## Problem Windows builds break with a current local toolchain (Scoop LLVM 22.1.8, CMake 4.4.0, VS 2026), in two independent ways: 1. The build stops at curl's deliberate guard: `#error "no non-blocking method was found/used/set"` in `third-party/curl/lib/nonblock.c`. 2. From the second configure onward, `cmake --build` re-runs CMake in an endless loop (observed 42 consecutive reconfigure cycles in a single build). Likely the same mechanism behind the "endlessly building" VS 2026 note in `docs/setup/dev/vs.md`. ## Root cause 1. `third-party/curl/CMake/CurlTests.c` passes `int *` to `ioctlsocket()`, whose third parameter is `u_long *`. Clang 22 promotes `-Wincompatible-pointer-types` to a hard error in C, so the `HAVE_IOCTLSOCKET_FIONBIO` try_compile silently fails and `curl_config.h` never defines it. Upstream CI does not see this because the windows-2022 runner image ships an older LLVM. GCC 14 promotes the same warning to a hard error, which is very likely the `CurlTests.c.obj` failure reported from MSYS2 in open-goal/jak-project#3551. Upstream curl hit the identical problem with GCC 14 and fixed the probe in curl 8.8.0 (curl/curl#13578). 2. The root CMakeLists copies the build tree's `compile_commands.json` into `<src>/build/` for clangd using `configure_file()`, which registers its input as a configure dependency. CMake rewrites `compile_commands.json` late in every generation, after `CTestTestfile.cmake` and `cmake_install.cmake` (outputs of the same Ninja regen rule), so once the dependency is registered the rule is deterministically dirty and every `ninja` invocation re-runs CMake. A pristine first configure is safe (the file does not exist yet, so the `if(EXISTS ...)` guard skips the copy), which is why the loop looks machine- or IDE-specific. ## Fix 1. Per review, re-vendor `third-party/curl` at the `curl-8_21_0` tag (previously `curl-8_3_0`), which carries the upstream probe fix plus two years of upstream development; `vendor.yaml` updated to match. Adjustments the version jump forced: - curl 8.15 removed the native macOS Secure Transport backend (`CURL_USE_SECTRANSP`), so macOS now builds curl against OpenSSL like Linux. The two macOS workflows install Homebrew `openssl@3` and export `OPENSSL_ROOT_DIR` (keg-only), and the macOS setup docs gained the same two lines. - `CURL_BROTLI` / `CURL_ZSTD` switched to AUTO-detection in curl 8.10; pinned OFF to keep the previous no-compression behavior and avoid silently linking whatever the CI images happen to have. - curl's new top-level `BUILD_EXAMPLES` cache option (default ON) leaked into discord-rpc's identically named option and broke configure at a nonexistent `examples/send-presence` directory; pinned OFF ahead of the third-party subdirectories. The diff is dominated by the mechanical tag-tree swap under `third-party/curl` (linguist-vendored, collapsed in review). The hand-written changes are `CMakeLists.txt`, the two macOS workflows, `docs/setup/system/macos.md`, and `vendor.yaml`. 2. Swap `configure_file()` for `file(COPY ...)`: the same clangd copy with no configure dependency registered. (`file(COPY_FILE ... ONLY_IF_DIFFERENT)` would be cleaner still but requires CMake 3.21, above the declared `cmake_minimum_required(VERSION 3.10)`.) ## Test plan - [x] Fresh `cmake --preset Release-windows-clang` (LLVM 22, no cache seeding) completes and logs `Enabled SSL backends: Schannel`; the FIONBIO probe passes without the previous `#error` - [x] Full Windows Release build from scratch in the branch worktree (all 1422 targets) - [x] goalc-test suite: 1509 passed, 0 failed - [x] Second consecutive configure with `compile_commands.json` present: the regen rule in `build.ninja` has no `compile_commands.json` input; `<src>/build/compile_commands.json` is still refreshed for clangd - [x] Repeated `ninja` invocations after a full build no longer re-run CMake - [x] macOS Intel and ARM CI green (first exercise of the OpenSSL backend switch) --- I work off a self-hosted forge, so this GitHub account is quiet; the configure logs and ninja dirty-node traces from the investigation are available if anyone wants the raw data. (AI-assisted)
Unit tests
The goal is to add tests for all functions in libcurl. If functions are too big and complicated, we should split them into smaller and testable ones.
Build Unit Tests
./configure --enable-debug is required for the unit tests to build. To
enable unit tests, there is a separate static libcurl built that is used
exclusively for linking unit test programs. Build everything as normal, and
then you can run the unit test cases as well.
Run Unit Tests
Unit tests are run as part of the regular test suite. If you have built
everything to run unit tests, to can do 'make test' at the root level. Or you
can cd tests and make and then invoke individual unit tests with
./runtests.pl NNNN where NNNN is the specific test number.
Debug Unit Tests
If a specific test fails you get told. The test case then has output left in
the %LOGDIR subdirectory, but most importantly you can re-run the test again
using gdb by doing ./runtests.pl -g NNNN. That is, add a -g to make it
start up gdb and run the same case using that.
Write Unit Tests
We put tests that focus on an area or a specific function into a single C
source file. The source file should be named unitNNNN.c where NNNN is a
previously unused number.
Add your test to tests/unit/Makefile.inc (if it is a unit test). Add your
test data filename to tests/data/Makefile.am
You also need a separate file called tests/data/testNNNN (using the same
number) that describes your test case. See the test1300 file for inspiration
and the tests/FILEFORMAT.md documentation.
For the actual C file, here's a simple example:
#include "unitcheck.h"
#include "a libcurl header.h" /* from the lib directory */
static CURLcode test_unit9998(const char *arg)
{
UNITTEST_BEGIN_SIMPLE
/* here you start doing things and checking that the results are good */
fail_unless(size == 0, "initial size should be zero");
fail_if(!head, "head should not be initiated to NULL");
/* you end the test code like this: */
UNITTEST_END_SIMPLE
}
Here's an example using optional initialization and cleanup:
#include "unitcheck.h"
#include "a libcurl header.h" /* from the lib directory */
static CURLcode t9999_setup(void)
{
/* whatever you want done first */
return CURLE_OK;
}
static void t9999_stop(void)
{
/* done before shutting down and exiting */
}
static CURLcode test_unit9999(const char *arg)
{
UNITTEST_BEGIN(t9999_setup())
/* here you start doing things and checking that the results are good */
fail_unless(size == 0, "initial size should be zero");
fail_if(!head, "head should not be initiated to NULL");
/* you end the test code like this: */
UNITTEST_END(t9999_stop())
}
Testing static functions
Lots of internal functions are made static, and they should be static if they are private within a single source file.
The curl build system provides a way to write unit tests that let us properly test these functions while keeping them static in release builds.
A function that is static in the build but should be provided for unit testing
needs to replace its static keyword with UNITTEST and it needs to have a
prototype provided immediately above it.
An example add_two_integers() function for unit testing:
UNITTEST int add_two_integers(int a, int b);
UNITTEST int add_two_integers(int a, int b)
{
return a + b;
}
Since the function is static and is private for this source file, it should not have its prototype in any header file.
When building unit tests, the build system automatically generates the
lib/unitprotos.h header file with all the prototypes for UNITTEST
functions provided in any libcurl C source code files. (This is done by the
scripts/extract-unit-protos script.)