fn, which should simply return '/' for an absolute path on (most) unix, and advance the path pointer to the first non-'/' character [if only Win32 and others were so simple ;-]

I don't know if this is the Right Way, but it certainly does make things work better right now. httpd-test is much much happier with this patch (ie, we can serve pages again :) I'm still seeing issues with mod_include, but that's probably a different problem... will investigate. Submitted by: Doug MacEachern

I'm still not 100% certain on this test. Regardless, it doesn't impact Unix, and this was the bug that Jeff's (none the less correct) fix masked.

Eliminated ap_os_[systemcase|[case_]canonical]_filename() and move ap_os_is_path_absolute() into util.c (now relies on apr.)

It doesn't get any simpler than this

The original answer was evil, return NULL if the file name has no path!

Reconsidered the wisdom of this change... back to the drawing board. The ap_ prefix is still good, IMHO

Clean up namespace badness with s/configfile_t/ap_configfile_t/

*) fix inline handling. we had: apr_inline, APR_INLINE, USE_GNU_INLINE, and INLINE. Now, we just have APR_INLINE and APR_HAS_INLINE. - convert all usage - note that apr_general messed up the defn (compared to apr.h) - simplify the inline decision logic in os/*/os.h - simplify the code in os/*/os-inline.c *) toss ap_checkconv() [no longer used]

Update copyright to 2001

write these in terms of strchr and strrchr so the compiler can possibly use builtin, optimized functions. [ they're stupid functions, but what the hell ]

Fix a potential segfault on all platforms. David Reid fixed this for BEOS, but this could actually happen on any platform, so we should be safe and protect ourselves. Submitted by: Cliff Woolley <cliffwoolley@yahoo.com> Reviewed by: Ryan Bloom

Clean up some of the includes: - explicitly include apr_lib.h since ap_config.h doesn't - use apr_want.h where possible - use APR_HAVE_ where possible - remove some unneeded includes

Fix more breakage on BeOS...

renaming various functions for consistency sake see: http://apr.apache.org/~dougm/apr_rename.pl PR: Obtained from: Submitted by: Reviewed by:

Get the correct IP address if ServerName isn't set and we can't find a fully-qualified domain name at startup. PR: 7170 Submitted by: Danek Duvall <dduvall@eng.sun.com>

Use the appropriate APR_FINFO_flags for the apr_stat/lstat/getfileinfo calls to avoid ownership and permissions on Win32 when they are not required, and until they are implemented.

BeOS BONE shouldn't be using this BeOS R5 check, so alter the check to make it work correctly. Shouldn't this be APR'ized?

The changes required for the APR_FINFO_wanted argument to apr_stat/lstat/getfileinfo. These are -NOT- optimal, they are simply the required changes to get the server working. The size of the patch is a warning about how we need to really look at what we are trying to accomplish with all of these stat/lstat calls.

Move initgroupgs, ap_uname2id and ap_gname2id from util.c to mpm_common.c. These functions are only valid on some platforms, so they should not be in the main-line code. These functions are also not portable to non-unix platforms, so they don't really belong in APR. Since they are only used in MPMs, for right now, I am moving them to mpm_common.c

Remove ap_chdir_file. This function is not thread-safe, and nobody is currently using it. The proper way to do this, is to use the apr_create_process API.

The big change. This is part 3 of the apr-util symbols rename, please see the first commit of srclib/apr-util/include (cvs apr-util/include) for the quick glance at symbols changed.

Fix a comment

add pool parameter to ap_is_directory and ap_is_rdirectory

As we now use pools in the logging of the failure to get a hostname, use ap_log_perror() as per Brians suggestion. Submitted by: Brian Havard <brianh@kheldar.apana.org.au>

This stops a segfault on my machine where I haven't got a hostname for the function to find.

Switch to the APR-provided APR_CHARSET_EBCDIC feature test macro.

Force all Apache functions to be linked into the executable, whether they are used or not. This uses the same mechanism that is used for APR and APR-util. This may not be the correct solution, but it works, and that is what I really care about. This also renames CHARSET_EBCDIC to AP_CHARSET_EBCDIC. This is for namespace correctness, but it also makes the exports script a bit easier.

Axe ap_get_virthost_addr(). This <address>[:port] parse routine is not used anymore and it doesn't handle IPv6, so it is best to drop it. If/when mod_tls (Netware) is ported to 2.0 it can use apr_parse_addr_port().

Fix two spelling and syntax problems in ap_get_local_host().

*) Compensate for recent changes in the APR headers. Specifically, some files need to specifically include stdio.h, or a particular apr_*.h header. *) Adjust callers of apr_create_process() to deal with the extra "const" *) Add "const" to args of ap_os_create_privileged_process()

Change references of AF_ to reflect APR_ so they should work on all systems.

Not all platforms have INADDR_NONE defined by default. Apache used to check for this and the define it if needed. Since APR also needs this check it makes more sense for APR to just check and export a symobl that is always available. Submitted by: Branko �ibej <brane@xbc.nu> Reviewed by: Ryan Bloom

Killing ap_os_is_filename_valid. Left actual win32 code, since it is moving into apr and the check_safe_file call.

This is always displayed to the console. There is no reason to print the date and time to the console, so we should use APLOG_STARTUP to suppress them

Start of moving to apr_port_t in the server code. This will probably the first of a few...

A more thorough example is appropriate. Also simplify the behavior and always return root (/) if n < 1, which is no worse than returning elems if n > elems. PR: Obtained from: Submitted by: Reviewed by:

somebody was a Very Bad Boy when they inserted casts into this function. casting away the const was absolutely wrong... the warnings were saying the return value type needed to be fixed. did that and torched the casts. who still thinks casts are a good idea? :-)

The final line of the config file was not being read if there was no \n at the end of it. This was caused by apr_fgets returning APR_EOF even though we had read valid data. This is solved by making cfg_getline check the buff that was returned from apr_fgets. If apr_fgets return APR_EOF, but there was data in the buf, then we return the buf, otherwise we return NULL.

Renamed all MODULE_EXPORT symbols to AP_MODULE_DECLARE and all symbols for CORE_EXPORT to AP_CORE_DECLARE (namespace protecting the wrapper) and retitled API_EXPORT as AP_DECLARE and APR_EXPORT as APR_DECLARE. All _VAR_ flavors changes to _DATA to be absolutely clear. Thank you Greg, for the most obvious suggestion.

Completes the port of the <Directory /> handling for Win32... there is still a patch needed for OS2 ... Brian?

Modify ap_make_dirstr_prefix, platforms with HAVE_DRIVE_LETTERS or NETWARE can request 0 elements, returning the '/' root. [William Rowe, Tim Costello] from 1.3, with documentation

Remove some deprecated functions

Port over the config directory stuff...

finish cleaning up after a change i made over 3 years ago.

A few more (last?) ap_xlate->apr_xlate changes covering stuff not completely handled in the big apr rename last week.

Remaining cleanup of ap_ -> apr_ and AP_ -> APR_ transformation... see src/lib/apr/apr_compat.h for most details. Also a few minor nits to get Win32 to build. PR: Obtained from: Submitted by: Reviewed by:

prefix libapr functions and types with apr_

Add APR_EOL_STR for a platform specific text delimiter, provided by apr.h (defined in apr.h.in and apr.hw). This is needed -only- in APR created files (true raw files) such as logs. It is not required in any splat to screen (stderr/stdout) formatting, nor any html markup. Some other modules slipped through in the prior apr_strings.h commit. Sorry 'bout that. PR: Obtained from: Submitted by: Reviewed by:

Move all APR functions related to strings to their own directory, and create a new header for those functions. This is the first step to removing the apr/lib directory completely, and moving those files/functions to descriptive directories.

Fix some bad ap_log_error() invocations. Comment on a bad ap_log_rerror() invocation. Almost all of this is in code never compiled.

Grab OtherBill's latest 1.3 code to provide a default ServerName and log the error. Submitted by: "William A. Rowe, Jr." <wrowe@lnd.com>

Include strings.h for strcasecmp(), strncasecmp(), and bzero(). Include time.h for time(). This removes a bunch of compiler warnings with gcc -Wall on AIX. Submitted by: Jeff Trawick, Victor Orlikowski

blast the old names for the status codes

Correct the problem where the only local host name that the IP stack can discover are 'undotted' private names. If no fully qualified domain name can be identified, the default ServerName will be set to the machine's IP address string.

More consification, correct command initialisation.

Constification.

Command handler revamp. Note that this makes the code produce a LOT of warnings!

Turn on buffering for config file reads. This is dependent on ap_fgets() doing the right thing. Brian Havard implemented buffering for ap_fgets() on Win32 recently; OS/2 had it already. This provides it for Unix. changes to ap_read(), ap_getc(), ap_fgets() for Unix: 1) this fixes a problem in where ap_open() where a lock is created for non-buffered files 2) this fixes problems setting rv correctly in the ap_read() buffered path 3) since ap_read() works as expected, it is possible to make ap_getc() and ap_fgets() even smaller 4) ap_fgets() no longer cares about '\r'

Get the ap_config.h cleanup working on FreeBSD by adding includes of additional system headers. A platform check in logresolve.c to determine whether or not we include <arpa/inet.h> was converted to a feature check.

Protect system header files with the appropriate macros.

Clean up a big chunk of ap_config.h. This basically stops ap_config.h from including any files. Because of this change, other files must include their own headers. I also cleaned up a couple of other bugs in some modules because I had to compile them all. I expect this to break multiple platforms, but this will be fixed over time. The massive configure cleanup is almost done. I will go through the files one more time after this commit.

Remove the final vestiges of stat.h from Apache 2.0. All calls are now to ap_stat. This also adds the new function ap_lstat(). This function is analogous to lstat.

Remove waitpid from the config checks and all calls to waitpid from the server.

Remove a bunch of string functions from Apache. These are basically standard string functions like strstr, strcasecmp, etc that Apache used to define for platforms that don't have them. These functions and the feature tests have moved down to APR where they really belong. In doing this, I am also able to remove a bunch of tests from the Apache configure process.

Remove all occurances of strerror from Apache. ap_strerror works just fine with standard errno values, and it is more portable. This also allows me to remove the check for strerror from Apache's configure script.

Remove all remaining instances in Apache code of ap_{block,unblock}_alarms.

The following is a patch to fix Dirk's patch from March 14 so that mod_rewrite will work without caveats. Basically Dirk was just dropping patterns of the form ${x} if x didn't resolve to an environment variable. That breaks mod_rewrite since mod_rewrite uses the same syntax for some of its non-environment directives. The patch below simply puts the behavior for unresolved values back to what it was before Dirk's patch. If it is unresolved, it is just left in the directive as is. Further processing will handle the value. Dirk's optimization still holds if x is resolved. A small price is paid if x is not resolved, but that shouldn't happen often. (Note: some tweaks to the comments were negotiated offline after Paul posted the patch to new-httpd.) Submitted by: Paul Reder <rederpj@raleigh.ibm.com> Reviewed by: Jeff Trawick

Cleanup ap_config.h fallout for Win32. Only a few very minor changes to ap_config.h and util.c, but they might hurt someone... please watch those two carefully.

APR-ize some EBCDIC logic which handles conversion between characters and hex digits

Pick up various EBCDIC fixes from 1.3 (from Martin Kraemer and Oliver Reh originally according to the change log).

Replace hsregex with PCRE. The PCRE library is always used in 2.0, regardless of what system we are on. Currently, we are using the POSIX wrappers that PCRE provides.

PR: Obtained from: Submitted by: William Rowe Reviewed by: Created ap_base64.h to extract those declarations for clarity from ap.h CORE_PRIVATE, httpd.h do not belong in library functions, removed from ap.lib Use apr headers for declarations in ap.lib stuff, kill AP_LONG from ap_sha1.h Move credit to lib/apr/lib/ap_snprintf.c from ap.h for authorship

clean up cmd_parms: config_file is no longer valid; end_token is bogus; add directive. move configfile_t and functions from httpd.h to http_config.h new signature for ap_build_config() (since config_file removed from cmd_parms) add "data" to ap_directive_t for future use by modules. add filename. syntax checking for section-close directives: a section-open must exist, the section-close must be </FOO>, and the open/close must match. the file as a whole must be properly balanced (issue errors for each unmatched section-open). </FOO> command_rec structures are obsolete. Remove from http_core.c. do not store </FOO> directives in the config tree. clean out section-close logic from http_core.c (and old, related comments) <Limit> and <LimitExcept> must walk their children. new mechanism in ap_check_cmd_context() for testing enclosure in a Directory/Location/File: find_parent() <IfModule> and <IfDefine> must pass cmd->context when walking the children several places: we had a walk followed by ap_get_module_config(). that assumed the walk would create a config that we could fetch, which is not true -- it is possible that the children are all from other modules (e.g. the <Files> section in httpd.conf-dist has no "core" directives). using ap_set_config_vectors() ensures we get a structure, and it returns it to us. [ note: when we had </Directory> (and friends) in the tree, the config would get created; removing the directive removed the config; this was a bitch to track down :-) ]

Change strncasecmp() definition to match Bill's prototype. Fixes compile break on OS/2.

Change ap_context_t to ap_pool_t. This compiles, runs, and serves pages on Linux, but probably breaks somewhere.

Make the server work without asking for buffered file I/O. This removes all APR_BUFFERED flags from the server, and it fixes a minor bug in the ungetchar patch (I forgot a file when comitting).

Update to Apache Software License version 1.1

Another one in the department of fairly useless patches which are best described as feature creep. Allows ${ENV} constructs in the config file. This avoids the need for mod_perl or m4 cleverness whilst mainting some of the usefullness. It does not do (of course) multiline things or anything that clever. Feel free to flame me. PR: Obtained from: Submitted by: Reviewed by:

Backport the CSS security fixes to Apache 2.0a. Or is that forward port? My sense of direction is all confused. PR: Obtained from: Submitted by: Reviewed by: