_ _ ____ _ ___| | | | _ \| | / __| | | | |_) | | | (__| |_| | _ <| |___ \___|\___/|_| \_\_____| Changelog Version 8.0.1 (20 Mar 2023) Daniel Stenberg (20 Mar 2023) - RELEASE-NOTES: synced curl 8.0.1 - Revert "multi: remove PENDING + MSGSENT handles from the main linked list" This reverts commit f6d6f3ce01e377932f1ce7c24ee34d45a36950b8. The commits caused issues in the 8.0.0 release. Needs a retake. Reported-by: Kamil Dudka Closes #10795 - include/curl/curlver.h: bump to 8.0.1 Version 8.0.0 (20 Mar 2023) Daniel Stenberg (20 Mar 2023) - RELEASE-NOTES: synced The curl 8.0.0 release - THANKS: from the 8.0.0 release - scripts/delta: fix "ambiguous argument" when used in branches - SECURITY-PROCESS.md: Busy-loops are not security problems Closes #10790 Stefan Eissing (17 Mar 2023) - tests/http: do not save files for downloads in scorecard testing Closes #10788 Daniel Stenberg (17 Mar 2023) - cf-socket: use port 80 when resolving name for local bind It turns out c-ares returns an error when asked to resolve a host name with ares_getaddrinfo using port number 0. Reported as a c-ares bug here: https://github.com/c-ares/c-ares/issues/517 The work-around is to simply use port 80 instead, as the number typically doe s not make a difference and a non-zero number works for c-ares. Fixes #10759 Reported-by: Matt Jolly Closes #10789 - curl.h: require gcc 12.1 for the deprecation magic Reported-by: kchow-FTNT on github Fixes #10726 Closes #10784 - Revert "rtsp: use dynbuf instead of custom reallocs" This reverts commit 1b9ea3239d22147e00d8 because of OSS-fuzz reports. I'll do another take after the pending release. Closes #10785 - test422: verify --next used without a prior URL Closes #10782 - tool_getparam: error if --next is used without a prior URL Reported-by: 積丹尼 Dan Jacobson Ref: https://github.com/curl/curl/pull/10769#discussion_r1137895629 Closes #10782 - libssh: use dynbuf instead of realloc When building lines to show for SFTP directory listings. Closes #10778 - lib2305: deal with CURLE_AGAIN The test does a slightly ugly busy-loop for this case but should be managable due to it likely being a very short moment. Mention CURLE_AGAIN in curl_ws_recv.3 Fixes #10760 Reported-by: Jay Satiro Closes #10781 - rtsp: use dynbuf instead of custom reallocs For the RTP buffering. Closes #10776 - libssh2: remove unused variable from libssh2's struct Closes #10777 - RELEASE-NOTES: synced - multi: remove PENDING + MSGSENT handles from the main linked list As they are not driving transfers or any socket activity, the main loop does not need to iterate over these handles. A performance improvement. They are instead only held in their own separate lists. Assisted-by: Stefan Eissing Ref: #10743 Closes #10762 - multi: turn link/unlinking easy handles into dedicated functions - http_aws_sigv4: fix scan-build "value stored to 'ret' is never read" Follow-up to 495d09810aa9a Closes #10766 - lib: skip Curl_llist_destroy calls Linked lists themselves do not carry any allocations, so for the lists that do not have have a set destructor we can just skip the Curl_llist_destroy() call and save CPU time. Closes #10764 - lib643: LIB644 is never defined, this is dead code Closes #10765 - libtest/Makefile.inc: remove superfluous variables Rely on the defaults when possible. Closes #10765 - tests/http: remove year ranges from copyrights Closes #10763 Casey Bodley (14 Mar 2023) - aws_sigv4: fall back to UNSIGNED-PAYLOAD for sign_as_s3 all s3 requests default to UNSIGNED-PAYLOAD and add the required x-amz-content-sha256 header. this allows CURLAUTH_AWS_SIGV4 to correctly sign s3 requests to amazon with no additional configuration Signed-off-by: Casey Bodley Closes #9995 Viktor Szakats (14 Mar 2023) - wolfssl: add quic/ngtcp2 detection in cmake, and fix builds - add QUIC/ngtcp2 detection in CMake with wolfSSL. Because wolfSSL uses zlib if available, move compression detection before TLS detection. (OpenSSL might also need this in the future.) - wolfSSL 5.5.0 started using C99 types in its `quic.h` header, but it doesn't #include the necessary C99 header itself, breaking builds (unless another dependency pulled it by chance.) Add local workaround for it. For this to work with all build tools, we had to fix our header detection first. Ref: #10745 Ref: https://github.com/curl/curl-for-win/commit/6ad5f6ecc15620c15625fc4434 76b3a1ecef4f3f Closes #10739 Stefan Eissing (14 Mar 2023) - secure-transport: fix recv return code handling Return code handling of recv calls were not always correct when an error occured or the connection was closed. Closes #10717 - http2: Use KEEP_SEND_HOLD for flow control in HTTP/2 - use the defined, but so far not used, KEEP_SEND_HOLD bit for flow control based suspend of sending in transfers. Prior to this change KEEP_SEND_PAUSE bit was used instead, but that can interfere with pausing streams from the user side via curl_easy_pause. Fixes https://github.com/curl/curl/issues/10751 Closes https://github.com/curl/curl/pull/10753 Dan Fandrich (13 Mar 2023) - tests: fix control code that hid some text in runtests.1 - tests: sync option lists in runtests.pl & its man page Daniel Stenberg (13 Mar 2023) - multi: make multi_perform ignore/unignore signals less often For improved performance Reported-by: Jerome St-Louis Ref: #10743 Closes #10750 Viktor Szakats (13 Mar 2023) - cmake: delete unused HAVE__STRTOI64 Also delete obsolete surrounding comments. Reviewed-by: Daniel Stenberg Closes #10756 - CI: fix copyright header Follow-up to 395b9175b7422d699fa93643973295c106cdf147 Daniel Stenberg (13 Mar 2023) - RELEASE-PROCEDURE.md: update coming release dates Stefan Eissing (13 Mar 2023) - tests/http: add pytest to GHA and improve tests - added to: ngtcp2-quictls, ngtcp2-gnutls and the linux varians quiche, bearssl, libressl, mbedtls, openssl3, rustls - added disabled in ngtcp2-wolfssl due to weird SSL_connect() errors not reproducable locally Improvements on pytest: - handling of systems with nghttpx in $PATH - configure will seach $PATH got nghttpx used in pytest - pytest fixes for managing nghttpx without h3 support - ngtcp2-wolfssl: use a fully enabled wolfssl build - lower parallel count for http/1.1 tests, since we do not want to test excessive connections. - check built curl for HTTPS-proxy support in proxy tests - bearssl does not like one of our critical cert extensions, making it non-critical now - bearssl is too slow for test_12, skipping - making sure we do h3 tests only when curl and server support is there Closes #10699 Marcel Raad (13 Mar 2023) - tool_operate: silence unused parameter warning `global` is only used in the `my_setopt` macro version without `CURL_DISABLE_LIBCURL_OPTION` since commit 4774decf10a. Closes https://github.com/curl/curl/pull/10752 Viktor Szakats (13 Mar 2023) - build: fix stdint/inttypes detection with non-autotools Fix `stdint.h` and `inttypes.h` detection with non-autotools builds on Windows. (autotools already auto-detected them accurately.) `lib/config-win32.h` builds (e.g. `Makefile.mk`): - set `HAVE_STDINT_H` where supported. - set `HAVE_INTTYPES_H` for MinGW. CMake: - auto-detect them on Windows. (They were both force-disabled.) - delete unused `CURL_PULL_STDINT_H`. - delete unused `CURL_PULL_INTTYPES_H`. - stop detecting `HAVE_STDINT_H` twice. Present since the initial CMake commit: 4c5307b45655ba75ab066564afdc0c111a8 b9291 curl doesn't use these C99 headers, we need them now to workaround broken wolfSSL builds. Ref: #10739 Once that clears up, we can delete these detections and macros (unless we want to keep them for future us.) Reviewed-by: Daniel Stenberg Closes #10745 Daniel Stenberg (13 Mar 2023) - RELEASE-NOTES: synced - ftp: add more conditions for connection reuse Reported-by: Harry Sintonen Closes #10730 Dan Fandrich (12 Mar 2023) - tests: make first.c the same for both lib tests and unit tests The only difference used to be global variable used in unittest tests. After cb7ed5a removed individual flag overrides for the unittests, first.c was no longer recompiled for unit tests to include the flag, so whether it worked or gave a link error depended on whether it was compiled in libtest or unittest first. This way also speeds up the build by eliminating 40 identical compile invocations. Fixes #10749 - tests: use AM_CPPFILES to modify flags in unit tests Using CPPFLAGS sometimes caused odd compile issues when building tests with parallel make and AM_CPPFILES is the right flag, anyway. Follow-up to cb7ed5a Ref #10749 Viktor Szakats (13 Mar 2023) - Makefile.mk: fix -g option in debug mode [ci skip] Add it to `CFLAGS` (was: `LDFLAGS`). Closes #10747 Jay Satiro (12 Mar 2023) - tool: improve --stderr handling - freopen stderr with the user-specified file (--stderr file) instead of using a separate 'errors' stream. - In tool_setup.h override stdio.h's stderr macro as global variable tool_stderr. Both freopen and overriding the stderr macro are necessary because if the user-specified filename is "-" then stdout is assigned to tool_stderr and no freopen takes place. See the PR for more information. Ref: https://github.com/curl/curl/issues/10491 Closes https://github.com/curl/curl/pull/10673 Dan Fandrich (11 Mar 2023) - CI: don't run CI jobs if only another CI was changed Also skip builds on non-Windows platforms when only Windows build files have changed. This should reduce the number of useless builds and the associated waiting time and chance of spurious failures, freeing resources for new PRs. Closes #10742 - http: don't send 100-continue for short PUT requests This is already how curl is documented to behave in Everything curl, but in actuality only short POSTs skip this. This should knock 30 seconds off a full run of the test suite since the 100-continue timeout will no longer be hit. Closes #10740 - tests: add DELAY keyword to more tests using waits - tests: hack to build most unit tests under cmake These are only built when a libcurl static library is available, since we're not building a special libcurlu library yet and these tests rely on private symbols that aren't available in the shared library. A few unit tests do require libcurlu, so those are not built. Closes #10722 - tests: fix MSVC unreachable code warnings in unit tests Switch unit1654 to use the proper test macros as well. - tests: make CPPFLAGS common to all unit tests There's no need to specify them individually. - tests: keep cmake unit tests names in sync Put only the test names into Makefile.inc so they can be used by both cmake and automake. This will prevent the list of tests from becoming out of date when they are also built under cmake. Viktor Szakats (11 Mar 2023) - src: silence wmain() warning for all build methods llvm/clang and gcc doesn't recognize the wmain() function in Unicode Windows builds: llvm/clang: ``` ../../src/tool_main.c:239:5: warning: no previous prototype for function 'wma in' [-Wmissing-prototypes] int wmain(int argc, wchar_t *argv[]) ^ 1 warning generated. ``` gcc: ``` ../../src/tool_main.c:239:5: warning: no previous prototype for 'wmain' [-Wmi ssing-prototypes] 239 | int wmain(int argc, wchar_t *argv[]) | ^~~~~ ``` Before this patch, we already silenced it with CMake. This patch moves the silencing to the source, so that it applies to all build tools. Bug: https://github.com/curl/curl/issues/7229#issuecomment-1464806651 Reviewed-by: Marcel Raad Closes #10744 Dan Fandrich (10 Mar 2023) - CI: fix retrying on brew failures The previous attempt didn't consider that the shell would exit immediately after the false statement in the retry case. Follow-up to dc141a37 Stefan Eissing (10 Mar 2023) - http2: fix error handling during parallel operations RST and connection close were not handled correctly during parallel transfers, leading to aborted response bodies being reported complete. Closes #10715 Daniel Stenberg (10 Mar 2023) - url: only reuse connections with same GSS delegation Reported-by: Harry Sintonen Closes #10731 Viktor Szakats (10 Mar 2023) - lib: silence clang/gcc -Wvla warnings in brotli headers brotli v1.0.0 throughout current latest v1.0.9 and latest master [1] trigger this warning. It happened with CMake and GNU Make. autotools builds avoid it with the `convert -I options to -isystem` macro. llvm/clang: ``` In file included from ./curl/lib/content_encoding.c:36: ./brotli/x64-ucrt/usr/include/brotli/decode.h:204:34: warning: variable lengt h array used [-Wvla] const uint8_t encoded_buffer[BROTLI_ARRAY_PARAM(encoded_size)], ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ./brotli/x64-ucrt/usr/include/brotli/port.h:253:34: note: expanded from macro 'BROTLI_ARRAY_PARAM' ^~~~~~ In file included from ./curl/lib/content_encoding.c:36: ./brotli/x64-ucrt/usr/include/brotli/decode.h:206:48: warning: variable lengt h array used [-Wvla] uint8_t decoded_buffer[BROTLI_ARRAY_PARAM(*decoded_size)]); ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~ ./brotli/x64-ucrt/usr/include/brotli/port.h:253:35: note: expanded from macro 'BROTLI_ARRAY_PARAM' ~^~~~~ ``` gcc: ``` In file included from ./curl/lib/content_encoding.c:36: ./brotli/x64-ucrt/usr/include/brotli/decode.h:204:5: warning: ISO C90 forbids variable length array 'encoded_buffer' [-Wvla] 204 | const uint8_t encoded_buffer[BROTLI_ARRAY_PARAM(encoded_size)], | ^~~~~ ./brotli/x64-ucrt/usr/include/brotli/decode.h:206:5: warning: ISO C90 forbids variable length array 'decoded_buffer' [-Wvla] 206 | uint8_t decoded_buffer[BROTLI_ARRAY_PARAM(*decoded_size)]); | ^~~~~~~ ``` [1] https://github.com/google/brotli/commit/ed1995b6bda19244070ab5d331111f16f 67c8054 Reviewed-by: Daniel Stenberg Reviewed-by: Marcel Raad Closes #10738 Daniel Stenberg (10 Mar 2023) - curl_path: create the new path with dynbuf Closes #10729 - url: remove dummy protocol handler Just two added checks were needed saves a whole handler struct. Closes #10727 Dan Fandrich (10 Mar 2023) - CI: retry a failed brew update too, not just brew install Also, make sure an eventual failure ends up returning a failure code so the job stops. Daniel Stenberg (10 Mar 2023) - url: fix the SSH connection reuse check Reported-by: Harry Sintonen Closes #10735 - CURLOPT_PROXY.3: curl+NSS does not handle HTTPS over unix domain socket It results in error "NSS error -5985 (PR_ADDRESS_NOT_SUPPORTED_ERROR)" Disabled test 1470 for NSS builds and documented the restriction. Reported-by: Dan Fandrich Fixes #10723 Closes #10734 - CURLSHOPT_SHARE.3: HSTS sharing is not thread-safe Reported-by: Hiroki Kurosawa Closes #10732 - telnet: only accept option arguments in ascii To avoid embedded telnet negotiation commands etc. Reported-by: Harry Sintonen Closes #10728 - test1903: test use of COOKIEFILE - reset - COOKIEFILE This also tests for the memory leak bug fixed by parent commit b559ef6f. Ref: #10694 Closes https://github.com/curl/curl/pull/10712 Jay Satiro (10 Mar 2023) - url: fix cookielist memleak when curl_easy_reset - Free set.cookelist in Curl_freeset instead of Curl_close. Prior to this change the cookielist linked list wasn't freed by curl_easy_reset which calls Curl_freeset to free all set. Bug: https://github.com/curl/curl/issues/10694#issuecomment-1458619157 Reported-by: Sergey Ryabinin Closes https://github.com/curl/curl/pull/10709 Dan Fandrich (10 Mar 2023) - tests: fix some keywords and unused sections - tests: fix test1301 to call the right binary It was refactored in commit 480ac6e5 but this step was missed. - tests: add timeout, SLOWDOWN and DELAY keywords to tests These are tests that are testing timing and end up being quite slow. Daniel Stenberg (10 Mar 2023) - RELEASE-NOTES: synced Stefan Eissing (10 Mar 2023) - wolfSSL: ressurect the BIO `io_result` In pytest'ing the situation occored that wolfSSL reported an IO error when the underlying BIO operation was returning an CURLE_AGAIN condition. Readding the `io_result` filter context member to detect such situations. Also, making sure that the returned CURLcode is initialized on all recv operations outcome. Closes #10716 - gssapi: align global `gss_OID_desc` vars to silence ld warnings on macOS vent ura Refs #9975 which first reported this. Closes #10718 Daniel Stenberg (10 Mar 2023) - libssh2: only set the memory callbacks when debugging This makes us debug libssh2 less and libcurl more when for example running torture tests that otherwise will spend a lot of time in libssh2 functions. We leave libssh2 to test libssh2. Closes #10721 - docs/SECURITY-PROCESS.md: updates - allow Low+Medium issues to be managed through plain PRs - update the bug-bounty part to reflect current reality Closes #10719 Dan Fandrich (9 Mar 2023) - tests: fix tag markup issues in some tests Marcel Raad (9 Mar 2023) - tests: add `cookies` features These tests don't work with `--disable-cookies`. Closes https://github.com/curl/curl/pull/10713 - test420: add cookies keyword It fails with `--disable-cookies`. Closes https://github.com/curl/curl/pull/10713 Dan Fandrich (8 Mar 2023) - CI: Add more labeler match patterns Also, add the CI, tests or libcurl API tags in conjunction with any others that might also apply. Andy Alt (9 Mar 2023) - GHA: minor improvements to spellcheck Closes #10640 Daniel Stenberg (9 Mar 2023) - test1671: fix after fix - test421: -w %{header_json} test with multiple same header names To reproduce the issue in #10704 - tool_writeout_json. fix the output for duplicate header names Header entries with index != 0 are handled at the index 0 level so they should then be skipped when iterated over. Reported-by: Boris Okunskiy Fixes #10704 Closes #10707 - headers: make curl_easy_header and nextheader return different buffers By letting curl_easy_header() and curl_easy_nextheader() store the header data in their own struct storage when they return a pointer to it, it makes it possible for applications to use them both in a loop. Like the curl tool does. Reported-by: Boris Okunskiy Fixes #10704 Closes #10707 rcombs (8 Mar 2023) - urlapi: take const args in _dup and _get functions Closes #10708 - urlapi: avoid mutating internals in getter routine This was not intended. Closes #10708 Daniel Stenberg (8 Mar 2023) - urlapi: '%' is illegal in host names Update test 1560 to verify Ref: #10708 Closes #10711 - ftp: make the 'ftpauth' a more normal 'char *'-array Closes #10703 Evgeny Grin (Karlson2k) (8 Mar 2023) - doc: fix compiler warning in libcurl.m4 Current test for curl_free() may produce warnings with strict compiler flags or even with default compiler flags with upcoming versions. These warning could turned into errors by -Werror or similar flags. Such warnings/errors are avoided by this patch. Closes #10710 Viktor Szakats (8 Mar 2023) - misc: fix typos Closes #10706 Stefan Eissing (7 Mar 2023) - ftp: active mode with SSL, add the damn filter - since 7.87.0 we lost adding the SSL filter for an active FTP connection that uses SSL. This leads to hangers and timeouts as reported in #10666. Reported-by: SandakovMM on github Fixes #10666 Closes #10669 Daniel Stenberg (7 Mar 2023) - docs: extend the URL API descriptions Closes #10701 Stefan Eissing (7 Mar 2023) - url: fix logic in connection reuse to deny reuse on "unclean" connections - add parameter to `conn_is_alive()` cfilter method that returns if there is input data waiting on the connection - refrain from re-using connnection from the cache that have input pending - adapt http/2 and http/3 alive checks to digest pending input to check the connection state - remove check_cxn method from openssl as that was just doing what the socket filter now does. - add tests for connection reuse with special server configs Closes #10690 Daniel Stenberg (6 Mar 2023) - x509asn1: use plain %x, not %lx, when the arg is an int Pointed out by Coverity. Closes #10689 Stefan Eissing (6 Mar 2023) - http2: fix handling of RST and GOAWAY to recognize partial transfers - a reset transfer (HTTP/2 RST) did not always lead to the proper error message on receiving its response, leading to wrong reports of a successful transfer - test_05_02 was able to trigger this condition with increased transfer count. The simulated response errors did not carry a 'Content-Length' so only proper RST handling could detect the abort - When doing such transfers in parallel, a connection could enter the state where a) it had been closed (GOAWAY received) b) the RST had not been "seen" for the transfer yet or c) the GOAWAY announced an error and the last successful stream id was not checked against ongoing transfers Closes #10693 - tests: use dynamic ports numbers in pytest suite - necessary ports are bound at start of test suite and then given to server fixtures for use. - this make parallel use of pytest (in separate directories), practically safe for use as OS tend to not reuse such port numbers for a while Closes #10692 - connect: fix time_connect and time_appconnect timer statistics - time_connect was not updated when the overall connection failed, e.g. when SSL verification was unsuccessful, refs #10670 - rework gather those values to interrogate involved filters, also from all eyeballing attempts, to report the maximum of those values. - added 3 test cases in test_06 to check reported values on successful, partially failed and totally failed connections. Reported-by: Master Inspire Fixes #10670 Closes #10671 Daniel Stenberg (6 Mar 2023) - test1905: update output cookie order After the #10685 update - test420: verify expiring cookies Cookies that are loaded fine from a jar but then are expired in headers. - cookie: don't load cookies again when flushing Reported-by: Sergio Mijatovic Fixes #10677 Closes #10685 - RELEASE-NOTES: synced Andy Alt (6 Mar 2023) - docs: note '--data-urlencode' option Closes #10687 Daniel Stenberg (6 Mar 2023) - DEPRECATE: the original legacy mingw version 1 Remove completely in September 2023 Closes #10667 Harry Sintonen (6 Mar 2023) - rand: use arc4random as fallback when available Normally curl uses cryptographically strong random provided by the selected SSL backend. If compiled without SSL support, a naive built-in function was used instead. Generally this was okay, but it will result in some downsides for non- SSL builds, such as predictable temporary file names. This change ensures that arc4random will be used instead, if available. Closes #10672 Grisha Levit (6 Mar 2023) - tool: dump headers even if file is write-only The fixes in #10079 brought a (seemingly unrelated) change of open mode from `wb`/`ab` to `wb+`/`ab+` for the headerfile. This makes it no longer possible to write the header file to e.g. a pipe, like: curl -D >(grep ...) file:///dev/null Which presently results in `Warning: Failed to open /dev/fd/63` See #10079 Closes #10675 Jay Satiro (6 Mar 2023) - tests: fix gnutls-serv check - If gnutls-serv doesn't exist then don't try to execute it. Follow-up to 2fdc1d81. Closes https://github.com/curl/curl/pull/10688 Daniel Stenberg (6 Mar 2023) - lib1560: fix enumerated type mixed with another type Follow-up to c84c0f9aa3bb006 Closes #10684 Viktor Szakats (5 Mar 2023) - cmake: fix enabling LDAPS on Windows Before this patch, enabling LDAPS required a manual C flag: https://github.com/curl/curl-for-win/blob/c1cfc31cfc04f24f7a4f946564d6f0e1b4d 7dd36/curl-cmake.sh#L105 Fix this and enable LDAPS automatically when using `wldap32` (and when not explicitly disabled). This matches autotools and `Makefile.mk` behavior. Also remove issue from KNOWN_BUGS. Add workaround for MSVS 2010 warning triggered by LDAPS now enabled in more CI tests: `ldap.c(360): warning C4306: 'type cast' : conversion from 'int' to 'void *' of greater size` Ref: https://ci.appveyor.com/project/curlorg/curl/builds/46408284/job/v8mwl9y fbmoeqwlr#L312 Reported-by: JackBoosY on github Reviewed-by: Jay Satiro Reviewed-by: Marcel Raad Fixes #6284 Closes #10674 - Makefile.mk: delete redundant `HAVE_LDAP_SSL` macro [ci skip] Since abebb2b8939c6b3e0f951eb2d9ec3729b569aa2c, we set this macro for all Windows `wldap32` builds using `Makefile.mk`. For OpenLDAP builds this macro is not enough to enable LDAPS, and OpenLDAP is not an option in `Makefile.mk`. For Novell LDAP it might have helped, but it's also not an option anymore in `Makefile.mk`. The future for LDAPS is that we should enable it by default without extra build knobs. Reviewed-by: Marcel Raad Closes #10681 - cmake: skip CA-path/bundle auto-detection in cross-builds Also remove issue from KNOWN_BUGS. Reported-by: Cristian Morales Vega Reviewed-by: Marcel Raad Fixes #6178 Closes #10676 Daniel Stenberg (3 Mar 2023) - schannel: loop over the algos to pick the selected one Avoid using the funny macro and the extra buffer copy. Closes #10647 - wildcard: remove files and move functions into ftplistparser.c - ftp: allocate the wildcard struct on demand The feature is rarely used so this frees up data for the vast majority of easy handles that don't use it. Rename "protdata" to "ftpwc" since it is always an FTP wildcard struct pointer. Made the state struct field an unsigned char to save space. Closes #10639 - lib1560: test parsing URLs with ridiculously large fields In the order of 120K. Closes #10665 Brad Spencer (3 Mar 2023) - urlapi: parse IPv6 literals without ENABLE_IPV6 This makes the URL parser API stable and working the same way independently of libcurl supporting IPv6 transfers or not. Closes #10660 Jan Engelhardt (3 Mar 2023) - build: drop the use of XC_AMEND_DISTCLEAN Because automake used to delete depdirs at once (.deps) and there was an issu e with portability, curl's XC_AMEND_DISTCLEAN greps the Makefiles in an attempt to build a list of all depfiles and delete them individually instead. Since commit 08849db866b44510f6b8fd49e313c91a43a3dfd3, automake switched from deleting directories to individual files. curl's custom logic now finds a lot more results with the grep (the filtering of these results isn't great), whic h causes a massive bloating of the Makefile in the order of O(n^2). Also remove now-unused XC_AMEND_DISTCLEAN macro group References: https://github.com/curl/curl/issues/9843 References: https://debbugs.gnu.org/cgi/bugreport.cgi?bug=59288 Reported-by: Ilmari Lauhakangas Fixes #9843 Closes #10661 Balakrishnan Balasubramanian (3 Mar 2023) - test1470: test socks proxy using unix sockets and connect to https Similar to test1468 except using https instead of http Closes #10662 Daniel Stenberg (3 Mar 2023) - test1960: verify CURL_SOCKOPT_ALREADY_CONNECTED When returned from the CURLOPT_SOCKOPTFUNCTION, like when we have a custom socket connected in the app, passed in to libcurl. Verifies the fix in #10648 Closes #10651 Stefan Eissing (2 Mar 2023) - tests: rename tests/tests-httpd to tests/http - httpd is only one server we test with - the suite coveres the HTTP protocol in general where the default test cases need a more beefy environment Closes #10654 - socket: detect "dead" connections better, e.g. not fit for reuse - refs #10646 where reuse was attempted on closed connections in the cache, leading to an exhaustion of retries on a transfer - the mistake was that poll events like POLLHUP, POLLERR, etc were regarded as "not dead". - change cf-socket filter check to regard such events as inidication of corpsiness. - vtls filter checks: fixed interpretation of backend check result when inconclusive to interrogate status further down the filter chain. Reported-by: SendSonS on github Fixes #10646 Closes #10652 - lib: give source files cf-http.* better fitting names Closes #10656 - http2: fix code indent Closes https://github.com/curl/curl/pull/10655 Shankar Jadhavar (1 Mar 2023) - cf-socket: if socket is already connected, return CURLE_OK In 7.87.0, if callback method for CURLOPT_SOCKOPTFUNCTION returns CURL_SOCKOPT_ALREADY_CONNECTED then curl library used to return CURLE_OK. n 7.88.0, now even if callback returns CURL_SOCKOPT_ALREADY_CONNECTED, curl library still tries to connect to socket by invoking method do_connect(). This is regression caused by commit https://github.com/curl/curl/commit/71b7e0161032927cdfb Fix: Check if we are already connected and return CURLE_OK. Fixes #10626 Closes #10648 Jay Satiro (1 Mar 2023) - DYNBUF.md: note Curl_dyn_add* calls Curl_dyn_free on failure This is the existing behavior and it has been widely assumed in the codebase. Closes https://github.com/curl/curl/pull/10645 Stefan Eissing (1 Mar 2023) - http2: fix upload busy loop - Set KEEP_SEND_PAUSE when exhausting remote HTTP/2 window size of a stream. - Clear KEEP_SEND_PAUSE when receiving HTTP/2 window updates on a paused stream. - Also fix http2 send compiler warnings reported in #10449. Prior to this change, starting in 71b7e016 which precedes 7.88.0, libcurl may eat CPU during HTTP/2 upload. Reported-by: Jay Satiro Fixes https://github.com/curl/curl/issues/10449 Fixes https://github.com/curl/curl/issues/10618 Closes https://github.com/curl/curl/pull/10627 Daniel Stenberg (1 Mar 2023) - sectransp: make read_cert() use a dynbuf when loading Closes #10632 Jay Satiro (1 Mar 2023) - transfer: limit Windows SO_SNDBUF updates to once a second - Change readwrite_upload() to call win_update_buffer_size() no more than once a second to update SO_SNDBUF (send buffer limit). Prior to this change during an upload readwrite_upload() could call win_update_buffer_size() anywhere from hundreds of times per second to an extreme test case of 100k per second (which is likely due to a bug, see #10618). In the latter case WPA profiler showed win_update_buffer_size was the highest capture count in readwrite_upload. In any case the calls were excessive and unnecessary. Ref: https://github.com/curl/curl/pull/2762 Closes https://github.com/curl/curl/pull/10611 Daniel Stenberg (28 Feb 2023) - RELEASE-NOTES: synced Stefan Eissing (28 Feb 2023) - http2: fix for http2-prior-knowledge when reusing connections - refs #10634 where errors in the HTTP/2 framing layer are observed. - the bug was that on connection reuse, the code attempted to switch in yet another layer of HTTP/2 handling instead of detecting that this was already in place. - added pytest testcase reproducing the issue. Reported-by: rwmjones on github Fixes #10634 Closes #10643 - cf-socket: fix handling of remote addr for accepted tcp sockets - do not try to determine the remote address of a listen socket. There is none. - Update remote address of an accepted socket by getpeername() if available. Reported-by: Harry Sintonen Fixes #10622 Closes #10642 - http: fix unix domain socket use in https connects - when h2/h3 eyeballing was involved, unix domain socket configurations were not honoured - configuring --unix-socket will disable HTTP/3 as candidate for eyeballing - combinatino of --unix-socket and --http3-only will fail during initialisati on - adding pytest test_11 to reproduce Reported-by: Jelle van der Waa Fixes #10633 Closes #10641 Daniel Stenberg (28 Feb 2023) - setopt: move the CURLOPT_CHUNK_DATA pointer to the set struct To make duphandle work etc Closes #10635 Viktor Szakats (28 Feb 2023) - quic/schannel: fix compiler warnings Fixes #10603 Closes #10616 Daniel Stenberg (28 Feb 2023) - page-footer: add explanation for three missing exit codes Added in 7.73.0, 7.77.0 and 7.84.0 Closes #10630 積丹尼 Dan Jacobson (28 Feb 2023) - rate.c: single URLs make no sense in --rate example Here somehow you need to put more than one URL in these examples, else they will make no sense, as --rate only affects the second and beyond URLs. The first URL will always finish the same time no matter what --rate is given. Closes #10638 Daniel Stenberg (28 Feb 2023) - libcurl-errors.3: add the CURLHcode errors from curl_easy_header.3 Closes #10629 - mqtt: on send error, return error Reported-by: Maciej Domanski Closes #10623 - ws: keep the socket non-blocking Reported-by: marski on github Fixes #10615 Closes #10625 - hostip: avoid sscanf and extra buffer copies Also made create_hostcache_id() return the id length. Closes #10601 - PARALLEL-TRANSFERS.md: not "early days" for this anymore Refresh the language as the support is now over three years old Closes #10624 - easy: remove infof() debug leftover from curl_easy_recv It said "reached [path]/easy.c:1231" Closes #10628 - idn: return error if the conversion ends up with a blank host Some IDN sequences are converted into "" (nothing), which can make this function end up with a zero length host name and we cannot consider that a valid host to continue with. Reported-by: Maciej Domanski Closes #10617 - examples/http3.c: use CURL_HTTP_VERSION_3 and update the comment Closes #10619 - x509asn1.c: use correct format specifier for infof() call Detected by Coverity Closes #10614 - Revert "GHA: add Microsoft C++ Code Analysis" This reverts commit e0db842b2a082dffad4a9fbe31321e9a75c74041. This tool seems very restricted in how often it might be used by a project and thus very quickly start to report fails simply because it refuses to run when "there are more runs than allowed". Closes #10613 Patrick Monnerat (25 Feb 2023) - tests: test secure mail protocols with explicit SSL requests New tests 987, 988 and 989, disabled for rustls (hanging). Closes #10077 - tests: support for imaps/pop3s/smtps protocols Closes #10077 - runtests: use a hash table for server port numbers Closes #10077 Andy Alt (25 Feb 2023) - INTERNALS.md: grammar Closes #10607 Daniel Stenberg (25 Feb 2023) - RELEASE-NOTES: synced Philip Heiduck (25 Feb 2023) - .cirrus.yml: Bump to FreeBSD 13.2 Closes #10270 - ngtcp2-gnutls.yml: bump to gnutls 3.8.0 Closes #10507 - CI: update ngtcp2 and nghttp2 for pytest Follow-up: https://github.com/curl/curl/commit/5c9ee8cef4b351a085b440f8178500 124647f8e6 Closes #10508 Andy Alt (25 Feb 2023) - GHA: use same flags for Slackbuild as Slack package Closes #10526 Daniel Stenberg (24 Feb 2023) - rtsp: avoid sscanf for parsing Closes #10605 - http_proxy: parse the status line without sscanf Closes #10602 - telnet: error correctly for WS set to "x[num]" Follow-up to e4f93be9d587 Reported-by: Harry Sintonen Closes #10606 - krb5: avoid sscanf for parsing Closes #10599 - misc: remove support for curl_off_t < 8 bytes Closes #10597 - telnet: parse NEW_ENVIRON without sscanf Closes #10596 - telnet: parse the WS= argument without sscanf Closes #10596 - telnet: parse telnet options without sscanf Closes #10596 - ftp: replace sscanf for MDTM 213 response parsing Closes #10590 - ftp: replace sscanf for PASV parsing Closes #10590 - ftp: make the EPSV response parser not use sscanf Closes #10590 Stefan Eissing (24 Feb 2023) - ngtcp2: fix unwanted close of file descriptor 0 ... causing macOS to hand out 0 as next socket handle and failing on further operations. Reported-by: Sergey Fionov Fixes #10593 Closes #10595 Daniel Stenberg (23 Feb 2023) - select: stop treating POLLRDBAND as an error POLLRDBAND does not seem to be an general error and on Windows the value for POLLIN is 768 and the value for POLLRDBAND is 512. Fixes #10501 Reported-by: opensslonzos-github on github Closes #10592 - test978: mark file as text mode Follow-up to 4ea5702980cb To fix test failures on Windows Closes #10594 - http: rewrite the status line parser without sscanf Closes #10585 - test978: verify that --stderr works for -w's stderr as well Jay Satiro (23 Feb 2023) - curl: make -w's %{stderr} use the file set with --stderr Reported-by: u20221022 on github Fixes #10491 Closes #10569 - winbuild: fix makefile clean - Fix and move 'clean' code that removes the output and obj directories trees from MakefileBuild.vc to Makefile.vc. Prior to this change the 'clean' code did not work right because the variables containing the directory names were not fully initialized and the rmdir syntax was sometimes incorrect (typos). DIRDIST for example was set to ..\builds\ and not ..\builds\$(CONFIG_NAME_LIB)\ so it would remove the former and not the latter. If WITH_PREFIX was set then that directory was removed instead. Also, DIRDIST (the output directory) even if initialized should not be removed by MakefileBuild.vc because by that time it could be set to a user directory that may contain other files if WITH_PREFIX is set (eg we don't want rmdir /s /q C:\usr\local). Therefore we remove from Makefile.vc before any of that happens. I added a comment in both makefiles explaining this. Closes https://github.com/curl/curl/pull/10576 - sectransp: fix compiler warning c89 mixed code/declaration Since cbf57176 the Cirrus CI 'macOS arm64 SecureTransport http2' has been failing due to c89 warnings mixed code/declaration. That commit is not the cause so I assume something has changed in the CI outside of our configuration. Anyway, we don't mix code/declaration so this is the fix for that. Closes https://github.com/curl/curl/pull/10574 Philipp Engel (22 Feb 2023) - BINDINGS: add Fortran binding Closes #10589 Stefan Eissing (22 Feb 2023) - test2600: detect when ALARM_TIMEOUT is in use and adjust - use higher timeout values > 1s - skip duration checks Assisted-by: Marcel Raad Closes #10513 Daniel Stenberg (22 Feb 2023) - RELEASE-NOTES: synced - test686: verify return code for no URL after --next - tool_operate: propagate error codes for missing URL after --next Fixes #10558 Reported-by: u20221022 on github Closes #10580 - test1278: verify that an extra --no-remote-name cause no warning - tool_getparam: don't add a new node for just --no-remote-name Unless --remote-name-all is used. Fixes #10564 Reported-by: u20221022 on github Closes #10582 - gen.pl: add '%GLOBALS' as a variable for mainpage And use it in page-header to list all global command line options. - docs/cmdline-opts: mark all global options gen.pl now outputs a generic explanations for them for each option Fixes #10566 Reported-by: u20221022 on github Closes #10584 - GHA: add Microsoft C++ Code Analysis Closes #10583 - tool_progress: shut off progress meter for --silent in parallel Reported-by: finkjsc on github Fixes #10573 Closes #10579 - lib1560: add a test using %25 in the userinfo in a URL Closes #10578 Stefan Eissing (21 Feb 2023) - CURLOPT_PIPEWAIT: allow waited reuse also for subsequent connections As tested in test_02_07, when firing off 200 urls with --parallel, 199 wait for the first connection to be established. if that is multiuse, urls are added up to its capacity. The first url over capacity opens another connection. But subsequent urls found the same situation and open a connection too. They should have waited for the second connection to actually connect and make its capacity known. This change fixes that by - setting `connkeep()` early in the HTTP setup handler. as otherwise a new connection is marked as closeit by default and not considered for multiuse at all - checking the "connected" status for a candidate always and continuing to PIPEWAIT if no alternative is found. pytest: - removed "skip" from test_02_07 - added test_02_07b to check that http/1.1 continues to work as before Closes #10456 Daniel Stenberg (21 Feb 2023) - test419: verify --dump-header to file that cannot be created Closes #10571 - tool_operate: avoid fclose(NULL) on bad header dump file Fixes #10570 Reported-by: Jérémy Rabasco Closes #10571 - RELEASE-NOTES: synced Starting the journey towards 8.0.0 - cookie: parse without sscanf() Saves us from using 2*4096 bytes buffers on stack, the extra copies and more. Closes #10550 - lib517: verify time stamps without leading zeroes plus some more - parsedate: replace sscanf( for time stamp parsing Closes #10547 - parsedate: parse strings without using sscanf() - sscanf is slow and complex, avoid it - give up already if the string is 12 bytes or longer as no valid string can be that long - this can now be done without copy Closes #10547 Matt Jolly (20 Feb 2023) - tests: HTTP server fixups - httpserver.pl -> http-server.pl for consistency - add http3-server.pl to EXTRA_DIST; alphabetise for maintainability - nghttpx proxy invocation scripts should not use getcwd Closes #10568 Version 7.88.1 (20 Feb 2023) Daniel Stenberg (20 Feb 2023) - RELEASE-NOTES: synced 7.88.1 release - THANKS: add contributors from 7.88.1 - socketpair: allow EWOULDBLOCK when reading the pair check bytes Reported-by: Gunamoi Software Co-authored-by: Jay Satiro Fixes #10561 Closes #10562 Jay Satiro (18 Feb 2023) - tool_operate: fix scanbuild compiler warning Prior to this change Azure CI scanbuild warned of a potential NULL pointer string passed to strtol when CURLDEBUG enabled, even though the way the code was written it wouldn't have happened. Bug: https://github.com/curl/curl/commit/5479d991#r101159711 Reported-by: Marcel Raad Closes https://github.com/curl/curl/pull/10559 - curl_setup: Suppress OpenSSL 3 deprecation warnings - Define OPENSSL_SUPPRESS_DEPRECATED. OpenSSL 3 has deprecated some of the functions libcurl uses such as those with DES, MD5 and ENGINE prefix. We don't have replacements for those functions so the warnings were disabled in autotools and cmake builds, but still showed in other builds. Closes https://github.com/curl/curl/pull/10543 - build-openssl.bat: keep OpenSSL 3 engine binaries Prior to this change copying the OpenSSL 3 engine binaries failed because 'engines-1_1' (OpenSSL 1.1.x folder name) was erroneously used instead of 'engines-3'. The OpenSSL 3 builds would complete successfully but without the engine binaries. Closes https://github.com/curl/curl/pull/10542 ALittleDruid (18 Feb 2023) - cmake: fix Windows check for CryptAcquireContext Check for CryptAcquireContext in windows.h and wincrypt.h only, since otherwise this check may fail due to third party headers not found. Closes https://github.com/curl/curl/pull/10353 Daniel Stenberg (19 Feb 2023) - remote-header-name.d: mention that filename* is not supported and that you can use --clobber to allow overwriting. Ref: #10533 Closes #10555 Co-authored-by: Jay Satiro Pierrick Charron (18 Feb 2023) - CURLOPT_WS_OPTIONS.3: fix the availability version Closes #10557 Jacob Hoffman-Andrews (18 Feb 2023) - GHA: update rustls dependency to 0.9.2 This allows re-enabling test 312 for the rustls backend. Closes #10553 Philip Heiduck (18 Feb 2023) - HTTP3.md: update git branches Closes #10554 Stefan Eissing (17 Feb 2023) - urldata: remove `now` from struct SingleRequest - not needed Closes #10549 Daniel Stenberg (17 Feb 2023) - lib1560: add IPv6 canonicalization tests Closes #10552 - RELEASE-NOTES: synced - urlapi: do the port number extraction without using sscanf() - sscanf() is rather complex and slow, strchr() much simpler - the port number function does not need to fully verify the IPv6 address anyway as it is done later in the hostname_check() function and doing it twice is unnecessary. Closes #10541 Stefan Eissing (17 Feb 2023) - setopt: allow HTTP3 when HTTP2 is not defined Reported-by: Karthikdasari0423 on github Fixes #10538 Closes #10544 Jon Rumsey (17 Feb 2023) - os400: correct Curl_os400_sendto() Add const qualifier to 5th argument of Curl_os400_sendto() Make OS400 wrapper for sendto match the normal prototype of sendto() with a const qualifier. Fixes #10539 Closes #10548 Stefan Eissing (17 Feb 2023) - tests-httpd: add proxy tests for direct and tunneling checks on http: and https: Closes #10519 Daniel Stenberg (17 Feb 2023) - curl: make --silent work stand-alone - renamed the struct field to 'silent' to match the cmdline option - make --show-error toggle independently of --silent - make --silent independent of ->noprogress as well By doing this, the three options --silent, --no-progress-meter and --show-error should work independently of each other and also work with and without '--no-' prefix as documented. Reported-by: u20221022 on github Fixes #10535 Closes #10536 - socks: allow using DoH to resolve host names For SOCKS modes where a local host resolve is done. It was previously disabled in 12d655d4561, but a few local tests seem to indicate that it works fine. Works now because of the SOCKS refactor of 4a4b63daaa01ef59 that made it non-blocking. Reported-by: roughtex on github Fixes #10537 Closes #10540 Stefan Eissing (17 Feb 2023) - test: add test for HTTP/2 corruption as reported in #10525 - adding test_02_20 for reproducing the situation - using recently released mod_h2 Apache module - skipping test if an older version is installed - adding installation of current mod_h2 to github pytest workflow This reproduces the error reliable (for me) on the lib/http2.c version of curl 7.88.0. And passes with the recent curl master. Closes #10534 Daniel Stenberg (16 Feb 2023) - tool_operate: allow debug builds to set buffersize Using the CURL_BUFFERSIZE environment variable. Closes #10532 Stefan Eissing (16 Feb 2023) - connnect: fix timeout handling to use full duration - connect timeout was used at half the configured value, if the destination had 1 ip version 4 and other version 6 addresses (or the other way around) - extended test2600 to reproduce these cases Reported-by: Michael Kaufmann Fixes #10514 Closes #10517 Daniel Stenberg (16 Feb 2023) - tool_getparam: make --get a true boolean To match how it is documented in the man page. Fixes #10527 Reported-by: u20221022 on github Closes #10531 Harry Sintonen (16 Feb 2023) - http:: include stdint.h more readily Closes #10516 Stefan Eissing (16 Feb 2023) - tests: make the telnet server shut down a socket gracefully - test 1452 failed occasionally with ECONNRESET errnos in curl when the server closed the connection in an unclean state. Closes #10509 Harry Sintonen (16 Feb 2023) - http2: set drain on stream end Ensure that on_frame_recv() stream end will trigger a read if there is pending data. Without this it could happen that the pending data is never consumed. This combined with https://github.com/curl/curl/pull/10529 should fix https://github.com/curl/curl/issues/10525 Ref: https://github.com/curl/curl/issues/10525 Closes #10530 Stefan Eissing (16 Feb 2023) - http2: buffer/pausedata and output flush fix. * do not process pending input data when copying pausedata to the caller * return CURLE_AGAIN if the output buffer could not be completely written out. Ref: #10525 Closes #10529 Marcel Raad (16 Feb 2023) - krb5: silence cast-align warning Add an intermediate cast to `void *`, as done everywhere else when casting from `sockaddr *` to `sockaddr_in *`. Closes https://github.com/curl/curl/pull/10528 Daniel Stenberg (15 Feb 2023) - RELEASE-NOTES: synced bumped to 7.88.1 - tests: make sure gnuserv-tls has SRP support before using it Reported-by: fundawang on github Fixes #10522 Closes #10524 - runtests: fix "uninitialized value $port" by using a more appropriate variable Reported-by: fundawang on github Fixes #10518 Closes #10520 Version 7.88.0 (15 Feb 2023) Daniel Stenberg (15 Feb 2023) - RELEASE-NOTES: synced 7.88.0 release - THANKS: added contributors from 7.88.0 - openssl: rename 'errcode_t' to 'sslerr_t' Turns out "/usr/include/et/com_err.h" typedefs this type (without proper variable scoping). comerr is the "common error description library" that apparently might be use d by krb5 code, which then makes this header get used in a curl build. Reported-by: Bruno Henrique Batista Cruz da Silva Fixed #10502 Closes #10500 Dan Fandrich (13 Feb 2023) - CONTRIBUTE: More formally specify the commit description This codifies what people have actually used in git commits over the past 6 years. I've left off some lesser-used headers that appear to duplicate others and tried to describe a consistent use for several others that were used more arbitrarily. This makes it easier for new committers to find out the kinds of things we want to acknowledge, makes it easier to perform statistical analysis on commits, and opens the possibility of performing lint checks on descriptions before submission. Reviewed-by: Daniel Stenberg Reviewed-by: Jay Satiro Closes #10478 Stefan Eissing (13 Feb 2023) - openssl: test and fix for forward proxy handling (non-tunneling). - adding pytest test_10 cases for proxy httpd setup tests - fixing openssl bug in https: proxy hostname verification that used the hostname of the request and not the proxy name. Closes #10498 Daniel Stenberg (13 Feb 2023) - cmdline-opts/Makefile: on error, do not leave a partial And support 'make V=1' to show the full command line Closes #10497 - curl.1: make help, version and manual sections "custom" Instead of using "multi: boolean", as these are slightly special as in they do are not enable/disable ones. Fixes #10490 Reported-by: u20221022 on github Closes #10497 Stefan Eissing (13 Feb 2023) - tests: add tests for HTTP/2 and HTTP/3 to verify the header API Test 2403 and 2503 check "header_json" output and therefore use of header-api Closes #10495 Philip Heiduck (13 Feb 2023) - CI: update wolfssl / wolfssh to 5.5.4 / 1.4.12 Closes #10493 Daniel Stenberg (13 Feb 2023) - KNOW_BUGS: cleanups with some changed to TODOs - remove "Excessive HTTP/2 packets with TCP_NODELAY" This is not a bug. Rather room for improvement. I believe these have been fixed: - 17.4 Connection failures with parallel HTTP/2 - 17.5 HTTP/2 connections through HTTPS proxy frequently stall - remove "FTPS needs session reuse" That is still true, but curl should also do session reuse now. - remove "ASCII FTP" It is documented behavior, and not single user has asked for extended functionality here the last decade or so. - remove "Passive transfer tries only one IP address" add as a TODO - remove "DoH leaks memory after followlocation" With a recipe on how to reproduce, this is pointless to keep around - remove "DoH does not inherit all transfer options" add it as a TODO Closes #10487 Tatsuhiro Tsujikawa (13 Feb 2023) - GHA: bump ngtcp2 workflow dependencies Closes #10494 Patrick Monnerat (13 Feb 2023) - content_encoding: do not reset stage counter for each header Test 418 verifies Closes #10492 Daniel Stenberg (13 Feb 2023) - RELEASE-NOTES: synced Jay Satiro (13 Feb 2023) - multi: stop sending empty HTTP/3 UDP datagrams on Windows - Limit the 0-sized send procedure that is used to reset a SOCKET's FD_WRITE to TCP sockets only. Prior to this change the reset was used on UDP sockets as well, but unlike TCP sockets a 0-sized send actually sends out a datagram. Assisted-by: Marc Hörsken Ref: https://github.com/curl/curl/pull/9203 Fixes https://github.com/curl/curl/issues/9086 Closes https://github.com/curl/curl/pull/10430 Viktor Szakats (12 Feb 2023) - h3: silence compiler warnings Reviewed-by: Daniel Stenberg Fixes #10485 Closes #10486 Daniel Stenberg (12 Feb 2023) - smb: return error on upload without size The protocol needs to know the size ahead of time, this is now a known restriction and not a bug. Also output a clearer error if the URL path does not contain proper share. Ref: #7896 Closes #10484 Viktor Szakats (12 Feb 2023) - windows: always use curl's basename() implementation The `basename()` [1][2] implementation provided by mingw-w64 [3] makes assumptions about input encoding and may break with non-ASCII strings. `basename()` was auto-detected with CMake, autotools and since 68fa9bf3f5d7b4fcbb57619f70cb4aabb79a51f6 (2022-10-13), also in `Makefile.mk` after syncing its behaviour with the mainline build methods. A similar patch for curl-for-win broke official Windows builds earlier, in release 7.83.1_4 (2022-06-15). This patch forces all Windows builds to use curl's internal `basename()` implementation to avoid such problems. [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/basename.html [2]: https://www.man7.org/linux/man-pages/man3/basename.3.html [3]: https://sourceforge.net/p/mingw-w64/mingw-w64/ci/master/tree/mingw-w64-c rt/misc/basename.c Reported-by: UnicornZhang on Github Assisted-by: Cherish98 on Github Reviewed-by: Daniel Stenberg Fixes #10261 Closes #10475 Philip Heiduck (12 Feb 2023) - Linux CI: Bump rustls-ffi to v0.9.1 Closes #10476 Daniel Stenberg (12 Feb 2023) - libtest: build lib2305 with multibyte as well Fixes a build regression. Follow-up to 5a9a04d5567 Reported-by: Viktor Szakats Ref: https://github.com/curl/curl/pull/10475#issuecomment-1426831800 Closes #10477 Dmitry Atamanov (12 Feb 2023) - cmake: fix dev warning due to mismatched arg The package name passed to find_package_handle_standard_args (BROTLI) does not match the name of the calling package (Brotli). This can lead to problems in calling code that expects find_package result variables (e.g., _FOUND) to follow a certain pattern. Closes https://github.com/curl/curl/pull/10471 James Keast (11 Feb 2023) - setopt: Address undefined behaviour by checking for null This addresses undefined behaviour found using clang's UBsan: curl/lib/setopt.c:177:14: runtime error: applying non-zero offset 1 to null p ointer SUMMARY: UndefinedBehaviorSanitizer: undefined-behavior curl/lib/setopt.c:177 :14 in Closes #10472 Jacob Hoffman-Andrews (11 Feb 2023) - rustls: improve error messages Changes numeric error codes into verbose error codes in two places. Adds a prefix indicating that the error came from rustls, and in some places which function it came from. Adds special handling for RUSTLS_RESULT_UNEXPECTED_EOF, since the default message of "Unexpected EOF" is insufficiently explanatory. Closes #10463 Daniel Stenberg (11 Feb 2023) - openssl: remove dead code Follow-up to e8b00fcd6a Due to the new 'if(!nonblocking)' check on the block a level above, there is no need to check for it again within the same conditional. Detected by Coverity Closes #10473 - ngtcp2: replace removed define and stop using removed function They were removed upstream. Reported-by: Karthikdasari0423 on github Fixes #10469 Closes #10474 - scripts/delta: show percent of number of files changed since last tag - RELEASE-NOTES: synced Stefan Eissing (10 Feb 2023) - pytest: add a test case for PUSH related things. - checking that "103 Early Hints" are visible in curl's header dump file Closes #10452 Gregory Panakkal (10 Feb 2023) - WEBSOCKET.md: typo Fixing missing slash for ws protocol scheme Closes #10464 Stefan Eissing (10 Feb 2023) - vquic: stabilization and improvements vquic stabilization - udp send code shared between ngtcp2 and quiche - quiche handling of data and events improved ngtcp2 and pytest improvements - fixes handling of "drain" situations, discovered in scorecard tests with the Caddy server. - improvements in handling transfers that have already data or are already closed to make an early return on recv pytest - adding caddy tests when available scorecard improvemnts. - using correct caddy port - allowing tests for only httpd or caddy Closes #10451 Philip Heiduck (10 Feb 2023) - Linux CI: update some dependecies to latest tag Closes #10458 Daniel Stenberg (10 Feb 2023) - test2305: send 3 frames, 4097 bytes each, as one message Receive them using a 256 bytes buffer in a loop. - ws: fix recv of larger frames + remove 'oleft' from the struct + deal with "overflow data" in a separate dynbuf Reported-by: Mike Duglas Fixes #10438 Closes #10447 - curl/websockets.h: extend the websocket frame struct - sws: fix typo, indentation add more ws logging - test2304: remove stdout verification This cripples the test somewhat but the check was bad since depending on timing it could exit before the output was done, making the test flaky. Dan Fandrich (9 Feb 2023) - CI: Add more labeler match patterns - CI: Retry failed downloads to reduce spurious failures A temporary error with a remote server shouldn't cause a CI run to fail. Also, put a cap on the time to download to fail faster on a misbehaving server or connection and use HTTP compression where possible to reduce download times. Daniel Stenberg (9 Feb 2023) - no-clobber.d: only use long form options in man page text ... since they are expanded and the short-form gets mentioned automatically so if the short form is mentioned as well, it gets repeated. Fixes #10461 Closes #10462 Reported-by: Dan Fandrich - GHA: enable websockets in the torture job Closes #10448 - header.d: add a header file example Closes #10455 Stefan Eissing (9 Feb 2023) - HTTP/[23]: continue upload when state.drain is set - as reported in #10433, HTTP/2 uploads may stall when a response is received before the upload is done. This happens when the data->state.drain is set for such a transfer, as the special handling in transfer.c from then on only cared about downloads. - add continuation of uploads, if applicable, in this case. - add pytest case test_07_12_upload_seq_large to reproduce this scenario (although, current nghttp2 implementation is using drain less often) Reported-by: Lucas Pardue Fixes #10433 Closes #10443 - http2: minor buffer and error path fixes - use memory buffer in full available size - fail receive of reset/errored streams early pytest: - make test_05 error cases more reliable Closes #10444 Federico Pellegrin (9 Feb 2023) - openldap: fix missing sasl symbols at build in specific configs If curl is built with openldap support (USE_OPENLDAP=1) but does not have also some other protocol (IMAP/SMTP/POP3) enabled that brings in Curl_sasl_* functions, then the build will fail with undefined references to various symbols: ld: ../lib/.libs/libcurl.so: undefined reference to `Curl_sasl_decode_mech' ld: ../lib/.libs/libcurl.so: undefined reference to `Curl_sasl_parse_url_auth _option' ld: ../lib/.libs/libcurl.so: undefined reference to `Curl_sasl_cleanup' ld: ../lib/.libs/libcurl.so: undefined reference to `Curl_sasl_can_authentica te' ld: ../lib/.libs/libcurl.so: undefined reference to `Curl_sasl_continue' ld: ../lib/.libs/libcurl.so: undefined reference to `Curl_sasl_start' ld: ../lib/.libs/libcurl.so: undefined reference to `Curl_sasl_init' This was tracked down to these functions bein used in openldap.c but defined in curl_sasl.c and then forward in two vauth/ files to have a guard against a set of #define configurations that was now extended to cover also this case. Example configuration targeted that could reproduce the problem: curl 7.87.1-DEV () libcurl/7.87.1-DEV .... OpenLDAP/2.6.3 Protocols: file ftp ftps http https ldap ldaps Closes #10445 Daniel Stenberg (9 Feb 2023) - ws: use %Ou for outputting curl_off_t with info() Reported-by: Mike Duglas Fixes #10439 Closes #10441 Jay Satiro (9 Feb 2023) - curl_setup: Disable by default recv-before-send in Windows Prior to this change a workaround for Windows to recv before every send was enabled by default. The way it works is a recv is called before every send and saves the received data, in case send fails because in Windows apparently that can wipe out the socket's internal received data buffer. This feature has led to several bugs because the way libcurl operates it waits on a socket to read or to write, and may not at all times check for buffered receive data. Two recent significant bugs this workaround caused: - Broken Schannel TLS 1.3 connections (#9431) - HTTP/2 arbitrary hangs (#10253) The actual code remains though it is disabled by default. Though future changes to connection filter buffering could improve the situation IMO it's just not tenable to manage this workaround. Ref: https://github.com/curl/curl/issues/657 Ref: https://github.com/curl/curl/pull/668 Ref: https://github.com/curl/curl/pull/720 Ref: https://github.com/curl/curl/issues/9431 Ref: https://github.com/curl/curl/issues/10253 Closes https://github.com/curl/curl/pull/10409 Stefan Eissing (8 Feb 2023) - http2: aggregate small SETTINGS/PRIO/WIN_UPDATE frames add a small buffer to nghttp2 session sending in order to aggregate small SETTINGS/PRIO/WIN_UPDATE frames that nghttp2 "writes" to the callback individually. Ref: #10389 Closes #10432 - openssl: store the CA after first send (ClientHello) move Curl_ssl_setup_x509_store() call after the first send (ClientHello) this gives time to parse CA anchors while waiting on the server reply Ref: #10389 Closes #10432 Daniel Stenberg (8 Feb 2023) - RELEASE-NOTES: synced Anthony Hu (8 Feb 2023) - wolfssl: remove deprecated post-quantum algorithms Closes #10440 John Bampton (8 Feb 2023) - misc: fix spelling Closes #10437 Daniel Stenberg (7 Feb 2023) - man pages: call the custom user pointer 'clientp' consistently The variable had a few different names. Now try to use 'clientp' consistently for all man pages using a custom pointer set by the application. Reported-by: Gerrit Renker Fixes #10434 Closes #10435 - vtls: infof using %.*s needs to provide the length as int Fixes a Coverity warning. Closes #10436 Stefan Eissing (7 Feb 2023) - vrls: addressing issues reported by coverity I believe the code was secure before this, but limiting the accepted name length to what is used in the structures should help Coverity's analysis. Closes #10431 Daniel Stenberg (7 Feb 2023) - tool_operate: move the 'updated' variable This was already done by Dan Fandrich in the previous PR but somehow I lost that fixup. Follow-up to 349c5391f2121e Dan Fandrich (7 Feb 2023) - tool_operate: Fix error codes during DOS filename sanitize It would return CURLE_URL_MALFORMAT in an OOM condition. Closes #10414 - tool_operate: Fix error codes on bad URL & OOM curl would erroneously report CURLE_OUT_OF_MEMORY in some cases instead of CURLE_URL_MALFORMAT. In other cases, it would erroneously return CURLE_URL_MALFORMAT instead of CURLE_OUT_OF_MEMORY. Add a test case to test the former condition. Fixes #10130 Closes #10414 Daniel Stenberg (6 Feb 2023) - setopt: use >, not >=, when checking if uarg is larger than uint-max Closes #10421 - vtls: fix failf() format argument type for %.*s handling Reported by Coverity Closes #10422 - openssl: fix "Improper use of negative value" By getting the socket first and returning error in case of bad socket. Detected by Coverity. Closes #10423 Dan Fandrich (6 Feb 2023) - packages: Remove Android.mk from makefile This was missed in commit #44141512 Ref: #10418 Daniel Stenberg (6 Feb 2023) - curl_ws_send.3: clarify how to send multi-frame messages Mike Duglas (6 Feb 2023) - ws: fix multiframe send handling Fixes #10413 Closes #10420 Daniel Stenberg (6 Feb 2023) - unit2600: make sure numerical curl_easy_setopt sets long Follow-up to 671158242db3203 Reported-by: Marcel Raad Fixes #10410 Closes #10419 Andy Alt (6 Feb 2023) - GHA: move Slackware test into matrix Closes #10412 Pronyushkin Petr (6 Feb 2023) - urlapi: fix part of conditional expression is always true: qlen Closes #10408 - url: fix part of conditional expression is always true Closes #10407 Daniel Stenberg (6 Feb 2023) - RELEASE-NOTES: synced Philip Heiduck (6 Feb 2023) - GHA/macos.yml: bump to gcc-12 Closes #10415 Daniel Stenberg (6 Feb 2023) - packages: remove Android, update README - Nobody builds curl for Android using this anymore - Refreshed the README and converted to markdown Reported-by: John Porter Fixes #10416 Closes #10418 Kvarec Lezki (5 Feb 2023) - fopen: remove unnecessary assignment [CWE-1164] V1048: The '* tempname' variable was assigned the same value. Ref: https://pvs-studio.com/en/docs/warnings/v1048/ Closes https://github.com/curl/curl/pull/10398 Gisle Vanem (5 Feb 2023) - libtest: add a sleep macro for Windows .. because sleep() is used in some libtests. Closes https://github.com/curl/curl/pull/10295 Kvarec Lezki (3 Feb 2023) - http_aws_sigv4: remove typecasts from HMAC_SHA256 macro V220: Suspicious sequence of types castings: memsize -> 32-bit integer -> mem size. https://pvs-studio.com/en/docs/warnings/v220/ Closes #10400 Daniel Stenberg (3 Feb 2023) - mailmap: Thomas1664 on github Thomas1664 on github (3 Feb 2023) - CURLOPT_WRITEFUNCTION.3: fix memory leak in example Closes #10390 Kvarec Lezki (3 Feb 2023) - doh: ifdef IPv6 code For disabled IPv6 a condition (conn->ip_version != CURL_IPRESOLVE_V4) is always false. https://pvs-studio.com/en/docs/warnings/v560/ Closes #10397 Daniel Stenberg (3 Feb 2023) - urlapi: remove pathlen assignment "Value stored to 'pathlen' is never read" Follow-up to 804d5293f89 Reported-by: Kvarec Lezki Closes #10405 Kvarec Lezki (3 Feb 2023) - http: fix "part of conditional expression is always false" [CWE-570] V560: A part of conditional expression is always false: conn->bits. authneg. [CWE-570] V560: A part of conditional expression is always false: conn->handl er->protocol & (0 | 0). https://pvs-studio.com/en/docs/warnings/v560/ Closes #10399 Daniel Stenberg (2 Feb 2023) - urlapi: skip the extra dedotdot alloc if no dot in path Saves an allocation for many/most URLs. Updates test 1395 accordingly Closes #10403 Stefan Eissing (2 Feb 2023) - connections: introduce http/3 happy eyeballs New cfilter HTTP-CONNECT for h3/h2/http1.1 eyeballing. - filter is installed when `--http3` in the tool is used (or the equivalent CURLOPT_ done in the library) - starts a QUIC/HTTP/3 connect right away. Should that not succeed after 100ms (subject to change), a parallel attempt is started for HTTP/2 and HTTP/1.1 via TCP - both attempts are subject to IPv6/IPv4 eyeballing, same as happens for other connections - tie timeout to the ip-version HAPPY_EYEBALLS_TIMEOUT - use a `soft` timeout at half the value. When the soft timeout expires, the HTTPS-CONNECT filter checks if the QUIC filter has received any data from the server. If not, it will start the HTTP/2 attempt. HTTP/3(ngtcp2) improvements. - setting call_data in all cfilter calls similar to http/2 and vtls filters for use in callback where no stream data is available. - returning CURLE_PARTIAL_FILE for prematurely terminated transfers - enabling pytest test_05 for h3 - shifting functionality to "connect" UDP sockets from ngtcp2 implementation into the udp socket cfilter. Because unconnected UDP sockets are weird. For example they error when adding to a pollset. HTTP/3(quiche) improvements. - fixed upload bug in quiche implementation, now passes 251 and pytest - error codes on stream RESET - improved debug logs - handling of DRAIN during connect - limiting pending event queue HTTP/2 cfilter improvements. - use LOG_CF macros for dynamic logging in debug build - fix CURLcode on RST streams to be CURLE_PARTIAL_FILE - enable pytest test_05 for h2 - fix upload pytests and improve parallel transfer performance. GOAWAY handling for ngtcp2/quiche - during connect, when the remote server refuses to accept new connections and closes immediately (so the local conn goes into DRAIN phase), the connection is torn down and a another attempt is made after a short grace period. This is the behaviour observed with nghttpx when we tell it to shut down gracefully. Tested in pytest test_03_02. TLS improvements - ALPN selection for SSL/SSL-PROXY filters in one vtls set of functions, repl aces copy of logic in all tls backends. - standardized the infof logging of offered ALPNs - ALPN negotiated: have common function for all backends that sets alpn propr ty and connection related things based on the negotiated protocol (or lack the reof). - new tests/tests-httpd/scorecard.py for testing h3/h2 protocol implementatio n. Invoke: python3 tests/tests-httpd/scorecard.py --help for usage. Improvements on gathering connect statistics and socket access. - new CF_CTRL_CONN_REPORT_STATS cfilter control for having cfilters report connection statistics. This is triggered when the connection has completely connected. - new void Curl_pgrsTimeWas(..) method to report a timer update with a timestamp of when it happend. This allows for updating timers "later", e.g. a connect statistic after full connectivity has been reached. - in case of HTTP eyeballing, the previous changes will update statistics only from the filter chain that "won" the eyeballing. - new cfilter query CF_QUERY_SOCKET for retrieving the socket used by a filter chain. Added methods Curl_conn_cf_get_socket() and Curl_conn_get_socket() for convenient use of this query. - Change VTLS backend to query their sub-filters for the socket when checks during the handshake are made. HTTP/3 documentation on how https eyeballing works. TLS improvements - ALPN selection for SSL/SSL-PROXY filters in one vtls set of functions, repl aces copy of logic in all tls backends. - standardized the infof logging of offered ALPNs - ALPN negotiated: have common function for all backends that sets alpn propr ty and connection related things based on the negotiated protocol (or lack the reof). Scorecard with Caddy. - configure can be run with `--with-test-caddy=path` to specify which caddy t o use for testing - tests/tests-httpd/scorecard.py now measures download speeds with caddy pytest improvements - adding Makfile to clean gen dir - adding nghttpx rundir creation on start - checking httpd version 2.4.55 for test_05 cases where it is needed. Skippin g with message if too old. - catch exception when checking for caddy existance on system. Closes #10349 Daniel Stenberg (2 Feb 2023) - CODEOWNERS: remove the peeps mentioned as CI owners These owners do not have the bandwidth/energy to do the reviews which makes PRs stall and this ownership claim flawed. We can bring people back when the situation is different. Follow-up to c04c78ac87c4d46737934345a Closes #10386 Martin D'Aloia (2 Feb 2023) - write-out.d: add 'since version' to %{header_json} documentation The documentation of `%{header_json}` missed to mention since which version this variable for `--write-out` is present. Based on commit https://github.com/curl/curl/commit/4133a69f2daa476bb we can determine from the tags were this commit is present that the first version to include it was `7.83.0`. This could be also checked with: `git tag --contains 4133a69f2daa476bb6d902687f1dd6660ea9c3c5` Closes #10395 Daniel Stenberg (1 Feb 2023) - urlapi: avoid Curl_dyn_addf() for hex outputs Inspired by the recent fixes to escape.c, we should avoid calling Curl_dyn_addf() in loops, perhaps in particular when adding something so simple as %HH codes - for performance reasons. This change makes the same thing for the URL parser's two URL-encoding loops. Closes #10384 - urlapi: skip path checks if path is just "/" As a miniscule optimization, treat a path of the length 1 as the same as non-existing, as it can only be a single leading slash, and that's what we do for no paths as well. Closes #10385 Philip Heiduck (1 Feb 2023) - GHA/macos: use Xcode_14.0.1 for cmake builds Fixes #10356 Closes #10381 Viktor Szakats (1 Feb 2023) - tls: fixes for wolfssl + openssl combo builds 1. Add `USE_WOLFSSL` to the TLS backend priority list in `lib/curl_ntlm_core.c`. 2. Fix `lib/curl_ntlm_core.h` to respect TLS backend priority, bringing it in sync with the above list and `lib/curl_ntlm_core.c` itself. Reported-by: Mark Roszko Ref: https://github.com/curl/curl/issues/10321 3. Allow enabling both wolfSSL and OpenSSL at the same time in `lib/Makefile.mk` bringing this in line with cmake/autotools builds. Update logic to select the crypto-specific lib for `ngtcp2`, which supports a single TLS backend at the same time. Closes #10322 Daniel Stenberg (1 Feb 2023) - RELEASE-NOTES: synced - docs/INSTALL: document how to use multiple TLS backends And document how OpenSSL forks and wolfSSL cannot be used at the same time. Reported-by: Mark Roszko Fixes #10321 Closes #10382 Kvarec Lezki (1 Feb 2023) - cookies: fp is always not NULL Closes #10383 Daniel Stenberg (31 Jan 2023) - escape: use table lookup when adding %-codes to output On my dev host, this code runs 7.8 times faster. Closes #10377 - unit2600: avoid error: ‘TEST_CASES’ defined but not used Follow-up to d55de24dce9d51 Closes #10379 - escape: hex decode with a lookup-table Makes the decoding 2.8 times faster in my tests. Closes #10376 - cf-socket: fix build error wo TCP_FASTOPEN_CONNECT Follow-up to 5651a36d1a Closes #10378 Reviewed-by: Stefan Eissing Stefan Eissing (31 Jan 2023) - CI: add pytest github workflow to CI test/tests-httpd on a HTTP/3 setup Closes #10317 - connect: fix strategy testing for attempts, timeouts and happy-eyeball - add test2600 as a unit test that triggers various connect conditions and monitors behaviour, available in a debug build only. - this exposed edge cases in connect.c that have been fixed Closes #10312 - cf-socket: improvements in socket I/O handling - Curl_write_plain/Curl_read_plain have been eliminated. Last code use now uses Curl_conn_send/recv so that requests use conn->send/revc callbacks which defaults to cfilters use. - Curl_recv_plain/Curl_send_plain have been internalized in cf-socket.c. - USE_RECV_BEFORE_SEND_WORKAROUND (active on Windows) has been moved into cf-socket.c. The pre_recv buffer is held at the socket filter context. `postponed_data` structures have been removed from `connectdata`. - the hanger in HTTP/2 request handling was a result of read buffering on all sends and the multi handling is not prepared for this. The following happens: - multi preforms on a HTTP/2 easy handle - h2 reads and processes data - this leads to a send of h2 data - which receives and buffers before the send - h2 returns - multi selects on the socket, but no data arrives (its in the buffer alre ady) the workaround now receives data in a loop as long as there is something i n the buffer. The real fix would be for multi to change, so that `data_pendi ng` is evaluated before deciding to wait on the socket. io_buffer, optional, in cf-socket.c, http/2 sets state.drain if lower filter have pending data. This io_buffer is only available/used when the -DUSE_RECV_BEFORE_SEND_WORKAROUND is active, e.g. on Windows configurations. It also maintains the original checks on protocol handler being HTTP and conn->send/recv not being replaced. The HTTP/2 (nghttp2) cfilter now sets data->state.drain when it finds out that the "lower" filter chain has still pending data at the end of its IO operation. This prevents the processing from becoming stalled. Closes #10280 Daniel Stenberg (31 Jan 2023) - openssl: only use CA_BLOB if verifying peer Reported-by: Paul Groke Bug: https://curl.se/mail/lib-2023-01/0070.html Fixes #10351 Closes #10359 Thomas1664 on github (31 Jan 2023) - curl_free.3: fix return type of `curl_free` Fixes #10373 Closes #10374 Daniel Stenberg (30 Jan 2023) - zuul: stop using this CI service The important jobs have already transitioned. The remaining ones we can skip for now. Closes #10368 - copyright: remove "m4/ax_compile_check_sizeof.m4" from skips and report if skipped files do not exist. Follow-up to 9e11c2791fb960758 which removed the file. Closes #10369 - ws: unstick connect-only shutdown As this mode uses blocking sockets, it must set them back to non-blocking in disconnect to avoid the risk of getting stuck. Closes #10366 - ws: remove bad assert Reported-by: Stanley Wucw Fixes #10347 Closes #10366 - openssl: adapt to boringssl's error code type BoringSSL uses uint32_t, OpenSSL uses 'unsigned 'long' Closes #10360 - tool_operate: repair --rate Regression from a55256cfb242 (7.87.0) Reported-by: highmtworks on github Fixes #10357 Closes #10358 - dict: URL decode the entire path always Reported-by: dekerser on github Fixes #10298 Closes #10354 Stefan Eissing (29 Jan 2023) - vtls: do not null-check when we already assume cf-ctx exists Fixes #10361 Closes #10362 Daniel Stenberg (29 Jan 2023) - RELEASE-NOTES: synced - CURLOPT_READFUNCTION.3: the callback 'size' arg is always 1 Reported-by: Brian Green Fixes #10328 Closes #10355 - copyright.pl: cease doing year verifications As we have (mostly) removed the copyright year ranges. Reported-by: Ryan Schmidt Fixes #10345 Closes #10352 Dan Fandrich (28 Jan 2023) - CI: Work around a labeler bug that removes labels Jay Satiro (26 Jan 2023) - write-out.d: clarify Windows % symbol escaping - Clarify that in Windows batch files the % must be escaped as %%, and at the command prompt it cannot be escaped which could lead to incorrect expansion. Prior to this change the doc implied % must be escaped as %% in win32 always. --- Examples showing how a write-out argument is received by curl: If curl --write-out "%{http_code}" is executed in a batch file: {http_code} If curl --write-out "%%{http_code}" is executed in a batch file: %{http_code} If curl --write-out "%{http_code}" is executed from the command prompt: %{http_code} If curl --write-out "%%{http_code}" is executed from the command prompt: %%{http_code} At the command prompt something like "%{speed_download}%{http_code}" would first be parsed by the command interpreter as %{speed_download}% and would be expanded as environment variable {speed_download} if it existed, though that's highly unlikely since Windows environment names don't use braces. --- Reported-by: Muhammad Hussein Ammari Ref: https://github.com/bagder/everything-curl/pull/279 Fixes https://github.com/curl/curl/issues/10323 Closes https://github.com/curl/curl/pull/10337 Ryan Schmidt (26 Jan 2023) - connect: Fix build when not ENABLE_IPV6 Check for ENABLE_IPV6 before accessing AF_INET6. Fixes build failure introduced in 1c5d8ac. Closes https://github.com/curl/curl/pull/10344 - cf-socket: Fix build when not HAVE_GETPEERNAME Remove remaining references to conn and sockfd, which were removed from the function signature when conninfo_remote was renamed to conn_set_primary_ip in 6a8d7ef. Closes https://github.com/curl/curl/pull/10343 Stefan Eissing (26 Jan 2023) - vtls: Manage current easy handle in nested cfilter calls The previous implementation cleared `data` so the outer invocation lost its data, which could lead to a crash. Bug: https://github.com/curl/curl/issues/10336 Reported-by: Fujii Hironori Closes https://github.com/curl/curl/pull/10340 Dan Fandrich (25 Jan 2023) - CI: Add even more paths to the labeler config (#10326) - scripts: Fix Appveyor job detection in cijobs.pl The reorganization in #9769 broke the script. This should probably be rewritten to use a YAML parser for better upward compatibility. - CI: Add a few more paths to the labeler config (#10326) - CI: Switch the labeler event to pull_request_target Otherwise, the action won't work on PRs from forked repositories (#10326). Viktor Szakats (25 Jan 2023) - cmake: delete redundant macro definition `SECURITY_WIN32` Stop explicitly defining `SECURITY_WIN32` in CMake builds. No other build systems define this macro, because it's unconditionally defined in `lib/curl_sspi.h` already. This is the only curl source using the `sspi.h` and `security.h` Win32 headers, and no other Win32 headers need this macro. Reviewed-by: Jay Satiro Closes #10341 Fredrik (24 Jan 2023) - winbuild: document that arm64 is supported Building an arm64 version works flawlessly with the VS arm64 toolset. Closes https://github.com/curl/curl/pull/10332 Cherish98 (24 Jan 2023) - openssl: don't log raw record headers - Skip content type SSL3_RT_HEADER in verbose TLS output. This commit prevents bogus and misleading verbose TLS header messages as discussed in #10299. Assisted-by: Peter Wu Closes https://github.com/curl/curl/pull/10299 Marc Aldorasi (24 Jan 2023) - cmake: use list APPEND syntax for CMAKE_REQUIRED_DEFINITIONS - Use list() instead of set() for CMAKE_REQUIRED_DEFINITIONS list since the former is clearer. Closes https://github.com/curl/curl/pull/10272 Dan Fandrich (23 Jan 2023) - CI: Add a workflow to automatically label pull requests The labeler language is quite restrictive right now so labels are added quite conservatively, meaning that many PRs won't get labels when it's "obvious" they should. It will still save some manual work on those that it can label. Jay Satiro (21 Jan 2023) - system.h: assume OS400 is always built with ILEC compiler Prior to this change the OS400 types were only defined when __ILEC400__. That symbol is only defined by IBM's C compiler and not their C++ compiler, which led to missing types when users on OS400 would compile a C++ application that included curl. The IBM C and C++ compilers are the only native compilers on the platform. Assisted-by: Jon Rumsey Reported-by: John Sherrill Fixes https://github.com/curl/curl/issues/10305 Closes https://github.com/curl/curl/pull/10329 xgladius (20 Jan 2023) - cmake: Remove deprecated symbols check curl stopped use of CMAKE_USE_ as a prefix for its own build symbols in 2021 and added a check, meant to last 1 year, to fatally error on those symbols. This commit removes that check. Closes https://github.com/curl/curl/pull/10314 Dan Fandrich (20 Jan 2023) - docs: POSTFIELDSIZE must be set to -1 with read function Reported-by: RanBarLavie on github Closes #10313 Stefan Eissing (20 Jan 2023) - vtls: fix hostname handling in filters - Copy the hostname and dispname to ssl_connect_data. Use a copy instead of referencing the `connectdata` instance since this may get free'ed on connection reuse. Reported-by: Stefan Talpalaru Reported-by: sergio-nsk@users.noreply.github.com Fixes https://github.com/curl/curl/issues/10273 Fixes https://github.com/curl/curl/issues/10309 Closes https://github.com/curl/curl/pull/10310 Sergey Bronnikov (17 Jan 2023) - lib: fix typos Closes https://github.com/curl/curl/pull/10307 - curl_version_info.3: fix typo Closes https://github.com/curl/curl/pull/10306 Jay Satiro (17 Jan 2023) - openssl: Don't ignore CA paths when using Windows CA store (redux) .. and remove 'experimental' designation from CURLSSLOPT_NATIVE_CA. This commit restores the behavior of CURLSSLOPT_NATIVE_CA so that it does not override CURLOPT_CAINFO / CURLOPT_CAPATH, or the hardcoded default locations. Instead the native Windows CA store can be used at the same time. --- This behavior was originally added over two years ago in abbc5d60 (#5585) but then 83393b1a (#7892) broke it over a year ago, I assume inadvertently. The CURLSSLOPT_NATIVE_CA feature was marked experimental and likely rarely used. Ref: https://github.com/curl/curl/pull/5585 Ref: https://github.com/curl/curl/pull/7892 Ref: https://curl.se/mail/lib-2023-01/0019.html Closes https://github.com/curl/curl/pull/10244 Daniel Stenberg (13 Jan 2023) - RELEASE-NOTES: synced - ws: fix autoping handling Reported-by: Alexey Savchuk Fixes #10289 Closes #10294 - curl_log: avoid printf() format checking with mingw Since it does not seem to like %zu and more Follow-up to db91dbbf2 Fixes #10291 Closes #10292 - tool_getparam: fix compiler warning when !HAVE_WRITABLE_ARGV Follow-up to 2ed0e1f70ee176edf3d2 Closes #10286 Stefan Eissing (12 Jan 2023) - openssl: make the BIO_METHOD a local variable in the connection filter This avoids UAF issues when `curl_global_cleanup()` is called before all transfers have been completely handled. Unfortunately this seems to be a more common pattern than we like. Closes #10285 Daniel Stenberg (12 Jan 2023) - curl: output warning at --verbose output for debug-enabled version + a libcurl warning in the debug output Assisted-by: Jay Satiro Ref: https://curl.se/mail/lib-2023-01/0039.html Closes #10278 - src: add --http3-only Warning: --http3 and --http3-only are subject to change again (or be removed) before HTTP/3 support goes non-experimental. Closes #10264 - curl.h: add CURL_HTTP_VERSION_3ONLY As the previous CURL_HTTP_VERSION_3 option gets a slightly altered meaning. Closes #10264 - connect: fix access of pointer before NULL check Detected by Coverity CID 1518992 Closes #10284 Daniel Gustafsson (12 Jan 2023) - easyoptions: Fix header printing in generation script The optiontable.pl script prints the header comment when generating easyoptions.c, but it wasn't escaping all characters which jumbled the curl ascii logo. Fix by escaping. Cloes #10275 Harry Sintonen (12 Jan 2023) - tool_getparam: fix hiding of command line secrets Closes #10276 Stefan Eissing (12 Jan 2023) - tests: document the cfilter debug logging options Closes #10283 - curl_log: for failf/infof and debug logging implementations - new functions and macros for cfilter debugging - set CURL_DEBUG with names of cfilters where debug logging should be enabled - use GNUC __attribute__ to enable printf format checks during compile Closes #10271 Daniel Stenberg (10 Jan 2023) - RELEASE-NOTES: synced Nick Banks (10 Jan 2023) - msh3: update to v0.6 Closes #10192 Stefan Eissing (10 Jan 2023) - ngtcp2: add CURLOPT_SSL_CTX_FUNCTION support for openssl+wolfssl Using common method for SSL_CTX initialization of verfiy peer and CA settings. This also provides X509_STORE sharing to become available for ngtcp2+openssl HTTP/3. Reported-by: violetlige on github Fixes #10222 Closes #10239 Daniel Stenberg (10 Jan 2023) - cf-socket: make infof() call use %zu for size_t output Detected by Coverity CID 1518986 and CID 1518984 Closes #10268 Jon Rumsey (10 Jan 2023) - os400: fixes to make-lib.sh and initscript.sh Adjust how exports list is generated from header files to account for declarations across multiple lines and CURL_DEPRECATED(...) tags. Update initscript.sh Specify qadrt_use_inline to prevent unistd.h in ASCII runtime defining close(a) -> close_a(a) Fixes #10266 Closes #10267 Stefan Eissing (9 Jan 2023) - tests-httpd: basic infra to run curl against an apache httpd plus nghttpx for h3 - adding '--with-test-httpd=' to configure non-standard apache2 install - python env and base classes for running httpd - basic tests for connectivity with h1/h2/h3 - adding test cases for truncated responses in http versions. - adding goaway test for HTTP/3. - adding "stuttering" tests with parallel downloads in chunks with varying delays between chunks. - adding a curltest module to the httpd server, adding GOAWAY test. - mod_curltest now installs 2 handlers - 'echo': writing as response body what came as request body - 'tweak': with query parameters to tweak response behaviour - marked known fails as skip for now Closes #10175 - quic: improve connect error message, debugging info, fix false connect report - ECONNECTREFUSED has not its own fail message in quic filters - Debug logging in connect eyballing improved - Fix bug in ngtcp2/quiche that could lead to false success reporting. Reported-by: Divy Le Ray Fixes #10245 Closes #10248 - quiche: fix build without any HTTP/2 implementation Fixes #10260 Closes #10263 Daniel Stenberg (9 Jan 2023) - .github/workflows/linux.yml: add a quiche CI job Move over from zuul Closes #10241 - curl.h: allow up to 10M buffer size Bump the limit from 512K. There might be reasons for applications using h3 to set larger buffers and there is no strong reason for curl to have a very small maximum. Ref: https://curl.se/mail/lib-2023-01/0026.html Closes #10256 Tatsuhiro Tsujikawa (8 Jan 2023) - GHA: use designated ngtcp2 and its dependencies versions Designate ngtcp2 and its dependency versions so that the CI build does not fail without our control. Closes #10257 Daniel Stenberg (8 Jan 2023) - docs/cmdline-opts/hsts.d: explain hsts more Closes #10258 Stefan Eissing (8 Jan 2023) - msh3: run again in its cfilter - test 2500, single GET works - test 2501, single POST stalls - test 2502, multiple, sequential GETs each use a new connection since MsH3ConnectionGetState(qconn) no longer reports CONNECTED after one GET. Closes #10204 Jay Satiro (8 Jan 2023) - sendf: fix build for Linux TCP fastopen - Fix the remote addr struct dereference. - Include cf-socket.h in urldata.h. Follow-up to 6a8d7ef9 which changed conn->ipaddr (Curl_addrinfo* ) member to conn->remote_addr (Curl_sockaddr_ex *) several days ago. Reported-by: Stephan Guilloux Fixes https://github.com/curl/curl/issues/10249 Closes https://github.com/curl/curl/pull/10250 Daniel Stenberg (7 Jan 2023) - RELEASE-NOTES: synced - setopt: move the SHA256 opt within #ifdef libssh2 Because only the libssh2 backend not supports it and thus this should return error if this option is used other backends. Reported-by: Harry Sintonen Closes #10255 Patrick Monnerat (7 Jan 2023) - nss: implement data_pending method NSS currently uses the default Curl_none_data_pending() method which always returns false, causing TLS buffered input data to be missed. The current commit implements the nss_data_pending() method that properly monitors the presence of available TLS data. Ref:#10077 Closes #10225 Jay Satiro (6 Jan 2023) - CURLOPT_HEADERDATA.3: warn DLL users must set write function - Warn that in Windows if libcurl is running from a DLL and if CURLOPT_HEADERDATA is set then CURLOPT_WRITEFUNCTION or CURLOPT_HEADERFUNCTION must be set as well, otherwise the user may experience crashes. We already have a similar warning in CURLOPT_WRITEDATA. Basically, in Windows libcurl could crash writing a FILE pointer that was created by a different C runtime. In Windows each DLL that is part of a program may or may not have its own C runtime. Ref: https://github.com/curl/curl/issues/10231 Closes https://github.com/curl/curl/pull/10233 Jon Rumsey (5 Jan 2023) - x509asn1: fix compile errors and warnings Various small issues when built for GSKit Closes #10238 Patrick Monnerat (5 Jan 2023) - runtests: fix detection of TLS backends Built-in TLS backends are detected at test time by scanning for their names in the version string line returned by the cli tool: as this line may also list the libssh configuration that mentions its own backend, the curl backend may be wrongly determined. In example, if the version line contains "libssh/0.10.4/openssl/zlib", OpenSSL is detected as a curl-configured backend even if not. This fix requires the backend names to appear as full words preceded by spacing in the version line to be recognized as curl TLS backends. Closes #10236 Andy Alt (5 Jan 2023) - GHA: add job on Slackware 15.0 Closes #10230 Daniel Stenberg (5 Jan 2023) - test363: make even smaller writes to loop more - http_proxy: do not assign data->req.p.http use local copy Avoid the tricky reusing of the data->req.p.http pointer for http proxy tunneling. Fixes #10194 Closes #10234 Stefan Eissing (5 Jan 2023) - quic: rename vquic implementations, fix for quiche build. - quiche in debug mode did not build, fixed. - moved all vquic implementation files to prefix curl_* to avoid the potential mixups between provided .h files and our own. - quich passes test 2500 and 2502. 2501, the POST, fail with the body being rejected. Quich bug? Closes #10242 - sectransp: fix for incomplete read/writes SecureTransport expects result code errSSLWouldBlock when the requested length could not be sent/recieved in full. The previous code returned noErr, which let SecureTransport to believe that the IO had terminated prematurely. Fixes #10227 Closes #10235 Andy Alt (5 Jan 2023) - GHA: Hacktoberfest CI: Update deprecated 'set-output' command Closes #10221 Jay Satiro (5 Jan 2023) - scripts: set file mode +x on all perl and shell scripts - Set all scripts +x, ie 644 => 755. Prior to this change some scripts were not executable and therefore could not be called directly. ~~~ git ls-files -s \*.{sh,pl,py} | grep -v 100755 ~~~ Closes https://github.com/curl/curl/pull/10219 Stefan Eissing (4 Jan 2023) - tool_operate: fix headerfile writing Do not rely on the first transfer started to be the first to get a response (remember -Z). All transfers now write the headefile (-D) in append mode, making sure that the order of transfer responses does not lead to overwrites of previous data. Closes #10224 Daniel Stenberg (4 Jan 2023) - misc: reduce struct and struct field sizes - by using BIT() instead of bool - imap: shrink struct - ftp: make state 'unsigned char' - ftp: sort ftp_conn struct entries on size - urldata: use smaller fields for SSL version info storage - pop3: reduce the pop3_conn struct size - smtp: reduce the size of the smtp structs Closes #10186 - noproxy: support for space-separated names is deprecated To be removed in July 2024. Assisted-by: Michael Osipov Fixes #10209 Closes #10215 Andrei Rybak (4 Jan 2023) - lib: fix typos in comments which repeat a word Remove erroneously duplicated words in code comments of files `lib.connect.c` and `lib/url.c`. Closes #10220 Radek Brich (3 Jan 2023) - cmake: set SOVERSION also for macOS Closes #10214 Jay Satiro (3 Jan 2023) - http2: fix compiler warning due to uninitialized variable Prior to this change http2_cfilter_add could return an uninitialized cfilter pointer in an OOM condition. In this case though, the pointer is discarded and not dereferenced so there was no risk of a crash. Stefan Eissing (3 Jan 2023) - cf-socket: keep sockaddr local in the socket filters - copy `struct Curl_addrinfo` on filter setup into context - remove `struct Curl_addrinfoi *` with `struct Curl_sockaddr_ex *` in connectdata that is set and NULLed by the socket filter - this means we have no reference to the resolver info in connectdata or its filters - trigger the CF_CTRL_CONN_INFO_UPDATE event when the complete filter chain reaches connected status - update easy handle connection information on CF_CTRL_DATA_SETUP event. Closes #10213 Daniel Stenberg (3 Jan 2023) - RELEASE-NOTES: synced - runtests: consider warnings fatal and error on them To help us detect and fix warnings in this script easier and faster. Assisted-by: Jakob Hirsch Ref: #10206 Closes #10208 - copyright: update all copyright lines and remove year ranges - they are mostly pointless in all major jurisdictions - many big corporations and projects already don't use them - saves us from pointless churn - git keeps history for us - the year range is kept in COPYING checksrc is updated to allow non-year using copyright statements Closes #10205 - docs/DEPRECATE.md: deprecate gskit Ref: #10163 - This is a niche TLS library, only running on some IBM systems - no regular curl contributors use this backend - no CI builds use or verify this backend - gskit, or the curl adaption for it, lacks many modern TLS features making it an inferior solution - build breakages in this code take weeks or more to get detected - fixing gskit code is mostly done "flying blind" Closes #10201 - Revert "x509asn1: avoid freeing unallocated pointers" This reverts commit 6b19247e794cfdf4ec63c5880d8f4f5485f653ab. Fixes #10163 Closes #10207 - ngtcp2: fix the build without 'sendmsg' Follow-up from 71b7e0161032 Closes #10210 - cmake: check for sendmsg Used by ngtcp2 Closes #10211 Timmy Schierling (2 Jan 2023) - runtest.pl: add expected fourth return value Fixes warning in autobild log: "Use of uninitialized value $HTTP2TLSPORT in substitution iterator at /tests/runtests.pl line 3516" Closes #10206 Daniel Stenberg (2 Jan 2023) - http2: when using printf %.*s, the length arg must be 'int' Detected by Coverity CID 1518341 Closes #10203 - cfilters: check for NULL before using pointer Detected by Coverity CID 1518343 Closes #10202 - http2: in connisdead check, attach the connection before reading Otherwise data->conn is NULL and things go wrong. This problem caused occastional failures in test 359, 1700 and more depending on timing and the alignment of various planets. Assisted-by: Stefan Eissing Closes #10199 Philip Heiduck (2 Jan 2023) - Linux CI: update some dependecies to latest tag Closes #10195 Daniel Stenberg (2 Jan 2023) - c-hyper: move down the Accept-Encoding header generation To match the internal HTTP request header order so that test 1277 works again. Closes #10200 - release-notes.pl: check fixes/closes lines better To better skip lines that just happen to mention those words at the start of a line without being instructions. - test1560: use a UTF8-using locale when run There are odd cases that don't use UTF8 and then the IDN handling goes wrong. Reported-by: Marcel Raad Fixes #10193 Closes #10196 - cf-socket: fix build regression Reported-by: Stephan Guilloux Fixes #10190 Closes #10191 - examples: remove the curlgtk.c example - it does not add a lot of value - we do not test-build it to verify because of its dependencies - unclear for what GTK versions it works or not Reported-by: odek86 on github Fixes #10197 Closes #10198 Andy Alt (2 Jan 2023) - docs: add link to GitHub Discussions Closes #10171 - GHA: ignore changes to md files for most workflows Closes #10176 Josh Brobst (2 Jan 2023) - http: decode transfer encoding first The unencoding stack is added to as Transfer-Encoding and Content-Encoding fields are encountered with no distinction between the two, meaning the stack will be incorrect if, e.g., the message has both fields and a non-chunked Transfer-Encoding comes first. This commit fixes this by ordering the stack with transfer encodings first. Reviewed-by: Patrick Monnerat Closes #10187 Daniel Stenberg (1 Jan 2023) - curl.h: mark CURLSSLBACKEND_MESALINK as deprecated Follow-up since 223f26c28a340b36 Deprecated since 7.82.0 Closes #10189 - curl_global_sslset.3: clarify the openssl situation and add rustls Closes #10188 Cameron Blomquist (1 Jan 2023) - http: add additional condition for including stdint.h stdint.h was only included in http.h when ENABLE_QUIC was defined, but symbols from stdint.h are also used when USE_NGHTTP2 is defined. This causes build errors when USE_NGHTTP2 is defined but ENABLE_QUIC is not. Closes #10185 Daniel Stenberg (31 Dec 2022) - urldata: cease storing TLS auth type The only TLS auth type libcurl ever supported is SRP and that is the default type. Since nobody ever sets any other type, there is no point in wasting space to store the set type and code to check the type. If TLS auth is used, SRP is now implied. Closes #10181 - vtls: use ALPN HTTP/1.0 when HTTP/1.0 is used Previously libcurl would use the HTTP/1.1 ALPN id even when the application specified HTTP/1.0. Reported-by: William Tang Ref: #10183 Marcel Raad (30 Dec 2022) - lib670: make test.h the first include As in all other lib tests. This avoids a macro redefinition warning for `_FILE_OFFSET_BITS` visible in the autobuilds. Closes https://github.com/curl/curl/pull/10182 Stefan Eissing (30 Dec 2022) - lib: connect/h2/h3 refactor Refactoring of connection setup and happy eyeballing. Move nghttp2. ngtcp2, quiche and msh3 into connection filters. - eyeballing cfilter that uses sub-filters for performing parallel connects - socket cfilter for all transport types, including QUIC - QUIC implementations in cfilter, can now participate in eyeballing - connection setup is more dynamic in order to adapt to what filter did really connect. Relevant to see if a SSL filter needs to be added or if SSL has already been provided - HTTP/3 test cases similar to HTTP/2 - multiuse of parallel transfers for HTTP/3, tested for ngtcp2 and quiche - Fix for data attach/detach in VTLS filters that could lead to crashes during parallel transfers. - Eliminating setup() methods in cfilters, no longer needed. - Improving Curl_conn_is_alive() to replace Curl_connalive() and integrated ssl alive checks into cfilter. - Adding CF_CNTRL_CONN_INFO_UPDATE to tell filters to update connection into and persist it at the easy handle. - Several more cfilter related cleanups and moves: - stream_weigth and dependency info is now wrapped in struct Curl_data_priority - Curl_data_priority members depend is available in HTTP2|HTTP3 - Curl_data_priority members depend on NGHTTP2 support - handling init/reset/cleanup of priority part of url.c - data->state.priority same struct, but shallow copy for compares only - PROTOPT_STREAM has been removed - Curl_conn_is_mulitplex() now available to check on capability - Adding query method to connection filters. - ngtcp2+quiche: implementing query for max concurrent transfers. - Adding is_alive and keep_alive cfilter methods. Adding DATA_SETUP event. - setting keepalive timestamp on connect - DATA_SETUP is called after the connection has been completely setup (but may not connected yet) to allow filters to initialize data members they use. - there is no socket to be had with msh3, it is unclear how select shall work - manual test via "curl --http3 https://curl.se" fail with "empty reply from server". - Various socket/conn related cleanups: - Curl_socket is now Curl_socket_open and in cf-socket.c - Curl_closesocket is now Curl_socket_close and in cf-socket.c - Curl_ssl_use has been replaced with Cur_conn_is_ssl - Curl_conn_tcp_accepted_set has been split into Curl_conn_tcp_listen_set and Curl_conn_tcp_accepted_set with a clearer purpose Closes #10141 Daniel Stenberg (30 Dec 2022) - RELEASE-NOTES: synced - docs/libcurl/curl_getdate.3: minor whitespace edit To avoid a fccp quirk that made it render wrongly on the website - transfer: break the read loop when RECV is cleared When the RECV bit is cleared because the response reading for this transfer is complete, the read loop should be stopped. data_pending() can otherwise still return TRUE and another read would be attempted. Reported-by: Hide Ishikawa Fixes #10172 Closes #10174 - multihandle: turn bool struct fields into bits Closes #10179 Stefan Eissing (30 Dec 2022) - ftpserver: lower the normal DATA connect timeout to speed up torture tests - tests/ftpserver.pl blocks when expecting a DATA connection from the client. - the previous 10 seconds were encountered repeatedly in torture tests and let to long waits. - 2 seconds should still be sufficient for current hw, but CI will show. Closes #10178 Nick Banks (28 Dec 2022) - msh3: add support for request payload Closes #10136 Stefan Eissing (28 Dec 2022) - openssl: remove attached easy handles from SSL instances - keeping the "current" easy handle registered at SSL* is no longer necessary, since the "calling" data object is already stored in the cfilter's context (and used by other SSL backends from there). - The "detach" of an easy handle that goes out of scope is then avoided. - using SSL_set0_wbio for clear reference counting where available. Closes #10151 Daniel Stenberg (28 Dec 2022) - socketpair: allow localhost MITM sniffers Windows allow programs to MITM connections to localhost. The previous check here would detect that and error out. This new method writes data to verify the pipe thus allowing MITM. Reported-by: SerusDev on github Fixes #10144 Closes #10169 - HTTP3: mention what needs to be in place to remove EXPERIMENTAL label Closes #10168 Andy Alt (28 Dec 2022) - MANUAL.md: add pipe to apt-key example Closes #10170 Daniel Stenberg (27 Dec 2022) - test417: verify %{certs} output - runtests: make 'mbedtls' a testable feature Also add to FILEFORMAT.md - writeout: add %{certs} and %{num_certs} Let users get the server certificate chain using the command line Closes #10019 Stefan Eissing (27 Dec 2022) - haxproxy: send before TLS handhshake - reverse order of haproxy and final ssl cfilter - make haproxy avaiable on PROXY builds, independent of HTTP support as it can be used with any protocol. Reported-by: Sergio-IME on github Fixes #10165 Closes #10167 Daniel Stenberg (27 Dec 2022) - RELEASE-NOTES: synced - test446: verify hsts with two URLs - runtests: support crlf="yes" for verify/proxy - hsts: handle adding the same host name again It will then use the largest expire time of the two entries. - tool_operate: share HSTS between handles - share: add sharing of HSTS cache among handles Closes #10138 Viktor Szakats (27 Dec 2022) - Makefile.mk: fix wolfssl and mbedtls default paths Fix the defaults for `WOLFSSL_PATH` and `MBEDTLS_PATH` to have meaningful values instead of the copy-pasted wrong ones. Ref: https://github.com/curl/curl/commit/66e68ca47f7fd00dff2cb7c45ba6725d4009 9585#r94275172 Reported-by: Ryan Schmidt Closes #10164 Daniel Stenberg (27 Dec 2022) - INTERNALS: cleanup - remove "operating systems" (mostly outdated) - upodate the "build tools" Closes #10162 - cmake: bump requirement to 3.7 Because this is the cmake version (released in November 2016) that introduced GREATER_EQUAL, which is used already. Reported-by: nick-telia on github Fixes #10128 Closes #10161 - cfilters:Curl_conn_get_select_socks: use the first non-connected filter When there are filters addded for both socket and SSL, the code previously checked the SSL sockets during connect when it *should* first check the socket layer until that has connected. Fixes #10157 Fixes #10146 Closes #10160 Reviewed-by: Stefan Eissing - urlapi: add CURLU_PUNYCODE Allows curl_url_get() get the punycode version of host names for the host name and URL parts. Extend test 1560 to verify. Closes #10109 - RELEASE-NOTES: synced - libssh2: try sha2 algos for hostkey methods As is supported by recent libssh2, but should just be ignored by older versions. Reported-by: norbertmm on github Assisted-by: norbertmm on github Fixes #10143 Closes #10145 Patrick Monnerat (26 Dec 2022) - typecheck: accept expressions for option/info parameters As expressions can have side effects, evaluate only once. To enable deprecation reporting only once, get rid of the __typeof__ use to define the local temporary variable and use the target type (CURLoption/CURLINFO). This also avoids multiple reports on type conflicts (if some) by the curlcheck_* macros. Note that CURLOPT_* and CURLINFO_* symbols may be deprecated, but not their values: a curl_easy_setopt call with an integer constant as option will never report a deprecation. Reported-by: Thomas Klausner Fixes #10148 Closes #10149 Paul Howarth (26 Dec 2022) - tests: avoid use of sha1 in certificates The SHA-1 algorithm is deprecated (particularly for security-sensitive applications) in a variety of OS environments. This already affects RHEL-9 and derivatives, which are not willing to use certificates using that algorithm. The fix is to use sha256 instead, which is already used for most of the other certificates in the test suite. Fixes #10135 This gets rid of issues related to sha1 signatures. Manual steps after "make clean-certs" and "make build-certs": - Copy tests/certs/stunnel-sv.pem to tests/stunnel.pem (make clean-certs does not remove the original tests/stunnel.pem) - Copy tests/certs/Server-localhost-sv.pubkey-pinned into --pinnedpubkey options of tests/data/test2041 and tests/data/test2087 Closes #10153 Yurii Rashkovskii (26 Dec 2022) - cmake: fix the snprintf detection I haven't had the time to check other configurations, but on my macOS Ventura 13.1 with XCode 14.2 cmake does not find `snprintf`. Solution: ensure stdio.h is checked for definitions Closes #10155 Radu Hociung (26 Dec 2022) - http: remove the trace message "Mark bundle... multiuse" The message "Mark bundle as not supporting multiuse" was added at commit 29364d93 when an http/2-related bug was fixed, and it appears to be a leftover trace message. This message should be removed because: * it conveys no information to the user * it is enabled in the default build (--enable-verbose) * it reads like a warning/unexpected condition * it is equivalent to "Detected http proto < 2", which is not a useful message. * it is a time-wasting red-herring for anyone who encounters it for the first time while investigating some other, real problem. This commit removes the trace message "Mark bundle as not supporting multiuse" Closes #10159 Hannah Schierling (26 Dec 2022) - url: fix build with `--disable-cookies` Struct `UserDefined` has no member `cookielist` if `CURL_DISABLE_COOKIES` is defined. Follow-up to af5999a Closes #10158 Stefan Eissing (23 Dec 2022) - runtests: also tear down http2/http3 servers when https server is stopped Closes #10114 - tests: add 3 new HTTP/2 test cases, plus https: support for nghttpx - a simple https get - a simple https post - a multi get of 4 requests and check that same connection was used Closes #10114 Daniel Stenberg (23 Dec 2022) - urldata: remove unused struct fields, made more conditional - source_quote, source_prequote and source_postquote have not been used since 5e0d9aea3; September 2006 - make several fields conditional on proxy support - make three quote struct fields conditional on FTP || SSH - make 'mime_options' depend on MIME - make trailer_* fields depend on HTTP - change 'gssapi_delegation' from long to unsigned char - make 'localportrange' unsigned short instead of int - conn->trailer now depends on HTTP Closes #10147 - urldata: make set.http200aliases conditional on HTTP being present And make a few SSH-only fields depend on SSH Closes #10140 - md4: fix build with GnuTLS + OpenSSL v1 Reported-by: Esdras de Morais da Silva Fixes #10110 Closes #10142 - urldata: make 'ftp_create_missing_dirs' depend on FTP || SFTP Closes #10139 John Bampton (22 Dec 2022) - misc: fix grammar and spelling Closes #10137 Daniel Stenberg (22 Dec 2022) - urldata: move the cookefilelist to the 'set' struct The cookiefile entries are set into the handle and should remain set for the lifetime of the handle so that duplicating it also duplicates the list. Therefore, the struct field is moved from 'state' to 'set'. Fixes #10133 Closes #10134 - strdup: name it Curl_strdup It does not belong in the curlx_ name space as it is never used externally. Closes #10132 Nick Banks (22 Dec 2022) - msh3: update to v0.5 Release Closes #10125 Andy Alt (22 Dec 2022) - workflows/linux.yml: merge 3 common packages Closes #10071 Daniel Stenberg (21 Dec 2022) - docs: mention indirect effects of --insecure Warn users that disabling certficate verification allows servers to "pollute" curl with data it trusts. Reported-by: Harry Sintonen Closes #10126 - SECURITY-PROCESS.md: document severity levels Closes #10118 - RELEASE_NOTES: synced bumped version for new cycle Marcel Raad (21 Dec 2022) - tool_operate: fix `CURLOPT_SOCKS5_GSSAPI_NEC` type `CURLOPT_SOCKS5_GSSAPI_NEC` is a long, while `socks5_gssapi_nec` was made a bool in commit 4ac64eadf60. Closes https://github.com/curl/curl/pull/10124 Version 7.87.0 (21 Dec 2022) Daniel Stenberg (21 Dec 2022) - RELEASE-NOTES: synced The curl 7.87.0 release - THANKS: 40 new contributors from 7.87.0 - http: fix the ::1 comparison for IPv6 localhost for cookies When checking if there is a "secure context", which it is if the connection is to localhost even if the protocol is HTTP, the comparison for ::1 was done incorrectly and included brackets. Reported-by: BratSinot on github Fixes #10120 Closes #10121 Philip Heiduck (19 Dec 2022) - CI/spell: actions/checkout@v2 > actions/checkout@v3 Daniel Stenberg (19 Dec 2022) - smb/telnet: do not free the protocol struct in *_done() It is managed by the generic layer. Reported-by: Trail of Bits Closes #10112 - http: use the IDN decoded name in HSTS checks Otherwise it stores the info HSTS into the persistent cache for the IDN name which will not match when the HSTS status is later checked for using the decoded name. Reported-by: Hiroki Kurosawa Closes #10111 - CURLOPT_DEBUGFUNCTION.3: emphasize that incoming data is "raw" Closes #10106 Xì Gà (16 Dec 2022) - socks: fix username max size is 255 (0xFF) Closes #10105 Reviewed-by: Daniel Gustafsson Daniel Stenberg (16 Dec 2022) - limit-rate.d: see also --rate - lib1560: add some basic IDN host name tests Closes #10094 - idn: rename the files to idn.[ch] and hold all IDN functions Closes #10094 - idn: remove Curl_win32_ascii_to_idn It was not used. Introduce a new IDN header for the prototype(s). Closes #10094 - RELEASE-NOTES: synced - curl_url_get.3: remove spurious backtick Put there by mistake. Follow-up from 9a8564a92 Closes #10101 - socks: fix infof() flag for outputing a char It used to be a 'long', %lu is no longer correct. Follow-up to 57d2d9b6bed33d Detected by Coverity CID 1517663 Closes #10100 - ssl-reqd.d: clarify that this is for upgrading connections only Closes #10093 - curl_url_set.3: document CURLU_DISALLOW_USER Closes #10099 - cmake: set the soname on the shared library Set SONAME and VERSION for platforms we think this works on. Remove issue from KNOWN_BUGS. Assisted-by: Jakub Zakrzewski Closes #10023 - tool_paramhlp: free the proto strings on exit And also make sure that repeated use of the options free the previous string before it stores a new. Follow-up from e6f8445edef8e7996d Closes #10098 - tool_cfgable: free the ssl_ec_curves on exit Follow-up to ede125b7b Closes #10097 - urlapi: reject more bad letters from the host name: &+() Follow-up from eb0167ff7d31d3a5 Extend test 1560 to verify Closes #10096 - altsvc: fix rejection of negative port numbers Follow-up to ac612dfeee95 strtoul() accepts a leading minus so better make sure there is none Extended test 356 somewhat to use a huge negative 64 bit number that otherwise becomes a low positive number. Closes #10095 - lib: use size_t or int etc instead of longs Since long is not using a consistent data size in curl builds, making it often "waste" 32 bits. Closes #10088 - azure: use "unversioned" clang and clang-tools for scanbuild job To make it less fragile Closes #10092 Daniel Gustafsson (14 Dec 2022) - x509asn1: avoid freeing unallocated pointers When utf8asn1str fails there is no allocation returned, so freeing the return pointer in **to is at best a no-op and at worst a double- free bug waiting to happen. The current coding isn't hiding any such bugs but to future proof, avoid freeing the return value pointer iff the function failed. Closes: #10087 Reviewed-by: Daniel Stenberg Emil Engler (13 Dec 2022) - curl_url_set.3: fix typo Closes: #10089 Reviewed-by: Daniel Gustafsson Daniel Stenberg (13 Dec 2022) - test2304: verify websocket handling when connection is closed - server/sws: if asked to close connection, skip the websocket handling - ws: if no connection is around, return error - curl_ws_send returns CURLE_SEND_ERROR if data->conn is gone - curl_ws_recv returns CURLE_GOT_NOTHING on connection close - curl_ws_recv.3: mention new return code for connection close + example embryo Closes #10084 Emil Engler (13 Dec 2022) - docs: extend the dump-header documentation This commit extends the documentation of the --dump-header command-line option to reflect the behavior introduced in 8b1e5df7. See #10079 Closes #10085 Daniel Stenberg (12 Dec 2022) - RELEASE-NOTES: synced - styled-output.d: this option does not work on Windows Reported-by: u20221022 on github Fixes #10082 Closes #10083 Emil Engler (12 Dec 2022) - tool: determine the correct fopen option for -D This commit fixes a bug in the dump-header feature regarding the determination of the second fopen(3) option. Reported-by: u20221022 on github See #4753 See #4762 Fixes #10074 Closes #10079 Christian Schmitz (11 Dec 2022) - docs/curl_ws_send: Fixed typo in websocket docs Replace as with is in relevant sentences. Closes: #10081 Reviewed-by: Daniel Gustafsson Prithvi MK (11 Dec 2022) - c-hyper: fix multi-request mechanism It makes test 565 run fine. Fixes #8896 Closes #10080 Assisted-by: Daniel Stenberg Andy Alt (11 Dec 2022) - page-header: grammar improvement (display transfer rate) Closes #10068 - docs/DEPRECATE.md: grammar improvement and sp correction The main thing I wanted to do was fix the spelling of "spent", but I think this rewording improves the flow of the paragraph. Closes #10067 Boris Verkhovskiy (11 Dec 2022) - tool_cfgable: make socks5_gssapi_nec a boolean Closes #10078 Frank Gevaerts (9 Dec 2022) - contributors.sh: actually use $CURLWWW instead of just setting it. The script was all set up for flexibility where curl-www is elsewhere in the filesystem, but then hard-coded ../curl-www anyway... Closes #10064 Daniel Stenberg (9 Dec 2022) - KNOWN_BUGS: remove items not considered bugs any more - CURL_GLOBAL_SSL This option was changed in libcurl 7.57.0 and clearly it has not caused too many issues and a lot of time has passed. - Store TLS context per transfer instead of per connection This is a possible future optimization. One that is much less important and interesting since the added support for CA caching. - Microsoft telnet server This bug was filed in May 2007 against curl 7.16.1 and we have not received further reports. - active FTP over a SOCKS Actually, proxies in general is not working with active FTP mode. This is now added in proxy documentation. - DICT responses show the underlying protocol curl still does this, but since this is now an established behavior since forever we cannot change it easily and adding an option for it seems crazy as this protocol is not so little its not worth it. Let's just live with it. - Secure Transport disabling hostname validation also disables SNI This is an already documented restriction in Secure Transport. - CURLOPT_SEEKFUNCTION not called with CURLFORM_STREAM The curl_formadd() function is marked and documented as deprecated. No point in collecting bugs for it. It should not be used further. - STARTTRANSFER time is wrong for HTTP POSTs After close source code inspection I cannot see how this is true or that there is any special treatment for different HTTP methods. We also have not received many further reports on this, making me strongly suspect that this is no (longer an) issue. - multipart formposts file name encoding The once proposed RFC 5987-encoding is since RFC 7578 documented as MUST NOT be used. The since then implemented MIME API allows the user to set the name on their own and can thus provide it encoded as it wants. - DoH is not used for all name resolves when enabled It is questionable if users actually want to use DoH for interface and FTP port name resolving. This restriction is now documented and we advice users against using name resolving at all for these functions. Closes #10043 - CURLOPT_COOKIEFILE.3: advice => advise Closes #10063 Reviewed-by: Daniel Gustafsson Daniel Gustafsson (9 Dec 2022) - curl.h: reword comment to not use deprecated option CURLOPT_INFILE was replaced by CURLOPT_READDATA in 7.9.7, reword the comment mentioning it to make code grepping easier as well as improve the documentation. Closes: #10062 Reviewed-by: Daniel Stenberg Ryan Schmidt (9 Dec 2022) - system.h: fix socklen_t, curl_off_t, long long for Classic Mac OS Change "__MWERKS__" to "macintosh". When this block was originally added in 3ac6929 it was probably intended to handle classic Mac OS since the previous classic Mac OS build procedure for curl (which was removed in bf327a9) used Metrowerks CodeWarrior. But there are other classic Mac OS compilers, such as the MPW compilers, that were not handled by this case. For classic Mac OS, CURL_TYPEOF_CURL_SOCKLEN_T needs to match what's provided by the third-party GUSI library, which does not vary by compiler. Meanwhile CodeWarrior works on platforms other than classic Mac OS, and they may need different definitions. Separate blocks could be added later for any of those platforms that curl doesn't already support. Closes #10049 - vms: remove SIZEOF_SHORT The rest of SIZEOF_SHORT was removed in d48dd15. See #9291 Closes #10061 Daniel Gustafsson (8 Dec 2022) - tool_formparse: avoid clobbering on function params While perfectly legal to do, clobbering function parameters and using them as local variables is confusing at best and rarely improves code readability. Fix by using a local variable instead, no functionality is changed. This also renames the parameter from data to mime_data since the term data is (soft) reserved for the easy handle struct. Closes: #10046 Reviewed-by: Daniel Stenberg - noproxy: guard against empty hostnames in noproxy check When checking for a noproxy setting we need to ensure that we get a hostname passed in. If there is no hostname then there cannot be a matching noproxy rule for it by definition. Closes: #10057 Reported-by: Geeknik Labs Reviewed-by: Daniel Stenberg Daniel Stenberg (8 Dec 2022) - c-hyper: CONNECT respones are not server responses Together with d31915a8dbbd it makes test 265 run fine. Fixes #8853 Assisted-by: Prithvi MK Assisted-by: Sean McArthur Closes #10060 - test265: Use "connection: keep-alive" response header When it answers as HTTP/1.0, so that clients (hyper) knows properly that the connection remains intact. - RELEASE-NOTES: synced Stefan Eissing (8 Dec 2022) - cfilter: improve SSL connection checks - fixes `Curl_ssl_cf_get_ssl()` to detect also the first filter instance as ssl (refs #10053) - replaces `Curl_ssl_use()` with the correct `Curl_conn_is_ssl()` Closes #10054 Fixes #10053 Reported-by: Patrick Monnerat Daniel Stenberg (8 Dec 2022) - runtests: silence nghttpx errors Also, move the output of the nghttpx_h3 info to the general "Env:" line in the test output header. Reported-by: Marcel Raad Ref: https://github.com/curl/curl/commit/ca15b7512e8d1199e55fbaa206ef01e64b8f 147d#commitcomment-92015094 Closes #10044 Ryan Schmidt (7 Dec 2022) - config-mac: define HAVE_SYS_IOCTL_H This is needed to compile nonblock.c on classic Mac OS with Grand Unified Socket Interface (GUSI) because nonblock.c uses FIONBIO which is defined in which is included by . Ref: https://sourceforge.net/projects/gusi/ Closes https://github.com/curl/curl/pull/10042 Philip Heiduck (7 Dec 2022) - CI: Change FreeBSD image from 12.3 to 12.4 Ref: https://www.phoronix.com/news/FreeBSD-12.4-Released Closes https://github.com/curl/curl/pull/10051 Ryan Schmidt (7 Dec 2022) - test1421: fix typo Closes https://github.com/curl/curl/pull/10055 Jay Satiro (7 Dec 2022) - build: assume errno.h is always available - Remove errno.h detection from all build configurations. errno.h is a standard header according to C89. Closes https://github.com/curl/curl/pull/9986 - build: assume assert.h is always available - Remove assert.h detection from all build configurations. assert.h is a standard header according to C89. I had proposed this several years ago as part of a larger change that was abandoned. Ref: https://github.com/curl/curl/issues/1237#issuecomment-277500720 Closes https://github.com/curl/curl/pull/9985 Philip Heiduck (7 Dec 2022) - CI: LGTM.com will be shut down in December 2022 Closes #10052 Daniel Stenberg (6 Dec 2022) - mailmap: Andy Alt Andy Alt (6 Dec 2022) - misc: Fix incorrect spelling Fix various uses of connnect by replacing them with connect. Closes: #10045 Reviewed-by: Daniel Stenberg Reviewed-by: Daniel Gustafsson Stefan Eissing (6 Dec 2022) - wolfssl: remove special BIO return code handling - rely solely on the retry flag in BIO, similar to OpenSSL vtls implementation. Ref: https://github.com/curl/curl/pull/10021#issuecomment-1336147053 Closes #10033 Daniel Stenberg (6 Dec 2022) - openssl: return -1 on error in the BIO callbacks BIO_read and BIO_write return negative numbers on error, including retryable ones. A regression from 55807e6. Both branches should be returning -1. The APIs are patterned after POSIX read and write which, similarly, return -1 on errors, not zero, with EAGAIN treated as an error. Bug: https://github.com/curl/curl/issues/10013#issuecomment-1335308146 Reported-by: David Benjamin Closes #10021 Ryan Schmidt (6 Dec 2022) - config-mac: remove HAVE_SYS_SELECT_H When compiling for classic Mac OS with GUSI, there is no sys/select.h. GUSI provides the "select" function prototype in sys/time.h. Closes #10039 - setup: do not require __MRC__ defined for Mac OS 9 builds Partially reverts "somewhat protect Mac OS X users from using Mac OS 9 config file", commit 62519bfe059251af2914199f284c736553ff0489. Do things that are specific to classic Mac OS (i.e. include config-mac.h in curl_setup.h and rename "main" to "curl_main" in tool_setup.h) when only "macintosh" is defined. Remove the additional condition that "__MRC__" should be defined since that would only be true with the MPW MrC compiler which prevents the use of other reasonable compilers like the MPW SC compiler and especially the Metrowerks CodeWarrior compilers. "macintosh" is only defined by classic Mac OS compilers so this change should not affect users of Mac OS X / OS X / macOS / any other OS. Closes #10037 - curl.h: name all public function parameters Most public function parameters already have names; this adds those that were missing. Closes #10036 Andy Alt (6 Dec 2022) - docs/examples: spell correction ('Retrieve') Closes #10040 Daniel Stenberg (6 Dec 2022) - unit1302: slightly extended To test more base64 decoding - base64: faster base64 decoding - by using a lookup table instead of strchr() - by doing full quantums first, then padding Closes #10032 Michael Musset (6 Dec 2022) - libssh2: return error when ssh_hostkeyfunc returns error return CURLE_PEER_FAILED_VERIFICATION if verification with the callback return a result different than CURLKHMATCH_OK Closes #10034 Viktor Szakats (5 Dec 2022) - Makefile.mk: improve a GNU Make hack [ci skip] Replace the hack of using `$() ` to represent a single space. The new method silences the `--warn-undefined-variables` debug warning and it's also a better-known form of solving this problem. Reviewed-by: Jay Satiro Closes #10031 Daniel Stenberg (5 Dec 2022) - tests/unit/.gitignore: ignore all unit + 4 digits files - base64: encode without using snprintf For speed. In some tests, this approch is 29 times faster! Closes #10026 - base64: better alloc size The previous algorithm allocated more bytes than necessary. Suggested-by: xtonik on github Fixes #10024 Closes #10025 Ryan Schmidt (5 Dec 2022) - config-mac: fix typo: size_T -> size_t Both MPW and CodeWarrior compilers complained about this. Closes #10029 Daniel Stenberg (3 Dec 2022) - RELEASE-NOTES: synced Jakub Zakrzewski (2 Dec 2022) - CMake: fix build with `CURL_USE_GSSAPI` CMAKE_*_LINKER_FLAGS must be a string but GSS_LINKER_FLAGS is a list, so we need to replace semicolons with spaces when setting those. Fixes #9017 Closes #1022 Max Dymond (2 Dec 2022) - ci: Reuse fuzzing snippet from curl-fuzzer project Diogo Teles Sant'Anna (2 Dec 2022) - GHA: clarify workflows permissions, set least possible privilege Set top-level permissions to None on all workflows, setting per-job permissions. This avoids that new jobs inherit unwanted permissions. Discussion: https://curl.se/mail/lib-2022-11/0028.html Signed-off-by: Diogo Teles Sant'Anna Closes #9928 Viktor Szakats (2 Dec 2022) - Makefile.mk: address minor issues - Fix `NROFF` auto-detection with certain shell/make-build combinations: When a non-MSYS2 GNU Make runs inside an MSYS2 shell, Make executes the detection command as-is via `CreateProcess()`. It fails because `command` is an `sh` built-in. Ensure to explicitly invoke the shell. - Initialize user-customizable variables: Silences a list of warnings when running GNU Make with the option `--warn-undefined-variables`. Another benefit is that it's now easy to look up all user-customizable `Makefile.mk` variables by grepping for ` ?=` in the curl source tree. Suggested-by: Gisle Vanem Ref: https://github.com/curl/curl/pull/9764#issuecomment-1330674433 - Fix `MKDIR` invocation: Avoid a warning and potential issue in envs without forward-slash support. Closes #10000 Rob de Wit (2 Dec 2022) - curl_get_line: allow last line without newline char improve backwards compatibility Test 3200 verifies Closes #9973 Daniel Stenberg (2 Dec 2022) - cookie: open cookie jar as a binary file On Windows there is a difference and for text files, ^Z means end of file which is not desirable. Ref: #9973 Closes #10017 - runtests: only do CRLF replacements for hyper if it is HTTP Closes #10016 Stefan Eissing (1 Dec 2022) - openssl: fix for BoringSSL BIO result interpretation mixups Reported-by: Robin Marx Fixes #10013 Closes #10015 Max Dymond (1 Dec 2022) - ci: Remove zuul fuzzing job as it's superseded by CIFuzz Daniel Stenberg (1 Dec 2022) - runtests: do CRLF replacements per section only The `crlf="yes"` attribute and "hyper mode" are now only applied on a subset of dedicated sections: data, datacheck, stdout and protocol. Updated test 2500 accordingly. Also made test1 use crlf="yes" for , mostly because it is often used as a template test case. Going forward, using this attribute we should be able to write test cases using linefeeds only and avoid mixed line ending encodings. Follow-up to ca15b7512e8d11 Fixes #10009 Closes #10010 Stefan Eissing (1 Dec 2022) - gnutls: use common gnutls init and verify code for ngtcp2 Closes #10007 Baitinq on github (1 Dec 2022) - aws_sigv4: fix typos in aws_sigv4.c Closes #10008 Kenneth Myhra (30 Nov 2022) - curl.h: include on SerenityOS Closes #10006 Daniel Stenberg (30 Nov 2022) - openssl: prefix errors with '[lib]/[version]: ' To help users understand where this (cryptic) error message comes from. Suggested-by: Philip Sanetra Ref: #10002 Closes #10004 Stefan Eissing (30 Nov 2022) - tests: add HTTP/3 test case, custom location for proper nghttpx - adding support for HTTP/3 test cases via a nghttpx server that is build with ngtcp2 and nghttp3. - test2500 is the first test case, performing a simple GET. - nghttpx is checked for support and the 'feature' nghttpx-h3 is set accordingly. test2500 will only run, when supported. - a specific nghttpx location can be given in the environment variable NGHTTPX or via the configure option --with-test-nghttpx= Extend NGHTTPX config to H2 tests as well * use $ENV{NGHTTPX} and the configured default also in http2 server starts * always provide the empty test/nghttpx.conf to nghttpx. as it defaults to reading /etc/nghttpx/nghttpx.conf otherwise. Added nghttpx to CI ngtcp2 jobs to run h3 tests. Closes #9031 Daniel Stenberg (30 Nov 2022) - RELEASE-NOTES: synced Removed duplicate after contributors.sh fix: 9967c10b6daa1 - scripts/contributors.sh: strip one OR MORE leading spaces From names found credited in commit logs - RELEASE-NOTES: synced - openssl/mbedtls: use %d for outputing port with failf (int) Coverity CID 1517100 Also, remove some int typecasts in vtls.c for the port number Closes #10001 - KNOWN_BUGS: remove "Multi perform hangs waiting for threaded resolver" We now offer a way to avoid that hang, using CURLOPT_QUICK_EXIT. Follow-up to 49798cac832ab1 fixed via #9147 Closes #9999 - KNOWN_BUGS: remove "--interface for ipv6 binds to unusable IP address" Since years back the "if2ip" function verifies that it binds to a local IPv6 address that uses the same scope as the remote address. This is not a bug. Fixes #686 Closes #9998 - test1276: verify lib/optiontable.pl Checks that it generates an output identical to the file. - lib/optiontable.pl: adapt to CURLOPTDEPRECATED() Follow-up from 6967571bf20624bc Reported-by: Gisle Vanem Fixes #9992 Closes #9993 - docs/INSTALL.md: list OSes and CPUs quoted to make them skip spellcheck. Also added a new CPU. Follow-up to 4506cbf7f24a2a Closes #9997 Ikko Ashimine (28 Nov 2022) - vtls: fix typo in vtls_int.h paramter -> parameter Closes: #9996 Reviewed-by: Daniel Gustafsson Daniel Stenberg (28 Nov 2022) - curl-openssl.m4: do not add $prefix/include/openssl to CPPFLAGS As OpenSSL's include files are all included using in curl source code, we just risk that existing openssl files will "shadow" include files without path if that path is provided. Fixes #9989 Closes #9988 - INSTALL: update operating systems and CPU archs Update after recent runs on Twitter/Mastodon and my blog Closes #9994 Stefan Eissing (28 Nov 2022) - tls: backends use connection filters for IO, enabling HTTPS-proxy - OpenSSL (and compatible) - BearSSL - gnutls - mbedtls - rustls - schannel - secure-transport - wolfSSL (v5.0.0 and newer) This leaves only the following without HTTPS-proxy support: - gskit - nss - wolfSSL (versions earlier than v5.0.0) Closes #9962 Daniel Stenberg (28 Nov 2022) - include/curl/curl.h: bump the deprecated requirements to gcc 6.1 Reported-by: Michael Kaufmann Fixes #9917 Closes #9987 Patrick Monnerat (28 Nov 2022) - mime: relax easy/mime structures binding Deprecation and removal of codeset conversion support from the library have released the strict need for an early binding of mime structures to an easy handle (https://github.com/curl/curl/commit/2610142). This constraint currently forces to create the handle before the mime structure and the latter cannot be attached to another handle once created (see https://curl.se/mail/lib-2022-08/0027.html). This commit removes the handle pointers from the mime structures allowing more flexibility on their use. When an easy handle is duplicated, bound mime structures must however still be duplicated too as their components hold send-time dynamic information. Closes #9927 fractal-access (26 Nov 2022) - test416: verify growing FTP file support Added setting: RETRSIZE [size] in the section. When set this will cause the test FTP server to return the size set (rather than the actual size) in the acknowledgement from a RETR request. Closes #9772 - ftp: support growing files with CURLOPT_IGNORE_CONTENT_LENGTH When using the option CURLOPT_IGNORE_CONTENT_LENGTH (set.ignorecl in code) to support growing files in FTP, the code should ignore the initial size it gets from the server as this will not be the final size of the file. This is done in ftp_state_quote() to prevent a size request being issued in the initial sequence. However, in a later call to ftp_state_get_resp() the code attempts to get the size of the content again if it doesn't already have it, by parsing the response from the RETR request. This fix prevents this parsing of the response to get the size when the set.ignorecl option is set. This should maintain the size value as -1, unknown, in this situation. Closes #9772 Stefan Eissing (26 Nov 2022) - cfilter: re-add `conn` as parameter to cfilter setup methods - `Curl_ssl_get_config()` now returns the first config if no SSL proxy filter is active - socket filter starts connection only on first invocation of its connect method Fixes #9982 Closes #9983 Daniel Stenberg (26 Nov 2022) - KNOWN_BUGS: remove five FTP related issues - "FTP with CONNECT and slow server" I believe this is not a problem these days. - "FTP with NULs in URL parts" The FTP protocol does not support them properly anyway. - remove "FTP and empty path parts in the URL" I don't think this has ever been reported as a real problem but was only a hypothetical one. - "Premature transfer end but healthy control channel" This is not a bug, this is an optimization that *could* be performed but is not an actual problem. - "FTP without or slow 220 response" Instead add to the documentation of the connect timeout that the connection is considered complete at TCP/TLS/QUIC layer. Closes #9979 Stefan Eissing (26 Nov 2022) - tests: add authorityInfoAccess to generated certs Generate stunnel.pem as well Closes #9980 Daniel Stenberg (25 Nov 2022) - runtests: --no-debuginfod now disables DEBUGINFOD_URLS Prior to this change, DEBUGINFOD_URLS was always disabled by runtests due to a report of it slowing down tests. However, some setups need it to fetch debug symbols, and if it is disabled on those systems then curl tests with valgrind will fail. Reported-by: Mark Gaiser Ref: #8805 Closes #9950 Casey Bodley (25 Nov 2022) - test/aws_sigv4: test cases for content-sha256 1956 adds the sha256 value corresponding to an empty buffer 1957 adds an arbitrary value and confirms that the signature differs from 195 6 1958 adds whitespace to 1957 and confirms that the signature matches 1957 1959 adds a value longer than 'char sha_hex[65]' in Curl_output_aws_sigv4() Signed-off-by: Casey Bodley Closes #9804 - aws_sigv4: consult x-%s-content-sha256 for payload hash `Curl_output_aws_sigv4()` doesn't always have the whole payload in memory to generate a real payload hash. this commit allows the user to pass in a header like `x-amz-content-sha256` to provide their desired payload hash some services like s3 require this header, and may support other values like s3's `UNSIGNED-PAYLOAD` and `STREAMING-AWS4-HMAC-SHA256-PAYLOAD` with special semantics. servers use this header's value as the payload hash during signature validation, so it must match what the client uses to generate the signature CURLOPT_AWS_SIGV4.3 now describes the content-sha256 interaction Signed-off-by: Casey Bodley Closes #9804 Philip Heiduck (25 Nov 2022) - GHA: NSS use clang instead of clang-9 Closes #9978 Daniel Stenberg (25 Nov 2022) - RELEASE-NOTES: synced - tool_operate: override the numeric locale and set "C" by force Makes curl always use dot as decimal separator for options, independently of what the locale says. Makes scripts and command lines portable. Updated docs accordingly. Reported-by: Daniel Faust Fixes #9969 Closes #9972 - test1662: verify formpost, 301 redirect, no rewind possible Reproduces #9735 and verifies the subsequent fix. The original issue uses a pipe that cannot be rewound, but this test case instead sets a callback without rewind ability to get roughly the same properties but being a much more portable test. - lib: rewind BEFORE request instead of AFTER previous This makes a big difference for cases when the rewind is not actually necessary to perofm (for example HTTP response code 301 converts to GET) and therefore the rewind can be avoided. In particular for situations when that rewind fails, for example when reading from a pipe or similar. Reported-by: Ali Utku Selen Fixes #9735 Closes #9958 - vtls: repair build with disabled proxy Closes #9974 Daniel Gustafsson (23 Nov 2022) - packaging: remove traces of deleted files Commit a8861b6cc removed packages/DOS but left a few traces of it which broke the distcheck CI. Remove all traces. Closes: #9971 Reviewed-by: Daniel Stenberg - openssl: silence compiler warning when not using IPv6 In non-IPv6 builds the conn parameter is unused, and compilers which run with "-Werror=unused-parameter" (or similar) warnings turned on fails to build. Below is an excerpt from a CI job: vtls/openssl.c: In function ‘Curl_ossl_verifyhost’: vtls/openssl.c:2016:75: error: unused parameter ‘conn’ [-Werror=unused- parameter] 2016 | CURLcode Curl_ossl_verifyhost(struct Curl_easy *data, struct connec tdata *conn, | ~~~~~~~~~~~~~ ~~~~~~~^~~~ Closes: #9970 Reviewed-by: Daniel Stenberg - netware: remove leftover traces Commit 3b16575ae938dec2a29454631a12aa52b6ab9c67 removed support for building on Novell Netware, but a few leftover traces remained. This removes the last bits. Closes: #9966 Reviewed-by: Daniel Stenberg Ryan Schmidt (23 Nov 2022) - curl_endian: remove Curl_write64_le from header The actual function was already removed in 4331c6dc. See #7280 Closes #9968 Daniel Stenberg (22 Nov 2022) - docs: add more "SEE ALSO" links to CA related pages Closes #9959 - examples: update descriptions Make them not say "this is an example showing..." and instead just say what the example shows. Closes #9960 Stefan Eissing (22 Nov 2022) - vtls: localization of state data in filters - almost all backend calls pass the Curl_cfilter intance instead of connectdata+sockindex - ssl_connect_data is remove from struct connectdata and made internal to vtls - ssl_connect_data is allocated in the added filter, kept at cf->ctx - added function to let a ssl filter access its ssl_primary_config and ssl_config_data this selects the propert subfields in conn and data, for filters added as plain or proxy - adjusted all backends to use the changed api - adjusted all backends to access config data via the exposed functions, no longer using conn or data directly cfilter renames for clear purpose: - methods `Curl_conn_*(data, conn, sockindex)` work on the complete filter chain at `sockindex` and connection `conn`. - methods `Curl_cf_*(cf, ...)` work on a specific Curl_cfilter instance. - methods `Curl_conn_cf()` work on/with filter instances at a connection. - rebased and resolved some naming conflicts - hostname validation (und session lookup) on SECONDARY use the same name as on FIRST (again). new debug macros and removing connectdata from function signatures where not needed. adapting schannel for new Curl_read_plain paramter. Closes #9919 Daniel Stenberg (22 Nov 2022) - examples/10-at-a-time: fix possible skipped final transfers Prior to this change if curl_multi_perform returned 0 running handles and then all remaining transfers were added, then the perform loop would end immediately without performing those transfers. Reported-by: Mikhail Kuznetsov Fixes https://github.com/curl/curl/issues/9953 Closes https://github.com/curl/curl/pull/9954 Viktor Szakats (22 Nov 2022) - Makefile.mk: portable Makefile.m32 Update bare GNU Make `Makefile.m32` to: - Move objects into a subdirectory. - Add support for MS-DOS. Tested with DJGPP. - Add support for Watt-32 (on MS-DOS). - Add support for AmigaOS. - Rename `Makefile.m32` to `Makefile.mk` - Replace `ARCH` with `TRIPLET`. - Build `tool_hugehelp.c` proper (when tools are available). - Drop MS-DOS compatibility macro `USE_ZLIB` (replaced by `HAVE_LIBZ`) - Add support for `ZLIB_LIBS` to override `-lz`. - Omit object files when building examples. - Default `CC` to `gcc` once again, for convenience. (Caveat: compiler name `cc` cannot be set now.) - Set `-DCURL_NO_OLDIES` for examples, like autotools does. - Delete `makefile.dj` files. Notice the configuration details and defaults are not retained with the new method. - Delete `makefile.amiga` files. A successful build needs a few custom options. We're also not retaining all build details from the existing Amiga make files. - Rename `Makefile.m32` to `Makefile.mk` to reflect that they are not Windows/MinGW32-specific anymore. - Add support for new `CFG` options: `-map`, `-debug`, `-trackmem` - Set `-DNDEBUG` by default. - Allow using `-DOS=...` in all `lib/config-*.h` headers, syncing this with `config-win32.h`. - Look for zlib parts in `ZLIB_PATH/include` and `ZLIB_PATH/lib` instead of bare `ZLIB_PATH`. Note that existing build configurations for MS-DOS and AmigaOS likely become incompatible with this change. Example AmigaOS configuration: ``` export CROSSPREFIX=/opt/amiga/bin/m68k-amigaos- export CC=gcc export CPPFLAGS='-DHAVE_PROTO_BSDSOCKET_H' export CFLAGS='-mcrt=clib2' export LDFLAGS="${CFLAGS}" export LIBS='-lnet -lm' make -C lib -f Makefile.mk make -C src -f Makefile.mk ``` Example MS-DOS configuration: ``` export CROSSPREFIX=/opt/djgpp/bin/i586-pc-msdosdjgpp- export WATT_PATH=/opt/djgpp/net/watt export ZLIB_PATH=/opt/djgpp export OPENSSL_PATH=/opt/djgpp export OPENSSL_LIBS='-lssl -lcrypt' export CFG=-zlib-ssl make -C lib -f Makefile.mk make -C src -f Makefile.mk ``` Closes #9764 Stefan Eissing (22 Nov 2022) - cfiler: filter types have flags indicating what they do - Adding Curl_conn_is_ip_connected() to check if network connectivity has been reached - having ftp wait for network connectivity before proceeding with transfers. Fixes test failures 1631 and 1632 with hyper. Closes #9952 Daniel Stenberg (21 Nov 2022) - RELEASE-NOTES: synced Jay Satiro (20 Nov 2022) - sendf: change Curl_read_plain to wrap Curl_recv_plain (take 2) Prior to this change Curl_read_plain would attempt to read the socket directly. On Windows that's a problem because recv data may be cached by libcurl and that data is only drained using Curl_recv_plain. Rather than rewrite Curl_read_plain to handle cached recv data, I changed it to wrap Curl_recv_plain, in much the same way that Curl_write_plain already wraps Curl_send_plain. Curl_read_plain -> Curl_recv_plain Curl_write_plain -> Curl_send_plain This fixes a bug in the schannel backend where decryption of arbitrary TLS records fails because cached recv data is never drained. We send data (TLS records formed by Schannel) using Curl_write_plain, which calls Curl_send_plain, and that may do a recv-before-send ("pre-receive") to cache received data. The code calls Curl_read_plain to read data (TLS records from the server), which prior to this change did not call Curl_recv_plain and therefore cached recv data wasn't retrieved, resulting in malformed TLS records and decryption failure (SEC_E_DECRYPT_FAILURE). The bug has only been observed during Schannel TLS 1.3 handshakes. Refer to the issue and PR for more information. -- This is take 2 of the original fix. It preserves the original behavior of Curl_read_plain to write 0 to the bytes read parameter on error, since apparently some callers expect that (SOCKS tests were hanging). The original fix which landed in 12e1def5 and was later reverted in 18383fbf failed to work properly because it did not do that. Also, it changes Curl_write_plain the same way to complement Curl_read_plain, and it changes Curl_send_plain to return -1 instead of 0 on CURLE_AGAIN to complement Curl_recv_plain. Behavior on error with these changes: Curl_recv_plain returns -1 and *code receives error code. Curl_send_plain returns -1 and *code receives error code. Curl_read_plain returns error code and *n (bytes read) receives 0. Curl_write_plain returns error code and *written receives 0. -- Ref: https://github.com/curl/curl/issues/9431#issuecomment-1312420361 Assisted-by: Joel Depooter Reported-by: Egor Pugin Fixes https://github.com/curl/curl/issues/9431 Closes https://github.com/curl/curl/pull/9949 Sean McArthur (19 Nov 2022) - hyper: classify headers as CONNECT and 1XX Closes #9947 Stefan Eissing (19 Nov 2022) - ftp: fix "AUTH TLS" on primary conn and for SSL in PASV second conn Follow-up to dafdb20a26d0c89 Reported-by: Anthony Hu Closes #9948 Jay Satiro (19 Nov 2022) - CURLOPT_POST.3: Explain setting to 0 changes request type Bug: https://github.com/curl/curl/issues/9849 Reported-by: MonkeybreadSoftware@users.noreply.github.com Closes https://github.com/curl/curl/pull/9942 Daniel Stenberg (19 Nov 2022) - docs/INSTALL.md: expand on static builds Remove from KNOWN_BUGS Closes #9944 Stefan Eissing (19 Nov 2022) - http: restore h3 to working condition after connection filter introduction Follow-up to dafdb20a26d0c HTTP/3 needs a special filter chain, since it does the TLS handling itself. This PR adds special setup handling in the HTTP protocol handler that takes are of it. When a handler, in its setup method, installs filters, the default behaviour for managing the filter chain is overridden. Reported-by: Karthikdasari0423 on github Fixes #9931 Closes #9945 Daniel Stenberg (18 Nov 2022) - urldata: change port num storage to int and unsigned short Instead of long. Closes #9946 - Revert "sendf: change Curl_read_plain to wrap Curl_recv_plain" This reverts commit 12e1def51a75392df62e65490416007d7e68dab9. It introduced SOCKS proxy fails, like test 700 never ending. Reopens #9431 - HTTP-COOKIES.md: update the 6265bis link to draft-11 Closes #9940 - docs/WEBSOCKET.md: explain the URL use Fixes #9936 Closes #9941 Jay Satiro (18 Nov 2022) - sendf: change Curl_read_plain to wrap Curl_recv_plain Prior to this change Curl_read_plain would attempt to read the socket directly. On Windows that's a problem because recv data may be cached by libcurl and that data is only drained using Curl_recv_plain. Rather than rewrite Curl_read_plain to handle cached recv data, I changed it to wrap Curl_recv_plain, in much the same way that Curl_write_plain already wraps Curl_send_plain. Curl_read_plain -> Curl_recv_plain Curl_write_plain -> Curl_send_plain This fixes a bug in the schannel backend where decryption of arbitrary TLS records fails because cached recv data is never drained. We send data (TLS records formed by Schannel) using Curl_write_plain, which calls Curl_send_plain, and that may do a recv-before-send ("pre-receive") to cache received data. The code calls Curl_read_plain to read data (TLS records from the server), which prior to this change did not call Curl_recv_plain and therefore cached recv data wasn't retrieved, resulting in malformed TLS records and decryption failure (SEC_E_DECRYPT_FAILURE). The bug has only been observed during Schannel TLS 1.3 handshakes. Refer to the issue and PR for more information. Ref: https://github.com/curl/curl/issues/9431#issuecomment-1312420361 Assisted-by: Joel Depooter Reported-by: Egor Pugin Fixes https://github.com/curl/curl/issues/9431 Closes https://github.com/curl/curl/pull/9904 - test3026: reduce runtime in legacy mingw builds - Load Windows system libraries secur32 and iphlpapi beforehand, so that libcurl's repeated global init/cleanup only increases/decreases the library's refcount rather than causing it to load/unload. Assisted-by: Marc Hoersken Closes https://github.com/curl/curl/pull/9412 Daniel Stenberg (18 Nov 2022) - url: move back the IDN conversion of proxy names Regression: in commit 53bcf55 we moved the IDN conversion calls to happen before the HSTS checks. But the HSTS checks are only done on the server host name, not the proxy names. By moving the proxy name IDN conversions, we accidentally broke the verbose output showing the proxy name. This change moves back the IDN conversions for the proxy names to the place in the code path they were before 53bcf55. Reported-by: Andy Stamp Fixes #9937 Closes #9939 Alexandre Ferrieux (18 Nov 2022) - CURLOPT_QUICK_EXIT: don't wait for DNS thread on exit Fixes #2975 Closes #9147 Daniel Stenberg (17 Nov 2022) - HTTP-COOKIES.md: mention that http://localhost is a secure context Reported-by: Trail of Bits Closes #9938 - lib: parse numbers with fixed known base 10 ... instead of using 0 argument that allows decimal, hex or octal when the number is documented and assumed to use base 10. Closes #9933 - RELEASE-NOTES: synced - scripts/delta: adapt to curl.h changes for the opt counter - cookie: expire cookies at once when max-age is negative Update test 329 to verify Reported-by: godmar on github Fixes #9930 Closes #9932 Stefan Eissing (17 Nov 2022) - proxy: haproxy filter is only available when PROXY and HTTP are Closes #9935 Daniel Stenberg (16 Nov 2022) - OtherTests.cmake: check for cross-compile, not for toolchain Build systems like vcpkg alway sets `CMAKE_TOOLCHAIN_FILE` so it should not be used as a sign that this is a cross-compile. Also indented the function correctly. Reported-by: Philip Chan Fixes #9921 Closes #9923 - ntlm: improve comment for encrypt_des Reported-by: Andrei Rybak Fixes #9903 Closes #9925 - include/curl/curl.h: bump the deprecated requirements to gcc 5.3 Reported-by: Stephan Guilloux Fixes #9917 Closes #9918 Stefan Eissing (15 Nov 2022) - proxy: refactor haproxy protocol handling as connection filter Closes #9893 Patrick Monnerat (15 Nov 2022) - lib: feature deprecation warnings in gcc >= 4.3 Add a deprecated attribute to functions and enum values that should not be used anymore. This uses a gcc 4.3 dialect, thus is only available for this version of gcc and newer. Note that the _Pragma() keyword is introduced by C99, but is available as part of the gcc dialect even when compiling in C89 mode. It is still possible to disable deprecation at a calling module compile time by defining CURL_DISABLE_DEPRECATION. Gcc type checking macros are made aware of possible deprecations. Some testing support Perl programs are adapted to the extended declaration syntax. Several test and unit test C programs intentionally use deprecated functions/options and are annotated to not generate a warning. New test 1222 checks the deprecation status in doc and header files. Closes #9667 Daniel Stenberg (15 Nov 2022) - log2changes.pl: wrap long lines at 80 columns Also, only use author names in the output. Fixes #9896 Reported-by: John Sherrill Closes #9897 - cfilters: use %zu for outputting size_t Detected by Coverity CID 1516894 Closes #9907 - Curl_closesocket: avoid using 'conn' if NULL ... in debug-only code. Reported by Coverity CID 1516896 Closes #9907 - url: only acknowledge fresh_reuse for non-followed transfers ... to make sure NTLM auth sticks to the connection it needs, as verified by 2032. Follow-up to fa0b9227616e Assisted-by: Stefan Eissing Closes #9905 - netrc.d: provide mutext info Reported-by: xianghongai on github Fixes #9899 Closes #9901 - cmdline-opts/page-footer: remove long option nroff formatting As gen.pl adds them - nroff-scan.pl: detect double highlights - cmdline-opts/gen.pl: fix the linkifier Improved logic for finding existing --options in text and replacing with the full version with nroff syntax. This also makes the web version link options better. Reported-by: xianghongai on github Fixes #9899 Closes #9902 Patrick Monnerat (14 Nov 2022) - tool: use feature names instead of bit mask, when possible If the run-time libcurl is too old to support feature names, the name array is created locally from the bit masks. This is the only sequence left that uses feature bit masks. Closes #9583 - docs: curl_version_info is not thread-safe before libcurl initialization Closes #9583 - version: add a feature names array to curl_version_info_data Field feature_names contains a null-terminated sorted array of feature names. Bitmask field features is deprecated. Documentation is updated. Test 1177 and tests/version-scan.pl updated to match new documentation format and extended to check feature names too. Closes #9583 Stefan Eissing (14 Nov 2022) - negtelnetserver.py: have it call its close() method Closes #9894 Nathan Moinvaziri (13 Nov 2022) - ntlm: silence ubsan warning about copying from null target_info pointer. runtime error: null pointer passed as argument 2, which is declared to never be null Closes #9898 Daniel Stenberg (12 Nov 2022) - RELEASE-NOTES: synced Stefan Eissing (12 Nov 2022) - Websocket: fixes for partial frames and buffer updates. - buffers updated correctly when handling partial frames - callbacks no longer invoked for incomplete payload data of 0 length - curl_ws_recv no longer returns with 0 length partial payload Closes #9890 Daniel Stenberg (12 Nov 2022) - tool_operate: provide better errmsg for -G with bad URL If the URL that -G would try to add a query to could not be parsed, it would display curl: (27) Out of memory It now instead shows: curl: (2) Could not parse the URL, failed to set query Reported-by: Alex Xu Fixes #9889 Closes #9892 - vtls: fix build without proxy support Follow-up to dafdb20a26d0c890 Closes #9895 - tool_getparam: make --no-get work as the opposite of --get ... as documented. Closes #9891 - http: mark it 'this_is_a_follow' in the Location: logic To make regular auth "reloads" to not count as redirects. Verified by test 3101 Fixes #9885 Closes #9887 Viktor Szakats (11 Nov 2022) - config-win32: fix SIZEOF_OFF_T for MSVC and old MinGW The previously set default value of 8 (64-bit) is only correct for mingw-w64 and only when we set `_FILE_OFFSET_BITS` to 64 (the default when building curl). For MSVC, old MinGW and other Windows compilers, the correct value is 4 (32-bit). Adjust condition accordingly. Also drop the manual override option. Regression in 7.86.0 (from 68fa9bf3f5d7b4fcbb57619f70cb4aabb79a51f6) Bug: https://github.com/curl/curl/pull/9712#issuecomment-1307330551 Reported-by: Peter Piekarski Reviewed-by: Jay Satiro Closes #9872 Daniel Stenberg (11 Nov 2022) - lib: remove bad set.opt_no_body assignments This struct field MUST remain what the application set it to, so that handle reuse and handle duplication work. Instead, the request state bit 'no_body' is introduced for code flows that need to change this in run-time. Closes #9888 Stefan Eissing (11 Nov 2022) - lib: connection filters (cfilter) addition to curl: - general construct/destroy in connectdata - default implementations of callback functions - connect: cfilters for connect and accept - socks: cfilter for socks proxying - http_proxy: cfilter for http proxy tunneling - vtls: cfilters for primary and proxy ssl - change in general handling of data/conn - Curl_cfilter_setup() sets up filter chain based on data settings, if none are installed by the protocol handler setup - Curl_cfilter_connect() boot straps filters into `connected` status, used by handlers and multi to reach further stages - Curl_cfilter_is_connected() to check if a conn is connected, e.g. all filters have done their work - Curl_cfilter_get_select_socks() gets the sockets and READ/WRITE indicators for multi select to work - Curl_cfilter_data_pending() asks filters if the have incoming data pending for recv - Curl_cfilter_recv()/Curl_cfilter_send are the general callbacks installed in conn->recv/conn->send for io handling - Curl_cfilter_attach_data()/Curl_cfilter_detach_data() inform filters and addition/removal of a `data` from their connection - adding vtl functions to prevent use of Curl_ssl globals directly in other parts of the code. Reviewed-by: Daniel Stenberg Closes #9855 - curl-rustls.m4: on macOS, rustls also needs the Security framework Closes #9883 Daniel Stenberg (10 Nov 2022) - rtsp: only store first_host once Suggested-by: Erik Janssen URL: https://github.com/curl/curl/pull/9870#issuecomment-1309499744 Closes #9882 Fata Nugraha (10 Nov 2022) - test3028: verify PROXY - http: do not send PROXY more than once Unlike `CONNECT`, currently we don't keep track whether `PROXY` is already sent or not. This causes `PROXY` header to be sent twice during `MSTATE_TUNNELING` and `MSTATE_PROTOCONNECT`. Closes #9878 Fixes #9442 Jay Satiro (10 Nov 2022) - lib: add CURL_WRITEFUNC_ERROR to signal write callback error Prior to this change if the user wanted to signal an error from their write callbacks they would have to use logic to return a value different from the number of bytes (nmemb) passed to the callback. Also, the inclination of some users has been to just return 0 to signal error, which is incorrect as that may be the number of bytes passed to the callback. To remedy this the user can now return CURL_WRITEFUNC_ERROR instead. Ref: https://github.com/curl/curl/issues/9873 Closes https://github.com/curl/curl/pull/9874 Daniel Stenberg (9 Nov 2022) - Revert "GHA: add scorecard.yml" This reverts commit ca76c79b34f9d90105674a2151bf228ff7b13bef. - GHA: add scorecard.yml add a "scorecard" scanner job Lorenzo Miniero (9 Nov 2022) - test3100: RTSP Basic authentication Closes #9449 Daniel Stenberg (9 Nov 2022) - rtsp: fix RTSP auth Verified with test 3100 Fixes #4750 Closes #9870 - KNOWN_BUGS: remove eight entries - 1.2 Multiple methods in a single WWW-Authenticate: header This is not considered a bug anymore but a restriction and one that we keep because we have NEVER gotten this reported by users in the wild and because of this I consider this a fringe edge case we don't need to support. - 1.6 Unnecessary close when 401 received waiting for 100 This is not a bug, but possibly an optimization that *can* be done. - 1.7 Deflate error after all content was received This is not a curl bug. This happens due to broken servers. - 2.1 CURLINFO_SSL_VERIFYRESULT has limited support This is not a bug. This is just the nature of the implementation. - 2.2 DER in keychain This is not a bug. - 5.7 Visual Studio project gaps This is not a bug. - 15.14 cmake build is not thread-safe Fixed in 109e9730ee5e2b - 11.3 Disconnects do not do verbose This is not a bug. Closes #9871 Hirotaka Tagawa (9 Nov 2022) - headers: add endif comments Closes #9853 Daniel Stenberg (8 Nov 2022) - test1221: verify --url-query - curl: add --url-query This option adds a piece of data, usually a name + value pair, to the end of the URL query part. The syntax is identical to that used for --data-urlencode with one extension: If the argument starts with a '+' (plus), the rest of the string is provided as-is unencoded. This allows users to "build" query parts with options and URL encoding even when not doing GET requests, which the already provided option -G (--get) is limited to. This idea was born in a Twitter thread. Closes #9691 - maketgz: set the right version in lib/libcurl.plist Follow-up to e498a9b1fe5964a18eb2a3a99dc52 Make sure the tarball gets a version of the libcurl.plist file that is updated with the new version string. Reported-by: jvreelanda on github Fixes #9866 Closes #9867 - RELEASE-NOTES: synced Bumped version to 7.87.0 Michael Drake (8 Nov 2022) - curl.h: add CURLOPT_CA_CACHE_TIMEOUT option Adds a new option to control the maximum time that a cached certificate store may be retained for. Currently only the OpenSSL backend implements support for caching certificate stores. Closes #9620 - openssl: reduce CA certificate bundle reparsing by caching Closes #9620 Rose (8 Nov 2022) - lib: fix some type mismatches and remove unneeded typecasts Many of these castings are unneeded if we change the variables to work better with each other. Ref: https://github.com/curl/curl/pull/9823 Closes https://github.com/curl/curl/pull/9835 Daniel Stenberg (8 Nov 2022) - cookie: compare cookie prefixes case insensitively Adapted to language in rfc6265bis draft-11. Closes #9863 Reviewed-by: Daniel Gustafsson - tool_operate: when aborting, make sure there is a non-NULL error buffer To store custom errors in. Or SIGSEGVs will follow. Reported-by: Trail of Bits Closes #9865 - WEBSOCKET.md: fix broken link Reported-by: Felipe Gasper Bug: https://curl.se/mail/lib-2022-10/0097.html Closes #9864 - CURLOPT_DEBUGFUNCTION.3: do not assume nul-termination in example Reported-by: Oskar Sigvardsson Bug: https://curl.se/mail/lib-2022-11/0016.html Closes #9862 Stefan Eissing (7 Nov 2022) - websockets: fix handling of partial frames buffer used and send length calculations are fixed when a partial websocket frame has been received. Closes #9861 Daniel Stenberg (7 Nov 2022) - mailmap: unify Stefan Eissing Stefan Eissing (7 Nov 2022) - hyper: fix handling of hyper_task's when reusing the same address Fixes #9840 Closes #9860 Jay Satiro (7 Nov 2022) - ws: return CURLE_NOT_BUILT_IN when websockets not built in - Change curl_ws_recv & curl_ws_send to return CURLE_NOT_BUILT_IN when websockets support is not built in. Prior to this change they returned CURLE_OK. Closes #9851 Daniel Stenberg (7 Nov 2022) - noproxy: tailmatch like in 7.85.0 and earlier A regfression in 7.86.0 (via 1e9a538e05c010) made the tailmatch work differently than before. This restores the logic to how it used to work: All names listed in NO_PROXY are tailmatched against the used domain name, if the lengths are identical it needs a full match. Update the docs, update test 1614. Reported-by: Stuart Henderson Fixes #9842 Closes #9858 - configure: require fork for NTLM-WB Reported-by: ウさん Fixes #9847 Closes #9856 - docs/EARLY-RELEASE.md: how to determine an early release URL: https://curl.se/mail/lib-2022-10/0079.html Closes #9820 - RELEASE-NOTES: synced Zespre Schmidt (3 Nov 2022) - docs: add missing parameters for --retry flag Closes #9848 Adam Averay (3 Nov 2022) - libcurl-errors.3: remove duplicate word Closes #9846 Eric Vigeant (3 Nov 2022) - cur_path: do not add '/' if homedir ends with one When using SFTP and a path relative to the user home, do not add a trailing '/' to the user home dir if it already ends with one. Closes #9844 Viktor Szakats (1 Nov 2022) - windows: fail early with a missing windres in autotools `windres` is not always auto-detected by autotools when building for Windows. When this happened, the build failed with a confusing error due to the empty `RC` command: ``` /bin/bash ../libtool --tag=RC --mode=compile -I../include -DCURL_EMBED_MANIF EST -i curl.rc -o curl.o [...] Usage: /sandbox/curl/libtool [OPTION]... [MODE-ARG]... Try 'libtool --help' for more information. libtool: error: unrecognised option: '-I../include' ``` Improve this by verifying if `RC` is set, and fail with a clear error otherwise. Follow-up to 6de7322c03d5b4d91576a7d9fc893e03cc9d1057 Ref: https://curl.se/mail/lib-2022-10/0049.html Reported-by: Thomas Glanzmann Closes #9781 - lib: sync guard for Curl_getaddrinfo_ex() definition and use `Curl_getaddrinfo_ex()` gets _defined_ with `HAVE_GETADDRINFO` set. But, `hostip4.c` _used_ it with `HAVE_GETADDRINFO_THREADSAFE` set alone. It meant a build with the latter, but without the former flag could result in calling this function but not defining it, and failing to link. Patch this by adding an extra check for `HAVE_GETATTRINFO` around the call. Before this patch, build systems prevented this condition. Now they don't need to. While here, simplify the related CMake logic on Windows by setting `HAVE_GETADDRINFO_THREADSAFE` to the detection result of `HAVE_GETADDRINFO`. This expresses the following intent clearer than the previous patch and keeps the logic in a single block of code: When we have `getaddrinfo()` on Windows, it's always threadsafe. Follow-up to 67d88626d44ec04b9e11dca4cfbf62cd29fe9781 Reviewed-by: Jay Satiro Closes #9734 - tidy-up: process.h detection and use This patch aims to cleanup the use of `process.h` header and the macro `HAVE_PROCESS_H` associated with it. - `process.h` is always available on Windows. In curl, it is required only for `_beginthreadex()` in `lib/curl_threads.c`. - `process.h` is also available in MS-DOS. In curl, its only use was in `lib/smb.c` for `getpid()`. But `getpid()` is in fact declared by `unistd.h`, which is always enabled via `lib/config-dos.h`. So the header is not necessary. - `HAVE_PROCESS_H` was detected by CMake, forced to 1 on Windows and left to real detection for other platforms. It was also set to always-on in `lib/config-win32.h` and `lib/config-dos.h`. In autotools builds, there was no detection and the macro was never set. Based on these observations, in this patch we: - Rework Windows `getpid` logic in `lib/smb.c` to always use the equivalent direct Win32 API function `GetCurrentProcessId()`, as we already did for Windows UWP apps. This makes `process.h` unnecessary here on Windows. - Stop #including `process.h` into files where it was not necessary. This is everywhere, except `lib/curl_threads.c`. > Strangely enough, `lib/curl_threads.c` compiled fine with autotools > because `process.h` is also indirecty included via `unistd.h`. This > might have been broken in autotools MSVC builds, where the latter > header is missing. - Delete all remaining `HAVE_PROCESS_H` feature guards, for they were unnecessary. - Delete `HAVE_PROCESS_H` detection from CMake and predefined values from `lib/config-*.h` headers. Reviewed-by: Jay Satiro Closes #9703 Daniel Stenberg (1 Nov 2022) - lib1301: unit103 turned into a libtest It is not a unit test so moved over to libtests. - strcase: use curl_str(n)equal for case insensitive matches No point in having two entry points for the same functions. Also merged the *safe* function treatment into these so that they can also be used when one or both pointers are NULL. Closes #9837 - README.md: remove badges and xmas-tree garnish URL: https://curl.se/mail/lib-2022-10/0050.html Closes #9833 Patrick Monnerat (1 Nov 2022) - gen.pl: do not generate CURLHELP bitmask lines > 79 characters If a command line option is in many help categories, there is a risk that CURLHELP bitmask source lines generated for listhelp are longer than 79 characters. This change takes care of folding such long lines. Cloes #9834 Marc Hoersken (30 Oct 2022) - CI/cirrus: remove superfluous double-quotes and sudo Follow up to #9565 and #9677 Closes #9738 - tests/sshserver.pl: re-enable ssh-rsa while using openssh 8.8+ Ref: #9738 Daniel Stenberg (30 Oct 2022) - style: use space after comment start and before comment end /* like this */ /*not this*/ checksrc is updated accordingly Closes #9828 Patrick Schlangen (30 Oct 2022) - docs: remove performance note in CURLOPT_SSL_VERIFYPEER This note became obsolete since PR #7892 (see also discussion in the PR comments). Closes #9832 Daniel Stenberg (30 Oct 2022) - tests/server: make use of strcasecompare from lib/ ... instead of having a second private implementation. Idea triggered by #9830 Closes #9831 - curl: timeout in the read callback The read callback can timeout if there's nothing to read within the given maximum period. Example use case is when doing "curl -m 3 telnet://example.com" or anything else that expects input on stdin or similar that otherwise would "hang" until something happens and then not respect the timeout. This fixes KNOWN_BUG 8.1, first filed in July 2009. Bug: https://sourceforge.net/p/curl/bugs/846/ Closes #9815 - noproxy: fix tail-matching Also ignore trailing dots in both host name and comparison pattern. Regression in 7.86.0 (from 1e9a538e05c0) Extended test 1614 to verify better. Reported-by: Henning Schild Fixes #9821 Closes #9822 - docs: explain the noproxy CIDR notation support Follow-up to 1e9a538e05c0107c Closes #9818 Jon Rumsey (27 Oct 2022) - os400: use platform socklen_t in Curl_getnameinfo_a Curl_getnameinfo_a() is prototyped before including curl.h as an ASCII'fied wrapper for getnameinfo(), which itself is prototyped with socklen_t arguments, so this should use the platform socklen_t and not curl_socklen_t too. Update setup-os400.h Fixes #9811 Closes #9812 Daniel Stenberg (27 Oct 2022) - noproxy: also match with adjacent comma If the host name is an IP address and the noproxy string contained that IP address with a following comma, it would erroneously not match. Extended test 1614 to verify this combo as well. Reported-by: Henning Schild Fixes #9813 Closes #9814 Randall S. Becker (27 Oct 2022) - build: fix for NonStop - Include arpa/inet.h in all units where htonl is called. Signed-off-by: Randall S. Becker Closes https://github.com/curl/curl/pull/9816 - system.h: support 64-bit curl_off_t for NonStop 32-bit - Correctly define curl_off_t on NonStop (ie __TANDEM) ia64 and x86 for 32-bit builds. Signed-off-by: Randall S. Becker Closes https://github.com/curl/curl/pull/9817 Daniel Stenberg (27 Oct 2022) - spellcheck.words: remove 'github' as an accepted word Prefer the properly cased version: GitHub Use markdown for links and GitHub in text. Closes #9810 Ayesh Karunaratne (27 Oct 2022) - misc: typo and grammar fixes - Replace `Github` with `GitHub`. - Replace `windows` with `Windows` - Replace `advice` with `advise` where a verb is used. - A few fixes on removing repeated words. - Replace `a HTTP` with `an HTTP` Closes #9802 Viktor Szakats (27 Oct 2022) - windows: fix linking .rc to shared curl with autotools `./configure --enable-shared --disable-static` fails when trying to link a shared `curl.exe`, due to `libtool` magically changing the output filename of `windres` to one that it doesn't find when linking: ``` /bin/sh ../libtool --tag=RC --mode=compile windres -I../../curl/include -DCUR L_EMBED_MANIFEST -i ../../curl/src/curl.rc -o curl.o libtool: compile: windres -I../../curl/include -DCURL_EMBED_MANIFEST -i ../. ./curl/src/curl.rc -o .libs/curl.o [...] CCLD curl.exe clang: error: no such file or directory: 'curl.o' ``` Let's resolve this by skipping `libtool` and calling `windres` directly when building `src` (aka `curl.exe`). Leave `lib` unchanged, as it does need the `libtool` magic. This solution is compatible with building a static `curl.exe`. This build scenario is not CI-tested. While here, delete an obsolete comment about a permanent `libtool` warning that we've resolved earlier. Regression from 6de7322c03d5b4d91576a7d9fc893e03cc9d1057 Reported-by: Christoph Reiter Fixes #9803 Closes #9805 - cmake: really enable warnings with clang Even though `PICKY_COMPILER=ON` is the default, warnings were not enabled when using llvm/clang, because `CMAKE_COMPILER_IS_CLANG` was always false (in my tests at least). This is the single use of this variable in curl, and in a different place we already use `CMAKE_C_COMPILER_ID MATCHES "Clang"`, which works as expected, so change the condition to use that instead. Also fix the warnings uncovered by the above: - lib: add casts to silence clang warnings - schannel: add casts to silence clang warnings in ALPN code Assuming the code is correct, solve the warnings with a cast. This particular build case isn't CI tested. There is a chance the warning is relevant for some platforms, perhaps Windows 32-bit ARM7. Closes #9783 Joel Depooter (26 Oct 2022) - sendf: remove unnecessary if condition At this point, the psnd->buffer will always exist. We have already allocated a new buffer if one did not previously exist, and returned from the function if the allocation failed. Closes #9801 Viktor Szakats (26 Oct 2022) - winidn: drop WANT_IDN_PROTOTYPES `WANT_IDN_PROTOTYPES` was necessary to avoid using a header that came via an optional package. MS stopped distributing this package some years ago and the winidn definitions are part of standard headers (via `windows.h`) since Vista. Auto-detect Vista inside `lib/idn_win32.c` and enable the manual definitions if building for an older Windows. This allows to delete this manual knob from all build-systems. Also drop the `_SAL_VERSION` sub-case: Our manual definitions are now only enabled with old systems. We assume that code analysis is not run on such systems, allowing us to delete the SAL-friendly flavour of these. Reviewed-by: Jay Satiro Closes #9793 Daniel Stenberg (26 Oct 2022) - misc: remove duplicated include files Closes #9796 - scripts/checksrc.pl: detect duplicated include files After an idea by Dan Fandrich in #9794 Closes #9796 - RELEASE-NOTES: synced And bumped version to 7.86.1 for now - CURLMOPT_SOCKETFUNCTION.3: clarify CURL_POLL_REMOVE The removal is brief or long, don't assume. Reported-by: Luca Niccoli Fixes #9799 Closes #9800 Version 7.86.0 (26 Oct 2022) Daniel Stenberg (26 Oct 2022) - RELEASE: synced The 7.86.0 release - THANKS: added from the 7.86.0 release Viktor Szakats (25 Oct 2022) - noproxy: include netinet/in.h for htonl() Solve the Amiga build warning by including `netinet/in.h`. `krb5.c` and `socketpair.c` are using `htonl()` too. This header is already included in those sources. Regression from 1e9a538e05c0107c54ef81d9de7cd0b27cd13309 Reviewed-by: Daniel Stenberg Closes #9787 Marc Hoersken (24 Oct 2022) - CI: fix AppVeyor status failing for starting jobs Daniel Stenberg (24 Oct 2022) - test445: verifies the protocols-over-http-proxy flaw and fix - http_proxy: restore the protocol pointer on error Reported-by: Trail of Bits Closes #9790 - multi: remove duplicate include of connect.h Reported-by: Martin Strunz Fixes #9794 Closes #9795 Daniel Gustafsson (24 Oct 2022) - idn: fix typo in test description s/enabked/enabled/i Daniel Stenberg (24 Oct 2022) - url: use IDN decoded names for HSTS checks Reported-by: Hiroki Kurosawa Closes #9791 - unit1614: fix disabled-proxy build Follow-up to 1e9a538e05c01 Closes #9792 Daniel Gustafsson (24 Oct 2022) - cookies: optimize control character check When checking for invalid octets the strcspn() call will return the position of the first found invalid char or the first NULL byte. This means that we can check the indicated position in the search- string saving a strlen() call. Closes: #9736 Reviewed-by: Jay Satiro Daniel Stenberg (24 Oct 2022) - netrc: replace fgets with Curl_get_line Make the parser only accept complete lines and avoid problems with overly long lines. Reported-by: Hiroki Kurosawa Closes #9789 - RELEASE-NOTES: add "Planned upcoming removals include" URL: https://curl.se/mail/archive-2022-10/0001.html Suggested-by: Dan Fandrich Viktor Szakats (23 Oct 2022) - ci: bump to gcc-11 for macos Ref: https://github.blog/changelog/2022-10-03-github-actions-jobs-running-on- macos-latest-are-now-running-on-macos-12/ Ref: https://github.com/actions/runner-images/blob/main/images/macos/macos-12 -Readme.md Reviewed-by: Max Dymond Closes #9785 - Makefile.m32: reintroduce CROSSPREFIX and -W -Wall [ci skip] - Reintroduce `CROSSPREFIX`: If set, we add it to the `CC` and `AR` values, and to the _default_ value of `RC`, which is `windres`. This allows to control each of these individidually, while also allowing to simplify configuration via `CROSSPREFIX`. This variable worked differently earlier. Hopefully this new solution hits a better compromise in usefulness/complexity/flexibility. Follow-up to: aa970c4c08775afcd0c2853be89b0a6f02582d50 - Enable warnings again: This time with an option to override it via `CFLAGS`. Warnings are also enabled by default in CMake, `makefile.dj` and `makefile.amiga` builds (not in autotools though). Follow-up to 10fbd8b4e3f83b967fd9ad9a41ab484c0e7e7ca3 Closes #9784 - noproxy: silence unused variable warnings with no ipv6 Follow-up to 36474f1050c7f4117e3c8de6cc9217cfebfc717d Reviewed-by: Daniel Stenberg Closes #9782 Daniel Stenberg (22 Oct 2022) - test644: verify --xattr (with redirect) - tool_xattr: save the original URL, not the final redirected one Adjusted test 1621 accordingly. Reported-by: Viktor Szakats Fixes #9766 Closes #9768 - docs: make sure libcurl opts examples pass in long arguments Reported-by: Sergey Fixes #9779 Closes #9780 Marc Hoersken (21 Oct 2022) - CI: fix AppVeyor job links only working for most recent build Ref: https://github.com/curl/curl/pull/9768#issuecomment-1286675916 Reported-by: Daniel Stenberg Follow up to #9769 Viktor Szakats (21 Oct 2022) - noproxy: fix builds without AF_INET6 Regression from 1e9a538e05c0107c54ef81d9de7cd0b27cd13309 Reviewed-by: Daniel Stenberg Closes #9778 Daniel Stenberg (21 Oct 2022) - noproxy: support proxies specified using cidr notation For both IPv4 and IPv6 addresses. Now also checks IPv6 addresses "correctly" and not with string comparisons. Split out the noproxy checks and functionality into noproxy.c Added unit test 1614 to verify checking functions. Reported-by: Mathieu Carbonneaux Fixes #9773 Fixes #5745 Closes #9775 - urlapi: remove two variable assigns To please scan-build: urlapi.c:1163:9: warning: Value stored to 'qlen' is never read qlen = Curl_dyn_len(&enc); ^ ~~~~~~~~~~~~~~~~~~ urlapi.c:1164:9: warning: Value stored to 'query' is never read query = u->query = Curl_dyn_ptr(&enc); ^ ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Follow-up to 7d6cf06f571d57 Closes #9777 Jeremy Maitin-Shepard (21 Oct 2022) - cmake: improve usability of CMake build as a sub-project - Renames `uninstall` -> `curl_uninstall` - Ensures all export rules are guarded by CURL_ENABLE_EXPORT_TARGET Closes #9638 Don J Olmstead (21 Oct 2022) - easy_lock: check for HAVE_STDATOMIC_H as well The check for `HAVE_STDATOMIC_H` looks to see if the `stdatomic.h` header is present. Closes #9755 Daniel Stenberg (21 Oct 2022) - RELEASE-NOTES: synced Brad Harder (20 Oct 2022) - CURLMOPT_PIPELINING.3: dedup manpage xref Closes #9776 Marc Hoersken (20 Oct 2022) - CI: report AppVeyor build status for each job Also give each job on AppVeyor CI a human-readable name. This aims to make job and therefore build failures more visible. Reviewed-by: Marcel Raad Closes #9769 Viktor Szakats (20 Oct 2022) - amiga: set SIZEOF_CURL_OFF_T=8 by default [ci skip] Reviewed-by: Daniel Stenberg Closes #9771 - connect: fix builds without AF_INET6 Regression from 2b309560c1e5d6ed5c0e542e6fdffa968b0521c9 Reviewed-by: Daniel Stenberg Reviewed-by: Jay Satiro Closes #9770 Daniel Stenberg (20 Oct 2022) - test1105: adjust to work with a hyper build Closes #9767 - urlapi: fix parsing URL without slash with CURLU_URLENCODE When CURLU_URLENCODE is set, the parser would mistreat the path component if the URL was specified without a slash like in http://local.test:80?-123 Extended test 1560 to reproduce and verify the fix. Reported-by: Trail of Bits Closes #9763 Marc Hoersken (19 Oct 2022) - tests: avoid CreateThread if _beginthreadex is available CreateThread is not threadsafe if mixed with CRT calls. _beginthreadex on the other hand can be mixed with CRT. Reviewed-by: Marcel Raad Closes #9705 Joel Depooter (19 Oct 2022) - schannel: Don't reset recv/send function pointers on renegotiation These function pointers will have been set when the initial TLS handshake was completed. If they are unchanged, there is no need to set them again. If they have been changed, as is the case with HTTP/2, we don't want to override that change. That would result in the http22_recv/send functions being completely bypassed. Prior to this change a connection that uses Schannel with HTTP/2 would fail on renegotiation with error "Received HTTP/0.9 when not allowed". Fixes https://github.com/curl/curl/issues/9451 Closes https://github.com/curl/curl/pull/9756 Viktor Szakats (18 Oct 2022) - hostip: guard PF_INET6 use Some platforms (e.g. Amiga OS) do not have `PF_INET6`. Adjust the code for these. ``` hostip.c: In function 'fetch_addr': hostip.c:308:12: error: 'PF_INET6' undeclared (first use in this function) pf = PF_INET6; ^~~~~~~~ ``` Regression from 1902e8fc511078fb5e26fc2b907b4cce77e1240d Reviewed-by: Daniel Stenberg Closes #9760 - amiga: do not hardcode openssl/zlib into the os config [ci skip] Enable them in `lib/makefile.amiga` and `src/makefile.amiga` instead. This allows builds without openssl and/or zlib. E.g. with the cross-compiler. Reviewed-by: Daniel Stenberg Closes #9762 - amigaos: add missing curl header [ci skip] Without it, `CURLcode` and `CURLE_*` are undefined. `lib/hostip.h` and conditional local code need them. Reviewed-by: Daniel Stenberg Closes #9761 Daniel Stenberg (18 Oct 2022) - cmdline/docs: add a required 'multi' keyword for each option The keyword specifies how option works when specified multiple times: - single: the last provided value replaces the earlier ones - append: it supports being provided multiple times - boolean: on/off values - mutex: flag-like option that disable anoter flag The 'gen.pl' script then outputs the proper and unified language for each option's multi-use behavior in the generated man page. The multi: header is requires in each .d file and will cause build error if missing or set to an unknown value. Closes #9759 - CURLOPT_AUTOREFERER.3: highlight the privacy leak risk Closes #9757 - mprintf: reject two kinds of precision for the same argument An input like "%.*1$.9999d" would first use the precision taken as an argument *and* then the precision specified in the string, which is confusing and wrong. pass1 will now instead return error on this double use. Adjusted unit test 1398 to verify Reported-by: Peter Goodman Closes #9754 - ftp: remove redundant if Reported-by: Trail of Bits Closes #9753 - tool_operate: more transfer cleanup after parallel transfer fail In some circumstances when doing parallel transfers, the single_transfer_cleanup() would not be called and then 'inglob' could leak. Test 496 verifies Reported-by: Trail of Bits Closes #9749 - mqtt: spell out CONNECT in comments Instead of calling it 'CONN' in several comments, use the full and correct protocol packet name. Suggested by Trail of Bits Closes #9751 - CURLOPT_POSTFIELDS.3: refer to CURLOPT_MIMEPOST Not the deprecated CURLOPT_HTTPPOST option. Also added two see-alsos. Reported-by: Trail of Bits Closes #9752 - RELEASE-NOTES: synced Jay Satiro (17 Oct 2022) - ngtcp2: Fix build errors due to changes in ngtcp2 library ngtcp2/ngtcp2@b0d86f60 changed: - ngtcp2_conn_get_max_udp_payload_size => ngtcp2_conn_get_max_tx_udp_payload_size - ngtcp2_conn_get_path_max_udp_payload_size => ngtcp2_conn_get_path_max_tx_udp_payload_size ngtcp2/ngtcp2@ec59b873 changed: - 'early_data_rejected' member added to ng_callbacks. Assisted-by: Daniel Stenberg Reported-by: jurisuk@users.noreply.github.com Fixes https://github.com/curl/curl/issues/9747 Closes https://github.com/curl/curl/pull/9748 Daniel Stenberg (16 Oct 2022) - curl_path: return error if given a NULL homedir Closes #9740 - libssh: if sftp_init fails, don't get the sftp error code This flow extracted the wrong code (sftp code instead of ssh code), and the code is sometimes (erroneously) returned as zero anyway, so skip getting it and set a generic error. Reported-by: David McLaughlin Fixes #9737 Closes #9740 - mqtt: return error for too long topic Closes #9744 Rickard Hallerbäck (16 Oct 2022) - tool_paramhlp: make the max argument a 'double' To fix compiler warnings "Implicit conversion from 'long' to 'double' may lose precision" Closes #9700 Philip Heiduck (15 Oct 2022) - cirrus-ci: add more macOS builds with m1 based on x86_64 builds Also refactor macOS builds to use task matrix. Assisted-by: Marc Hörsken Closes #9565 Viktor Szakats (14 Oct 2022) - cmake: set HAVE_SOCKADDR_IN6_SIN6_SCOPE_ID on Windows `lib/config-win32.h` enables this configuration option unconditionally. Make it apply to CMake builds as well. While here, delete a broken check for `HAVE_SOCKADDR_IN6_SIN6_SCOPE_ID` from `CMakeLists.txt`. This came with the initial commit [1], but did not include the actual verification code inside `CMake/CurlTests.c`, so it always failed. A later commit [2] added a second test, for non-Windows platforms. Enabling this flag causes test 1056 to fail with CMake builds, as they do with autotools builds. Let's apply the same solution and ignore the results here as well. [1] 4c5307b45655ba75ab066564afdc0c111a8b9291 [2] aec7c5a87c8482b6ddffa352d7d220698652262e Reviewed-by: Daniel Stenberg Assisted-by: Marcel Raad Closes #9726 - cmake: set HAVE_GETADDRINFO_THREADSAFE on Windows autotools enables this configuration option unconditionally for Windows [^1]. Do the same in CMake. The above will make this work for all reasonably recent environments. The logic present in `lib/config-win32.h` [^2] has the following exceptions which we did not cover in this CMake update: - Builds targeting Windows 2000 and earlier - MS Visual C++ 5.0 (1997) and earlier Also make sure to disable this feature when `HAVE_GETADDRINFO` isn't set, to avoid a broken build. We might want to handle that in the C sources in a future commit. [^1]: https://github.com/curl/curl/blob/68fa9bf3f5d7b4fcbb57619f70cb4aabb79a5 1f6/m4/curl-functions.m4#L2067-L2070 [^2]: https://github.com/curl/curl/blob/68fa9bf3f5d7b4fcbb57619f70cb4aabb79a5 1f6/lib/config-win32.h#L511-L528 Closes #9727 - cmake: sync HAVE_SIGNAL detection with autotools `HAVE_SIGNAL` means the availability of the `signal()` function in autotools, while in CMake it meant the availability of that function _and_ the symbol `SIGALRM`. The latter is not available on Windows, but the function is, which means on Windows, autotools did define `HAVE_SIGNAL`, but CMake did not, introducing a slight difference into the binaries. This patch syncs CMake behaviour with autotools to look for the function only. The logic came with the initial commit adding CMake support to curl, so the commit history doesn't reveal the reason behind it. In any case, it's best to check the existence of `SIGALRM` directly in the source before use. For now, curl builds fine with `HAVE_SIGNAL` enabled and `SIGALRM` missing. Follow-up to 68fa9bf3f5d7b4fcbb57619f70cb4aabb79a51f6 Closes #9725 - cmake: delete duplicate HAVE_GETADDRINFO test A custom `HAVE_GETADDRINFO` check came with the initial CMake commit [1]. A later commit [2] added a standard check for it as well. The standard check run before the custom one, so CMake ignored the latter. The custom check was also non-portable, so this patch deletes it in favor of the standard check. [1] 4c5307b45655ba75ab066564afdc0c111a8b9291 [2] aec7c5a87c8482b6ddffa352d7d220698652262e Closes #9731 Daniel Stenberg (14 Oct 2022) - tool_formparse: unroll the NULL_CHECK and CONST_FREE macros To make the code read more obvious Assisted-by: Jay Satiro Closes #9710 Christopher Sauer (14 Oct 2022) - docs/INSTALL: update Android Instructions for newer NDKs Closes #9732 Daniel Stenberg (14 Oct 2022) - markdown-uppercase: ignore quoted sections Sections within the markdown ~~~ or ``` are now ignored. Closes #9733 - RELEASE-NOTES: synced - test8: update as cookies no longer can have "embedded" TABs in content - test1105: extend to verify TAB in name/content discarding cookies - cookie: reject cookie names or content with TAB characters TABs in name and content seem allowed by RFC 6265: "the algorithm strips leading and trailing whitespace from the cookie name and value (but maintains internal whitespace)" Cookies with TABs in the names are rejected by Firefox and Chrome. TABs in content are stripped out by Firefox, while Chrome discards the whole cookie. TABs in cookies also cause issues in saved netscape cookie files. Reported-by: Trail of Bits URL: https://curl.se/mail/lib-2022-10/0032.html URL: https://github.com/httpwg/http-extensions/issues/2262 Closes #9659 - curl/add_parallel_transfers: better error handling 1 - consider the transfer handled at once when in the function, to avoid the same list entry to get added more than once in rare error situations 2 - set the ERRORBUFFER for the handle first after it has been added successfully Reported-by: Trail of Bits Closes #9729 - netrc: remove the two 'changed' arguments As no user of these functions used the returned content. - test495: verify URL encoded user name + netrc-optional Reproduced issue #9709 - netrc: use the URL-decoded user When the user name is provided in the URL it is URL encoded there, but when used for authentication the encoded version should be used. Regression introduced after 7.83.0 Reported-by: Jonas Haag Fixes #9709 Closes #9715 Shaun Mirani (13 Oct 2022) - url: allow non-HTTPS HSTS-matching for debug builds Closes #9728 Daniel Stenberg (13 Oct 2022) - test1275: remove the check of stderr To avoid the mysterious test failures on Windows, instead rely on the error code returned on failure. Fixes #9716 Closes #9723 Viktor Szakats (13 Oct 2022) - lib: set more flags in config-win32.h The goal is to add any flag that affect the created binary, to get in sync with the ones built with CMake and autotools. I took these flags from curl-for-win [0], where they've been tested with mingw-w64 and proven to work well. This patch brings them to curl as follows: - Enable unconditionally those force-enabled via `CMake/WindowsCache.cmake`: - `HAVE_SETJMP_H` - `HAVE_STRING_H` - `HAVE_SIGNAL` (CMake equivalent is `HAVE_SIGNAL_FUNC`) - Expand existing guards with mingw-w64: - `HAVE_STDBOOL_H` - `HAVE_BOOL_T` - Enable Win32 API functions for Windows Vista and later: - `HAVE_INET_NTOP` - `HAVE_INET_PTON` - Set sizes, if not already set: - `SIZEOF_OFF_T = 8` - `_FILE_OFFSET_BITS = 64` when `USE_WIN32_LARGE_FILES` is set, and using mingw-w64. - Add the remaining for mingw-w64 only. Feel free to expand as desired: - `HAVE_LIBGEN_H` - `HAVE_FTRUNCATE` - `HAVE_BASENAME` - `HAVE_STRTOK_R` Future TODO: - `HAVE_SIGNAL` has a different meaning in CMake. It's enabled when both the `signal()` function and the `SIGALRM` macro are found. In autotools and this header, it means the function only. For the function alone, CMake uses `HAVE_SIGNAL_FUNC`. [0] https://github.com/curl/curl-for-win/blob/c9b9a5f273c94c73d2b565ee892c4df f0ca97a8c/curl-m32.sh#L53-L58 Reviewed-by: Daniel Stenberg Closes #9712 Daniel Stenberg (13 Oct 2022) - tests: add tests/markdown-uppercase.pl to dist tarball Follow-up to aafb06c5928183d Closes #9722 - tool_paramhelp: asserts verify maximum sizes for string loading The two defines MAX_FILE2MEMORY and MAX_FILE2STRING define the largest strings accepted when loading files into memory, but as the size is later used as input to functions that take the size as 'int' as argument, the sizes must not be larger than INT_MAX. These two new assert()s make the code error out if someone would bump the sizes without this consideration. Reported-by Trail of Bits Closes #9719 - http: try parsing Retry-After: as a number first Since the date parser allows YYYYMMDD as a date format (due to it being a bit too generic for parsing this particular header), a large integer number could wrongly match that pattern and cause the parser to generate a wrong value. No date format accepted for this header starts with a decimal number, so by reversing the check and trying a number first we can deduct that if that works, it was not a date. Reported-by Trail of Bits Closes #9718 Patrick Monnerat (13 Oct 2022) - doc: fix deprecation versions inconsistencies Ref: https://curl.se/mail/lib-2022-10/0026.html Closes #9711 Daniel Stenberg (13 Oct 2022) - http_aws_sigv4: fix strlen() check The check was off-by-one leading to buffer overflow. Follow-up to 29c4aa00a16872 Detected by OSS-Fuzz Closes #9714 - curl/main_checkfds: check the fcntl return code better fcntl() can (in theory) return a non-zero number for success, so a better test for error is checking for -1 explicitly. Follow-up to 41e1b30ea1b77e9ff Mentioned-by: Dominik Klemba Closes #9708 Viktor Szakats (12 Oct 2022) - tidy-up: delete unused HAVE_STRUCT_POLLFD It was only defined in `lib/config-win32.h`, when building for Vista. It was only used in `select.h`, in a condition that also included a check for `POLLIN` which is a superior choice for this detection and which was already used by cmake and autotools builds. Delete both instances of this macro. Closes #9707 Daniel Stenberg (12 Oct 2022) - test1275: verify upercase after period in markdown Script based on the #9474 pull-request logic, but implemented in perl. Updated docs/URL-SYNTAX.md accordingly. Suggested-by: Dan Fandrich Closes #9697 12932 (12 Oct 2022) - misc: nitpick grammar in comments/docs because the 'u' in URL is actually a consonant *sound* it is only correct to write "a URL" sorry this is a bit nitpicky :P https://english.stackexchange.com/questions/152/when-should-i-use-a-vs-an https://www.techtarget.com/whatis/feature/Which-is-correct-a-URL-or-an-URL Closes #9699 Viktor Szakats (11 Oct 2022) - Makefile.m32: drop CROSSPREFIX and our CC/AR defaults [ci skip] This patch aimed to fix a regression [0], where `CC` initialization moved beyond its first use. But, on closer inspection it turned out that the `CC` initialization does not work as expected due to GNU Make filling it with `cc` by default. So unless implicit values were explicitly disabled via a GNU Make option, the default value of `$CROSSPREFIX` + `gcc` was never used. At the same time the implicit value `cc` maps to `gcc` in (most/all?) MinGW envs. `AR` has the same issue, with a default value of `ar`. We could reintroduce a separate variable to fix this without ill effects, but for simplicity and flexibility, it seems better to drop support for `CROSSPREFIX`, along with our own `CC`/`AR` init logic, and require the caller to initialize `CC`, `AR` and `RC` to the full (prefixed if necessary) names of these tools, as desired. We keep `RC ?= windres` because `RC` is empty by default. Also fix grammar in a comment. [0] 10fbd8b4e3f83b967fd9ad9a41ab484c0e7e7ca3 Closes #9698 - smb: replace CURL_WIN32 with WIN32 PR #9255 aimed to fix a Cygwin/MSYS issue (#8220). It used the `CURL_WIN32` macro, but that one is not defined here, while compiling curl itself. This patch changes this to `WIN32`, assuming this was the original intent. Regression from 1c52e8a3795ccdf8ec9c308f4f8f19cf10ea1f1a Reviewed-by: Marcel Raad Closes #9701 Matthias Gatto (11 Oct 2022) - aws_sigv4: fix header computation Handle canonical headers and signed headers creation as explained here: https://docs.aws.amazon.com/general/latest/gr/sigv4-create-canonical-request. html The algo tells that signed and canonical must contain at last host and x-amz-date. So we check whatever thoses are present in the curl http headers list. If they are, we use the one enter by curl user, otherwise we generate them. then we to lower, and remove space from each http headers plus host and x-amz-date, then sort them all by alphabetical order. This patch also fix a bug with host header, which was ignoring the port. Closes #7966 Aftab Alam (11 Oct 2022) - README.md: link the curl logo to the website - Link the curl:// image to https://curl.se/ Closes https://github.com/curl/curl/pull/9675 Dustin Howett (11 Oct 2022) - schannel: when importing PFX, disable key persistence By default, the PFXImportCertStore API persists the key in the user's key store (as though the certificate was being imported for permanent, ongoing use.) The documentation specifies that keys that are not to be persisted should be imported with the flag PKCS12_NO_PERSIST_KEY. NOTE: this flag is only supported on versions of Windows newer than XP and Server 2003. -- This is take 2 of the original fix. It extends the lifetime of the client certificate store to that of the credential handle. The original fix which landed in 70d010d and was later reverted in aec8d30 failed to work properly because it did not do that. Minor changes were made to the schannel credential context to support closing the client certificate store handle at the end of an SSL session. -- Reported-by: ShadowZzj@users.noreply.github.com Fixes https://github.com/curl/curl/issues/9300 Supersedes https://github.com/curl/curl/pull/9363 Closes https://github.com/curl/curl/pull/9460 Viktor Szakats (11 Oct 2022) - Makefile.m32: support more options [ci skip] - Add support for these options: `-wolfssl`, `-wolfssh`, `-mbedtls`, `-libssh`, `-psl` Caveats: - `-wolfssh` requires `-wolfssl`. - `-wolfssl` cannot be used with OpenSSL backends in parallel. - `-libssh` has build issues with BoringSSL and LibreSSL, and also what looks like a world-writable-config vulnerability on Windows. Consider it experimental. - `-psl` requires `-idn2` and extra libs passed via `LIBS=-liconv -lunistring`. - Detect BoringSSL/wolfSSL and set ngtcp2 crypto lib accordingly. - Generalize MultiSSL detection. - Use else-if syntax. Requires GNU Make 3.81 (2006-04-01). - Document more customization options. This brings over some configuration logic from `curl-for-win`. Closes #9680 - cmake: enable more detection on Windows Enable `HAVE_UNISTD_H`, `HAVE_STRTOK_R` and `HAVE_STRCASECMP` detection on Windows, instead of having predefined values. With these features detected correctly, CMake Windows builds get closer to the autotools and `config-win32.h` ones. This also fixes detecting `HAVE_FTRUNCATE` correctly, which required `unistd.h`. Fixing `ftruncate()` in turn causes a build warning/error with legacy MinGW/MSYS1 due to an offset type size mismatch. This env misses to detect `HAVE_FILE_OFFSET_BITS`, which may be a reason. This patch force-disables `HAVE_FTRUNCATE` for this platform. Reviewed-by: Daniel Stenberg Closes #9687 - autotools: allow unix sockets on Windows Fixes: https://github.com/curl/curl-for-win/blob/73a070d96fd906fdee929e2f1f00 a9149fb39239/curl-autotools.sh#L44-L47 On Windows this feature is present, but not the header used in the detection logic. It also requires an elaborate enabler logic (as seen in `lib/curl_setup.h`). Let's always allow it and let the lib code deal with the details. Closes #9688 - cmake: add missing inet_ntop check This adds the missing half of the check, next to the other half already present in `lib/curl_config.h.cmake`. Force disable `HAVE_INET_NTOP` for old MSVC where it caused compiler warnings. Reviewed-by: Daniel Stenberg Closes #9689 Daniel Stenberg (11 Oct 2022) - RELEASE-NOTES: synced bsergean on github (11 Oct 2022) - asyn-ares: set hint flags when calling ares_getaddrinfo The hint flag is ARES_AI_NUMERICSERV, and it will save a call to getservbyname or getservbyname_r to set it. Closes #9694 Daniel Stenberg (11 Oct 2022) - header.d: add category smtp and imap They were previously (erroneously) added manually to tool_listhelp.c which would make them get removed again when the file is updated next time, unless added correctly here in header.d Follow-up to 2437fac01 Closes #9690 - curl/get_url_file_name: use libcurl URL parser To avoid URL tricks, use the URL parser for this. This update changes curl's behavior slightly in that it will ignore the possible query part from the URL and only use the file name from the actual path from the URL. I consider it a bugfix. "curl -O localhost/name?giveme-giveme" will now save the output in the local file named 'name' Updated test 1210 to verify Assisted-by: Jay Satiro Closes #9684 Martin Ågren (11 Oct 2022) - docs: fix grammar around needing pass phrase "You never needed a pass phrase" reads like it's about to be followed by something like "until version so-and-so", but that is not what is intended. Change to "You never need a pass phrase". There are two instances of this text, so make sure to update both. Xiang Xiao (10 Oct 2022) - cmake: add the check of HAVE_SOCKETPAIR which is used by Curl_socketpair Signed-off-by: Xiang Xiao Closes #9686 Daniel Stenberg (10 Oct 2022) - curl/add_file_name_to_url: use the libcurl URL parser instead of the custom error-prone parser, to extract and update the path of the given URL Closes #9683 - single_transfer: use the libcurl URL parser when appending query parts Instead of doing "manual" error-prone parsing in another place. Used when --data contents is added to the URL query when -G is provided. Closes #9681 - ws: fix buffer pointer use in the callback loop Closes #9678 Petr Štetiar (10 Oct 2022) - curl-wolfssl.m4: error out if wolfSSL is not usable When I explicitly declare, that I would like to have curl built with wolfSSL support using `--with-wolfssl` configure option, then I would expect, that either I endup with curl having that support, for example in form of https support or it wouldn't be available at all. Downstream projects like for example OpenWrt build curl wolfSSL variant with `--with-wolfssl` already, but in certain corner cases it does fail: configure:25299: checking for wolfSSL_Init in -lwolfssl configure:25321: x86_64-openwrt-linux-musl-gcc -o conftest [snip] In file included from target-x86_64_musl/usr/include/wolfssl/wolfcrypt/dsa. h:33, from target-x86_64_musl/usr/include/wolfssl/wolfcrypt/asn_ public.h:35, from target-x86_64_musl/usr/include/wolfssl/ssl.h:35, from conftest.c:47: target-x86_64_musl/usr/include/wolfssl/wolfcrypt/integer.h:37:14: fatal err or: wolfssl/wolfcrypt/sp_int.h: No such file or directory #include ^~~~~~~~~~~~~~~~~~~~~~~~~~~~ compilation terminated. and in the end thus produces curl without https support: curl: (1) Protocol "https" not supported or disabled in libcurl So fix it, by making the working wolfSSL mandatory and error out in configure step when that's not the case: checking for wolfSSL_Init in -lwolfssl... no configure: error: --with-wolfssl but wolfSSL was not found or doesn't work References: https://github.com/openwrt/packages/issues/19005 References: https://github.com/openwrt/packages/issues/19547 Signed-off-by: Petr Štetiar Closes #9682 Daniel Stenberg (10 Oct 2022) - tool_getparam: pass in the snprintf("%.*s") string length as 'int' Reported by Coverity CID 1515928 Closes #9679 Paul Seligman (9 Oct 2022) - ws: minor fixes for web sockets without the CONNECT_ONLY flag - Fixed an issue where is_in_callback was getting cleared when using web sockets with debug logging enabled - Ensure the handle is is_in_callback when calling out to fwrite_func - Change the write vs. send_data decision to whether or not the handle is in CONNECT_ONLY mode. - Account for buflen not including the header length in curl_ws_send Closes #9665 Marc Hoersken (8 Oct 2022) - CI/cirrus: merge existing macOS jobs into a job matrix Ref: #9627 Reviewed-by: Philip H. Closes #9672 Daniel Stenberg (8 Oct 2022) - strcase: add and use Curl_timestrcmp This is a strcmp() alternative function for comparing "secrets", designed to take the same time no matter the content to not leak match/non-match info to observers based on how fast it is. The time this function takes is only a function of the shortest input string. Reported-by: Trail of Bits Closes #9658 - tool_getparam: split out data_urlencode() into its own function Closes #9673 - connect: fix Curl_updateconninfo for TRNSPRT_UNIX Reported-by: Vasiliy Ulyanov Fixes #9664 Closes #9670 - ws: fix Coverity complaints Coverity pointed out several flaws where variables remained uninitialized after forks. Follow-up to e3f335148adc6742728f Closes #9666 Marc Hoersken (7 Oct 2022) - CI/GHA: merge msh3 and openssl3 builds into linux workflow Continue work on merging all Linux workflows into one file. Follow up to #9501 Closes #9646 Daniel Stenberg (7 Oct 2022) - curl_ws_send.3: call the argument 'fragsize' Since WebSocket works with "fragments" not "frames" Closes #9668 - easy: avoid Intel error #2312: pointer cast involving 64-bit pointed-to type Follow-up to e3f335148adc6742728ff8 Closes #9669 - tool_main: exit at once if out of file descriptors If the main_checkfds function cannot create new file descriptors in an attempt to detect of stdin, stdout or stderr are closed. Also changed the check to use fcntl() to check if the descriptors are open, which avoids superfluously calling pipe() if they all already are. Follow-up to facfa19cdd4d0094 Reported-by: Trail of Bits Closes #9663 - websockets: remodeled API to support 63 bit frame sizes curl_ws_recv() now receives data to fill up the provided buffer, but can return a partial fragment. The function now also get a pointer to a curl_ws_frame struct with metadata that also mentions the offset and total size of the fragment (of which you might be receiving a smaller piece). This way, large incoming fragments will be "streamed" to the application. When the curl_ws_frame struct field 'bytesleft' is 0, the final fragment piece has been delivered. curl_ws_recv() was also adjusted to work with a buffer size smaller than the fragment size. (Possibly needless to say as the fragment size can now be 63 bit large). curl_ws_send() now supports sending a piece of a fragment, in a streaming manner, in addition to sending the entire fragment in a single call if it is small enough. To send a huge fragment, curl_ws_send() can be used to send it in many small calls by first telling libcurl about the total expected fragment size, and then send the payload in N number of separate invokes and libcurl will stream those over the wire. The struct curl_ws_meta() returns is now called 'curl_ws_frame' and it has been extended with two new fields: *offset* and *bytesleft*. To help describe the passed on data chunk when a fragment is delivered in many smaller pieces. The documentation has been updated accordingly. Closes #9636 Patrick Monnerat (7 Oct 2022) - docs/examples: avoid deprecated options in examples where possible Example programs targeting a deprecated feature/option are commented with a warning about it. Other examples are adapted to not use deprecated options. Closes #9661 Viktor Szakats (6 Oct 2022) - cmake: fix enabling websocket support Follow-up from 664249d095275ec532f55dd1752d80c8c1093a77 Closes #9660 - tidy-up: delete parallel/unused feature flags Detecting headers and lib separately makes sense when headers come in variations or with extra ones, but this wasn't the case here. These were duplicate/parallel macros that we had to keep in sync with each other for a working build. This patch leaves a single macro for each of these dependencies: - Rely on `HAVE_LIBZ`, delete parallel `HAVE_ZLIB_H`. Also delete CMake logic making sure these two were in sync, along with a toggle to turn off that logic, called `CURL_SPECIAL_LIBZ`. Also delete stray `HAVE_ZLIB` defines. There is also a `USE_ZLIB` variant in `lib/config-dos.h`. This patch retains it for compatibility and deprecates it. - Rely on `USE_LIBSSH2`, delete parallel `HAVE_LIBSSH2_H`. Also delete `LIBSSH2_WIN32`, `LIBSSH2_LIBRARY` from `winbuild/MakefileBuild.vc`, these have a role when building libssh2 itself. And `CURL_USE_LIBSSH`, which had no use at all. Also delete stray `HAVE_LIBSSH2` defines. - Rely on `USE_LIBSSH`, delete parallel `HAVE_LIBSSH_LIBSSH_H`. Also delete `LIBSSH_WIN32`, `LIBSSH_LIBRARY` and `HAVE_LIBSSH` from `winbuild/MakefileBuild.vc`, these were the result of copy-pasting the libssh2 line, and were not having any use. - Delete unused `HAVE_LIBPSL_H` and `HAVE_LIBPSL`. Reviewed-by: Daniel Stenberg Closes #9652 Daniel Stenberg (6 Oct 2022) - netrc: compare user name case sensitively User name comparisions in netrc need to match the case. Closes #9657 - CURLOPT_COOKIEFILE: insist on "" for enable-without-file The former way that also suggested using a non-existing file to just enable the cookie engine could lead to developers maybe a bit carelessly guessing a file name that will not exist, and then in a future due to circumstances, such a file could be made to exist and then accidentally libcurl would read cookies not actually meant to. Reported-by: Trail of bits Closes #9654 - tests/Makefile: remove run time stats from ci-test The ci-test is the normal makefile target invoked in CI jobs. This has been using the -r option to runtests.pl since a long time, but I find that it mostly just adds many lines to the test output report without anyone caring much about those stats. Remove it. Closes #9656 Patrick Monnerat (6 Oct 2022) - tool: reorganize function c_escape around a dynbuf This is a bit shorter and a lot safer. Substrings of unescaped characters are added by a single call to reduce overhead. Extend test 1465 to handle more kind of escapes. Closes #9653 Jay Satiro (5 Oct 2022) - CURLOPT_HTTPPOST.3: bolden the deprecation notice Ref: https://github.com/curl/curl/pull/9621 Closes https://github.com/curl/curl/pull/9637 John Bampton (5 Oct 2022) - misc: fix spelling in docs and comments also: remove outdated sentence Closes #9644 Patrick Monnerat (5 Oct 2022) - tool: avoid generating ambiguous escaped characters in --libcurl C string hexadecimal-escaped characters may have more than 2 digits. This results in a wrong C compiler interpretation of a 2-digit escaped character when followed by an hex digit character. The solution retained here is to represent such characters as 3-digit octal escapes. Adjust and extend test 1465 for this case. Closes #9643 Daniel Stenberg (5 Oct 2022) - configure: the ngtcp2 option should default to 'no' While still experimental. Bug: https://curl.se/mail/lib-2022-10/0007.html Reported-by: Daniel Hallberg Closes #9650 - CURLOPT_MIMEPOST.3: add an (inline) example Reported-by: Jay Satiro Bug: https://github.com/curl/curl/pull/9637#issuecomment-1268070723 Closes #9649 Viktor Szakats (5 Oct 2022) - Makefile.m32: exclude libs & libpaths for shared mode exes [ci skip] Exclude linker flags specifying depedency libs and libpaths, when building against `libcurl.dll`. In such case these options are not necessary (but may cause errors if not/wrongly configured.) Also move and reword a comment on `CPPFLAGS` to not apply to `UNICODE` options. These are necessary for all build targets. Closes #9651 Jay Satiro (5 Oct 2022) - runtests: fix uninitialized value on ignored tests - Don't show TESTFAIL message (ie tests failed which aren't ignored) if only ignored tests failed. Before: IGNORED: failed tests: 571 612 1056 TESTDONE: 1214 tests out of 1217 reported OK: 99% Use of uninitialized value $failed in concatenation (.) or string at ./runtests.pl line 6290. TESTFAIL: These test cases failed: After: IGNORED: failed tests: 571 612 1056 TESTDONE: 1214 tests out of 1217 reported OK: 99% Closes https://github.com/curl/curl/pull/9648 - cirrus: use make LDFLAGS=-all-static instead of curl_LDFLAGS - Correct the use of -all-static for static Windows CI builds. curl_LDFLAGS was removed from the makefile when metalink support was removed. LDFLAGS=-all-static is passed to make only, because it is not a valid option for configure compilation tests. Closes https://github.com/curl/curl/pull/9633 Viktor Szakats (4 Oct 2022) - Makefile.m32: fix regression with tool_hugehelp [ci skip] In a recent commit I mistakenly deleted this logic, after seeing a reference to a filename ending with `.cvs` and thinking it must have been long gone. Turns out this is an existing file. Restore the rule and the necessary `COPY` definitions with it. The restored logic is required for a successful build on a bare source tree (as opposed to a source release tarball). Also shorten an existing condition similar to the one added in this patch. Regression since 07a0047882dd3f1fbf73486c5dd9c15370877ad6 Closes #9645 - Makefile.m32: deduplicate build rules [ci skip] After this patch, we reduce the three copies of most `Makefile.m32` logic to one. This now resides in `lib/Makefile.m32`. It makes future updates easier, the code shorter, with a small amount of added complexity. `Makefile.m32` reduction: | | bytes | LOC total | blank | comment | code | |-------------------|-------:|----------:|-------:|---------:|------:| | 7.85.0 | 34772 | 1337 | 79 | 192 | 1066 | | before this patch | 17601 | 625 | 62 | 106 | 457 | | after this patch | 11680 | 392 | 52 | 104 | 236 | Details: - Change rules to create objects for the `v*` subdirs in the `lib` dir. This allows to use a shared compile rule and assumes that filenames are not (and will not be) colliding across these directories. `Makefile.m32` now also stores a list of these subdirs. They are changing rarely though. - Sync as much as possible between the three `Makefile.m32` scripts' rules and their source/target sections. - After this patch `CPPFLAGS` are all applied to the `src` sources once again. This matches the behaviour of cmake/autotools. Only zlib ones are actually required there. - Use `.rc` names from `Makefile.inc` instead of keeping a duplicate. - Change examples to link `libcurl.dll` by default. This makes building trivial, even as a cross-build: `CC=x86_64-w64-mingw32-gcc make -f Makefile.m32` To run them, you need to move/copy or add-to-path `libcurl.dll`. You can select static mode via `CFG=-static`. - List more of the `Makefile.m32` config variables. - Drop `.rc` support from examples. It made it fragile without much benefit. - Include a necessary system lib for the `externalsocket.c` example. - Exclude unnecessary systems libs when building in `-dyn` mode. Closes #9642 Daniel Stenberg (4 Oct 2022) - RELEASE-NOTES: synced - CURLOPT_COOKIELIST.3: fix formatting mistake Also, updated manpage-syntax.pl to make it detect this error in test 1173. Reported-by: ProceduralMan on github Fixes #9639 Closes #9640 Jay Satiro (4 Oct 2022) - connect: change verbose IPv6 address:port to [address]:port - Use brackets for the IPv6 address shown in verbose message when the format is address:port so that it is less confusing. Before: Trying 2606:4700:4700::1111:443... After: Trying [2606:4700:4700::1111]:443... Bug: https://curl.se/mail/archive-2022-02/0041.html Reported-by: David Hu Closes #9635 Viktor Szakats (3 Oct 2022) - Makefile.m32: major rework [ci skip] This patch overhauls `Makefile.m32` scripts, fixing a list of quirks, making its behaviour and customization envvars align better with other build systems, aiming for less code, that is easier to read, use and maintain. Details: - Rename customization envvars: `CURL_CC` -> `CC` `CURL_RC` -> `RC` `CURL_AR` -> `AR` `CURL_LDFLAG_EXTRAS_DLL` -> `CURL_LDFLAGS_LIB` `CURL_LDFLAG_EXTRAS_EXE` -> `CURL_LDFLAGS_BIN` - Drop `CURL_STRIP` and `CURL_RANLIB`. These tools are no longer used. - Accept `CFLAGS`, `CPPFLAGS`, `RCFLAGS`, `LDFLAGS` and `LIBS` envvars. - Drop `CURL_CFLAG_EXTRAS`, `CURL_LDFLAG_EXTRAS`, `CURL_RCFLAG_EXTRAS` in favor of the above. - Do not automatically enable `zlib` with `libssh2`. `zlib` is optional with `libssh2`. - Omit unnecessary `CPPFLAGS` options when building `curl.exe` and examples. - Drop support for deprecated `-winssl` `CFG` option. Use `-schannel` instead. - Avoid late evaluation where not necessary (`=` -> `:=`). - Drop support for `CURL_DLL_A_SUFFIX` to override the implib suffix. Instead, use the standard naming scheme by default: `libcurl.dll.a`. The toolchain recognizes the name, and selects it automatically when asking for a `-shared` vs. `-static` build. - Stop applying `strip` to `libcurl.a`. Follow-up from 16a58e9f93c7e89e1f87720199388bcfcfa148a4. There was no debug info to strip since then. - Stop setting `-O3`, `-W`, `-Wall` options. You can add these to `CFLAGS` as desired. - Always enable `-DCURL_DISABLE_OPENSSL_AUTO_LOAD_CONFIG` with OpenSSL, to avoid that vulnerability on Windows. - Add `-lbrotlicommon` to `LIBS` when using `brotli`. - Do not enable `-nghttp3` without `-ngtcp2`. - `-ssh2` and `-rtmp` options no longer try to auto-select a TLS-backend. You need to set the backend explicitly. This scales better and avoids issues with certain combinations (e.g. `libssh2` + `wolfssl` with no `schannel`). - Default to OpenSSL TLS-backend with `ngtcp2`. Possible to override via `NGTCP2_LIBS`. - Old, alternate method of enabling components (e.g. `SSH2=1`) no longer supported. - Delete `SPNEGO` references. They were no-ops. - Drop support for Win9x environments. - Allow setting `OPENSSL_LIBS` independently from `OPENSSL_LIBPATH`. - Support autotools/CMake `libssh2` builds by default. - Respect `CURL_DLL_SUFFIX` in `-dyn` mode when building `curl.exe` and examples. - Assume standard directory layout with `LIBCARES_PATH`. (Instead of the long gone embedded one.) - Stop static linking with c-ares by default. Add `CPPFLAGS=-DCARES_STATICLIB` to enable it. - Reorganize internal layout to avoid redundancy and emit clean diffs between src/lib and example make files. - Delete unused variables. - Code cleanups/rework. - Comment and indentation fixes. Closes #9632 - scripts/release-notes.pl: strip ci skip tag [ci skip] Ref: https://github.com/curl/curl/commit/e604a82cae922bf86403a94f5803ac5e4303 ae97#commitcomment-85637701 Reviewed-by: Daniel Stenberg Closes #9634 - Makefile.m32: delete legacy component bits [ci skip] - Drop auto-detection of OpenSSL 1.0.2 and earlier. Now always defaulting to OpenSSL 1.1.0 and later, LibreSSL and BoringSSL. - Drop `Invalid path to OpenSSL package` detection. OpenSSL has been using a standard file layout since 1.1.0, so this seems unnecessary now. - Drop special logic to enable Novell LDAP SDK support. - Drop special logic to enable OpenLDAP LDAP SDK support. This seems to be distinct from native OpenLDAP, with support implemented inside `lib/ldap.c` (vs. `lib/openldap.c`) back when the latter did not exist yet in curl. - Add `-lwldap32` only if there is no other LDAP library (either native OpenLDAP, or SDKs above) present. - Update `doc/INSTALL.md` accordingly. After this patch, it's necessary to make configration changes when using OpenSSL 1.0.2 or earlier, or the two LDAP SDKs. OpenSSL 1.0.2 and earlier: ``` export OPENSSL_INCLUDE = /outinc export OPENSSL_LIBPATH = /out export OPENSSL_LIBS = -lssl32 -leay32 -lgdi32 ``` Novell LDAP SDK, previously enabled via `USE_LDAP_NOVELL=1`: ``` export CURL_CFLAG_EXTRAS = -I/inc -DCURL_HAS_NOVELL_LDAPSDK export CURL_LDFLAG_EXTRAS = -L/lib/mscvc -lldapsdk -lldapssl -ll dapx ``` OpenLDAP LDAP SDK, previously enabled via `USE_LDAP_OPENLDAP=1`: ``` export CURL_CFLAG_EXTRAS = -I/include -DCURL_HAS_OPENLDAP_LDAPSD K export CURL_LDFLAG_EXTRAS = -L/lib -lldap -llber ``` I haven't tested these scenarios, and in general we recommend using a recent OpenSSL release. Also, WinLDAP (the Windows default) and OpenLDAP (via `-DUSE_OPENLDAP`) are the LDAP options actively worked on in curl. Closes #9631 Daniel Stenberg (2 Oct 2022) - vauth/ntlm.h: make line shorter than 80 columns Follow-up from 265fbd937 Viktor Szakats (1 Oct 2022) - docs: update sourceforge project links [ci skip] SourceForge projects can now choose between two hostnames, with .io and .net ending. Both support HTTPS by default now. Opening the other variant will perm-redirected to the one chosen by the project. The .io -> .net redirection is done insecurely. Let's update the URLs to point to the current canonical endpoints to avoid any redirects. Closes #9630 Daniel Stenberg (1 Oct 2022) - curl_url_set.3: document CURLU_APPENDQUERY proper Listed among the other supported flags. Reported-by: Robby Simpson Fixes #9628 Closes #9629 Viktor Szakats (1 Oct 2022) - Makefile.m32: cleanups and fixes [ci skip] - Add `-lcrypt32` once, and add it always for simplicity. - Delete broken link and reference to the pre-Vista WinIDN add-on. MS no longer distribute it. - Delete related `WINIDN_PATH` option. IDN is a system lib since Vista. - Sync `LIBCARES_PATH` default with the rest of dependencies. - Delete version numbers from dependency path defaults. - `libgsasl` package is now called `gsasl`. - Delete `libexpat` and `libxml2` references. No longer used by curl. - Delete `Edit the path below...` comments. We recommend to predefine those envvars instead. - `libcares.a` is not an internal dependency anymore. Stop using it as such. - `windres` `--include-dir` -> `-I`, `-F` -> `--target=` for readability. - Delete `STRIP`, `CURL_STRIP`, `AR` references from `src/Makefile.m32`. They were never used. - Stop to `clean` some objects twice in `src/Makefile.m32`. - Delete cvs-specific leftovers. - Finish resource support in examples make file. - Delete `-I/lib` from examples make file. - Fix copyright start year in examples make file. - Delete duplicate `ftpuploadresume` input in examples make file. - Sync OpenSSL lib order, `SYNC` support, `PROOT` use, dependency path defaults, variables names and other internal bits between the three make files. - `lib/Makefile.m32` accepted custom options via `DLL_LIBS` envvar. This was lib-specific and possibly accidental. Use `CURL_LDFLAG_EXTRAS_DLL` envvar for the same effect. - Fix linking `curl.exe` and examples to wrong static libs with auto-detected OpenSSL 1.0.2 or earlier. - Add `-lgdi32` for OpenSSL 1.0.2 and earlier only. - Add link to Novell LDAP SDK and use a relative default path. Latest version is from 2016, linked to an outdated OpenSSL 1.0.1. - Whitespace and comment cleanups. TODO in a next commit: Delete built-in detection/logic for OpenSSL 1.0.2 and earlier, the Novell LDAP SDK and the other LDAP SDK (which is _not_ OpenLDAP). Write up the necessary custom envvars to configure them. Closes #9616 Daniel Stenberg (30 Sep 2022) - RELEASE-NOTES: synced Matt Holt (30 Sep 2022) - HTTP3.md: update Caddy example Closes #9623 Daniel Stenberg (30 Sep 2022) - easy: fix the altsvc init for curl_easy_duphandle It was using the old #ifdef which nothing sets anymore Closes #9624 - GHA: build tests in a separate step from the running of them ... to make the output smaller for when you want to look at test failures. Removed the examples build from msh3 Closes #9619 Viktor Szakats (29 Sep 2022) - ldap: delete stray CURL_HAS_MOZILLA_LDAP reference Added in 68b215157fdf69612edebdb220b3804822277822, while adding openldap support. This is also the single mention of this constant in the source tree and also in that commit. Based on these, it seems like an accident. Delete this reference. Reviewed-by: Daniel Stenberg Closes #9625 - docs: spelling nits - MingW -> MinGW (Minimalist GNU for Windows) - f.e. -> e.g. - some whitespace and punctuation. Reviewed-by: Daniel Stenberg Closes #9622 Philip Heiduck (29 Sep 2022) - cirrus-ci: add macOS build with m1 Signed-off-by: Philip H <47042125+pheiduck@users.noreply.github.com> Closes #9565 Patrick Monnerat (29 Sep 2022) - lib: sanitize conditional exclusion around MIME The introduction of CURL_DISABLE_MIME came with some additional bugs: - Disabled MIME is compiled-in anyway if SMTP and/or IMAP is enabled. - CURLOPT_MIMEPOST, CURLOPT_MIME_OPTIONS and CURLOPT_HTTPHEADER are conditioned on HTTP, although also needed for SMTP and IMAP MIME mail uploads. In addition, the CURLOPT_HTTPHEADER and --header documentation does not mention their use for MIME mail. This commit fixes the problems above. Closes #9610 Thiago Suchorski (29 Sep 2022) - docs: minor grammar fixes Closes #9609 Daniel Stenberg (28 Sep 2022) - CURLSHOPT_UNLOCKFUNC.3: the callback as no 'access' argument Probably a copy and paste error from the lock function man page. Reported-by: Robby Simpson Fixes #9612 Closes #9613 - CURLOPT_ACCEPT_ENCODING.3: remove "four" as they are five ... instead just list the supported encodings. Reported-by: ProceduralMan on github Fixes #9614 Closes #9615 Dan Fandrich (28 Sep 2022) - tests: Remove a duplicated keyword - docs: document more server names for test files Daniel Stenberg (28 Sep 2022) - altsvc: reject bad port numbers The existing code tried but did not properly reject alternative services using negative or too large port numbers. With this fix, the logic now also flushes the old entries immediately before adding a new one, making a following header with an illegal entry not flush the already stored entry. Report from the ongoing source code audit by Trail of Bits. Adjusted test 356 to verify. Closes #9607 - functypes: provide the recv and send arg and return types This header is for providing the argument types for recv() and send() when built to not use a dedicated config-[platfor].h file. Remove the slow brute-force checks from configure and cmake. This change also removes the use of the types for select, as they were not used in code. Closes #9592 - urlapi: reject more bad characters from the host name field Extended test 1560 to verify Report from the ongoing source code audit by Trail of Bits. Closes #9608 - configure: deprecate builds with small curl_off_t If curl_off_t turns out to be smaller than 8 bytes, --with-n64-deprecated needs to be used to allow the build to continue. This is to highlight the fact that support for such builds is going away next year. Also mentioned in DEPRECATED.md Closes #9605 Patrick Monnerat (27 Sep 2022) - http, vauth: always provide Curl_allow_auth_to_host() functionality This function is currently located in the lib/http.c module and is therefore disabled by the CURL_DISABLE_HTTP conditional token. As it may be called by TLS backends, disabling HTTP results in an undefined reference error at link time. Move this function to vauth/vauth.c to always provide it and rename it as Curl_auth_allowed_to_host() to respect the vauth module naming convention. Closes #9600 Daniel Stenberg (27 Sep 2022) - ngtcp2: fix C89 compliance nit - openssl: make certinfo available for QUIC Curl_ossl_certchain() is now an exported function in lib/vtls/openssl.c that can also be used from quiche.c and ngtcp2.c to get the cert chain for QUIC connections as well. The *certchain function was moved to the top of the file for this reason. Reported-by: Eloy Degen Fixes #9584 Closes #9597 - RELEASE-NOTES: synced - DEPRECATE.md: Support for systems without 64 bit data types Closes #9604 Patrick Monnerat (27 Sep 2022) - tests: skip mime/form tests when mime is not built-in Closes #9596 Daniel Stenberg (27 Sep 2022) - url: rename function due to name-clash in Watt-32 Follow-up to 2481dbe5f4f58 and applies the change the way it was intended. Viktor Szakats (26 Sep 2022) - windows: adjust name of two internal public functions According to `docs/INTERNALS.md`, internal function names spanning source files start with uppercase `Curl_`. Bring these two functions in alignment with this. This also stops exporting them from `libcurl.dll` in autotools builds. Reviewed-by: Daniel Stenberg Closes #9598 Gisle Vanem (26 Sep 2022) - url: rename function due to name-clash in Watt-32 Since the commit 764c958c52edb427f39, there was a new function called resolve_ip(). This clashes with an internal function in Watt-32. Closes #9585 Jay Satiro (26 Sep 2022) - schannel: ban server ALPN change during recv renegotiation By the time schannel_recv is renegotiating the connection, libcurl has already decided on a protocol and it is too late for the server to select a protocol via ALPN except for the originally selected protocol. Ref: https://github.com/curl/curl/issues/9451 Closes https://github.com/curl/curl/pull/9463 Daniel Stenberg (26 Sep 2022) - url: a zero-length userinfo part in the URL is still a (blank) user Adjusted test 1560 to verify Reported-by: Jay Satiro Fixes #9088 Closes #9590 Viktor Szakats (25 Sep 2022) - autotools: allow --enable-symbol-hiding with windows This local autotools logic was put in place in 9e24b9c7afbcb81120af4cf3f6cdee49a06d8224 (in 2012) which disabled it for Windows unconditionally. Testing reveals that it actually works with tested toolchains (mingw-w64 and CI ones), so let's allow this build feature on that platform. Bringing this in sync with CMake, which already supported this. Reviewed-by: Jay Satiro Closes #9586 - autotools: reduce brute-force when detecting recv/send arg list autotools uses brute-force to detect `recv`/`send`/`select` argument lists, by interating through _all_ argument type combinations on each `./configure` run. This logic exists since 01fa02d0b545e1433dced2430561f8c0c72b74a9 (from 2006) and was a bit later extended with Windows support. This results in a worst-case number of compile + link cycles as below: - `recv`: 96 - `send`: 192 - `select`: 60 Total: 348 (the number of curl C source files is 195, for comparison) Notice that e.g. curl-for-win autotools builds require two `./configure` invocations, doubling these numbers. `recv` on Windows was especially unlucky because `SOCKET` (the correct choice there) was listed _last_ in one of the outer trial loops. This resulted in lengthy waits while autotools was trying all invalid combinations first, wasting cycles, disk writes and slowing down iteration. This patch reduces the amount of idle work by reordering the tests in a way to succeed first on a well-known platform such as Windows, and also on non-Windows by testing for POSIX prototypes first, on the assumption that these are the most likely candidates these days. (We do not touch `select`, where the order was already optimal for these platforms.) For non-Windows, this means to try a return value of `ssize_t` first, then `int`, reordering the buffer argument type to try `void *` first, then `byte *`, and prefer the `const` flavor with `send`. If we are here, also stop testing for `SOCKET` type in non-Windows builds. After the patch, detection on Windows is instantaneous. It should also be faster on popular platforms such as Linux and BSD-based ones. If there are known-good variations for other platforms, they can also be fast-tracked like above, given a way to check for that platform inside the autotools logic. Reviewed-by: Daniel Stenberg Closes #9591 Daniel Stenberg (23 Sep 2022) - TODO: Provide the error body from a CONNECT response Spellchecked-by: Jay Satiro Closes #9513 Closes #9581 Viktor Szakats (23 Sep 2022) - windows: autotools .rc warnings fixup Move `LT_LANG([Windows Resource])` after `XC_LIBTOOL`, fixing: - Warnings when running `autoreconf -fi`. - Warning when compiling .rc files: libtool: compile: unable to infer tagged configuration libtool: error: specify a tag with '--tag' Follow up to 6de7322c03d5b4d91576a7d9fc893e03cc9d1057 Ref: https://github.com/curl/curl/pull/9521#issuecomment-1256291156 Suggested-by: Patrick Monnerat Closes #9582 Randall S. Becker (23 Sep 2022) - curl_setup: disable use of FLOSS for 64-bit NonStop builds Older 32-bit builds currently need FLOSS. This dependency may be removed in future OS releases. Signed-off-by: Randall S. Becker Closes #9575 Patrick Monnerat (23 Sep 2022) - tool: remove dead code Add a debug assertion to verify protocols included/excluded in a set are always tokenized. Follow-up to commit 677266c. Closes #9576 - lib: prepare the incoming of additional protocols Move the curl_prot_t to its own conditional block. Introduce symbol PROTO_TYPE_SMALL to control it. Fix a cast in a curl_prot_t assignment. Remove an outdated comment. Follow-up to cd5ca80. Closes #9534 Daniel Stenberg (23 Sep 2022) - msh3: change the static_assert to make the code C89 - bearssl: make it proper C89 compliant - curl-compilers.m4: for gcc + want warnings, set gnu89 standard To better verify that the code is C89 Closes #9542 Patrick Monnerat (22 Sep 2022) - lib517: fix C89 constant signedness In C89, positive integer literals that overflow an int but not an unsigned int may be understood as a negative int. lib517.c:129:3: warning: this decimal constant is unsigned only in ISO C90 {"Sun, 06 Nov 2044 08:49:37 GMT", 2362034977 }, ^ Closes #9572 Daniel Stenberg (22 Sep 2022) - mprintf: use snprintf if available This is the single place in libcurl code where it uses the "native" s(n)printf() function. Used for writing floats. The use has been reviewed and vetted and uses a HUGE target buffer, but switching to snprintf() still makes this safer and removes build-time warnings. Reported-by: Philip Heiduck Fixes #9569 Closes #9570 - docs: tag curl options better in man pages As it makes them links in the HTML versions. Verified by the extended test 1176 - symbols-in-versions: CURLOPT_ENCODING is deprecated since 7.21.6 - manpage-syntax.pl: all libcurl option symbols should be \fI-tagged ... as that makes them links to their corresponding man page. This script is used for test 1173. Closes #9574 - RELEASE-NOTES: synced Patrick Monnerat (22 Sep 2022) - tool: remove protocol count limitation Replace bit mask protocol sets by null-terminated arrays of protocol tokens. These are the addresses of the protocol names returned by curl_version_info(). Protocol names are sorted case-insensitively before output to satisfy CI tests matches consistency. The protocol list returned by curl_version_info() is augmented with all RTMP protocol variants. Test 1401 adjusted for new alpha ordered output. Closes #9546 Daniel Stenberg (22 Sep 2022) - test972: verify the output without using external tool It seems too restrictive to assume and use an external tool to verify the JSON. This now verifies the outut byte per byte. We could consider building a local "JSON verifyer" in a future. Remove 'jsonlint' from the CI job. Reported-by: Marcel Raad Fixes #9563 Closes #9564 - hostip: lazily wait to figure out if IPv6 works until needed The check may take many milliseconds, so now it is performed once the value is first needed. Also, this change makes sure that the value is not used if the resolve is set to be IPv4-only. Closes #9553 - curl.h: fix mention of wrong error code in comment The same error and comment were also used and is now corrected in CURLOPT_SSH_KEYFUNCTION.3 - symbol-scan.pl: scan and verify .3 man pages This script now also finds all .3 man pages in docs/include and docs/include/opts, extracts all uses of CURL* symbols and verifies that all symbols mentioned in docs are defined in public headers. A "global symbol" is one of those matching a known prefix and the script make s an attempt to check all/most of them. Just using *all* symbols that match CURL* proved matching a little too many other references as well and turned difficult turning into something useful. Closes #9544 - symbols-in-versions: add missing LIBCURL* symbols - symbol-scan.pl: also check for LIBCURL* symbols Closes #9544 - docs/libcurl/symbols-in-versions: add several missing symbols - test1119: scan all public headers Previously this test only scanned a subset of the headers, which made us accidentally miss symbols that were provided in the others. Now, the script iterates over all headers present in include/curl. Closes #9544 Patrick Monnerat (21 Sep 2022) - examples/chkspeed: improve portability The example program chkspeed uses strncasecmp() which is not portable across systems. Replace calls to this function by tests on characters. Closes #9562 Daniel Stenberg (21 Sep 2022) - easy: fix the #include order The mentioned "last 3 includes" order should be respected. easy_lock.h should be included before those three. Reported-by: Yuriy Chernyshov Fixes #9560 Closes #9561 - docs: spellfixes Pointed by the new CI job - GHA: spellcheck This spellchecker checks markdown files. For this reason this job converts all man pages in the repository to markdown with pandoc before the check runs. The perl script 'cleanspell' filters out details from the man page in the process, to avoid the spellchecker trying to spellcheck things it can't. Like curl specific symbols and the SYNOPSIS and EXAMPLE sections of libcurl man pages. The spell checker does not check words in sections that are within pre, strong and em tags. 'spellcheck.words' is a custom word list with additional accepted words. Closes #9523 - connect: fix the wrong error message on connect failures The "Failed to connect to" message after a connection failure would include the strerror message based on the presumed previous socket error, but in times it seems that error number is not set when reaching this code and therefore it would include the wrong error message. The strerror message is now removed from here and the curl_easy_strerror error is used instead. Reported-by: Edoardo Lolletti Fixes #9549 Closes #9554 - httpput-postfields.c: shorten string for C89 compliance httpput-postfields.c:41:3: error: string length ‘522’ is greater than the length ‘509’ ISO C90 compilers are required to support [-Woverlength-str ings] 41 | "this chapter."; | ^~~~~~~~~~~~~~~ Closes #9555 - ws: fix a C89 compliance nit Closes #9541 Patrick Monnerat (21 Sep 2022) - unit test 1655: make it C89-compliant Initializations performed in unit test 1655 use automatic variables in aggregates and thus can only be computed at run-time. Using gcc in C89 dialect mode produces warning messages like: unit1655.c:96:7: warning: initializer element is not computable at load time [-Wpedantic] 96 | { toolong, DOH_DNS_NAME_TOO_LONG }, /* expect early failure */ | ^~~~~~~ Fix the problem by converting these automatic pointer variables to static arrays. Closes #9551 Tobias Schaefer (20 Sep 2022) - curl_strequal.3: fix typo Closes #9548 Dmitry Karpov (20 Sep 2022) - resolve: make forced IPv4 resolve only use A queries This protects IPv4-only transfers from undesired bad IPv6-related side effects and make IPv4 transfers in dual-stack libcurl behave the same way as in IPv4 single-stack libcurl. Closes #9540 Daniel Stenberg (20 Sep 2022) - RELEASE-NOTES: synced - winbuild/MakefileBuild.vc: handle spaces in libssh(2) include paths Patched-by: Mark Itzcovitz Bug: https://curl.se/mail/lib-2022-09/0038.html Closes #9536 - TODO: Reduce CA certificate bundle reparsing By adding some sort of cache. Reported-by: Michael Drake Closes #9379 Closes #9538 Marc Hoersken (19 Sep 2022) - CI/GHA: cancel outdated CI runs on new PR changes Avoid letting outdated CI runs continue if a PR receives new changes. Outside a PR we let them continue running by tying the concurrency to the commit hash instead. Also only let one CodeQL or Hacktoberfest job run at a time. Other CI platforms we use have this build in, but GitHub unfortunately neither by default nor with a simple option. This saves CI resources and therefore a little energy. Approved-by: Daniel Stenberg Approved-by: Max Dymond Closes #9533 Daniel Stenberg (19 Sep 2022) - docs: fix proselint complaints - GHA: run proselint on markdown files Co-authored-by: Marc Hörsken Closes #9520 - lib: the number four in a sequence is the "fourth" Spelling is hard Closes #9535 John Bampton (19 Sep 2022) - misc: fix spelling in two source files Closes #9529 Viktor Szakats (18 Sep 2022) - windows: add .rc support to autotools builds After this update autotools builds will compile and link `.rc` resources to Windows executables. Bringing this feature on par with CMake and Makefile.m32 builds. And also making it unnecessary to improvise these steps manually, while monkey patching build files, e.g. [0]. You can customize the resource compiler via the `RC` envvar, and its options via `RCFLAGS`. This harmless warning may appear throughout the build, even though the autotools manual documents [1] `RC` as a valid tag, and it fails when omitting one: `libtool: error: ignoring unknown tag RC` [0] https://github.com/curl/curl-for-win/blob/535f19060d4b708f72e75dd849409ce 50baa1b84/curl-autotools.sh#L376-L382 [1] https://www.gnu.org/software/libtool/manual/html_node/Tags.html Closes #9521 Marc Hoersken (18 Sep 2022) - CI/linkcheck: only run if a Markdown file is changed This saves CI resources and therefore a little energy. Reviewed-by: Max Dymond Closes #9531 - README.md: add GHA status badges for Linux and macOS builds This makes sense now that Linux builds are being consolidated. Approved-by: Daniel Stenberg Closes #9530 [skip ci] Daniel Stenberg (17 Sep 2022) - misc: null-terminate Make use of this term consistently. Closes #9527 Marc Hoersken (17 Sep 2022) - CI/GHA: merge intel CC and more TLS libs into linux workflow Continue work on merging all Linux workflows into one file. Reviewed-by: Max Dymond Follow up to #9501 Closes #9514 Patrick Monnerat (17 Sep 2022) - lib1597: make it C89-compliant again Automatic variable addresses cannot be used in an initialisation aggregate. Follow-up to 9d51329 Reported-by: Daniel Stenberg Fixes: #9524 Closes #9525 Daniel Stenberg (17 Sep 2022) - tool_libinfo: silence "different 'const' qualifiers" in qsort() MSVC 15.0.30729.1 warned about it Follow-up to dd2a024323dcc Closes #9522 Patrick Monnerat (16 Sep 2022) - docs: tell about disabled protocols in CURLOPT_*PROTOCOLS_STR. Disabled protocols are now handled as if they were unknown. Also update the possible protocol list. - cli tool: do not use disabled protocols As they are now rejected by the library, take care of not passing disabled protocol names to CURLOPT_PROTOCOLS_STR and CURLOPT_REDIR_PROTOCOLS_STR. Rather than using the CURLPROTO_* constants, dynamically assign protocol numbers based on the order they are listed by curl_version_info(). New type proto_set_t implements prototype bit masks: it should therefore be large enough to accomodate all library-enabled protocols. If not, protocol numbers beyond the bit count of proto_set_t are recognized but "inaccessible": when used, a warning is displayed and the value is ignored. Should proto_set_t overflows, enabled protocols are reordered to force those having a public CURLPROTO_* representation to be accessible. Code has been added to subordinate RTMP?* protocols to the presence of RTMP in the enabled protocol list, being returned by curl_version_info() or not. - setopt: use the handler table for protocol name to number conversions This also returns error CURLE_UNSUPPORTED_PROTOCOL rather than CURLE_BAD_FUNCTION_ARGUMENT when a listed protocol name is not found. A new schemelen parameter is added to Curl_builtin_scheme() to support this extended use. Note that disabled protocols are not recognized anymore. Tests adapted accordingly. Closes #9472 Daniel Stenberg (16 Sep 2022) - altsvc: use 'h3' for h3 Since the official and real version has been out for a while now and servers are deployed out there using it, there is no point in sticking to h3-29. Reported-by: ウさん Fixes #9515 Closes #9516 chemodax (16 Sep 2022) - winbuild: Use NMake batch-rules for compilation - Invoke cl compiler once for each group of .c files. This is significantly improves compilation time. For example in my environment: 40 s --> 20 s. Prior to this change cl was invoked per .c file. Closes https://github.com/curl/curl/pull/9512 Daniel Stenberg (16 Sep 2022) - ws: the infof() flags should be %zu Follow-up to e5e9e0c5e49ae0 Closes #9518 - curl: warn for --ssl use, considered insecure Closes #9519 Sergey Bronnikov (16 Sep 2022) - curl_escape.3: fix typo lengthf -> length Closes #9517 Daniel Stenberg (16 Sep 2022) - mailmap: merge Philip Heiduck's two addresses into one - test1948: verify PUT + POST reusing the same handle Reproduced #9507, verifies the fix - setopt: when POST is set, reset the 'upload' field Reported-by: RobBotic1 on github Fixes #9507 Closes #9511 Marc Hoersken (15 Sep 2022) - github: initial CODEOWNERS setup for CI configuration Reviewed-by: Daniel Stenberg Reviewed-by: Marcel Raad Reviewed-by: Max Dymond Closes #9505 [skip ci] Philip Heiduck (15 Sep 2022) - CI: optimize some more dependencies install Signed-off-by: Philip Heiduck Closes #9500 Marc Hoersken (15 Sep 2022) - CI/GHA: merge event-based and NSS into new linux workflow Continue work on merging all Linux workflows into one file. Follow up to #9501 Closes #9506 Daniel Stenberg (15 Sep 2022) - include/curl/websockets.h: add extern "C" for C++ Reported-by: n0name321 on github Fixes #9509 Closes #9510 - lib1560: extended to verify detect/reject of unknown schemes ... when no guessing is allowed. - urlapi: detect scheme better when not guessing When the parser is not allowed to guess scheme, it should consider the word ending at the first colon to be the scheme, independently of number of slashes. The parser now checks that the scheme is known before it counts slashes, to improve the error messge for URLs with unknown schemes and maybe no slashes. When following redirects, no scheme guessing is allowed and therefore this change effectively prevents redirects to unknown schemes such as "data". Fixes #9503 - strerror: improve two URL API error messages Marc Hoersken (14 Sep 2022) - CI/GHA: merge bearssl and hyper into initial linux workflow Begin work on merging all Linux workflows into one file. Closes #9501 Daniel Stenberg (14 Sep 2022) - RELEASE-NOTES: synced - cmake: define BUILDING_LIBCURL in lib/CMakeLists, not config.h Since the config file might also get included by the tool code at times. This syncs with how other builds do it. Closes #9498 - tool_hugehelp: make hugehelp a blank macro when disabled Closes #9485 - getparameter: return PARAM_MANUAL_REQUESTED for -M even when disabled ... to improve the output in this situation. Now it doesn't say "option unknown" anymore. Closes #9485 - setopt: fix compiler warning Follow-up to cd5ca80f00d2 closes #9502 Philip Heiduck (13 Sep 2022) - CI: skip make, do make install at once for dependencies Signed-off-by: Philip Heiduck Closes #9477 Daniel Stenberg (13 Sep 2022) - formdata: typecast the va_arg return value To avoid "enumerated type mixed with another type" warnings Follow-up from 0f52dd5fd5aa3592691a Closes #9499 - RELEASE-PROCEDURE.md: mention patch releases - When to make them and how to argue for them - Refreshed the release date list Closes #9495 - urldata: use a curl_prot_t type for storing protocol bits This internal-use-only storage type can be bumped to a curl_off_t once we need to use bit 32 as the previous 'unsigned int' can no longer hold them all then. The websocket protocols take bit 30 and 31 so they are the last ones that fit within 32 bits - but cannot properly be exported through APIs since those use *signed* 32 bit types (long) in places. Closes #9481 zhanghu on xiaomi (13 Sep 2022) - formdata: fix warning: 'CURLformoption' is promoted to 'int' curl/lib/formdata.c: In function 'FormAdd': curl/lib/formdata.c:249:31: warning: 'CURLformoption' is promoted to 'int' wh en passed through '...' 249 | option = va_arg(params, CURLformoption); | ^ curl/lib/formdata.c:249:31: note: (so you should pass 'int' not 'CURLformopti on' to 'va_arg') curl/lib/formdata.c:249:31: note: if this code is reached, the program will a bort Closes #9484 Daniel Stenberg (13 Sep 2022) - CURLOPT_CONNECT_ONLY.3: for ws(s) as well and correct the version number for when that support comes. Even if it is still experimental for WebSocket. Closes #9487 - tool_operate: avoid a few #ifdefs for disabled-libcurl builds By providing empty macros in the header file instead, the code gets easier to read and yet is disabled on demand. Closes #9486 a1346054 on github (13 Sep 2022) - scripts: use `grep -E` instead of `egrep` egrep is deprecated Closes #9491 Hayden Roche (13 Sep 2022) - wolfSSL: fix session management bug. Prior to this commit, non-persistent pointers were being used to store sessions. When a WOLFSSL object was then freed, that freed the session it owned, and thus invalidated the pointer held in curl's cache. This commit makes it so we get a persistent (deep copied) session pointer that we then add to the cache. Accordingly, wolfssl_session_free, which was previously a no-op, now needs to actually call SSL_SESSION_free. This bug was discovered by a wolfSSL customer. Closes #9492 Daniel Stenberg (13 Sep 2022) - docs: use "WebSocket" in singular This is how the RFC calls the protocol. Also rename the file in docs/ to WEBSOCKET.md in uppercase to match how we have done it for many other protocol docs in similar fashion. Add the WebSocket docs to the tarball. Closes #9496 Marcel Raad (12 Sep 2022) - ws: fix build without `USE_WEBSOCKETS` The curl.h include is required unconditionally. - ws: add missing curl.h include A conflict between commits 664249d0952 and e5839f4ee70 broke the build. Daniel Stenberg (12 Sep 2022) - ws: fix an infof() call to use %uz for size_t output Detected by Coverity, CID 1514665. Closes #9480 Marcel Raad (12 Sep 2022) - curl_setup: include only system.h instead of curl.h As done before commit 9506d01ee50. Ref: https://github.com/curl/curl/pull/9375#discussion_r957010158 Closes https://github.com/curl/curl/pull/9453 - lib: add missing limits.h includes Closes https://github.com/curl/curl/pull/9453 - lib and tests: add missing curl.h includes Closes https://github.com/curl/curl/pull/9453 - curl_setup: include curl.h after platform setup headers The platform setup headers might set definitions required for the includes in curl.h. Ref: https://github.com/curl/curl/pull/9375#discussion_r956998269 Closes https://github.com/curl/curl/pull/9453 Benjamin Loison (12 Sep 2022) - docs: correct missing uppercase in Markdown files To detect these typos I used: ``` clear && grep -rn '\. [a-z]' . | uniq | grep -v '\. lib' | grep -v '[0-9]\. [ a-z]' | grep -v '\.\. [a-z]' | grep -v '\. curl' | grep -v 'e.g. [a-z]' | gre p -v 'eg. [a-z]' | grep -v '\etc. [a-z]' | grep -v 'i.e\. [a-z]' | grep --col or=always '\. [a-z]' | grep '\.md' ``` Closes #9474 Daniel Stenberg (12 Sep 2022) - tool_setopt: use better English in --libcurl source comments Like this: XYZ was set to an object pointer ABC was set to a function pointer Closes #9475 - setopt: make protocol2num use a curl_off_t for the protocol bit ... since WSS does not fit within 32 bit. Bug: https://github.com/curl/curl/pull/9467#issuecomment-1243014887 Closes #9476 - RELEASE-NOTES: synced - configure: polish the grep -E message a bit further Suggested-by: Emanuele Torre Closes #9473 - GHA: add a gcc-11 -O3 build using OpenSSL Since -O3 might trigger other warnings Closes #9454 Patrick Monnerat (11 Sep 2022) - content_encoding: use writer struct subclasses for different encodings The variable-sized encoding-specific storage of a struct contenc_writer currently relies on void * alignment that may be insufficient with regards to the specific storage fields, although having not caused any problems yet. In addition, gcc 11.3 issues a warning on access to fields of partially allocated structures that can occur when the specific storage size is 0: content_encoding.c: In function ‘Curl_build_unencoding_stack’: content_encoding.c:980:21: warning: array subscript ‘struct contenc_write r[0]’ is partly outside array bounds of ‘unsigned char[16]’ [-Warray-bo unds] 980 | writer->handler = handler; | ~~~~~~~~~~~~~~~~^~~~~~~~~ In file included from content_encoding.c:49: memdebug.h:115:29: note: referencing an object of size 16 allocated by ‘c url_dbg_calloc’ 115 | #define calloc(nbelem,size) curl_dbg_calloc(nbelem, size, __LINE__, __FILE__) | ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ ~~~~~~~~~~ content_encoding.c:977:60: note: in expansion of macro ‘calloc’ 977 | struct contenc_writer *writer = (struct contenc_writer *)calloc(1 , sz); To solve both these problems, the current commit replaces the contenc_writer/params structure pairs by "subclasses" of struct contenc_writer. These are structures that contain a contenc_writer at offset 0. Proper field alignment is therefore handled by the compiler and full structure allocation is performed, silencing the warnings. Closes #9455 Daniel Stenberg (11 Sep 2022) - configure: correct the wording when checking grep -E The check first checks that grep -E works, and only as a fallback tries to find and use egrep. egrep is deprecated. This change only corrects the output wording, not the checks themselves. Closes #9471 Viktor Szakats (10 Sep 2022) - websockets: sync prototypes in docs with implementation [ci skip] Docs for the new send/recv functions synced with the committed versions of these. Closes #9470 Daniel Stenberg (10 Sep 2022) - setopt: make protocols2num() work with websockets So that CURLOPT_PROTOCOLS_STR and CURLOPT_REDIR_PROTOCOLS_STR can specify those as well. Reported-by: Patrick Monnerat Bug: https://curl.se/mail/lib-2022-09/0016.html Closes #9467 - curl/websockets.h: remove leftover bad typedef Just a leftover trace of a development thing that did not stay like that. Reported-by: Marc Hörsken Fixes #9465 Cloes #9466 Orgad Shaneh (10 Sep 2022) - fix Cygwin/MSYS compilation _getpid is Windows API. On Cygwin variants it should remain getpid. Fixes #8220 Closes #9255 Marc Hoersken (10 Sep 2022) - GHA: prepare workflow merge by aligning structure again Closes #9413 Daniel Stenberg (9 Sep 2022) - docs: the websockets symbols are added in 7.86.0 Nothing else Closes #9459 - tests/libtest/Makefile.inc: fixup merge conflict mistake - EXPERIMENTAL.md: add WebSockets - appveyor: enable websockets - cirrus: enable websockets in the windows builds - GHA: add websockets to macos, openssl3 and hyper builds - tests: add websockets tests - add websockets support to sws - 2300: first very basic websockets test - 2301: first libcurl test for ws (not working yet) - 2302: use the ws callback - 2303: test refused upgrade - curl_ws_meta: initial implementation - curl_ws_meta.3: added docs - ws: initial websockets support Closes #8995 - version: add ws + wss - libtest/lib1560: test basic websocket URL parsing - configure: add --enable-websockets - docs/WebSockets.md: docs - test415: verify Content-Length parser with control code + negative value - strtoofft: after space, there cannot be a control code With the change from ISSPACE() to ISBLANK() this function no longer deals with (ignores) control codes the same way, which could lead to this function returning unexpected values like in the case of "Content-Length: \r-12354". Follow-up to 6f9fb7ec2d7cb389a0da5 Detected by OSS-fuzz Bug: https://bugs.chromium.org/p/oss-fuzz/issues/detail?id=51140 Assisted-by: Max Dymond Closes #9458 - headers: reset the requests counter at transfer start If not, reusing an easy handle to do a subsequent transfer would continue the counter from the previous invoke, which then would make use of the header API difficult/impossible as the request counter mismatched. Add libtest 1947 to verify. Reported-by: Andrew Lambert Fixes #9424 Closes #9447 Jay Satiro (8 Sep 2022) - header: define public API functions as extern c Prior to this change linker errors would occur if curl_easy_header or curl_easy_nextheader was called from a C++ unit. Bug: https://github.com/curl/curl/issues/9424#issuecomment-1238818007 Reported-by: Andrew Lambert Closes https://github.com/curl/curl/pull/9446 Daniel Stenberg (8 Sep 2022) - http2: make nghttp2 less picky about field whitespace In nghttp2 1.49.0 it returns error on leading and trailing whitespace in header fields according to language in the recently shipped RFC 9113. nghttp2 1.50.0 introduces an option to switch off this strict check and this change enables this option by default which should make curl behave more similar to how it did with nghttp2 1.48.0 and earlier. We might want to consider making this an option in the future. Closes #9448 - RELEASE-NOTES: synced And bump to 7.86.0 for the pending next release