- Parse the input string without modifying it.
Prior to this change a segfault could occur if the input string was
const because the tokenizer modified the input string. For example if
the user set CURLOPT_COOKIELIST to a const string then libcurl would
likely cause a crash when modifying that string. Even if the string was
not const or a crash did not occur there was still the incorrect and
unexpected modification of the user's input string.
This issue was caused by 30da1f59 (precedes 8.11.0) which refactored
some options parsing and eliminated the copy of the input string. Also,
an earlier commit f88cc654 incorrectly cast the input pointer when
passing it to strtok.
Co-authored-by: Daniel Stenberg
Closes https://github.com/curl/curl/pull/15826
build:
- autotools: fix to build generated sources for the `tidy` target.
- autotools: allow passing custom clang-tidy options via
`CURL_CLANG_TIDYFLAGS` env.
- cmake: add `CURL_CLANG_TIDY` option to configure for `clang-tidy`.
Also add:
- `CLANG_TIDY` variable to customize the `clang-tidy` tool.
- `CURL_CLANG_TIDYFLAGS` to pass custom options to `clang-tidy`.
- apply `--enable-werror` and `-DCURL_WERROR=ON` to `clang-tidy`.
CI/GHA:
- add clang-tidy job for Linux, using autotools and clang-tidy v18.
This one needs to disable `clang-analyzer-valist.Uninitialized`
to avoid false positives:
https://github.com/llvm/llvm-project/issues/40656
Duration: 5.5 minutes
- add clang-tidy job for macOS, using cmake and clang-tidy v19.
This one also covers tests and examples, and doesn't hit the false
positives seen with llvm v18 and earlier.
Duration: 4.5 minutes
- Linux/macOS: skip installing test dependencies when not building or
running tests.
fix fallouts reported by `clang-tidy`:
- lib:
- cf-h2-proxy: unused assignment in non-debug builds.
- cf-socket: silence warning.
FIXME: https://github.com/curl/curl/pull/15825#issuecomment-2561867769
- ftp: NULL passed to `strncmp()`.
- http2: NULL-ptr deref.
- mprintf: silence warning.
- src/tool_writeout: NULL passed to `fputs()`.
- examples:
- invalid file pointers.
- missing `fclose()`.
- tests:
- http/clients/hx-download: memory leaks on error.
- http/clients/hx-download: memory leak on repeat `-r` option.
- server: double `fclose()`.
https://www.man7.org/linux/man-pages/man3/fclose.3.html
- server: invalid file pointer/handle.
- server/getpart: unused assignments.
- server/mqttd: leak on failed `realloc()`.
- server/tftpd: NULL passed to `strcmp()`.
Closes#15825
The new detection method also allows to enable librtmp without using
OpenSSL as a curl TLS backend at the same time.
Also:
- implement manual version detection for librtmp.
Version info is in hex. With CMake 3.13 and newer, extract it as a hex
number. With earlier CMake version, just strip the leading zeroes.
Doing more here seems overkill because librtmp has been standing
at 2.3/2.4 for a decade now. Bumping into hex digits seems unlikely
before deprecating CMake 3.13 support.
librtmp advertises v2.4 via its `pkg-config` module, and v2.3 via
its public header. The latter shows up in `curl -V` and either can
be shown at configure-time depending on detection method.
This isn't a curl bug.
- GHA/macos: enable rtmp in a job.
- apply the "half-detection" fix to the Find module.
`librtmp` is also affected (in CI too), because it depends on libssl and
libcrypto.
Closes#15832
Make the Find modules set and return their respective `pkg-config`
module name(s) to the CMake build process, which then adds those
to the `Requires:` list.
Before this patch, `pkg-config` module names were maintainted in two
separate places. After this patch, they are maintained in the Find
modules for dependencies that have one (most do).
Re-align existing modules with this change: msh3, mbedtls, rustls.
These modules return their `pkg-config` module name only when
detected via `pkg-config`.
Follow-up to d511ec8b0a#15573Closes#15800
Before this patch `CURL_LTO` only applied to builds explicitly
configured for `Release` or `RelWithDebInfo` via `CURL_BUILD_TYPE`.
After this patch it enables LTO regardess of build type.
Also:
- GHA/linux: enable LTO in an existing job.
Follow-up to 4ccf7622db#4799Closes#15829
Same issue as seen before with libssh2: `libpsl`'s pkg-config module
depends on another module, but that's not found. CMake ends up reporting
`LIBPSL_FOUND=YES`, while leaving `LIBPSL_INCLUDE_DIRS` empty. Then
the build fails to find `psl.h`.
The missing dependency in this case is `icu4c`, which is "keg-only",
meaning it's not exposed in the default Homebrew header, pkg-config,
lib, etc locations. It must be added to the `PKG_CONFIG_PATH` env, as
suggested by the warnings messages of `pkgconf`.
To avoid this fallout, let's ensure that `LIBPSL_INCLUDE_DIRS` is
non-empty when detecting via `pkg-config` and fall back to the CMake
detection method otherwise.
This was an issue till Homebrew libpsl 0.21.5_1, fixed in 0.21.5_2, that
no longer depends on `icu4c`.
Example log:
```
-- Checking for module 'libpsl'
-- Found libpsl, version 0.21.5
Package icu-uc was not found in the pkg-config search path.
Perhaps you should add the directory containing `icu-uc.pc'
to the PKG_CONFIG_PATH environment variable
Package 'icu-uc', required by 'libpsl', not found
[...]
-- Found Libpsl (via pkg-config): (found version "0.21.5")
[...]
In file included from curl/_bld/lib/CMakeFiles/libcurl_static.dir/Unity/unity_0_c.c:4:
In file included from curl/lib/altsvc.c:32:
In file included from curl/lib/urldata.h:145:
curl/lib/psl.h:28:10: fatal error: 'libpsl.h' file not found
^~~~~~~~~~
1 error generated.
```
Follow-up to 39c741b7b0#15408Closes#15827
`list(APPEND ...)` does this automatically, but we're _prepending_ LDAP
(to follow historical code and also autotools). `list(PREPEND ...)`
would likely do that the same, but it requires CMake 3.15 so we do this
manually and the manual method needs to skip the empty value manually.
Follow-up to 49f2a23d50#15273Closes#15828
- optimize out `cppflag_curl_staticlib` variable.
- optimize out `CPPFLAG_CURL_STATICLIB` variable and simplify logic.
- lowercase local variable name `CURLLIBDIR`.
Closes#15810
To make them run clang-tidy correctly. clang-tidy occasionally finds
mistakes none of the other static code analyzers we use finds.
Also added the
-clang-analyzer-security.insecureAPI.DeprecatedOrUnsafeBufferHandling
flag, to make it not complain about memcpy()
"make tidy" in the build root works fine now. The previous clang-tidy CI
job was removed in e43c3b3e3e. It is probably time to bring it
back.
Closes#15813
To avoid this in certain debug build cases:
```
./lib/easy_lock.h:72:20: error: inlining failed in call to 'curl_simple_lock_lock': function not considered for inlining [-Werror=inline]
```
Ref: #13596
Reported-by: Rudi Heitbaum
Fixes#15815Closes#15819
Add `CURL_STRICMP()` macro that works on all platforms depending on
which lib C function is available.
Make sure to always use `_stricmp()` on Windows, which is the
non-deprecated, official API for this on this platform. Before this
patch it used a MinGW-specific call, or a deprecated compatibility
wrapper with MSVC.
Drop `stricmp` variant detections on Windows with autotools.
https://learn.microsoft.com/cpp/c-runtime-library/reference/stricmp-wcsicmp-mbsicmp-stricmp-l-wcsicmp-l-mbsicmp-l
Ref: #15652Closes#15788
When a QUIC TLS session announced early data support and
'CURLSSLOPT_EARLYDATA' is set for the transfer, send initial request and
body (up to the 128k we buffer) as 0RTT when curl is built with
ngtcp2+gnutls.
QUIC 0RTT needs not only the TLS session but the QUIC transport
paramters as well. Store those and the earlydata max value together with
the session in the cache.
Add test case for h3 use of this. Enable quic early data in nghttpx for
testing.
Closes#15667
We started using codeql for static code analysis in 7183f5acc3,
June 2020.
Since then, not a single commit has been merged into the source code
repository citing codeql as source or reason. Yet, it keeps getting
updated and we get constant reminders to upgrade the pinning it to the
latest hash.
During 4.5 years with intense development and significant code churn.
While Coverity, scan-build and CodeSonar have belped us point out many
mistakes, codeql has remained silent (or had false positives).
For this little gain, I think we spend a disproportionate amount of work
on codeql maintanance.
We can try again in a future if we think it improves.
Assisted-by: Viktor Szakats
Closes#15798
This is debug code, but since the caller might use the value even when
this function returns error, it needs to be cleared properly here.
Spotted by Clang-tidy
Closes#15807
- When converting Curl_addrinfo to Curl_sockaddr_ex, if the address
length is too large then return error CURLE_TOO_LARGE.
Prior to this change the address structure was truncated on copy, and
the length shortened which I think is incorrect.
AFAICS the only time it could conceivably happen is when a UNIX socket
path is too long, and even then curl should've accounted for that by
having a structure that is large enough to store it. This is why I added
a DEBUGASSERT for debug builds, because I don't think it should ever
happen.
Closes https://github.com/curl/curl/pull/15784
- Remove wolfSSL from the legacy projects for Visual Studio 2010 - 2013.
It's no longer possible to maintain the custom build configuration of
wolfSSL for these old versions of Visual Studio.
Note support for wolfSSL was recently added to the winbuild build system
in 4de627ab and the user could possibly make their own wolfSSL build and
attempt to link it using an old version of Visual Studio that way.
Ref: https://gist.github.com/jay/5f6d8d5ba15c12c7457e3216a94da72d
Closes https://github.com/curl/curl/pull/15468
Make Curl_node_uremove() and Curl_node_take_elem() properly survive
run-time when the ->list field has been cleared previously. Like when
Curl_node_take_elem() is called twice.
We have asserts to catch those situations to make sure we avoid them if
we can, but if they still happen in a non-debug build we should make
sure the functions survive proper.
Pointed out by CodeSonar.
Closes#15791
And drop the prefix. This function was not use elsewhere and it should
certainly not be present in libcurl code when not used in the library.
Closes#15796
And use it from src and tests.
Syncing this functionality between platforms and build targets.
Also: Stop redefining `O_BINARY` in src, and use a local macro with
the same effect. `O_BINARY` is used in `CURL_SET_BINMODE()` to decide
if this functionality is supported, and redefining it makes this check
pass always in unity builds. The check is required for Apple OS, because
it offers a `setmode()` function, successfully detected by both CMake
and autotools, but that function has a different functionality and
signature than that expected by `CURL_SET_BINMODE()`.
Also:
- drop MetaWare High C (MS-DOS) support for set binmode.
- tests/libtest/Makefile.inc: dedupe comments.
- lib/curl_setup_once.h: tidy up feature guards for `io.h`, `fcntl.h`.
Ref: #15652Closes#15787
Let CURLINFO_QUEUE_TIME_T count only the time a transfer spends queued,
including possible redirect requests.
Add var 'time_queue' for reporting the time in write outs.
Add test for verifying correct reporting.
Closes#15512
Allowing --variable read a portion of provided files, makes curl work on
partial files for any options that accepts strings. Like --data and others.
The byte offset is provided within brackets, with a semicolon separator
like: --variable name@file;[100-200]"
Inspired by #14479
Assisted-by: Manuel Einfalt
Test 784 - 789. Documentation update provided.
Closes#15739
lib : remove all hyper code
configure: stop detecting hyper
docs: no more mention of hyper
tests: mo more special-handling of hyper builds
CI: no jobs using hyper
Closes#15120
For MinGW this is `-DNO_OLDNAMES`, with MSVC it is
`-D_CRT_DECLARE_NONSTDC_NAMES=0`.
There have been some support for this before this patch.
After this patch this is extended to all examples.
(And also the standalone http/client programs, if here.)
Cherry-picked from #15652Closes#15789
Described in detail in internal doc TLS-SESSIONS.md
Main points:
- use a new `ssl_peer_key` for cache lookups by connection filters
- recognize differences between TLSv1.3 and other tickets
* TLSv1.3 tickets are single-use, cache can hold several of them for a peer
* TLSv1.2 are reused, keep only a single one per peer
- differentiate between ticket BLOB to store (that could be persisted) and object instances
- use put/take/return pattern for cache access
- remember TLS version, ALPN protocol, time received and lifetime of ticket
- auto-expire tickets after their lifetime
Closes#15774
