cozis/cHTTP
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Basic test infrastructure

Browse Source
This commit is contained in:
cozis
2025-04-20 14:11:52 +02:00
parent a4aafb1ee8
commit e1dca84a44
9 changed files with 1407 additions and 260 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/main.c
+290
View File
@@ -0,0 +1,290 @@
#include <stdio.h>
#include <string.h>
#include <stdlib.h>
#include "picohttpparser.h"
#include "../tinyhttp.h"
//////////////////////////////////////////////////////////////////////////////////////
// TYPES, MACROS, PROTOTYPES
//////////////////////////////////////////////////////////////////////////////////////
typedef struct {
int minor;
int status_code;
TinyHTTPString status_text;
int num_headers;
TinyHTTPHeader headers[TINYHTTP_HEADER_LIMIT];
char *body;
int body_len;
} Response;
static void *memfunc(TinyHTTPMemoryFuncTag tag, void *ptr, int len, void *data);
static void send_request(TinyHTTPStream *stream, const char *str);
static void recv_response(TinyHTTPStream *stream, Response *res, char *dst, int cap);
#define TEST(X) {if (!(X)) { printf("Test failed at %s:%d\n", __FILE__, __LINE__); fflush(stdout); __builtin_trap(); }}
//////////////////////////////////////////////////////////////////////////////////////
// TEST CASES
//////////////////////////////////////////////////////////////////////////////////////
static void basic_request(void)
{
TinyHTTPStream stream;
tinyhttp_stream_init(&stream, memfunc, NULL);
// Send request
send_request(&stream,
"GET / HTTP/1.1\r\n"
"Host: 127.0.0.1:8080\r\n"
"User-Agent: curl/7.81.0\r\n"
"Accept: */*\r\n"
"\r\n");
// Build response
tinyhttp_stream_response_status(&stream, 200);
tinyhttp_stream_response_send(&stream);
// Receive response
char buf[1<<12];
Response res;
recv_response(&stream, &res, buf, sizeof(buf));
// TODO
tinyhttp_stream_free(&stream);
}
//////////////////////////////////////////////////////////////////////////////////////
// ENTRY POINT
//////////////////////////////////////////////////////////////////////////////////////
int main(void)
{
basic_request();
return 0;
}
//////////////////////////////////////////////////////////////////////////////////////
// Helper Functions
//////////////////////////////////////////////////////////////////////////////////////
static void*
memfunc(TinyHTTPMemoryFuncTag tag, void *ptr, int len, void *data)
{
(void) data;
switch (tag) {
case TINYHTTP_MEM_MALLOC:
return malloc(len);
case TINYHTTP_MEM_FREE:
free(ptr);
return NULL;
}
return NULL;
}
static int
buffer_into_stream(TinyHTTPStream *stream, const char *src, int len)
{
int state = tinyhttp_stream_state(stream);
ptrdiff_t copied = 0;
while (copied < len && (state & TINYHTTP_STREAM_RECV)) {
char *dst;
ptrdiff_t cap;
ptrdiff_t cpy;
dst = tinyhttp_stream_recv_buf(stream, &cap);
cpy = len - copied;
if (cpy > cap) cpy = cap;
memcpy(dst, src + copied, cpy);
copied += cpy;
tinyhttp_stream_recv_ack(stream, cpy);
state = tinyhttp_stream_state(stream);
}
return copied;
}
static int
stream_into_buffer(TinyHTTPStream *stream, char *dst, int cap)
{
int state = tinyhttp_stream_state(stream);
int copied = 0;
while (copied < cap && (state & TINYHTTP_STREAM_SEND)) {
char *src;
ptrdiff_t len;
ptrdiff_t cpy;
src = tinyhttp_stream_send_buf(stream, &len);
cpy = cap - copied;
if (cpy > len) cpy = len;
memcpy(dst + copied, src, cpy);
copied += cpy;
tinyhttp_stream_send_ack(stream, cpy);
state = tinyhttp_stream_state(stream);
}
return copied;
}
static void
expect_request(TinyHTTPStream *stream, TinyHTTPRequest expreq)
{
int state = tinyhttp_stream_state(stream);
TEST(state & TINYHTTP_STREAM_READY);
TinyHTTPRequest *req = tinyhttp_stream_request(stream);
TEST(req);
TEST(req->method == expreq.method);
TEST(req->minor == expreq.minor);
TEST(tinyhttp_streq(req->path, expreq.path));
TEST(req->num_headers == expreq.num_headers);
for (int i = 0; i < expreq.num_headers; i++) {
TEST(tinyhttp_streq(req->headers[i].name, expreq.headers[i].name));
TEST(tinyhttp_streq(req->headers[i].value, expreq.headers[i].value));
}
TEST(req->body_len == expreq.body_len);
if (expreq.body_len == 0) {
TEST(req->body == NULL);
} else {
TEST(!memcmp(req->body, expreq.body, expreq.body_len));
}
}
static int
parse_request(TinyHTTPString txt, TinyHTTPRequest *req)
{
const char *method;
size_t method_len;
const char *path;
size_t path_len;
int minor;
struct phr_header headers[TINYHTTP_HEADER_LIMIT];
size_t num_headers = TINYHTTP_HEADER_LIMIT;
int ret = phr_parse_request(
txt.ptr, txt.len,
&method, &method_len,
&path, &path_len,
&minor,
headers, &num_headers,
0);
TEST(ret == txt.len);
if (method_len == 3 && !memcmp("GET", method, 3)) {
req->method = TINYHTTP_METHOD_GET;
} else if (method_len == 4 && !memcmp("POST", method, 4)) {
req->method = TINYHTTP_METHOD_POST;
} else {
return -1;
}
req->minor = minor;
req->path = (TinyHTTPString) { path, path_len };
req->num_headers = num_headers;
for (int i = 0; i < (int) num_headers; i++) {
req->headers[i].name = (TinyHTTPString) {
headers[i].name,
headers[i].name_len,
};
req->headers[i].value = (TinyHTTPString) {
headers[i].value,
headers[i].value_len,
};
}
req->body = NULL; // TODO
req->body_len = 0; // TODO
return 0;
}
static void
parse_response(TinyHTTPString txt, Response *res)
{
int minor;
int status_code;
const char *status_text;
size_t status_text_len;
struct phr_header headers[TINYHTTP_HEADER_LIMIT];
size_t num_headers = TINYHTTP_HEADER_LIMIT;
int ret = phr_parse_response(
txt.ptr, txt.len,
&minor,
&status_code, &status_text, &status_text_len,
headers, &num_headers,
0);
TEST(ret == txt.len);
res->minor = minor;
res->status_code = status_code;
res->status_text = (TinyHTTPString) { status_text, status_text_len };
res->num_headers = num_headers;
for (int i = 0; i < (int) num_headers; i++) {
res->headers[i].name = (TinyHTTPString) {
headers[i].name,
headers[i].name_len
};
res->headers[i].value = (TinyHTTPString) {
headers[i].value,
headers[i].value_len
};
}
res->body = NULL; // TODO
res->body_len = 0; // TODO
}
static void
send_request(TinyHTTPStream *stream, const char *str)
{
int received = buffer_into_stream(stream, str, strlen(str));
TEST(received == (int) strlen(str));
TinyHTTPRequest req;
TEST(parse_request((TinyHTTPString) {str, strlen(str)}, &req) == 0);
expect_request(stream, req);
}
static void
recv_response(TinyHTTPStream *stream, Response *res, char *dst, int cap)
{
int len = stream_into_buffer(stream, dst, cap);
int state = tinyhttp_stream_state(stream);
TEST((state & TINYHTTP_STREAM_SEND) == 0);
parse_response((TinyHTTPString) { dst, len }, res);
}
//////////////////////////////////////////////////////////////////////////////////////
// THE END
//////////////////////////////////////////////////////////////////////////////////////
tests/picohttpparser.c
+685
View File
@@ -0,0 +1,685 @@
/*
* Copyright (c) 2009-2014 Kazuho Oku, Tokuhiro Matsuno, Daisuke Murase,
* Shigeo Mitsunari
*
* The software is licensed under either the MIT License (below) or the Perl
* license.
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to
* deal in the Software without restriction, including without limitation the
* rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
* sell copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#include <assert.h>
#include <stddef.h>
#include <string.h>
#ifdef __SSE4_2__
#ifdef _MSC_VER
#include <nmmintrin.h>
#else
#include <x86intrin.h>
#endif
#endif
#include "picohttpparser.h"
#if __GNUC__ >= 3
#define likely(x) __builtin_expect(!!(x), 1)
#define unlikely(x) __builtin_expect(!!(x), 0)
#else
#define likely(x) (x)
#define unlikely(x) (x)
#endif
#ifdef _MSC_VER
#define ALIGNED(n) _declspec(align(n))
#else
#define ALIGNED(n) __attribute__((aligned(n)))
#endif
#define IS_PRINTABLE_ASCII(c) ((unsigned char)(c)-040u < 0137u)
#define CHECK_EOF() \
if (buf == buf_end) { \
*ret = -2; \
return NULL; \
}
#define EXPECT_CHAR_NO_CHECK(ch) \
if (*buf++ != ch) { \
*ret = -1; \
return NULL; \
}
#define EXPECT_CHAR(ch) \
CHECK_EOF(); \
EXPECT_CHAR_NO_CHECK(ch);
#define ADVANCE_TOKEN(tok, toklen) \
do { \
const char *tok_start = buf; \
static const char ALIGNED(16) ranges2[16] = "\000\040\177\177"; \
int found2; \
buf = findchar_fast(buf, buf_end, ranges2, 4, &found2); \
if (!found2) { \
CHECK_EOF(); \
} \
while (1) { \
if (*buf == ' ') { \
break; \
} else if (unlikely(!IS_PRINTABLE_ASCII(*buf))) { \
if ((unsigned char)*buf < '\040' || *buf == '\177') { \
*ret = -1; \
return NULL; \
} \
} \
++buf; \
CHECK_EOF(); \
} \
tok = tok_start; \
toklen = buf - tok_start; \
} while (0)
static const char *token_char_map = "\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"
"\0\1\0\1\1\1\1\1\0\0\1\1\0\1\1\0\1\1\1\1\1\1\1\1\1\1\0\0\0\0\0\0"
"\0\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\0\0\0\1\1"
"\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\1\0\1\0\1\0"
"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"
"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"
"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"
"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0";
static const char *findchar_fast(const char *buf, const char *buf_end, const char *ranges, size_t ranges_size, int *found)
{
*found = 0;
#if __SSE4_2__
if (likely(buf_end - buf >= 16)) {
__m128i ranges16 = _mm_loadu_si128((const __m128i *)ranges);
size_t left = (buf_end - buf) & ~15;
do {
__m128i b16 = _mm_loadu_si128((const __m128i *)buf);
int r = _mm_cmpestri(ranges16, ranges_size, b16, 16, _SIDD_LEAST_SIGNIFICANT | _SIDD_CMP_RANGES | _SIDD_UBYTE_OPS);
if (unlikely(r != 16)) {
buf += r;
*found = 1;
break;
}
buf += 16;
left -= 16;
} while (likely(left != 0));
}
#else
/* suppress unused parameter warning */
(void)buf_end;
(void)ranges;
(void)ranges_size;
#endif
return buf;
}
static const char *get_token_to_eol(const char *buf, const char *buf_end, const char **token, size_t *token_len, int *ret)
{
const char *token_start = buf;
#ifdef __SSE4_2__
static const char ALIGNED(16) ranges1[16] = "\0\010" /* allow HT */
"\012\037" /* allow SP and up to but not including DEL */
"\177\177"; /* allow chars w. MSB set */
int found;
buf = findchar_fast(buf, buf_end, ranges1, 6, &found);
if (found)
goto FOUND_CTL;
#else
/* find non-printable char within the next 8 bytes, this is the hottest code; manually inlined */
while (likely(buf_end - buf >= 8)) {
#define DOIT() \
do { \
if (unlikely(!IS_PRINTABLE_ASCII(*buf))) \
goto NonPrintable; \
++buf; \
} while (0)
DOIT();
DOIT();
DOIT();
DOIT();
DOIT();
DOIT();
DOIT();
DOIT();
#undef DOIT
continue;
NonPrintable:
if ((likely((unsigned char)*buf < '\040') && likely(*buf != '\011')) || unlikely(*buf == '\177')) {
goto FOUND_CTL;
}
++buf;
}
#endif
for (;; ++buf) {
CHECK_EOF();
if (unlikely(!IS_PRINTABLE_ASCII(*buf))) {
if ((likely((unsigned char)*buf < '\040') && likely(*buf != '\011')) || unlikely(*buf == '\177')) {
goto FOUND_CTL;
}
}
}
FOUND_CTL:
if (likely(*buf == '\015')) {
++buf;
EXPECT_CHAR('\012');
*token_len = buf - 2 - token_start;
} else if (*buf == '\012') {
*token_len = buf - token_start;
++buf;
} else {
*ret = -1;
return NULL;
}
*token = token_start;
return buf;
}
static const char *is_complete(const char *buf, const char *buf_end, size_t last_len, int *ret)
{
int ret_cnt = 0;
buf = last_len < 3 ? buf : buf + last_len - 3;
while (1) {
CHECK_EOF();
if (*buf == '\015') {
++buf;
CHECK_EOF();
EXPECT_CHAR('\012');
++ret_cnt;
} else if (*buf == '\012') {
++buf;
++ret_cnt;
} else {
++buf;
ret_cnt = 0;
}
if (ret_cnt == 2) {
return buf;
}
}
*ret = -2;
return NULL;
}
#define PARSE_INT(valp_, mul_) \
if (*buf < '0' || '9' < *buf) { \
buf++; \
*ret = -1; \
return NULL; \
} \
*(valp_) = (mul_) * (*buf++ - '0');
#define PARSE_INT_3(valp_) \
do { \
int res_ = 0; \
PARSE_INT(&res_, 100) \
*valp_ = res_; \
PARSE_INT(&res_, 10) \
*valp_ += res_; \
PARSE_INT(&res_, 1) \
*valp_ += res_; \
} while (0)
/* returned pointer is always within [buf, buf_end), or null */
static const char *parse_token(const char *buf, const char *buf_end, const char **token, size_t *token_len, char next_char,
int *ret)
{
/* We use pcmpestri to detect non-token characters. This instruction can take no more than eight character ranges (8*2*8=128
* bits that is the size of a SSE register). Due to this restriction, characters `|` and `~` are handled in the slow loop. */
static const char ALIGNED(16) ranges[] = "\x00 " /* control chars and up to SP */
"\"\"" /* 0x22 */
"()" /* 0x28,0x29 */
",," /* 0x2c */
"//" /* 0x2f */
":@" /* 0x3a-0x40 */
"[]" /* 0x5b-0x5d */
"{\xff"; /* 0x7b-0xff */
const char *buf_start = buf;
int found;
buf = findchar_fast(buf, buf_end, ranges, sizeof(ranges) - 1, &found);
if (!found) {
CHECK_EOF();
}
while (1) {
if (*buf == next_char) {
break;
} else if (!token_char_map[(unsigned char)*buf]) {
*ret = -1;
return NULL;
}
++buf;
CHECK_EOF();
}
*token = buf_start;
*token_len = buf - buf_start;
return buf;
}
/* returned pointer is always within [buf, buf_end), or null */
static const char *parse_http_version(const char *buf, const char *buf_end, int *minor_version, int *ret)
{
/* we want at least [HTTP/1.<two chars>] to try to parse */
if (buf_end - buf < 9) {
*ret = -2;
return NULL;
}
EXPECT_CHAR_NO_CHECK('H');
EXPECT_CHAR_NO_CHECK('T');
EXPECT_CHAR_NO_CHECK('T');
EXPECT_CHAR_NO_CHECK('P');
EXPECT_CHAR_NO_CHECK('/');
EXPECT_CHAR_NO_CHECK('1');
EXPECT_CHAR_NO_CHECK('.');
PARSE_INT(minor_version, 1);
return buf;
}
static const char *parse_headers(const char *buf, const char *buf_end, struct phr_header *headers, size_t *num_headers,
size_t max_headers, int *ret)
{
for (;; ++*num_headers) {
CHECK_EOF();
if (*buf == '\015') {
++buf;
EXPECT_CHAR('\012');
break;
} else if (*buf == '\012') {
++buf;
break;
}
if (*num_headers == max_headers) {
*ret = -1;
return NULL;
}
if (!(*num_headers != 0 && (*buf == ' ' || *buf == '\t'))) {
/* parsing name, but do not discard SP before colon, see
* http://www.mozilla.org/security/announce/2006/mfsa2006-33.html */
if ((buf = parse_token(buf, buf_end, &headers[*num_headers].name, &headers[*num_headers].name_len, ':', ret)) == NULL) {
return NULL;
}
if (headers[*num_headers].name_len == 0) {
*ret = -1;
return NULL;
}
++buf;
for (;; ++buf) {
CHECK_EOF();
if (!(*buf == ' ' || *buf == '\t')) {
break;
}
}
} else {
headers[*num_headers].name = NULL;
headers[*num_headers].name_len = 0;
}
const char *value;
size_t value_len;
if ((buf = get_token_to_eol(buf, buf_end, &value, &value_len, ret)) == NULL) {
return NULL;
}
/* remove trailing SPs and HTABs */
const char *value_end = value + value_len;
for (; value_end != value; --value_end) {
const char c = *(value_end - 1);
if (!(c == ' ' || c == '\t')) {
break;
}
}
headers[*num_headers].value = value;
headers[*num_headers].value_len = value_end - value;
}
return buf;
}
static const char *parse_request(const char *buf, const char *buf_end, const char **method, size_t *method_len, const char **path,
size_t *path_len, int *minor_version, struct phr_header *headers, size_t *num_headers,
size_t max_headers, int *ret)
{
/* skip first empty line (some clients add CRLF after POST content) */
CHECK_EOF();
if (*buf == '\015') {
++buf;
EXPECT_CHAR('\012');
} else if (*buf == '\012') {
++buf;
}
/* parse request line */
if ((buf = parse_token(buf, buf_end, method, method_len, ' ', ret)) == NULL) {
return NULL;
}
do {
++buf;
CHECK_EOF();
} while (*buf == ' ');
ADVANCE_TOKEN(*path, *path_len);
do {
++buf;
CHECK_EOF();
} while (*buf == ' ');
if (*method_len == 0 || *path_len == 0) {
*ret = -1;
return NULL;
}
if ((buf = parse_http_version(buf, buf_end, minor_version, ret)) == NULL) {
return NULL;
}
if (*buf == '\015') {
++buf;
EXPECT_CHAR('\012');
} else if (*buf == '\012') {
++buf;
} else {
*ret = -1;
return NULL;
}
return parse_headers(buf, buf_end, headers, num_headers, max_headers, ret);
}
int phr_parse_request(const char *buf_start, size_t len, const char **method, size_t *method_len, const char **path,
size_t *path_len, int *minor_version, struct phr_header *headers, size_t *num_headers, size_t last_len)
{
const char *buf = buf_start, *buf_end = buf_start + len;
size_t max_headers = *num_headers;
int r;
*method = NULL;
*method_len = 0;
*path = NULL;
*path_len = 0;
*minor_version = -1;
*num_headers = 0;
/* if last_len != 0, check if the request is complete (a fast countermeasure
againt slowloris */
if (last_len != 0 && is_complete(buf, buf_end, last_len, &r) == NULL) {
return r;
}
if ((buf = parse_request(buf, buf_end, method, method_len, path, path_len, minor_version, headers, num_headers, max_headers,
&r)) == NULL) {
return r;
}
return (int)(buf - buf_start);
}
static const char *parse_response(const char *buf, const char *buf_end, int *minor_version, int *status, const char **msg,
size_t *msg_len, struct phr_header *headers, size_t *num_headers, size_t max_headers, int *ret)
{
/* parse "HTTP/1.x" */
if ((buf = parse_http_version(buf, buf_end, minor_version, ret)) == NULL) {
return NULL;
}
/* skip space */
if (*buf != ' ') {
*ret = -1;
return NULL;
}
do {
++buf;
CHECK_EOF();
} while (*buf == ' ');
/* parse status code, we want at least [:digit:][:digit:][:digit:]<other char> to try to parse */
if (buf_end - buf < 4) {
*ret = -2;
return NULL;
}
PARSE_INT_3(status);
/* get message including preceding space */
if ((buf = get_token_to_eol(buf, buf_end, msg, msg_len, ret)) == NULL) {
return NULL;
}
if (*msg_len == 0) {
/* ok */
} else if (**msg == ' ') {
/* Remove preceding space. Successful return from `get_token_to_eol` guarantees that we would hit something other than SP
* before running past the end of the given buffer. */
do {
++*msg;
--*msg_len;
} while (**msg == ' ');
} else {
/* garbage found after status code */
*ret = -1;
return NULL;
}
return parse_headers(buf, buf_end, headers, num_headers, max_headers, ret);
}
int phr_parse_response(const char *buf_start, size_t len, int *minor_version, int *status, const char **msg, size_t *msg_len,
struct phr_header *headers, size_t *num_headers, size_t last_len)
{
const char *buf = buf_start, *buf_end = buf + len;
size_t max_headers = *num_headers;
int r;
*minor_version = -1;
*status = 0;
*msg = NULL;
*msg_len = 0;
*num_headers = 0;
/* if last_len != 0, check if the response is complete (a fast countermeasure
against slowloris */
if (last_len != 0 && is_complete(buf, buf_end, last_len, &r) == NULL) {
return r;
}
if ((buf = parse_response(buf, buf_end, minor_version, status, msg, msg_len, headers, num_headers, max_headers, &r)) == NULL) {
return r;
}
return (int)(buf - buf_start);
}
int phr_parse_headers(const char *buf_start, size_t len, struct phr_header *headers, size_t *num_headers, size_t last_len)
{
const char *buf = buf_start, *buf_end = buf + len;
size_t max_headers = *num_headers;
int r;
*num_headers = 0;
/* if last_len != 0, check if the response is complete (a fast countermeasure
against slowloris */
if (last_len != 0 && is_complete(buf, buf_end, last_len, &r) == NULL) {
return r;
}
if ((buf = parse_headers(buf, buf_end, headers, num_headers, max_headers, &r)) == NULL) {
return r;
}
return (int)(buf - buf_start);
}
enum {
CHUNKED_IN_CHUNK_SIZE,
CHUNKED_IN_CHUNK_EXT,
CHUNKED_IN_CHUNK_DATA,
CHUNKED_IN_CHUNK_CRLF,
CHUNKED_IN_TRAILERS_LINE_HEAD,
CHUNKED_IN_TRAILERS_LINE_MIDDLE
};
static int decode_hex(int ch)
{
if ('0' <= ch && ch <= '9') {
return ch - '0';
} else if ('A' <= ch && ch <= 'F') {
return ch - 'A' + 0xa;
} else if ('a' <= ch && ch <= 'f') {
return ch - 'a' + 0xa;
} else {
return -1;
}
}
ssize_t phr_decode_chunked(struct phr_chunked_decoder *decoder, char *buf, size_t *_bufsz)
{
size_t dst = 0, src = 0, bufsz = *_bufsz;
ssize_t ret = -2; /* incomplete */
decoder->_total_read += bufsz;
while (1) {
switch (decoder->_state) {
case CHUNKED_IN_CHUNK_SIZE:
for (;; ++src) {
int v;
if (src == bufsz)
goto Exit;
if ((v = decode_hex(buf[src])) == -1) {
if (decoder->_hex_count == 0) {
ret = -1;
goto Exit;
}
/* the only characters that may appear after the chunk size are BWS, semicolon, or CRLF */
switch (buf[src]) {
case ' ':
case '\011':
case ';':
case '\012':
case '\015':
break;
default:
ret = -1;
goto Exit;
}
break;
}
if (decoder->_hex_count == sizeof(size_t) * 2) {
ret = -1;
goto Exit;
}
decoder->bytes_left_in_chunk = decoder->bytes_left_in_chunk * 16 + v;
++decoder->_hex_count;
}
decoder->_hex_count = 0;
decoder->_state = CHUNKED_IN_CHUNK_EXT;
/* fallthru */
case CHUNKED_IN_CHUNK_EXT:
/* RFC 7230 A.2 "Line folding in chunk extensions is disallowed" */
for (;; ++src) {
if (src == bufsz)
goto Exit;
if (buf[src] == '\012')
break;
}
++src;
if (decoder->bytes_left_in_chunk == 0) {
if (decoder->consume_trailer) {
decoder->_state = CHUNKED_IN_TRAILERS_LINE_HEAD;
break;
} else {
goto Complete;
}
}
decoder->_state = CHUNKED_IN_CHUNK_DATA;
/* fallthru */
case CHUNKED_IN_CHUNK_DATA: {
size_t avail = bufsz - src;
if (avail < decoder->bytes_left_in_chunk) {
if (dst != src)
memmove(buf + dst, buf + src, avail);
src += avail;
dst += avail;
decoder->bytes_left_in_chunk -= avail;
goto Exit;
}
if (dst != src)
memmove(buf + dst, buf + src, decoder->bytes_left_in_chunk);
src += decoder->bytes_left_in_chunk;
dst += decoder->bytes_left_in_chunk;
decoder->bytes_left_in_chunk = 0;
decoder->_state = CHUNKED_IN_CHUNK_CRLF;
}
/* fallthru */
case CHUNKED_IN_CHUNK_CRLF:
for (;; ++src) {
if (src == bufsz)
goto Exit;
if (buf[src] != '\015')
break;
}
if (buf[src] != '\012') {
ret = -1;
goto Exit;
}
++src;
decoder->_state = CHUNKED_IN_CHUNK_SIZE;
break;
case CHUNKED_IN_TRAILERS_LINE_HEAD:
for (;; ++src) {
if (src == bufsz)
goto Exit;
if (buf[src] != '\015')
break;
}
if (buf[src++] == '\012')
goto Complete;
decoder->_state = CHUNKED_IN_TRAILERS_LINE_MIDDLE;
/* fallthru */
case CHUNKED_IN_TRAILERS_LINE_MIDDLE:
for (;; ++src) {
if (src == bufsz)
goto Exit;
if (buf[src] == '\012')
break;
}
++src;
decoder->_state = CHUNKED_IN_TRAILERS_LINE_HEAD;
break;
default:
assert(!"decoder is corrupt");
}
}
Complete:
ret = bufsz - src;
Exit:
if (dst != src)
memmove(buf + dst, buf + src, bufsz - src);
*_bufsz = dst;
/* if incomplete but the overhead of the chunked encoding is >=100KB and >80%, signal an error */
if (ret == -2) {
decoder->_total_overhead += bufsz - dst;
if (decoder->_total_overhead >= 100 * 1024 && decoder->_total_read - decoder->_total_overhead < decoder->_total_read / 4)
ret = -1;
}
return ret;
}
int phr_decode_chunked_is_in_data(struct phr_chunked_decoder *decoder)
{
return decoder->_state == CHUNKED_IN_CHUNK_DATA;
}
#undef CHECK_EOF
#undef EXPECT_CHAR
#undef ADVANCE_TOKEN
tests/picohttpparser.h
+90
View File
@@ -0,0 +1,90 @@
/*
* Copyright (c) 2009-2014 Kazuho Oku, Tokuhiro Matsuno, Daisuke Murase,
* Shigeo Mitsunari
*
* The software is licensed under either the MIT License (below) or the Perl
* license.
*
* Permission is hereby granted, free of charge, to any person obtaining a copy
* of this software and associated documentation files (the "Software"), to
* deal in the Software without restriction, including without limitation the
* rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
* sell copies of the Software, and to permit persons to whom the Software is
* furnished to do so, subject to the following conditions:
*
* The above copyright notice and this permission notice shall be included in
* all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
* IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
* FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
* AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
* LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
* FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
* IN THE SOFTWARE.
*/
#ifndef picohttpparser_h
#define picohttpparser_h
#include <stdint.h>
#include <sys/types.h>
#ifdef _MSC_VER
#define ssize_t intptr_t
#endif
#ifdef __cplusplus
extern "C" {
#endif
/* contains name and value of a header (name == NULL if is a continuing line
* of a multiline header */
struct phr_header {
const char *name;
size_t name_len;
const char *value;
size_t value_len;
};
/* returns number of bytes consumed if successful, -2 if request is partial,
* -1 if failed */
int phr_parse_request(const char *buf, size_t len, const char **method, size_t *method_len, const char **path, size_t *path_len,
int *minor_version, struct phr_header *headers, size_t *num_headers, size_t last_len);
/* ditto */
int phr_parse_response(const char *_buf, size_t len, int *minor_version, int *status, const char **msg, size_t *msg_len,
struct phr_header *headers, size_t *num_headers, size_t last_len);
/* ditto */
int phr_parse_headers(const char *buf, size_t len, struct phr_header *headers, size_t *num_headers, size_t last_len);
/* should be zero-filled before start */
struct phr_chunked_decoder {
size_t bytes_left_in_chunk; /* number of bytes left in current chunk */
char consume_trailer; /* if trailing headers should be consumed */
char _hex_count;
char _state;
uint64_t _total_read;
uint64_t _total_overhead;
};
/* the function rewrites the buffer given as (buf, bufsz) removing the chunked-
* encoding headers. When the function returns without an error, bufsz is
* updated to the length of the decoded data available. Applications should
* repeatedly call the function while it returns -2 (incomplete) every time
* supplying newly arrived data. If the end of the chunked-encoded data is
* found, the function returns a non-negative number indicating the number of
* octets left undecoded, that starts from the offset returned by `*bufsz`.
* Returns -1 on error.
*/
ssize_t phr_decode_chunked(struct phr_chunked_decoder *decoder, char *buf, size_t *bufsz);
/* returns if the chunked decoder is in middle of chunked data */
int phr_decode_chunked_is_in_data(struct phr_chunked_decoder *decoder);
#ifdef __cplusplus
}
#endif
#endif