Files
jak-project/third-party/curl/lib/headers.c
T
Alexander J. Semenuk 5d5e35fb9b fix: Windows toolchain compatibility (curl 8.21 re-vendor, endless reconfigure loop) (#4355)
## 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)
2026-07-27 19:19:18 -04:00

395 lines
11 KiB
C
Vendored
Generated

/***************************************************************************
* _ _ ____ _
* Project ___| | | | _ \| |
* / __| | | | |_) | |
* | (__| |_| | _ <| |___
* \___|\___/|_| \_\_____|
*
* Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
*
* This software is licensed as described in the file COPYING, which
* you should have received as part of this distribution. The terms
* are also available at https://curl.se/docs/copyright.html.
*
* You may opt to use, copy, modify, merge, publish, distribute and/or sell
* copies of the Software, and permit persons to whom the Software is
* furnished to do so, under the terms of the COPYING file.
*
* This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
* KIND, either express or implied.
*
* SPDX-License-Identifier: curl
*
***************************************************************************/
#include "curl_setup.h"
#include "urldata.h"
#include "sendf.h"
#include "curl_trc.h"
#include "headers.h"
#if !defined(CURL_DISABLE_HTTP) && !defined(CURL_DISABLE_HEADERS_API)
/* Generate the curl_header struct for the user. This function MUST assign all
struct fields in the output struct. */
static void copy_header_external(struct Curl_header_store *hs,
size_t index,
size_t amount,
struct Curl_llist_node *e,
struct curl_header *hout)
{
struct curl_header *h = hout;
h->name = hs->name;
h->value = hs->value;
h->amount = amount;
h->index = index;
/* this will randomly OR a reserved bit for the sole purpose of making it
impossible for applications to do == comparisons, as that would otherwise
be tempting and then lead to the reserved bits not being reserved
anymore. */
h->origin = (unsigned int)(hs->type | (1 << 27));
h->anchor = e;
}
/* public API */
CURLHcode curl_easy_header(CURL *curl,
const char *name,
size_t nameindex,
unsigned int origin,
int request,
struct curl_header **hout)
{
struct Curl_llist_node *e;
struct Curl_llist_node *e_pick = NULL;
struct Curl_easy *data = curl;
size_t match = 0;
size_t amount = 0;
struct Curl_header_store *hs = NULL;
struct Curl_header_store *pick = NULL;
if(!name || !hout || !data ||
(origin > (CURLH_HEADER | CURLH_TRAILER | CURLH_CONNECT | CURLH_1XX |
CURLH_PSEUDO)) || !origin || (request < -1))
return CURLHE_BAD_ARGUMENT;
if(!Curl_llist_count(&data->state.httphdrs))
return CURLHE_NOHEADERS; /* no headers available */
if(request > data->state.requests)
return CURLHE_NOREQUEST;
if(request == -1)
request = data->state.requests;
/* we need a first round to count amount of this header */
for(e = Curl_llist_head(&data->state.httphdrs); e; e = Curl_node_next(e)) {
hs = Curl_node_elem(e);
if(curl_strequal(hs->name, name) &&
(hs->type & origin) &&
(hs->request == request)) {
amount++;
pick = hs;
e_pick = e;
}
}
if(!amount)
return CURLHE_MISSING;
else if(nameindex >= amount)
return CURLHE_BADINDEX;
if(nameindex == amount - 1)
/* if the last or only occurrence is what's asked for, then we know it */
hs = pick;
else {
for(e = Curl_llist_head(&data->state.httphdrs); e; e = Curl_node_next(e)) {
hs = Curl_node_elem(e);
if(curl_strequal(hs->name, name) &&
(hs->type & origin) &&
(hs->request == request) &&
(match++ == nameindex)) {
e_pick = e;
break;
}
}
if(!e) /* this should not happen */
return CURLHE_MISSING;
}
/* this is the name we want */
copy_header_external(hs, nameindex, amount, e_pick,
&data->state.headerout[0]);
*hout = &data->state.headerout[0];
return CURLHE_OK;
}
/* public API */
struct curl_header *curl_easy_nextheader(CURL *curl,
unsigned int origin,
int request,
struct curl_header *prev)
{
struct Curl_easy *data = curl;
struct Curl_llist_node *pick;
struct Curl_llist_node *e;
struct Curl_header_store *hs;
size_t amount = 0;
size_t index = 0;
if(request > data->state.requests)
return NULL;
if(request == -1)
request = data->state.requests;
if(prev) {
pick = prev->anchor;
if(!pick)
/* something is wrong */
return NULL;
pick = Curl_node_next(pick);
}
else
pick = Curl_llist_head(&data->state.httphdrs);
if(pick) {
/* make sure it is the next header of the desired type */
do {
hs = Curl_node_elem(pick);
if((hs->type & origin) && (hs->request == request))
break;
pick = Curl_node_next(pick);
} while(pick);
}
if(!pick)
/* no more headers available */
return NULL;
hs = Curl_node_elem(pick);
/* count number of occurrences of this name within the mask and figure out
the index for the currently selected entry */
for(e = Curl_llist_head(&data->state.httphdrs); e; e = Curl_node_next(e)) {
struct Curl_header_store *check = Curl_node_elem(e);
if(curl_strequal(hs->name, check->name) &&
(check->request == request) &&
(check->type & origin))
amount++;
if(e == pick)
index = amount - 1;
}
copy_header_external(hs, index, amount, pick,
&data->state.headerout[1]);
return &data->state.headerout[1];
}
static CURLcode namevalue(char *header, size_t hlen, unsigned int type,
char **name, char **value)
{
char *end = header + hlen - 1; /* point to the last byte */
DEBUGASSERT(hlen);
*name = header;
if(type == CURLH_PSEUDO) {
if(*header != ':')
return CURLE_BAD_FUNCTION_ARGUMENT;
header++;
}
/* Find the end of the header name */
while(*header && (*header != ':'))
++header;
if(*header)
/* Skip over colon, null it */
*header++ = 0;
else
return CURLE_BAD_FUNCTION_ARGUMENT;
/* skip all leading blank letters */
while(ISBLANK(*header))
header++;
*value = header;
/* skip all trailing space letters */
while((end > header) && ISBLANK(*end))
*end-- = 0; /* null-terminate */
return CURLE_OK;
}
/*
* Curl_headers_push() gets passed a full HTTP header to store. It gets called
* immediately before the header callback. The header is CRLF, CR or LF
* terminated.
*/
CURLcode Curl_headers_push(struct Curl_easy *data, const char *header,
size_t hlen, /* length of header */
unsigned char type)
{
char *value = NULL;
char *name = NULL;
struct Curl_header_store *hs;
CURLcode result = CURLE_OUT_OF_MEMORY;
const size_t ilen = hlen;
if((header[0] == '\r') || (header[0] == '\n'))
/* ignore the body separator */
return CURLE_OK;
/* trim off newline characters */
if(hlen && (header[hlen - 1] == '\n'))
hlen--;
if(hlen && (header[hlen - 1] == '\r'))
hlen--;
if(hlen == ilen)
/* neither CR nor LF as terminator is not a valid header */
return CURLE_WEIRD_SERVER_REPLY;
if(ISBLANK(header[0])) {
/* pass leading blanks */
while(hlen && ISBLANK(*header)) {
header++;
hlen--;
}
if(!hlen)
return CURLE_WEIRD_SERVER_REPLY;
}
if(Curl_llist_count(&data->state.httphdrs) >= MAX_HTTP_RESP_HEADER_COUNT) {
failf(data, "Too many response headers, %d is max",
MAX_HTTP_RESP_HEADER_COUNT);
return CURLE_TOO_LARGE;
}
hs = curlx_calloc(1, sizeof(*hs) + hlen);
if(!hs)
return CURLE_OUT_OF_MEMORY;
memcpy(hs->buffer, header, hlen);
hs->buffer[hlen] = 0; /* null-terminate */
result = namevalue(hs->buffer, hlen, type, &name, &value);
if(!result) {
hs->name = name;
hs->value = value;
hs->type = type;
hs->request = data->state.requests;
/* insert this node into the list of headers */
Curl_llist_append(&data->state.httphdrs, hs, &hs->node);
data->state.prevhead = hs;
}
else {
failf(data, "Invalid response header");
curlx_free(hs);
}
return result;
}
/*
* Curl_headers_reset(). Reset the headers subsystem.
*/
static void headers_reset(struct Curl_easy *data)
{
Curl_llist_init(&data->state.httphdrs, NULL);
data->state.prevhead = NULL;
}
struct hds_cw_collect_ctx {
struct Curl_cwriter super;
};
static CURLcode hds_cw_collect_write(struct Curl_easy *data,
struct Curl_cwriter *writer, int type,
const char *buf, size_t blen)
{
if((type & CLIENTWRITE_HEADER) && !(type & CLIENTWRITE_STATUS)) {
unsigned char htype = (unsigned char)
(type & CLIENTWRITE_CONNECT ? CURLH_CONNECT :
(type & CLIENTWRITE_1XX ? CURLH_1XX :
(type & CLIENTWRITE_TRAILER ? CURLH_TRAILER :
CURLH_HEADER)));
CURLcode result = Curl_headers_push(data, buf, blen, htype);
CURL_TRC_WRITE(data, "header_collect pushed(type=%x, len=%zu) -> %d",
htype, blen, (int)result);
if(result)
return result;
}
return Curl_cwriter_write(data, writer->next, type, buf, blen);
}
static const struct Curl_cwtype hds_cw_collect = {
"hds-collect",
NULL,
Curl_cwriter_def_init,
hds_cw_collect_write,
Curl_cwriter_def_close,
sizeof(struct hds_cw_collect_ctx)
};
CURLcode Curl_headers_init(struct Curl_easy *data)
{
struct Curl_cwriter *writer;
CURLcode result;
if(data->conn && (data->conn->scheme->protocol & PROTO_FAMILY_HTTP)) {
/* avoid installing it twice */
if(Curl_cwriter_get_by_name(data, hds_cw_collect.name))
return CURLE_OK;
result = Curl_cwriter_create(&writer, data, &hds_cw_collect,
CURL_CW_PROTOCOL);
if(result)
return result;
result = Curl_cwriter_add(data, writer);
if(result) {
Curl_cwriter_free(data, writer);
return result;
}
}
return CURLE_OK;
}
/*
* Curl_headers_cleanup(). Free all stored headers and associated memory.
*/
CURLcode Curl_headers_cleanup(struct Curl_easy *data)
{
struct Curl_llist_node *e;
struct Curl_llist_node *n;
for(e = Curl_llist_head(&data->state.httphdrs); e; e = n) {
struct Curl_header_store *hs = Curl_node_elem(e);
n = Curl_node_next(e);
curlx_free(hs);
}
headers_reset(data);
return CURLE_OK;
}
#else /* HTTP-disabled builds below */
CURLHcode curl_easy_header(CURL *easy,
const char *name,
size_t index,
unsigned int origin,
int request,
struct curl_header **hout)
{
(void)easy;
(void)name;
(void)index;
(void)origin;
(void)request;
(void)hout;
return CURLHE_NOT_BUILT_IN;
}
struct curl_header *curl_easy_nextheader(CURL *easy,
unsigned int type,
int request,
struct curl_header *prev)
{
(void)easy;
(void)type;
(void)request;
(void)prev;
return NULL;
}
#endif