1 Changelog for the c-ares project
3 * Aug 4 2008 (Daniel Stenberg)
7 * Users (usually, but not always) on 2-Wire routers and the Comcast service
8 and a wired connection to their router would find that the second and
9 subsequent DNS lookups from fresh processes using c-ares to resolve the same
10 address would cause the process to never see a reply (it keeps polling for
11 around 1m15s before giving up).
14 * On such a machine (and yeah, it took us a lot of QA to find the systems
15 that reproduce such a specific problem!), do 'ahost www.secondlife.com',
16 then do it again. The first process's lookup will work, subsequent lookups
17 will time-out and fail.
20 * init_id_key() was calling randomize_key() *before* it initialized
21 key->state, meaning that the randomness generated by randomize_key() is
22 immediately overwritten with deterministic values. (/dev/urandom was also
23 being read incorrectly in the c-ares version we were using, but this was
24 fixed in a later version.)
25 * This makes the stream of generated query-IDs from any new c-ares process
26 be an identical and predictable sequence of IDs.
27 * This makes the 2-Wire's default built-in DNS server detect these queries
28 as probable-duplicates and (erroneously) not respond at all.
31 * Aug 4 2008 (Yang Tse)
32 - Autoconf 2.62 has changed the behaviour of the AC_AIX macro which we use.
33 Prior versions of autoconf defined _ALL_SOURCE if _AIX was defined. 2.62
34 version of AC_AIX defines _ALL_SOURCE and other four preprocessor symbols
35 no matter if the system is AIX or not. To keep the traditional behaviour,
36 and an uniform one across autoconf versions AC_AIX is replaced with our
37 own internal macro CARES_CHECK_AIX_ALL_SOURCE.
39 * Aug 1 2008 (Yang Tse)
40 - Configure process now checks if the preprocessor _REENTRANT symbol is already
41 defined. If it isn't currently defined a set of checks are performed to test
42 if its definition is required to make visible to the compiler a set of *_r
43 functions. Finally, if _REENTRANT is already defined or needed it takes care
44 of making adjustments necessary to ensure that it is defined equally for the
45 configure process tests and generated config file.
47 * Jul 20 2008 (Yang Tse)
48 - When recvfrom prototype uses a void pointer for arguments 2, 5 or 6 this will
49 now cause the definition, as appropriate, of RECVFROM_TYPE_ARG2_IS_VOID,
50 RECVFROM_TYPE_ARG5_IS_VOID or RECVFROM_TYPE_ARG6_IS_VOID.
52 * Jul 17 2008 (Yang Tse)
53 - RECVFROM_TYPE_ARG2, RECVFROM_TYPE_ARG5 and RECVFROM_TYPE_ARG6 are now defined
54 to the data type pointed by its respective argument and not the pointer type.
56 * Jul 16 2008 (Yang Tse)
57 - Improved configure detection of number of arguments for getservbyport_r.
58 Detection is now based on compilation checks instead of linker ones.
60 - Configure process now checks availability of recvfrom() socket function and
61 finds out its return type and the types of its arguments. Added definitions
62 for non-configure systems config files, and introduced macro sreadfrom which
63 will be used on udp sockets as a recvfrom() wrapper in the future.
65 * Jul 15 2008 (Yang Tse)
66 - Introduce definition of _REENTRANT symbol in setup.h to improve library
67 usability. Previously the configure process only used the AC_SYS_LARGEFILE
68 macro for debug builds, now it is also used for non-debug ones enabling the
69 use of configure options --enable-largefile and --disable-largefile which
70 might be needed for library compatibility. Remove checking the size of
71 curl_off_t, it is no longer needed.
73 * Jul 3 2008 (Daniel Stenberg)
74 - Phil Blundell: If you ask ares_gethostbyname() to do an AF_INET6 lookup and
75 the target host has only A records, it automatically falls back to an
76 AF_INET lookup and gives you the A results. However, if the target host has
77 a CNAME record, this behaviour is defeated since the original query does
78 return some data even though ares_parse_aaa_reply() doesn't consider it
79 relevant. Here's a small patch to make it behave the same with and without
82 * Jul 2 2008 (Yang Tse)
83 - Fallback to gettimeofday when monotonic clock is unavailable at run-time.
85 * Jun 30 2008 (Daniel Stenberg)
87 - As was pointed out to me by Andreas Schuldei, the MAXHOSTNAMELEN define is
88 not posix or anything and thus c-ares failed to build on hurd (and possibly
89 elsewhere). The define was also somewhat artificially used in the windows
90 port. Now, I instead rewrote the use of gethostbyname to enlarge the host
91 name buffer in case of need and totally avoid the use of the MAXHOSTNAMELEN
92 define. I thus also removed the defien from the namser.h file where it was
93 once added for the windows build.
95 I also fixed init_by_defaults() function to not leak memory in case if
98 * Jun 9 2008 (Yang Tse)
100 - Make libcares.pc generated file for pkg-config include information relative
101 to the libraries needed for the static linking of c-ares.
103 * May 30 2008 (Yang Tse)
105 - Brad House fixed a missing header file inclusion in adig sample program.
107 Version 1.5.2 (May 29, 2008)
109 * May 13 2008 (Daniel Stenberg)
111 - Introducing millisecond resolution support for the timeout option. See
112 ares_init_options()'s ARES_OPT_TIMEOUTMS.
114 * May 9 2008 (Yang Tse)
116 - Use monotonic time source if available, for private function ares__tvnow()
118 * May 7 2008 (Daniel Stenberg)
120 - Sebastian made c-ares able to return all PTR-records when doing reverse
121 lookups. It is not common practice to have multiple PTR-Records for a single
122 IP, but its perfectly legal and some sites have those.
124 - Doug Goldstein provided a configure patch: updates autoconf 2.13 usage to
125 autoconf 2.57 usage (which is the version you have specified as the minimum
126 version). It's a minor change but it does clean up some warnings with newer
127 autoconf (specifically 2.62).
129 * May 5 2008 (Yang Tse)
131 - Improved parsing of resolver configuration files.
133 * April 4 2008 (Daniel Stenberg)
135 - Eino Tuominen improved the code when a file is used to seed the randomizer.
137 - Alexey Simak made adig support NAPTR records
139 - Alexey Simak fixed the VC dsp file by adding the missing source file
142 * December 11 2007 (Gisle Vanem)
144 - Added another sample application; acountry.c which converts an
145 IPv4-address(es) and/or host-name(s) to country-name and country-code.
146 This uses the service of the DNSBL at countries.nerd.dk.
148 * December 3 2007 (Daniel Stenberg)
150 - Brad Spencer fixed the configure script to assume that there's no
151 /dev/urandom when built cross-compiled as then the script cannot check for
154 - Erik Kline cleaned up ares_gethostbyaddr.c:next_lookup() somewhat
156 Version 1.5.1 (Nov 21, 2007)
158 * November 21 2007 (Daniel Stenberg)
160 - Robin Cornelius pointed out that ares_llist.h was missing in the release
163 Version 1.5.0 (Nov 21, 2007)
165 * October 2 2007 (Daniel Stenberg)
167 - ares_strerror() segfaulted if the input error number was out of the currently
170 - Yang Tse: Avoid a segfault when generating a DNS "Transaction ID" in
171 internal function init_id_key() under low memory conditions.
173 * September 28 2007 (Daniel Stenberg)
175 - Bumped version to 1.5.0 for next release and soname bumped to 2 due to ABI
176 and API changes in the progress callback (and possibly more coming up from
179 * September 28 2007 (Steinar H. Gunderson)
181 - Don't skip a server if it's the only one. (Bugfix from the Google tree.)
183 - Made the query callbacks receive the number of timeouts that happened during
184 the execution of a query, and updated documentation accordingly. (Patch from
187 - Support a few more socket options: ARES_OPT_SOCK_SNDBUF and
190 - Always register for TCP events even if there are no outstanding queries, as
191 the other side could always close the connection, which is a valid event
192 which should be responded to.
194 * September 22 2007 (Daniel Stenberg)
196 - Steinar H. Gunderson fixed: Correctly clear sockets from the fd_set on in
197 several functions (write_tcp_data, read_tcp_data, read_udp_packets) so that
198 if it fails and the socket is closed the following code doesn't try to use
201 - Steinar H. Gunderson modified c-ares to now also do to DNS retries even when
202 TCP is used since there are several edge cases where it still makes sense.
204 - Brad House provided a fix for ares_save_options():
206 Apparently I overlooked something with the ares_save_options() where it
207 would try to do a malloc(0) when no options of that type needed to be saved.
208 On most platforms, this was fine because malloc(0) doesn't actually return
209 NULL, but on AIX it does, so ares_save_options would return ARES_ENOMEM.
211 * July 14 2007 (Daniel Stenberg)
213 - Vlad Dinulescu fixed two outstanding valgrind reports:
215 1. In ares_query.c , in find_query_by_id we compare q->qid (which is a short
216 int variable) with qid, which is declared as an int variable. Moreover,
217 DNS_HEADER_SET_QID is used to set the value of qid, but DNS_HEADER_SET_QID
218 sets only the first two bytes of qid. I think that qid should be declared as
219 "unsigned short" in this function.
221 2. The same problem occurs in ares_process.c, process_answer() . query->qid
222 (an unsigned short integer variable) is compared with id, which is an
223 integer variable. Moreover, id is initialized from DNS_HEADER_QID which sets
224 only the first two bytes of id. I think that the id variable should be
225 declared as "unsigned short" in this function.
227 Even after declaring these variables as "unsigned short", the valgrind
228 errors are still there. Which brings us to the third problem.
230 3. The third problem is that Valgrind assumes that query->qid is not
231 initialised correctly. And it does that because query->qid is set from
232 DNS_HEADER_QID(qbuf); Valgrind says that qbuf has unitialised bytes. And
233 qbuf has uninitialised bytes because of channel->next_id . And next_id is
234 set by ares_init.c:ares__generate_new_id() . I found that putting short r=0
235 in this function (instead of short r) makes all Valgrind warnings go away.
236 I have studied ares__rc4() too, and this is the offending line:
238 buffer_ptr[counter] ^= state[xorIndex]; (ares_query.c:62)
240 This is what triggers Valgrind.. buffer_ptr is unitialised in this function,
241 and by applying ^= on it, it remains unitialised.
243 Version 1.4.0 (June 8, 2007)
245 * June 4 2007 (Daniel Stenberg)
247 - James Bursa reported a major memory problem when resolving multi-IP names
248 and I found and fixed the problem. It was added by Ashish Sharma's patch
251 When I then tried to verify multiple entries in /etc/hosts after my fix, I
252 got another segfault and decided this code was not ripe for inclusion and I
257 - Brad Spencer found and fixed three flaws in the code, found with the new
258 gcc 4.2.0 warning: -Waddress
260 - Brad House fixed VS2005 compiler warnings due to time_t being 64bit.
261 He also made recent Microsoft compilers use _strdup() instead of strdup().
263 - Brad House's man pages for ares_save_options() and ares_destroy_options()
266 - Ashish Sharma provided a patch for supporting multiple entries in the
267 /etc/hosts file. Patch edited for coding style and functionality by me
272 - Shmulik Regev brought cryptographically secure transaction IDs:
274 The c-ares library implementation uses a DNS "Transaction ID" field that is
275 seeded with a pseudo random number (based on gettimeofday) which is
276 incremented (++) between consecutive calls and is therefore rather
277 predictable. In general, predictability of DNS Transaction ID is a well
278 known security problem (e.g.
279 http://bak.spc.org/dms/archive/dns_id_attack.txt) and makes a c-ares based
280 implementation vulnerable to DNS poisoning. Credit goes to Amit Klein
281 (Trusteer) for identifying this problem.
283 The patch I wrote changes the implementation to use a more secure way of
284 generating unique IDs. It starts by obtaining a key with reasonable entropy
285 which is used with an RC4 stream to generate the cryptographically secure
288 Note that the key generation code (in ares_init:randomize_key) has two
289 versions, the Windows specific one uses a cryptographically safe function
290 provided (but undocumented :) by the operating system (described at
291 http://blogs.msdn.com/michael_howard/archive/2005/01/14/353379.aspx). The
292 default implementation is a bit naive and uses the standard 'rand'
293 function. Surely a better way to generate random keys exists for other
296 The patch can be tested by using the adig utility and using the '-s' option.
298 - Brad House added ares_save_options() and ares_destroy_options() that can be
299 used to keep options for later re-usal when ares_init_options() is used.
301 Problem: Calling ares_init() for each lookup can be unnecessarily resource
302 intensive. On windows, it must LoadLibrary() or search the registry
303 on each call to ares_init(). On unix, it must read and parse
304 multiple files to obtain the necessary configuration information. In
305 a single-threaded environment, it would make sense to only
306 ares_init() once, but in a heavily multi-threaded environment, it is
307 undesirable to ares_init() and ares_destroy() for each thread created
310 Solution: Create ares_save_options() and ares_destroy_options() functions to
311 retrieve and free options obtained from an initialized channel. The
312 options populated can be used to pass back into ares_init_options(),
313 it should populate all needed fields and not retrieve any information
314 from the system. Probably wise to destroy the cache every minute or
315 so to prevent the data from becoming stale.
317 - Daniel S added ares_process_fd() to allow applications to ask for processing
318 on specific sockets and thus avoiding select() and associated
319 functions/macros. This function will be used by upcoming libcurl releases
320 for this very reason. It also made me export the ares_socket_t type in the
321 public ares.h header file, since ares_process_fd() uses that type for two of
326 - Ravi Pratap fixed a flaw in the init_by_resolv_conf() function for windows
327 that could cause it to return a bad return code.
331 - Yang Tse: Provide ares_getopt() command-line parser function as a source
332 code helper function, not belonging to the actual c-ares library.
336 - Vlad Dinulescu added ares_parse_ns_reply().
340 - Yang Tse: Fix failure to get the search sequence of /etc/hosts and
341 DNS from /etc/nsswitch.conf, /etc/host.conf or /etc/svc.conf when
342 /etc/resolv.conf did not exist or was unable to read it.
346 - Install ares_dns.h too
348 - Michael Wallner fixed this problem: When I set domains in the options
349 struct, and there are domain/search entries in /etc/resolv.conf, the domains
350 of the options struct will be overridden.
354 - Yang Tse removed a couple of potential zero size memory allocations.
356 - Andreas Rieke fixed the line endings in the areslib.dsp file that I (Daniel)
357 broke in the 1.3.2 release. We should switch to a system where that file is
358 auto-generated. We could rip some code for that from curl...
360 Version 1.3.2 (November 3, 2006)
364 - Prevent ares_getsock() to overflow if more than 16 sockets are used.
368 - Guilherme Balena Versiani: I noted a strange BUG in Win32 port
369 (ares_init.c/get_iphlpapi_dns_info() function): when I disable the network
370 by hand or disconnect the network cable in Windows 2000 or Windows XP, my
371 application gets 127.0.0.1 as the only name server. The problem comes from
372 'GetNetworkParams' function, that returns the empty string "" as the only
373 name server in that case. Moreover, the Windows implementation of
374 inet_addr() returns INADDR_LOOPBACK instead of INADDR_NONE.
380 o made ares_version.h use extern "C" for c++ compilers
381 o fixed compiler warnings in ares_getnameinfo.c
382 o fixed a buffer position init for TCP reads
386 - Ravi Pratap fixed ares_getsock() to actually return the proper bitmap and
389 Version 1.3.1 (June 24, 2006)
393 - Gisle Vanem added getopt() to the ahost program. Currently accepts
394 only [-t {a|aaaa}] to specify address family in ares_gethostbyname().
398 - (wahern) Removed "big endian" DNS section and RR data integer parser
399 macros from ares_dns.h, which break c-ares on my Sparc64. Bit-wise
400 operations in C operate on logical values. And in any event the octets are
401 already in big-endian (aka network) byte order so they're being reversed
402 (thus the source of the breakage).
406 - William Ahern handles EAGAIN/EWOULDBLOCK errors in most of the I/O calls
409 TODO: Handle one last EAGAIN for a UDP socket send(2) in
414 - Bram Matthys brought my attention to a libtool peculiarity where detecting
415 things such as C++ compiler actually is a bad thing and since we don't need
416 that detection I added a work-around, much inspired by a previous patch by
417 Paolo Bonzini. This also shortens the configure script quite a lot.
421 - Nick Mathewson added the ARES_OPT_SOCK_STATE_CB option that when set makes
422 c-ares call a callback on socket state changes. A better way than the
423 ares_getsock() to get full control over the socket state.
427 - Alexander Lazic improved the getservbyport_r() configure check.
431 - Alexander Lazic pointed out that the buildconf should use the ACLOCAL_FLAGS
432 variable for easier controlling what it does and how it runs.
436 - James Bursa fixed c-ares to find the hosts file on RISC OS, and made it
437 build with newer gcc versions that no longer defines "riscos".
441 - Daniel Stenberg added ares_getsock() that extracts the set of sockets to
442 wait for action on. Similar to ares_fds() but not restricted to using
443 select() for the waiting.
447 - Yang Tse fixed some send() / recv() compiler warnings
451 - Added constants that will be used by ares_getaddrinfo
453 - Made ares_getnameinfo use the reentrant getservbyport (getservbyport_r) if it
454 is available to ensure it works properly in a threaded environment.
458 - configure fix for detecting a member in the sockaddr_in6 struct which failed
459 on ipv6-enabled HP-UX 11.00
461 Version 1.3.0 (August 29, 2005)
465 - Alfredo Tupone provided a fix for the Windows code in get_iphlpapi_dns_info()
466 when getting the DNS server etc.
470 - Added some checks for the addrinfo structure.
476 Make UDP sockets non-blocking. I've confirmed that at least on Linux 2.4 a
477 read event can come back from poll() on a valid SOCK_DGRAM socket but
478 recv(2) will still block. This patch doesn't ignore EAGAIN in
479 read_udp_packets(), though maybe it should. (This patch was edited by Daniel
480 Stenberg and a new configure test was added (imported from curl's configure)
481 to properly detect what non-blocking socket approach to use.)
483 I'm not quite sure how this was happening, but I've been seeing PTR queries
484 which seem to return empty responses. At least, they were empty when calling
485 ares_expand_name() on the record. Here's a patch which guarantees to
486 NUL-terminate the expanded name. The old behavior failed to NUL-terminate if
487 len was 0, and this was causing strlen() to run past the end of the buffer
488 after calling ares_expand_name() and getting ARES_SUCCESS as the return
489 value. If q is not greater than *s then it's equal and *s is always
490 allocated with at least one byte.
494 - Added ares_getnameinfo which mimics the getnameinfo API (another feature
495 that could use testing).
499 - Added an inet_ntop function from BIND for systems that do not have it.
503 - Made sortlist support IPv6 (this can probably use some testing).
505 - Made sortlist support CIDR matching for IPv4.
509 - Added preliminary IPv6 support to ares_gethostbyname. Currently, sortlist
510 does not work with IPv6. Also provided an implementation of bitncmp from
511 BIND for systems that do not supply this function. This will be used to add
512 IPv6 support to sortlist.
514 - Made ares_gethostbyaddr support IPv6 by specifying AF_INET6 as the family.
515 The function can lookup IPv6 addresses both from files (/etc/hosts) and
520 - Tupone Alfredo fixed includes of arpa/nameser_compat.h to build fine on Mac
525 - Dominick Meglio: Provided implementations of inet_net_pton and inet_pton
526 from BIND for systems that do not include these functions.
530 - Dominick Meglio added ares_parse_aaaa_reply.c and did various
531 adjustments. The first little steps towards IPv6 support!
535 - Fixed the VC project and makefile to use ares_cancel and ares_version
539 - The released ares_version.h from 1.2.1 says 1.2.0 due to a maketgz flaw.
542 Version 1.2.1 (October 20, 2004)
546 - Henrik Stoerner fix: got a report that Tru64 Unix (the unix from Digital
547 when they made Alpha's) uses /etc/svc.conf for the purpose fixed below for
548 other OSes. He made c-ares check for and understand it if present.
550 - Now c-ares will use local host name lookup _before_ DNS resolving by default
551 if nothing else is told.
555 - Henrik Stoerner: found out that c-ares does not look at the /etc/host.conf
556 file to determine the sequence in which to search /etc/hosts and DNS. So on
557 systems where this order is defined by /etc/host.conf instead of a "lookup"
558 entry in /etc/resolv.conf, c-ares will always default to looking in DNS
559 first, and /etc/hosts second.
563 1) resolv.conf (for the "lookup" line);
564 2) nsswitch.fon (for the "hosts:" line);
565 3) host.conf (for the "order" line).
569 - Dominick Meglio patched: C-ares on Windows assumed that the HOSTS file is
570 located in a static location. It assumed
571 C:\Windows\System32\Drivers\Etc. This is a poor assumption to make. In fact,
572 the location of the HOSTS file can be changed via a registry setting.
574 There is a key called DatabasePath which specifies the path to the HOSTS
576 http://www.microsoft.com/technet/itsolutions/network/deploy/depovg/tcpip2k.mspx
578 The patch will make c-ares correctly consult the registry for the location
583 - Gisle Vanem fixed the MSVC build files.
587 - Gisle Vanem made c-ares build and work with his Watt-32 TCP/IP stack.
591 - Harshal Pradhan made a minor syntax change in ares_init.c to make it build
596 - Made the lib get built static only if --enable-debug is used.
600 Basically in loops like handle_errors(), 'query->next' was assigned a local
601 variable and then query was referenced after the memory was freed by
602 next_server(). I've changed that so next_server() and end_query() returns
603 the next query. So callers should use this ret-value.
605 The next problem was that 'server->tcp_buffer_pos' had a random value at
606 entry to 1st recv() (luckily causing Winsock to return ENOBUFS).
608 I've also added a ares_writev() for Windows to streamline the code a bit
612 - Fixed a few variable return types for some system calls. Made configure
613 check for ssize_t to make it possible to use that when receiving the send()
614 error code. This is necessary to prevent compiler warnings on some systems.
616 - Made configure create config.h, and all source files now include setup.h that
617 might include the proper config.h (or a handicrafted alternative).
619 - Switched to 'ares_socket_t' type for sockets in ares, since Windows don't
622 - automake-ified and libool-ified c-ares. Now it builds libcares as a shared
623 lib on most platforms if wanted. (This bloated the size of the release
624 archive with another 200K!)
626 - Makefile.am now uses Makefile.inc for the c sources, h headers and man
627 pages, to make it easier for other makefiles to use the exact same set of
630 - Adjusted 'maketgz' to use the new automake magic when building distribution
633 - Anyone desires HTML and/or PDF versions of the man pages in the release
637 - Günter Knauf made c-ares build and run on Novell Netware.
640 - Gisle Vanem provided Makefile.dj to build with djgpp, added a few more djgpp
641 fixes and made ares not use 'errno' to provide further info on Windows.
644 - Gisle Vanem made it build with djgpp and run fine with the Watt-32 stack.
647 - Gisle Vanem's init patch for Windows:
649 The init_by_resolv_conf() function fetches the DNS-server(s)
650 from a series of registry branches.
652 This can be wrong in the case where DHCP has assigned nameservers, but the
653 user has overridden these servers with other prefered settings. Then it's
654 wrong to use the DHCPNAMESERVER setting in registry.
656 In the case of no global DHCP-assigned or fixed servers, but DNS server(s)
657 per adapter, one has to query the adapter branches. But how can c-ares know
658 which adapter is valid for use? AFAICS it can't. There could be one adapter
659 that is down (e.g. a VPN adapter).
661 So it's better to leave this to the IP Helper API (iphlapi) available in
662 Win-98/2000 and later. My patch falls-back to the old way if not available.
665 - James Bursa fixed an init issue for RISC OS.
668 - Nico Stappenbelt reported that when processing domain and search lines in
669 the resolv.conf file, the first entry encountered is processed and used as
670 the search list. According to the manual pages for both Linux, Solaris and
671 Tru64, the last entry of either a domain or a search field is used.
673 This is now adjusted in the code
675 Version 1.2.0 (April 13, 2004)
678 - Updated various man pages to look nicer when converted to HTML on the web
682 - Dirk Manske provided a new function that is now named ares_cancel(). It is
683 used to cancel/cleanup a resolve/request made using ares functions on the
684 given ares channel. It does not destroy/kill the ares channel itself.
686 - Dominick Meglio cleaned up the formatting in several man pages.
689 - Dominick Meglio's new ares_expand_string. A helper function when decoding
690 incoming DNS packages.
692 - Daniel Stenberg modified the Makefile.in to use a for loop for the man page
693 installation to improve overview and make it easier to add man pages.
695 Version 1.1.0 (March 11, 2004)
698 - Gisle Vanem improved build on Windows.
701 - Dan Fandrich found a flaw in the Feb 22 fix.
703 - Added better configure --enable-debug logic (taken from the curl configure
704 script). Added acinclude.m4 to the tarball.
707 - Removed ares_free_errmem(), the function, the file and the man page. It was
708 not used and it did nothing.
710 - Fixed a lot of code that wasn't "64bit clean" and thus caused a lot of
711 compiler warnings on picky compilers.
714 - Dominick Meglio made ares init support multiple name servers in the
715 NameServer key on Windows.
718 - Modified ares_private.h to include libcurl's memory debug header if
719 CURLDEBUG is set. This makes all the ares-functions supervised properly by
720 the curl test suite. This also forced me to add inclusion of the
721 ares_private.h header in a few more files that are using some kind of
722 memory-related resources.
724 - Made the makefile only build ahost and adig if 'make demos' is used.
727 - Dirk Manske made ares_version.h installed with 'make install'
730 - ares_free_errmem() is subject for removal, it is simply present for future
731 purposes, and since we removed the extra parameter in strerror() it won't
733 - configure --enable-debug now enables picky compiler options if gcc is used
734 - fixed several compiler warnings --enable-debug showed and Joerg Mueller-Tolk
737 Version 1.0.0 (February 3, 2004)
740 - now we produce the libcares.a library instead of the previous libares.a
741 since we are no longer compatible
745 - ares_strerror() has one argument less. This is the first official
746 modification of the existing provided ares API.
750 - Dirk Manske fixed how the socket is set non-blocking.
754 - Dominick Meglio made the private gettimeofday() become ares_gettimeofday()
755 instead in order to not pollute the name space and risk colliding with
756 other libraries' versions of this function.
758 * October 24, 2003. Daniel Stenberg
760 Added ares_version().
762 Version 1.0-pre1 (8 October 2003)
764 - James Bursa made it run on RISC OS
766 - Dominick Meglio made it run fine on NT4
768 - Duncan Wilcox made it work fine on Mac OS X
770 - Daniel Stenberg adjusted the windows port
772 - liren at vivisimo.com made the initial windows port
774 * Imported the sources from ares 1.1.1