aboutsummaryrefslogtreecommitdiff
path: root/src
diff options
context:
space:
mode:
authorOskari Timperi <oskari.timperi@iki.fi>2014-05-24 21:32:07 +0300
committerOskari Timperi <oskari.timperi@iki.fi>2014-05-24 21:32:07 +0300
commitf3d7fe4ca0679b199860d883b086bc443d1c29f8 (patch)
treee7854ab60244bfa4ef0fb97274e844de86e355e6 /src
downloadlibuvh-f3d7fe4ca0679b199860d883b086bc443d1c29f8.tar.gz
libuvh-f3d7fe4ca0679b199860d883b086bc443d1c29f8.zip
initial commit
Diffstat (limited to 'src')
-rw-r--r--src/http_parser.c2257
-rw-r--r--src/http_parser.h325
-rw-r--r--src/sds.c907
-rw-r--r--src/sds.h101
-rw-r--r--src/uvh.c473
-rw-r--r--src/uvh.h129
6 files changed, 4192 insertions, 0 deletions
diff --git a/src/http_parser.c b/src/http_parser.c
new file mode 100644
index 0000000..df1b696
--- /dev/null
+++ b/src/http_parser.c
@@ -0,0 +1,2257 @@
+/* Based on src/http/ngx_http_parse.c from NGINX copyright Igor Sysoev
+ *
+ * Additional changes are licensed under the same terms as NGINX and
+ * copyright Joyent, Inc. and other Node contributors. All rights reserved.
+ *
+ * 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 "http_parser.h"
+#include <assert.h>
+#include <stddef.h>
+#include <ctype.h>
+#include <stdlib.h>
+#include <string.h>
+#include <limits.h>
+
+#ifndef ULLONG_MAX
+# define ULLONG_MAX ((uint64_t) -1) /* 2^64-1 */
+#endif
+
+#ifndef MIN
+# define MIN(a,b) ((a) < (b) ? (a) : (b))
+#endif
+
+#ifndef ARRAY_SIZE
+# define ARRAY_SIZE(a) (sizeof(a) / sizeof((a)[0]))
+#endif
+
+#ifndef BIT_AT
+# define BIT_AT(a, i) \
+ (!!((unsigned int) (a)[(unsigned int) (i) >> 3] & \
+ (1 << ((unsigned int) (i) & 7))))
+#endif
+
+#ifndef ELEM_AT
+# define ELEM_AT(a, i, v) ((unsigned int) (i) < ARRAY_SIZE(a) ? (a)[(i)] : (v))
+#endif
+
+#define SET_ERRNO(e) \
+do { \
+ parser->http_errno = (e); \
+} while(0)
+
+
+/* Run the notify callback FOR, returning ER if it fails */
+#define CALLBACK_NOTIFY_(FOR, ER) \
+do { \
+ assert(HTTP_PARSER_ERRNO(parser) == HPE_OK); \
+ \
+ if (settings->on_##FOR) { \
+ if (0 != settings->on_##FOR(parser)) { \
+ SET_ERRNO(HPE_CB_##FOR); \
+ } \
+ \
+ /* We either errored above or got paused; get out */ \
+ if (HTTP_PARSER_ERRNO(parser) != HPE_OK) { \
+ return (ER); \
+ } \
+ } \
+} while (0)
+
+/* Run the notify callback FOR and consume the current byte */
+#define CALLBACK_NOTIFY(FOR) CALLBACK_NOTIFY_(FOR, p - data + 1)
+
+/* Run the notify callback FOR and don't consume the current byte */
+#define CALLBACK_NOTIFY_NOADVANCE(FOR) CALLBACK_NOTIFY_(FOR, p - data)
+
+/* Run data callback FOR with LEN bytes, returning ER if it fails */
+#define CALLBACK_DATA_(FOR, LEN, ER) \
+do { \
+ assert(HTTP_PARSER_ERRNO(parser) == HPE_OK); \
+ \
+ if (FOR##_mark) { \
+ if (settings->on_##FOR) { \
+ if (0 != settings->on_##FOR(parser, FOR##_mark, (LEN))) { \
+ SET_ERRNO(HPE_CB_##FOR); \
+ } \
+ \
+ /* We either errored above or got paused; get out */ \
+ if (HTTP_PARSER_ERRNO(parser) != HPE_OK) { \
+ return (ER); \
+ } \
+ } \
+ FOR##_mark = NULL; \
+ } \
+} while (0)
+
+/* Run the data callback FOR and consume the current byte */
+#define CALLBACK_DATA(FOR) \
+ CALLBACK_DATA_(FOR, p - FOR##_mark, p - data + 1)
+
+/* Run the data callback FOR and don't consume the current byte */
+#define CALLBACK_DATA_NOADVANCE(FOR) \
+ CALLBACK_DATA_(FOR, p - FOR##_mark, p - data)
+
+/* Set the mark FOR; non-destructive if mark is already set */
+#define MARK(FOR) \
+do { \
+ if (!FOR##_mark) { \
+ FOR##_mark = p; \
+ } \
+} while (0)
+
+
+#define PROXY_CONNECTION "proxy-connection"
+#define CONNECTION "connection"
+#define CONTENT_LENGTH "content-length"
+#define TRANSFER_ENCODING "transfer-encoding"
+#define UPGRADE "upgrade"
+#define CHUNKED "chunked"
+#define KEEP_ALIVE "keep-alive"
+#define CLOSE "close"
+
+
+static const char *method_strings[] =
+ {
+#define XX(num, name, string) #string,
+ HTTP_METHOD_MAP(XX)
+#undef XX
+ };
+
+
+/* Tokens as defined by rfc 2616. Also lowercases them.
+ * token = 1*<any CHAR except CTLs or separators>
+ * separators = "(" | ")" | "<" | ">" | "@"
+ * | "," | ";" | ":" | "\" | <">
+ * | "/" | "[" | "]" | "?" | "="
+ * | "{" | "}" | SP | HT
+ */
+static const char tokens[256] = {
+/* 0 nul 1 soh 2 stx 3 etx 4 eot 5 enq 6 ack 7 bel */
+ 0, 0, 0, 0, 0, 0, 0, 0,
+/* 8 bs 9 ht 10 nl 11 vt 12 np 13 cr 14 so 15 si */
+ 0, 0, 0, 0, 0, 0, 0, 0,
+/* 16 dle 17 dc1 18 dc2 19 dc3 20 dc4 21 nak 22 syn 23 etb */
+ 0, 0, 0, 0, 0, 0, 0, 0,
+/* 24 can 25 em 26 sub 27 esc 28 fs 29 gs 30 rs 31 us */
+ 0, 0, 0, 0, 0, 0, 0, 0,
+/* 32 sp 33 ! 34 " 35 # 36 $ 37 % 38 & 39 ' */
+ 0, '!', 0, '#', '$', '%', '&', '\'',
+/* 40 ( 41 ) 42 * 43 + 44 , 45 - 46 . 47 / */
+ 0, 0, '*', '+', 0, '-', '.', 0,
+/* 48 0 49 1 50 2 51 3 52 4 53 5 54 6 55 7 */
+ '0', '1', '2', '3', '4', '5', '6', '7',
+/* 56 8 57 9 58 : 59 ; 60 < 61 = 62 > 63 ? */
+ '8', '9', 0, 0, 0, 0, 0, 0,
+/* 64 @ 65 A 66 B 67 C 68 D 69 E 70 F 71 G */
+ 0, 'a', 'b', 'c', 'd', 'e', 'f', 'g',
+/* 72 H 73 I 74 J 75 K 76 L 77 M 78 N 79 O */
+ 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o',
+/* 80 P 81 Q 82 R 83 S 84 T 85 U 86 V 87 W */
+ 'p', 'q', 'r', 's', 't', 'u', 'v', 'w',
+/* 88 X 89 Y 90 Z 91 [ 92 \ 93 ] 94 ^ 95 _ */
+ 'x', 'y', 'z', 0, 0, 0, '^', '_',
+/* 96 ` 97 a 98 b 99 c 100 d 101 e 102 f 103 g */
+ '`', 'a', 'b', 'c', 'd', 'e', 'f', 'g',
+/* 104 h 105 i 106 j 107 k 108 l 109 m 110 n 111 o */
+ 'h', 'i', 'j', 'k', 'l', 'm', 'n', 'o',
+/* 112 p 113 q 114 r 115 s 116 t 117 u 118 v 119 w */
+ 'p', 'q', 'r', 's', 't', 'u', 'v', 'w',
+/* 120 x 121 y 122 z 123 { 124 | 125 } 126 ~ 127 del */
+ 'x', 'y', 'z', 0, '|', 0, '~', 0 };
+
+
+static const int8_t unhex[256] =
+ {-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,-1,-1,-1
+ ,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1,-1
+ , 0, 1, 2, 3, 4, 5, 6, 7, 8, 9,-1,-1,-1,-1,-1,-1
+ ,-1,10,11,12,13,14,15,-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
+ ,-1,10,11,12,13,14,15,-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
+ };
+
+
+#if HTTP_PARSER_STRICT
+# define T(v) 0
+#else
+# define T(v) v
+#endif
+
+
+static const uint8_t normal_url_char[32] = {
+/* 0 nul 1 soh 2 stx 3 etx 4 eot 5 enq 6 ack 7 bel */
+ 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0,
+/* 8 bs 9 ht 10 nl 11 vt 12 np 13 cr 14 so 15 si */
+ 0 | T(2) | 0 | 0 | T(16) | 0 | 0 | 0,
+/* 16 dle 17 dc1 18 dc2 19 dc3 20 dc4 21 nak 22 syn 23 etb */
+ 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0,
+/* 24 can 25 em 26 sub 27 esc 28 fs 29 gs 30 rs 31 us */
+ 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0,
+/* 32 sp 33 ! 34 " 35 # 36 $ 37 % 38 & 39 ' */
+ 0 | 2 | 4 | 0 | 16 | 32 | 64 | 128,
+/* 40 ( 41 ) 42 * 43 + 44 , 45 - 46 . 47 / */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 48 0 49 1 50 2 51 3 52 4 53 5 54 6 55 7 */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 56 8 57 9 58 : 59 ; 60 < 61 = 62 > 63 ? */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 0,
+/* 64 @ 65 A 66 B 67 C 68 D 69 E 70 F 71 G */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 72 H 73 I 74 J 75 K 76 L 77 M 78 N 79 O */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 80 P 81 Q 82 R 83 S 84 T 85 U 86 V 87 W */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 88 X 89 Y 90 Z 91 [ 92 \ 93 ] 94 ^ 95 _ */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 96 ` 97 a 98 b 99 c 100 d 101 e 102 f 103 g */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 104 h 105 i 106 j 107 k 108 l 109 m 110 n 111 o */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 112 p 113 q 114 r 115 s 116 t 117 u 118 v 119 w */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 128,
+/* 120 x 121 y 122 z 123 { 124 | 125 } 126 ~ 127 del */
+ 1 | 2 | 4 | 8 | 16 | 32 | 64 | 0, };
+
+#undef T
+
+enum state
+ { s_dead = 1 /* important that this is > 0 */
+
+ , s_start_req_or_res
+ , s_res_or_resp_H
+ , s_start_res
+ , s_res_H
+ , s_res_HT
+ , s_res_HTT
+ , s_res_HTTP
+ , s_res_first_http_major
+ , s_res_http_major
+ , s_res_first_http_minor
+ , s_res_http_minor
+ , s_res_first_status_code
+ , s_res_status_code
+ , s_res_status_start
+ , s_res_status
+ , s_res_line_almost_done
+
+ , s_start_req
+
+ , s_req_method
+ , s_req_spaces_before_url
+ , s_req_schema
+ , s_req_schema_slash
+ , s_req_schema_slash_slash
+ , s_req_server_start
+ , s_req_server
+ , s_req_server_with_at
+ , s_req_path
+ , s_req_query_string_start
+ , s_req_query_string
+ , s_req_fragment_start
+ , s_req_fragment
+ , s_req_http_start
+ , s_req_http_H
+ , s_req_http_HT
+ , s_req_http_HTT
+ , s_req_http_HTTP
+ , s_req_first_http_major
+ , s_req_http_major
+ , s_req_first_http_minor
+ , s_req_http_minor
+ , s_req_line_almost_done
+
+ , s_header_field_start
+ , s_header_field
+ , s_header_value_discard_ws
+ , s_header_value_discard_ws_almost_done
+ , s_header_value_discard_lws
+ , s_header_value_start
+ , s_header_value
+ , s_header_value_lws
+
+ , s_header_almost_done
+
+ , s_chunk_size_start
+ , s_chunk_size
+ , s_chunk_parameters
+ , s_chunk_size_almost_done
+
+ , s_headers_almost_done
+ , s_headers_done
+
+ /* Important: 's_headers_done' must be the last 'header' state. All
+ * states beyond this must be 'body' states. It is used for overflow
+ * checking. See the PARSING_HEADER() macro.
+ */
+
+ , s_chunk_data
+ , s_chunk_data_almost_done
+ , s_chunk_data_done
+
+ , s_body_identity
+ , s_body_identity_eof
+
+ , s_message_done
+ };
+
+
+#define PARSING_HEADER(state) (state <= s_headers_done)
+
+
+enum header_states
+ { h_general = 0
+ , h_C
+ , h_CO
+ , h_CON
+
+ , h_matching_connection
+ , h_matching_proxy_connection
+ , h_matching_content_length
+ , h_matching_transfer_encoding
+ , h_matching_upgrade
+
+ , h_connection
+ , h_content_length
+ , h_transfer_encoding
+ , h_upgrade
+
+ , h_matching_transfer_encoding_chunked
+ , h_matching_connection_keep_alive
+ , h_matching_connection_close
+
+ , h_transfer_encoding_chunked
+ , h_connection_keep_alive
+ , h_connection_close
+ };
+
+enum http_host_state
+ {
+ s_http_host_dead = 1
+ , s_http_userinfo_start
+ , s_http_userinfo
+ , s_http_host_start
+ , s_http_host_v6_start
+ , s_http_host
+ , s_http_host_v6
+ , s_http_host_v6_end
+ , s_http_host_port_start
+ , s_http_host_port
+};
+
+/* Macros for character classes; depends on strict-mode */
+#define CR '\r'
+#define LF '\n'
+#define LOWER(c) (unsigned char)(c | 0x20)
+#define IS_ALPHA(c) (LOWER(c) >= 'a' && LOWER(c) <= 'z')
+#define IS_NUM(c) ((c) >= '0' && (c) <= '9')
+#define IS_ALPHANUM(c) (IS_ALPHA(c) || IS_NUM(c))
+#define IS_HEX(c) (IS_NUM(c) || (LOWER(c) >= 'a' && LOWER(c) <= 'f'))
+#define IS_MARK(c) ((c) == '-' || (c) == '_' || (c) == '.' || \
+ (c) == '!' || (c) == '~' || (c) == '*' || (c) == '\'' || (c) == '(' || \
+ (c) == ')')
+#define IS_USERINFO_CHAR(c) (IS_ALPHANUM(c) || IS_MARK(c) || (c) == '%' || \
+ (c) == ';' || (c) == ':' || (c) == '&' || (c) == '=' || (c) == '+' || \
+ (c) == '$' || (c) == ',')
+
+#if HTTP_PARSER_STRICT
+#define TOKEN(c) (tokens[(unsigned char)c])
+#define IS_URL_CHAR(c) (BIT_AT(normal_url_char, (unsigned char)c))
+#define IS_HOST_CHAR(c) (IS_ALPHANUM(c) || (c) == '.' || (c) == '-')
+#else
+#define TOKEN(c) ((c == ' ') ? ' ' : tokens[(unsigned char)c])
+#define IS_URL_CHAR(c) \
+ (BIT_AT(normal_url_char, (unsigned char)c) || ((c) & 0x80))
+#define IS_HOST_CHAR(c) \
+ (IS_ALPHANUM(c) || (c) == '.' || (c) == '-' || (c) == '_')
+#endif
+
+
+#define start_state (parser->type == HTTP_REQUEST ? s_start_req : s_start_res)
+
+
+#if HTTP_PARSER_STRICT
+# define STRICT_CHECK(cond) \
+do { \
+ if (cond) { \
+ SET_ERRNO(HPE_STRICT); \
+ goto error; \
+ } \
+} while (0)
+# define NEW_MESSAGE() (http_should_keep_alive(parser) ? start_state : s_dead)
+#else
+# define STRICT_CHECK(cond)
+# define NEW_MESSAGE() start_state
+#endif
+
+
+/* Map errno values to strings for human-readable output */
+#define HTTP_STRERROR_GEN(n, s) { "HPE_" #n, s },
+static struct {
+ const char *name;
+ const char *description;
+} http_strerror_tab[] = {
+ HTTP_ERRNO_MAP(HTTP_STRERROR_GEN)
+};
+#undef HTTP_STRERROR_GEN
+
+int http_message_needs_eof(const http_parser *parser);
+
+/* Our URL parser.
+ *
+ * This is designed to be shared by http_parser_execute() for URL validation,
+ * hence it has a state transition + byte-for-byte interface. In addition, it
+ * is meant to be embedded in http_parser_parse_url(), which does the dirty
+ * work of turning state transitions URL components for its API.
+ *
+ * This function should only be invoked with non-space characters. It is
+ * assumed that the caller cares about (and can detect) the transition between
+ * URL and non-URL states by looking for these.
+ */
+static enum state
+parse_url_char(enum state s, const char ch)
+{
+ if (ch == ' ' || ch == '\r' || ch == '\n') {
+ return s_dead;
+ }
+
+#if HTTP_PARSER_STRICT
+ if (ch == '\t' || ch == '\f') {
+ return s_dead;
+ }
+#endif
+
+ switch (s) {
+ case s_req_spaces_before_url:
+ /* Proxied requests are followed by scheme of an absolute URI (alpha).
+ * All methods except CONNECT are followed by '/' or '*'.
+ */
+
+ if (ch == '/' || ch == '*') {
+ return s_req_path;
+ }
+
+ if (IS_ALPHA(ch)) {
+ return s_req_schema;
+ }
+
+ break;
+
+ case s_req_schema:
+ if (IS_ALPHA(ch)) {
+ return s;
+ }
+
+ if (ch == ':') {
+ return s_req_schema_slash;
+ }
+
+ break;
+
+ case s_req_schema_slash:
+ if (ch == '/') {
+ return s_req_schema_slash_slash;
+ }
+
+ break;
+
+ case s_req_schema_slash_slash:
+ if (ch == '/') {
+ return s_req_server_start;
+ }
+
+ break;
+
+ case s_req_server_with_at:
+ if (ch == '@') {
+ return s_dead;
+ }
+
+ /* FALLTHROUGH */
+ case s_req_server_start:
+ case s_req_server:
+ if (ch == '/') {
+ return s_req_path;
+ }
+
+ if (ch == '?') {
+ return s_req_query_string_start;
+ }
+
+ if (ch == '@') {
+ return s_req_server_with_at;
+ }
+
+ if (IS_USERINFO_CHAR(ch) || ch == '[' || ch == ']') {
+ return s_req_server;
+ }
+
+ break;
+
+ case s_req_path:
+ if (IS_URL_CHAR(ch)) {
+ return s;
+ }
+
+ switch (ch) {
+ case '?':
+ return s_req_query_string_start;
+
+ case '#':
+ return s_req_fragment_start;
+ }
+
+ break;
+
+ case s_req_query_string_start:
+ case s_req_query_string:
+ if (IS_URL_CHAR(ch)) {
+ return s_req_query_string;
+ }
+
+ switch (ch) {
+ case '?':
+ /* allow extra '?' in query string */
+ return s_req_query_string;
+
+ case '#':
+ return s_req_fragment_start;
+ }
+
+ break;
+
+ case s_req_fragment_start:
+ if (IS_URL_CHAR(ch)) {
+ return s_req_fragment;
+ }
+
+ switch (ch) {
+ case '?':
+ return s_req_fragment;
+
+ case '#':
+ return s;
+ }
+
+ break;
+
+ case s_req_fragment:
+ if (IS_URL_CHAR(ch)) {
+ return s;
+ }
+
+ switch (ch) {
+ case '?':
+ case '#':
+ return s;
+ }
+
+ break;
+
+ default:
+ break;
+ }
+
+ /* We should never fall out of the switch above unless there's an error */
+ return s_dead;
+}
+
+size_t http_parser_execute (http_parser *parser,
+ const http_parser_settings *settings,
+ const char *data,
+ size_t len)
+{
+ char c, ch;
+ int8_t unhex_val;
+ const char *p = data;
+ const char *header_field_mark = 0;
+ const char *header_value_mark = 0;
+ const char *url_mark = 0;
+ const char *body_mark = 0;
+ const char *status_mark = 0;
+
+ /* We're in an error state. Don't bother doing anything. */
+ if (HTTP_PARSER_ERRNO(parser) != HPE_OK) {
+ return 0;
+ }
+
+ if (len == 0) {
+ switch (parser->state) {
+ case s_body_identity_eof:
+ /* Use of CALLBACK_NOTIFY() here would erroneously return 1 byte read if
+ * we got paused.
+ */
+ CALLBACK_NOTIFY_NOADVANCE(message_complete);
+ return 0;
+
+ case s_dead:
+ case s_start_req_or_res:
+ case s_start_res:
+ case s_start_req:
+ return 0;
+
+ default:
+ SET_ERRNO(HPE_INVALID_EOF_STATE);
+ return 1;
+ }
+ }
+
+
+ if (parser->state == s_header_field)
+ header_field_mark = data;
+ if (parser->state == s_header_value)
+ header_value_mark = data;
+ switch (parser->state) {
+ case s_req_path:
+ case s_req_schema:
+ case s_req_schema_slash:
+ case s_req_schema_slash_slash:
+ case s_req_server_start:
+ case s_req_server:
+ case s_req_server_with_at:
+ case s_req_query_string_start:
+ case s_req_query_string:
+ case s_req_fragment_start:
+ case s_req_fragment:
+ url_mark = data;
+ break;
+ case s_res_status:
+ status_mark = data;
+ break;
+ }
+
+ for (p=data; p != data + len; p++) {
+ ch = *p;
+
+ if (PARSING_HEADER(parser->state)) {
+ ++parser->nread;
+ /* Don't allow the total size of the HTTP headers (including the status
+ * line) to exceed HTTP_MAX_HEADER_SIZE. This check is here to protect
+ * embedders against denial-of-service attacks where the attacker feeds
+ * us a never-ending header that the embedder keeps buffering.
+ *
+ * This check is arguably the responsibility of embedders but we're doing
+ * it on the embedder's behalf because most won't bother and this way we
+ * make the web a little safer. HTTP_MAX_HEADER_SIZE is still far bigger
+ * than any reasonable request or response so this should never affect
+ * day-to-day operation.
+ */
+ if (parser->nread > (HTTP_MAX_HEADER_SIZE)) {
+ SET_ERRNO(HPE_HEADER_OVERFLOW);
+ goto error;
+ }
+ }
+
+ reexecute_byte:
+ switch (parser->state) {
+
+ case s_dead:
+ /* this state is used after a 'Connection: close' message
+ * the parser will error out if it reads another message
+ */
+ if (ch == CR || ch == LF)
+ break;
+
+ SET_ERRNO(HPE_CLOSED_CONNECTION);
+ goto error;
+
+ case s_start_req_or_res:
+ {
+ if (ch == CR || ch == LF)
+ break;
+ parser->flags = 0;
+ parser->content_length = ULLONG_MAX;
+
+ if (ch == 'H') {
+ parser->state = s_res_or_resp_H;
+
+ CALLBACK_NOTIFY(message_begin);
+ } else {
+ parser->type = HTTP_REQUEST;
+ parser->state = s_start_req;
+ goto reexecute_byte;
+ }
+
+ break;
+ }
+
+ case s_res_or_resp_H:
+ if (ch == 'T') {
+ parser->type = HTTP_RESPONSE;
+ parser->state = s_res_HT;
+ } else {
+ if (ch != 'E') {
+ SET_ERRNO(HPE_INVALID_CONSTANT);
+ goto error;
+ }
+
+ parser->type = HTTP_REQUEST;
+ parser->method = HTTP_HEAD;
+ parser->index = 2;
+ parser->state = s_req_method;
+ }
+ break;
+
+ case s_start_res:
+ {
+ parser->flags = 0;
+ parser->content_length = ULLONG_MAX;
+
+ switch (ch) {
+ case 'H':
+ parser->state = s_res_H;
+ break;
+
+ case CR:
+ case LF:
+ break;
+
+ default:
+ SET_ERRNO(HPE_INVALID_CONSTANT);
+ goto error;
+ }
+
+ CALLBACK_NOTIFY(message_begin);
+ break;
+ }
+
+ case s_res_H:
+ STRICT_CHECK(ch != 'T');
+ parser->state = s_res_HT;
+ break;
+
+ case s_res_HT:
+ STRICT_CHECK(ch != 'T');
+ parser->state = s_res_HTT;
+ break;
+
+ case s_res_HTT:
+ STRICT_CHECK(ch != 'P');
+ parser->state = s_res_HTTP;
+ break;
+
+ case s_res_HTTP:
+ STRICT_CHECK(ch != '/');
+ parser->state = s_res_first_http_major;
+ break;
+
+ case s_res_first_http_major:
+ if (ch < '0' || ch > '9') {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ parser->http_major = ch - '0';
+ parser->state = s_res_http_major;
+ break;
+
+ /* major HTTP version or dot */
+ case s_res_http_major:
+ {
+ if (ch == '.') {
+ parser->state = s_res_first_http_minor;
+ break;
+ }
+
+ if (!IS_NUM(ch)) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ parser->http_major *= 10;
+ parser->http_major += ch - '0';
+
+ if (parser->http_major > 999) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ break;
+ }
+
+ /* first digit of minor HTTP version */
+ case s_res_first_http_minor:
+ if (!IS_NUM(ch)) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ parser->http_minor = ch - '0';
+ parser->state = s_res_http_minor;
+ break;
+
+ /* minor HTTP version or end of request line */
+ case s_res_http_minor:
+ {
+ if (ch == ' ') {
+ parser->state = s_res_first_status_code;
+ break;
+ }
+
+ if (!IS_NUM(ch)) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ parser->http_minor *= 10;
+ parser->http_minor += ch - '0';
+
+ if (parser->http_minor > 999) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ break;
+ }
+
+ case s_res_first_status_code:
+ {
+ if (!IS_NUM(ch)) {
+ if (ch == ' ') {
+ break;
+ }
+
+ SET_ERRNO(HPE_INVALID_STATUS);
+ goto error;
+ }
+ parser->status_code = ch - '0';
+ parser->state = s_res_status_code;
+ break;
+ }
+
+ case s_res_status_code:
+ {
+ if (!IS_NUM(ch)) {
+ switch (ch) {
+ case ' ':
+ parser->state = s_res_status_start;
+ break;
+ case CR:
+ parser->state = s_res_line_almost_done;
+ break;
+ case LF:
+ parser->state = s_header_field_start;
+ break;
+ default:
+ SET_ERRNO(HPE_INVALID_STATUS);
+ goto error;
+ }
+ break;
+ }
+
+ parser->status_code *= 10;
+ parser->status_code += ch - '0';
+
+ if (parser->status_code > 999) {
+ SET_ERRNO(HPE_INVALID_STATUS);
+ goto error;
+ }
+
+ break;
+ }
+
+ case s_res_status_start:
+ {
+ if (ch == CR) {
+ parser->state = s_res_line_almost_done;
+ break;
+ }
+
+ if (ch == LF) {
+ parser->state = s_header_field_start;
+ break;
+ }
+
+ MARK(status);
+ parser->state = s_res_status;
+ parser->index = 0;
+ break;
+ }
+
+ case s_res_status:
+ if (ch == CR) {
+ parser->state = s_res_line_almost_done;
+ CALLBACK_DATA(status);
+ break;
+ }
+
+ if (ch == LF) {
+ parser->state = s_header_field_start;
+ CALLBACK_DATA(status);
+ break;
+ }
+
+ break;
+
+ case s_res_line_almost_done:
+ STRICT_CHECK(ch != LF);
+ parser->state = s_header_field_start;
+ break;
+
+ case s_start_req:
+ {
+ if (ch == CR || ch == LF)
+ break;
+ parser->flags = 0;
+ parser->content_length = ULLONG_MAX;
+
+ if (!IS_ALPHA(ch)) {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+
+ parser->method = (enum http_method) 0;
+ parser->index = 1;
+ switch (ch) {
+ case 'C': parser->method = HTTP_CONNECT; /* or COPY, CHECKOUT */ break;
+ case 'D': parser->method = HTTP_DELETE; break;
+ case 'G': parser->method = HTTP_GET; break;
+ case 'H': parser->method = HTTP_HEAD; break;
+ case 'L': parser->method = HTTP_LOCK; break;
+ case 'M': parser->method = HTTP_MKCOL; /* or MOVE, MKACTIVITY, MERGE, M-SEARCH */ break;
+ case 'N': parser->method = HTTP_NOTIFY; break;
+ case 'O': parser->method = HTTP_OPTIONS; break;
+ case 'P': parser->method = HTTP_POST;
+ /* or PROPFIND|PROPPATCH|PUT|PATCH|PURGE */
+ break;
+ case 'R': parser->method = HTTP_REPORT; break;
+ case 'S': parser->method = HTTP_SUBSCRIBE; /* or SEARCH */ break;
+ case 'T': parser->method = HTTP_TRACE; break;
+ case 'U': parser->method = HTTP_UNLOCK; /* or UNSUBSCRIBE */ break;
+ default:
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+ parser->state = s_req_method;
+
+ CALLBACK_NOTIFY(message_begin);
+
+ break;
+ }
+
+ case s_req_method:
+ {
+ const char *matcher;
+ if (ch == '\0') {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+
+ matcher = method_strings[parser->method];
+ if (ch == ' ' && matcher[parser->index] == '\0') {
+ parser->state = s_req_spaces_before_url;
+ } else if (ch == matcher[parser->index]) {
+ ; /* nada */
+ } else if (parser->method == HTTP_CONNECT) {
+ if (parser->index == 1 && ch == 'H') {
+ parser->method = HTTP_CHECKOUT;
+ } else if (parser->index == 2 && ch == 'P') {
+ parser->method = HTTP_COPY;
+ } else {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+ } else if (parser->method == HTTP_MKCOL) {
+ if (parser->index == 1 && ch == 'O') {
+ parser->method = HTTP_MOVE;
+ } else if (parser->index == 1 && ch == 'E') {
+ parser->method = HTTP_MERGE;
+ } else if (parser->index == 1 && ch == '-') {
+ parser->method = HTTP_MSEARCH;
+ } else if (parser->index == 2 && ch == 'A') {
+ parser->method = HTTP_MKACTIVITY;
+ } else {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+ } else if (parser->method == HTTP_SUBSCRIBE) {
+ if (parser->index == 1 && ch == 'E') {
+ parser->method = HTTP_SEARCH;
+ } else {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+ } else if (parser->index == 1 && parser->method == HTTP_POST) {
+ if (ch == 'R') {
+ parser->method = HTTP_PROPFIND; /* or HTTP_PROPPATCH */
+ } else if (ch == 'U') {
+ parser->method = HTTP_PUT; /* or HTTP_PURGE */
+ } else if (ch == 'A') {
+ parser->method = HTTP_PATCH;
+ } else {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+ } else if (parser->index == 2) {
+ if (parser->method == HTTP_PUT) {
+ if (ch == 'R') {
+ parser->method = HTTP_PURGE;
+ } else {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+ } else if (parser->method == HTTP_UNLOCK) {
+ if (ch == 'S') {
+ parser->method = HTTP_UNSUBSCRIBE;
+ } else {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+ } else {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+ } else if (parser->index == 4 && parser->method == HTTP_PROPFIND && ch == 'P') {
+ parser->method = HTTP_PROPPATCH;
+ } else {
+ SET_ERRNO(HPE_INVALID_METHOD);
+ goto error;
+ }
+
+ ++parser->index;
+ break;
+ }
+
+ case s_req_spaces_before_url:
+ {
+ if (ch == ' ') break;
+
+ MARK(url);
+ if (parser->method == HTTP_CONNECT) {
+ parser->state = s_req_server_start;
+ }
+
+ parser->state = parse_url_char((enum state)parser->state, ch);
+ if (parser->state == s_dead) {
+ SET_ERRNO(HPE_INVALID_URL);
+ goto error;
+ }
+
+ break;
+ }
+
+ case s_req_schema:
+ case s_req_schema_slash:
+ case s_req_schema_slash_slash:
+ case s_req_server_start:
+ {
+ switch (ch) {
+ /* No whitespace allowed here */
+ case ' ':
+ case CR:
+ case LF:
+ SET_ERRNO(HPE_INVALID_URL);
+ goto error;
+ default:
+ parser->state = parse_url_char((enum state)parser->state, ch);
+ if (parser->state == s_dead) {
+ SET_ERRNO(HPE_INVALID_URL);
+ goto error;
+ }
+ }
+
+ break;
+ }
+
+ case s_req_server:
+ case s_req_server_with_at:
+ case s_req_path:
+ case s_req_query_string_start:
+ case s_req_query_string:
+ case s_req_fragment_start:
+ case s_req_fragment:
+ {
+ switch (ch) {
+ case ' ':
+ parser->state = s_req_http_start;
+ CALLBACK_DATA(url);
+ break;
+ case CR:
+ case LF:
+ parser->http_major = 0;
+ parser->http_minor = 9;
+ parser->state = (ch == CR) ?
+ s_req_line_almost_done :
+ s_header_field_start;
+ CALLBACK_DATA(url);
+ break;
+ default:
+ parser->state = parse_url_char((enum state)parser->state, ch);
+ if (parser->state == s_dead) {
+ SET_ERRNO(HPE_INVALID_URL);
+ goto error;
+ }
+ }
+ break;
+ }
+
+ case s_req_http_start:
+ switch (ch) {
+ case 'H':
+ parser->state = s_req_http_H;
+ break;
+ case ' ':
+ break;
+ default:
+ SET_ERRNO(HPE_INVALID_CONSTANT);
+ goto error;
+ }
+ break;
+
+ case s_req_http_H:
+ STRICT_CHECK(ch != 'T');
+ parser->state = s_req_http_HT;
+ break;
+
+ case s_req_http_HT:
+ STRICT_CHECK(ch != 'T');
+ parser->state = s_req_http_HTT;
+ break;
+
+ case s_req_http_HTT:
+ STRICT_CHECK(ch != 'P');
+ parser->state = s_req_http_HTTP;
+ break;
+
+ case s_req_http_HTTP:
+ STRICT_CHECK(ch != '/');
+ parser->state = s_req_first_http_major;
+ break;
+
+ /* first digit of major HTTP version */
+ case s_req_first_http_major:
+ if (ch < '1' || ch > '9') {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ parser->http_major = ch - '0';
+ parser->state = s_req_http_major;
+ break;
+
+ /* major HTTP version or dot */
+ case s_req_http_major:
+ {
+ if (ch == '.') {
+ parser->state = s_req_first_http_minor;
+ break;
+ }
+
+ if (!IS_NUM(ch)) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ parser->http_major *= 10;
+ parser->http_major += ch - '0';
+
+ if (parser->http_major > 999) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ break;
+ }
+
+ /* first digit of minor HTTP version */
+ case s_req_first_http_minor:
+ if (!IS_NUM(ch)) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ parser->http_minor = ch - '0';
+ parser->state = s_req_http_minor;
+ break;
+
+ /* minor HTTP version or end of request line */
+ case s_req_http_minor:
+ {
+ if (ch == CR) {
+ parser->state = s_req_line_almost_done;
+ break;
+ }
+
+ if (ch == LF) {
+ parser->state = s_header_field_start;
+ break;
+ }
+
+ /* XXX allow spaces after digit? */
+
+ if (!IS_NUM(ch)) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ parser->http_minor *= 10;
+ parser->http_minor += ch - '0';
+
+ if (parser->http_minor > 999) {
+ SET_ERRNO(HPE_INVALID_VERSION);
+ goto error;
+ }
+
+ break;
+ }
+
+ /* end of request line */
+ case s_req_line_almost_done:
+ {
+ if (ch != LF) {
+ SET_ERRNO(HPE_LF_EXPECTED);
+ goto error;
+ }
+
+ parser->state = s_header_field_start;
+ break;
+ }
+
+ case s_header_field_start:
+ {
+ if (ch == CR) {
+ parser->state = s_headers_almost_done;
+ break;
+ }
+
+ if (ch == LF) {
+ /* they might be just sending \n instead of \r\n so this would be
+ * the second \n to denote the end of headers*/
+ parser->state = s_headers_almost_done;
+ goto reexecute_byte;
+ }
+
+ c = TOKEN(ch);
+
+ if (!c) {
+ SET_ERRNO(HPE_INVALID_HEADER_TOKEN);
+ goto error;
+ }
+
+ MARK(header_field);
+
+ parser->index = 0;
+ parser->state = s_header_field;
+
+ switch (c) {
+ case 'c':
+ parser->header_state = h_C;
+ break;
+
+ case 'p':
+ parser->header_state = h_matching_proxy_connection;
+ break;
+
+ case 't':
+ parser->header_state = h_matching_transfer_encoding;
+ break;
+
+ case 'u':
+ parser->header_state = h_matching_upgrade;
+ break;
+
+ default:
+ parser->header_state = h_general;
+ break;
+ }
+ break;
+ }
+
+ case s_header_field:
+ {
+ c = TOKEN(ch);
+
+ if (c) {
+ switch (parser->header_state) {
+ case h_general:
+ break;
+
+ case h_C:
+ parser->index++;
+ parser->header_state = (c == 'o' ? h_CO : h_general);
+ break;
+
+ case h_CO:
+ parser->index++;
+ parser->header_state = (c == 'n' ? h_CON : h_general);
+ break;
+
+ case h_CON:
+ parser->index++;
+ switch (c) {
+ case 'n':
+ parser->header_state = h_matching_connection;
+ break;
+ case 't':
+ parser->header_state = h_matching_content_length;
+ break;
+ default:
+ parser->header_state = h_general;
+ break;
+ }
+ break;
+
+ /* connection */
+
+ case h_matching_connection:
+ parser->index++;
+ if (parser->index > sizeof(CONNECTION)-1
+ || c != CONNECTION[parser->index]) {
+ parser->header_state = h_general;
+ } else if (parser->index == sizeof(CONNECTION)-2) {
+ parser->header_state = h_connection;
+ }
+ break;
+
+ /* proxy-connection */
+
+ case h_matching_proxy_connection:
+ parser->index++;
+ if (parser->index > sizeof(PROXY_CONNECTION)-1
+ || c != PROXY_CONNECTION[parser->index]) {
+ parser->header_state = h_general;
+ } else if (parser->index == sizeof(PROXY_CONNECTION)-2) {
+ parser->header_state = h_connection;
+ }
+ break;
+
+ /* content-length */
+
+ case h_matching_content_length:
+ parser->index++;
+ if (parser->index > sizeof(CONTENT_LENGTH)-1
+ || c != CONTENT_LENGTH[parser->index]) {
+ parser->header_state = h_general;
+ } else if (parser->index == sizeof(CONTENT_LENGTH)-2) {
+ parser->header_state = h_content_length;
+ }
+ break;
+
+ /* transfer-encoding */
+
+ case h_matching_transfer_encoding:
+ parser->index++;
+ if (parser->index > sizeof(TRANSFER_ENCODING)-1
+ || c != TRANSFER_ENCODING[parser->index]) {
+ parser->header_state = h_general;
+ } else if (parser->index == sizeof(TRANSFER_ENCODING)-2) {
+ parser->header_state = h_transfer_encoding;
+ }
+ break;
+
+ /* upgrade */
+
+ case h_matching_upgrade:
+ parser->index++;
+ if (parser->index > sizeof(UPGRADE)-1
+ || c != UPGRADE[parser->index]) {
+ parser->header_state = h_general;
+ } else if (parser->index == sizeof(UPGRADE)-2) {
+ parser->header_state = h_upgrade;
+ }
+ break;
+
+ case h_connection:
+ case h_content_length:
+ case h_transfer_encoding:
+ case h_upgrade:
+ if (ch != ' ') parser->header_state = h_general;
+ break;
+
+ default:
+ assert(0 && "Unknown header_state");
+ break;
+ }
+ break;
+ }
+
+ if (ch == ':') {
+ parser->state = s_header_value_discard_ws;
+ CALLBACK_DATA(header_field);
+ break;
+ }
+
+ if (ch == CR) {
+ parser->state = s_header_almost_done;
+ CALLBACK_DATA(header_field);
+ break;
+ }
+
+ if (ch == LF) {
+ parser->state = s_header_field_start;
+ CALLBACK_DATA(header_field);
+ break;
+ }
+
+ SET_ERRNO(HPE_INVALID_HEADER_TOKEN);
+ goto error;
+ }
+
+ case s_header_value_discard_ws:
+ if (ch == ' ' || ch == '\t') break;
+
+ if (ch == CR) {
+ parser->state = s_header_value_discard_ws_almost_done;
+ break;
+ }
+
+ if (ch == LF) {
+ parser->state = s_header_value_discard_lws;
+ break;
+ }
+
+ /* FALLTHROUGH */
+
+ case s_header_value_start:
+ {
+ MARK(header_value);
+
+ parser->state = s_header_value;
+ parser->index = 0;
+
+ c = LOWER(ch);
+
+ switch (parser->header_state) {
+ case h_upgrade:
+ parser->flags |= F_UPGRADE;
+ parser->header_state = h_general;
+ break;
+
+ case h_transfer_encoding:
+ /* looking for 'Transfer-Encoding: chunked' */
+ if ('c' == c) {
+ parser->header_state = h_matching_transfer_encoding_chunked;
+ } else {
+ parser->header_state = h_general;
+ }
+ break;
+
+ case h_content_length:
+ if (!IS_NUM(ch)) {
+ SET_ERRNO(HPE_INVALID_CONTENT_LENGTH);
+ goto error;
+ }
+
+ parser->content_length = ch - '0';
+ break;
+
+ case h_connection:
+ /* looking for 'Connection: keep-alive' */
+ if (c == 'k') {
+ parser->header_state = h_matching_connection_keep_alive;
+ /* looking for 'Connection: close' */
+ } else if (c == 'c') {
+ parser->header_state = h_matching_connection_close;
+ } else {
+ parser->header_state = h_general;
+ }
+ break;
+
+ default:
+ parser->header_state = h_general;
+ break;
+ }
+ break;
+ }
+
+ case s_header_value:
+ {
+
+ if (ch == CR) {
+ parser->state = s_header_almost_done;
+ CALLBACK_DATA(header_value);
+ break;
+ }
+
+ if (ch == LF) {
+ parser->state = s_header_almost_done;
+ CALLBACK_DATA_NOADVANCE(header_value);
+ goto reexecute_byte;
+ }
+
+ c = LOWER(ch);
+
+ switch (parser->header_state) {
+ case h_general:
+ break;
+
+ case h_connection:
+ case h_transfer_encoding:
+ assert(0 && "Shouldn't get here.");
+ break;
+
+ case h_content_length:
+ {
+ uint64_t t;
+
+ if (ch == ' ') break;
+
+ if (!IS_NUM(ch)) {
+ SET_ERRNO(HPE_INVALID_CONTENT_LENGTH);
+ goto error;
+ }
+
+ t = parser->content_length;
+ t *= 10;
+ t += ch - '0';
+
+ /* Overflow? Test against a conservative limit for simplicity. */
+ if ((ULLONG_MAX - 10) / 10 < parser->content_length) {
+ SET_ERRNO(HPE_INVALID_CONTENT_LENGTH);
+ goto error;
+ }
+
+ parser->content_length = t;
+ break;
+ }
+
+ /* Transfer-Encoding: chunked */
+ case h_matching_transfer_encoding_chunked:
+ parser->index++;
+ if (parser->index > sizeof(CHUNKED)-1
+ || c != CHUNKED[parser->index]) {
+ parser->header_state = h_general;
+ } else if (parser->index == sizeof(CHUNKED)-2) {
+ parser->header_state = h_transfer_encoding_chunked;
+ }
+ break;
+
+ /* looking for 'Connection: keep-alive' */
+ case h_matching_connection_keep_alive:
+ parser->index++;
+ if (parser->index > sizeof(KEEP_ALIVE)-1
+ || c != KEEP_ALIVE[parser->index]) {
+ parser->header_state = h_general;
+ } else if (parser->index == sizeof(KEEP_ALIVE)-2) {
+ parser->header_state = h_connection_keep_alive;
+ }
+ break;
+
+ /* looking for 'Connection: close' */
+ case h_matching_connection_close:
+ parser->index++;
+ if (parser->index > sizeof(CLOSE)-1 || c != CLOSE[parser->index]) {
+ parser->header_state = h_general;
+ } else if (parser->index == sizeof(CLOSE)-2) {
+ parser->header_state = h_connection_close;
+ }
+ break;
+
+ case h_transfer_encoding_chunked:
+ case h_connection_keep_alive:
+ case h_connection_close:
+ if (ch != ' ') parser->header_state = h_general;
+ break;
+
+ default:
+ parser->state = s_header_value;
+ parser->header_state = h_general;
+ break;
+ }
+ break;
+ }
+
+ case s_header_almost_done:
+ {
+ STRICT_CHECK(ch != LF);
+
+ parser->state = s_header_value_lws;
+ break;
+ }
+
+ case s_header_value_lws:
+ {
+ if (ch == ' ' || ch == '\t') {
+ parser->state = s_header_value_start;
+ goto reexecute_byte;
+ }
+
+ /* finished the header */
+ switch (parser->header_state) {
+ case h_connection_keep_alive:
+ parser->flags |= F_CONNECTION_KEEP_ALIVE;
+ break;
+ case h_connection_close:
+ parser->flags |= F_CONNECTION_CLOSE;
+ break;
+ case h_transfer_encoding_chunked:
+ parser->flags |= F_CHUNKED;
+ break;
+ default:
+ break;
+ }
+
+ parser->state = s_header_field_start;
+ goto reexecute_byte;
+ }
+
+ case s_header_value_discard_ws_almost_done:
+ {
+ STRICT_CHECK(ch != LF);
+ parser->state = s_header_value_discard_lws;
+ break;
+ }
+
+ case s_header_value_discard_lws:
+ {
+ if (ch == ' ' || ch == '\t') {
+ parser->state = s_header_value_discard_ws;
+ break;
+ } else {
+ /* header value was empty */
+ MARK(header_value);
+ parser->state = s_header_field_start;
+ CALLBACK_DATA_NOADVANCE(header_value);
+ goto reexecute_byte;
+ }
+ }
+
+ case s_headers_almost_done:
+ {
+ STRICT_CHECK(ch != LF);
+
+ if (parser->flags & F_TRAILING) {
+ /* End of a chunked request */
+ parser->state = NEW_MESSAGE();
+ CALLBACK_NOTIFY(message_complete);
+ break;
+ }
+
+ parser->state = s_headers_done;
+
+ /* Set this here so that on_headers_complete() callbacks can see it */
+ parser->upgrade =
+ (parser->flags & F_UPGRADE || parser->method == HTTP_CONNECT);
+
+ /* Here we call the headers_complete callback. This is somewhat
+ * different than other callbacks because if the user returns 1, we
+ * will interpret that as saying that this message has no body. This
+ * is needed for the annoying case of recieving a response to a HEAD
+ * request.
+ *
+ * We'd like to use CALLBACK_NOTIFY_NOADVANCE() here but we cannot, so
+ * we have to simulate it by handling a change in errno below.
+ */
+ if (settings->on_headers_complete) {
+ switch (settings->on_headers_complete(parser)) {
+ case 0:
+ break;
+
+ case 1:
+ parser->flags |= F_SKIPBODY;
+ break;
+
+ default:
+ SET_ERRNO(HPE_CB_headers_complete);
+ return p - data; /* Error */
+ }
+ }
+
+ if (HTTP_PARSER_ERRNO(parser) != HPE_OK) {
+ return p - data;
+ }
+
+ goto reexecute_byte;
+ }
+
+ case s_headers_done:
+ {
+ STRICT_CHECK(ch != LF);
+
+ parser->nread = 0;
+
+ /* Exit, the rest of the connect is in a different protocol. */
+ if (parser->upgrade) {
+ parser->state = NEW_MESSAGE();
+ CALLBACK_NOTIFY(message_complete);
+ return (p - data) + 1;
+ }
+
+ if (parser->flags & F_SKIPBODY) {
+ parser->state = NEW_MESSAGE();
+ CALLBACK_NOTIFY(message_complete);
+ } else if (parser->flags & F_CHUNKED) {
+ /* chunked encoding - ignore Content-Length header */
+ parser->state = s_chunk_size_start;
+ } else {
+ if (parser->content_length == 0) {
+ /* Content-Length header given but zero: Content-Length: 0\r\n */
+ parser->state = NEW_MESSAGE();
+ CALLBACK_NOTIFY(message_complete);
+ } else if (parser->content_length != ULLONG_MAX) {
+ /* Content-Length header given and non-zero */
+ parser->state = s_body_identity;
+ } else {
+ if (parser->type == HTTP_REQUEST ||
+ !http_message_needs_eof(parser)) {
+ /* Assume content-length 0 - read the next */
+ parser->state = NEW_MESSAGE();
+ CALLBACK_NOTIFY(message_complete);
+ } else {
+ /* Read body until EOF */
+ parser->state = s_body_identity_eof;
+ }
+ }
+ }
+
+ break;
+ }
+
+ case s_body_identity:
+ {
+ uint64_t to_read = MIN(parser->content_length,
+ (uint64_t) ((data + len) - p));
+
+ assert(parser->content_length != 0
+ && parser->content_length != ULLONG_MAX);
+
+ /* The difference between advancing content_length and p is because
+ * the latter will automaticaly advance on the next loop iteration.
+ * Further, if content_length ends up at 0, we want to see the last
+ * byte again for our message complete callback.
+ */
+ MARK(body);
+ parser->content_length -= to_read;
+ p += to_read - 1;
+
+ if (parser->content_length == 0) {
+ parser->state = s_message_done;
+
+ /* Mimic CALLBACK_DATA_NOADVANCE() but with one extra byte.
+ *
+ * The alternative to doing this is to wait for the next byte to
+ * trigger the data callback, just as in every other case. The
+ * problem with this is that this makes it difficult for the test
+ * harness to distinguish between complete-on-EOF and
+ * complete-on-length. It's not clear that this distinction is
+ * important for applications, but let's keep it for now.
+ */
+ CALLBACK_DATA_(body, p - body_mark + 1, p - data);
+ goto reexecute_byte;
+ }
+
+ break;
+ }
+
+ /* read until EOF */
+ case s_body_identity_eof:
+ MARK(body);
+ p = data + len - 1;
+
+ break;
+
+ case s_message_done:
+ parser->state = NEW_MESSAGE();
+ CALLBACK_NOTIFY(message_complete);
+ break;
+
+ case s_chunk_size_start:
+ {
+ assert(parser->nread == 1);
+ assert(parser->flags & F_CHUNKED);
+
+ unhex_val = unhex[(unsigned char)ch];
+ if (unhex_val == -1) {
+ SET_ERRNO(HPE_INVALID_CHUNK_SIZE);
+ goto error;
+ }
+
+ parser->content_length = unhex_val;
+ parser->state = s_chunk_size;
+ break;
+ }
+
+ case s_chunk_size:
+ {
+ uint64_t t;
+
+ assert(parser->flags & F_CHUNKED);
+
+ if (ch == CR) {
+ parser->state = s_chunk_size_almost_done;
+ break;
+ }
+
+ unhex_val = unhex[(unsigned char)ch];
+
+ if (unhex_val == -1) {
+ if (ch == ';' || ch == ' ') {
+ parser->state = s_chunk_parameters;
+ break;
+ }
+
+ SET_ERRNO(HPE_INVALID_CHUNK_SIZE);
+ goto error;
+ }
+
+ t = parser->content_length;
+ t *= 16;
+ t += unhex_val;
+
+ /* Overflow? Test against a conservative limit for simplicity. */
+ if ((ULLONG_MAX - 16) / 16 < parser->content_length) {
+ SET_ERRNO(HPE_INVALID_CONTENT_LENGTH);
+ goto error;
+ }
+
+ parser->content_length = t;
+ break;
+ }
+
+ case s_chunk_parameters:
+ {
+ assert(parser->flags & F_CHUNKED);
+ /* just ignore this shit. TODO check for overflow */
+ if (ch == CR) {
+ parser->state = s_chunk_size_almost_done;
+ break;
+ }
+ break;
+ }
+
+ case s_chunk_size_almost_done:
+ {
+ assert(parser->flags & F_CHUNKED);
+ STRICT_CHECK(ch != LF);
+
+ parser->nread = 0;
+
+ if (parser->content_length == 0) {
+ parser->flags |= F_TRAILING;
+ parser->state = s_header_field_start;
+ } else {
+ parser->state = s_chunk_data;
+ }
+ break;
+ }
+
+ case s_chunk_data:
+ {
+ uint64_t to_read = MIN(parser->content_length,
+ (uint64_t) ((data + len) - p));
+
+ assert(parser->flags & F_CHUNKED);
+ assert(parser->content_length != 0
+ && parser->content_length != ULLONG_MAX);
+
+ /* See the explanation in s_body_identity for why the content
+ * length and data pointers are managed this way.
+ */
+ MARK(body);
+ parser->content_length -= to_read;
+ p += to_read - 1;
+
+ if (parser->content_length == 0) {
+ parser->state = s_chunk_data_almost_done;
+ }
+
+ break;
+ }
+
+ case s_chunk_data_almost_done:
+ assert(parser->flags & F_CHUNKED);
+ assert(parser->content_length == 0);
+ STRICT_CHECK(ch != CR);
+ parser->state = s_chunk_data_done;
+ CALLBACK_DATA(body);
+ break;
+
+ case s_chunk_data_done:
+ assert(parser->flags & F_CHUNKED);
+ STRICT_CHECK(ch != LF);
+ parser->nread = 0;
+ parser->state = s_chunk_size_start;
+ break;
+
+ default:
+ assert(0 && "unhandled state");
+ SET_ERRNO(HPE_INVALID_INTERNAL_STATE);
+ goto error;
+ }
+ }
+
+ /* Run callbacks for any marks that we have leftover after we ran our of
+ * bytes. There should be at most one of these set, so it's OK to invoke
+ * them in series (unset marks will not result in callbacks).
+ *
+ * We use the NOADVANCE() variety of callbacks here because 'p' has already
+ * overflowed 'data' and this allows us to correct for the off-by-one that
+ * we'd otherwise have (since CALLBACK_DATA() is meant to be run with a 'p'
+ * value that's in-bounds).
+ */
+
+ assert(((header_field_mark ? 1 : 0) +
+ (header_value_mark ? 1 : 0) +
+ (url_mark ? 1 : 0) +
+ (body_mark ? 1 : 0) +
+ (status_mark ? 1 : 0)) <= 1);
+
+ CALLBACK_DATA_NOADVANCE(header_field);
+ CALLBACK_DATA_NOADVANCE(header_value);
+ CALLBACK_DATA_NOADVANCE(url);
+ CALLBACK_DATA_NOADVANCE(body);
+ CALLBACK_DATA_NOADVANCE(status);
+
+ return len;
+
+error:
+ if (HTTP_PARSER_ERRNO(parser) == HPE_OK) {
+ SET_ERRNO(HPE_UNKNOWN);
+ }
+
+ return (p - data);
+}
+
+
+/* Does the parser need to see an EOF to find the end of the message? */
+int
+http_message_needs_eof (const http_parser *parser)
+{
+ if (parser->type == HTTP_REQUEST) {
+ return 0;
+ }
+
+ /* See RFC 2616 section 4.4 */
+ if (parser->status_code / 100 == 1 || /* 1xx e.g. Continue */
+ parser->status_code == 204 || /* No Content */
+ parser->status_code == 304 || /* Not Modified */
+ parser->flags & F_SKIPBODY) { /* response to a HEAD request */
+ return 0;
+ }
+
+ if ((parser->flags & F_CHUNKED) || parser->content_length != ULLONG_MAX) {
+ return 0;
+ }
+
+ return 1;
+}
+
+
+int
+http_should_keep_alive (const http_parser *parser)
+{
+ if (parser->http_major > 0 && parser->http_minor > 0) {
+ /* HTTP/1.1 */
+ if (parser->flags & F_CONNECTION_CLOSE) {
+ return 0;
+ }
+ } else {
+ /* HTTP/1.0 or earlier */
+ if (!(parser->flags & F_CONNECTION_KEEP_ALIVE)) {
+ return 0;
+ }
+ }
+
+ return !http_message_needs_eof(parser);
+}
+
+
+const char *
+http_method_str (enum http_method m)
+{
+ return ELEM_AT(method_strings, m, "<unknown>");
+}
+
+
+void
+http_parser_init (http_parser *parser, enum http_parser_type t)
+{
+ void *data = parser->data; /* preserve application data */
+ memset(parser, 0, sizeof(*parser));
+ parser->data = data;
+ parser->type = t;
+ parser->state = (t == HTTP_REQUEST ? s_start_req : (t == HTTP_RESPONSE ? s_start_res : s_start_req_or_res));
+ parser->http_errno = HPE_OK;
+}
+
+const char *
+http_errno_name(enum http_errno err) {
+ assert(err < (sizeof(http_strerror_tab)/sizeof(http_strerror_tab[0])));
+ return http_strerror_tab[err].name;
+}
+
+const char *
+http_errno_description(enum http_errno err) {
+ assert(err < (sizeof(http_strerror_tab)/sizeof(http_strerror_tab[0])));
+ return http_strerror_tab[err].description;
+}
+
+static enum http_host_state
+http_parse_host_char(enum http_host_state s, const char ch) {
+ switch(s) {
+ case s_http_userinfo:
+ case s_http_userinfo_start:
+ if (ch == '@') {
+ return s_http_host_start;
+ }
+
+ if (IS_USERINFO_CHAR(ch)) {
+ return s_http_userinfo;
+ }
+ break;
+
+ case s_http_host_start:
+ if (ch == '[') {
+ return s_http_host_v6_start;
+ }
+
+ if (IS_HOST_CHAR(ch)) {
+ return s_http_host;
+ }
+
+ break;
+
+ case s_http_host:
+ if (IS_HOST_CHAR(ch)) {
+ return s_http_host;
+ }
+
+ /* FALLTHROUGH */
+ case s_http_host_v6_end:
+ if (ch == ':') {
+ return s_http_host_port_start;
+ }
+
+ break;
+
+ case s_http_host_v6:
+ if (ch == ']') {
+ return s_http_host_v6_end;
+ }
+
+ /* FALLTHROUGH */
+ case s_http_host_v6_start:
+ if (IS_HEX(ch) || ch == ':' || ch == '.') {
+ return s_http_host_v6;
+ }
+
+ break;
+
+ case s_http_host_port:
+ case s_http_host_port_start:
+ if (IS_NUM(ch)) {
+ return s_http_host_port;
+ }
+
+ break;
+
+ default:
+ break;
+ }
+ return s_http_host_dead;
+}
+
+static int
+http_parse_host(const char * buf, struct http_parser_url *u, int found_at) {
+ enum http_host_state s;
+
+ const char *p;
+ size_t buflen = u->field_data[UF_HOST].off + u->field_data[UF_HOST].len;
+
+ u->field_data[UF_HOST].len = 0;
+
+ s = found_at ? s_http_userinfo_start : s_http_host_start;
+
+ for (p = buf + u->field_data[UF_HOST].off; p < buf + buflen; p++) {
+ enum http_host_state new_s = http_parse_host_char(s, *p);
+
+ if (new_s == s_http_host_dead) {
+ return 1;
+ }
+
+ switch(new_s) {
+ case s_http_host:
+ if (s != s_http_host) {
+ u->field_data[UF_HOST].off = p - buf;
+ }
+ u->field_data[UF_HOST].len++;
+ break;
+
+ case s_http_host_v6:
+ if (s != s_http_host_v6) {
+ u->field_data[UF_HOST].off = p - buf;
+ }
+ u->field_data[UF_HOST].len++;
+ break;
+
+ case s_http_host_port:
+ if (s != s_http_host_port) {
+ u->field_data[UF_PORT].off = p - buf;
+ u->field_data[UF_PORT].len = 0;
+ u->field_set |= (1 << UF_PORT);
+ }
+ u->field_data[UF_PORT].len++;
+ break;
+
+ case s_http_userinfo:
+ if (s != s_http_userinfo) {
+ u->field_data[UF_USERINFO].off = p - buf ;
+ u->field_data[UF_USERINFO].len = 0;
+ u->field_set |= (1 << UF_USERINFO);
+ }
+ u->field_data[UF_USERINFO].len++;
+ break;
+
+ default:
+ break;
+ }
+ s = new_s;
+ }
+
+ /* Make sure we don't end somewhere unexpected */
+ switch (s) {
+ case s_http_host_start:
+ case s_http_host_v6_start:
+ case s_http_host_v6:
+ case s_http_host_port_start:
+ case s_http_userinfo:
+ case s_http_userinfo_start:
+ return 1;
+ default:
+ break;
+ }
+
+ return 0;
+}
+
+int
+http_parser_parse_url(const char *buf, size_t buflen, int is_connect,
+ struct http_parser_url *u)
+{
+ enum state s;
+ const char *p;
+ enum http_parser_url_fields uf, old_uf;
+ int found_at = 0;
+
+ u->port = u->field_set = 0;
+ s = is_connect ? s_req_server_start : s_req_spaces_before_url;
+ uf = old_uf = UF_MAX;
+
+ for (p = buf; p < buf + buflen; p++) {
+ s = parse_url_char(s, *p);
+
+ /* Figure out the next field that we're operating on */
+ switch (s) {
+ case s_dead:
+ return 1;
+
+ /* Skip delimeters */
+ case s_req_schema_slash:
+ case s_req_schema_slash_slash:
+ case s_req_server_start:
+ case s_req_query_string_start:
+ case s_req_fragment_start:
+ continue;
+
+ case s_req_schema:
+ uf = UF_SCHEMA;
+ break;
+
+ case s_req_server_with_at:
+ found_at = 1;
+
+ /* FALLTROUGH */
+ case s_req_server:
+ uf = UF_HOST;
+ break;
+
+ case s_req_path:
+ uf = UF_PATH;
+ break;
+
+ case s_req_query_string:
+ uf = UF_QUERY;
+ break;
+
+ case s_req_fragment:
+ uf = UF_FRAGMENT;
+ break;
+
+ default:
+ assert(!"Unexpected state");
+ return 1;
+ }
+
+ /* Nothing's changed; soldier on */
+ if (uf == old_uf) {
+ u->field_data[uf].len++;
+ continue;
+ }
+
+ u->field_data[uf].off = p - buf;
+ u->field_data[uf].len = 1;
+
+ u->field_set |= (1 << uf);
+ old_uf = uf;
+ }
+
+ /* host must be present if there is a schema */
+ /* parsing http:///toto will fail */
+ if ((u->field_set & ((1 << UF_SCHEMA) | (1 << UF_HOST))) != 0) {
+ if (http_parse_host(buf, u, found_at) != 0) {
+ return 1;
+ }
+ }
+
+ /* CONNECT requests can only contain "hostname:port" */
+ if (is_connect && u->field_set != ((1 << UF_HOST)|(1 << UF_PORT))) {
+ return 1;
+ }
+
+ if (u->field_set & (1 << UF_PORT)) {
+ /* Don't bother with endp; we've already validated the string */
+ unsigned long v = strtoul(buf + u->field_data[UF_PORT].off, NULL, 10);
+
+ /* Ports have a max value of 2^16 */
+ if (v > 0xffff) {
+ return 1;
+ }
+
+ u->port = (uint16_t) v;
+ }
+
+ return 0;
+}
+
+void
+http_parser_pause(http_parser *parser, int paused) {
+ /* Users should only be pausing/unpausing a parser that is not in an error
+ * state. In non-debug builds, there's not much that we can do about this
+ * other than ignore it.
+ */
+ if (HTTP_PARSER_ERRNO(parser) == HPE_OK ||
+ HTTP_PARSER_ERRNO(parser) == HPE_PAUSED) {
+ SET_ERRNO((paused) ? HPE_PAUSED : HPE_OK);
+ } else {
+ assert(0 && "Attempting to pause parser in error state");
+ }
+}
+
+int
+http_body_is_final(const struct http_parser *parser) {
+ return parser->state == s_message_done;
+}
+
+unsigned long
+http_parser_version(void) {
+ return HTTP_PARSER_VERSION_MAJOR * 0x10000 |
+ HTTP_PARSER_VERSION_MINOR * 0x00100 |
+ HTTP_PARSER_VERSION_PATCH * 0x00001;
+}
diff --git a/src/http_parser.h b/src/http_parser.h
new file mode 100644
index 0000000..d150f26
--- /dev/null
+++ b/src/http_parser.h
@@ -0,0 +1,325 @@
+/* Copyright Joyent, Inc. and other Node contributors. All rights reserved.
+ *
+ * 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 http_parser_h
+#define http_parser_h
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/* Also update SONAME in the Makefile whenever you change these. */
+#define HTTP_PARSER_VERSION_MAJOR 2
+#define HTTP_PARSER_VERSION_MINOR 3
+#define HTTP_PARSER_VERSION_PATCH 0
+
+#include <sys/types.h>
+#if defined(_WIN32) && !defined(__MINGW32__) && (!defined(_MSC_VER) || _MSC_VER<1600)
+#include <BaseTsd.h>
+#include <stddef.h>
+typedef __int8 int8_t;
+typedef unsigned __int8 uint8_t;
+typedef __int16 int16_t;
+typedef unsigned __int16 uint16_t;
+typedef __int32 int32_t;
+typedef unsigned __int32 uint32_t;
+typedef __int64 int64_t;
+typedef unsigned __int64 uint64_t;
+#else
+#include <stdint.h>
+#endif
+
+/* Compile with -DHTTP_PARSER_STRICT=0 to make less checks, but run
+ * faster
+ */
+#ifndef HTTP_PARSER_STRICT
+# define HTTP_PARSER_STRICT 1
+#endif
+
+/* Maximium header size allowed. If the macro is not defined
+ * before including this header then the default is used. To
+ * change the maximum header size, define the macro in the build
+ * environment (e.g. -DHTTP_MAX_HEADER_SIZE=<value>). To remove
+ * the effective limit on the size of the header, define the macro
+ * to a very large number (e.g. -DHTTP_MAX_HEADER_SIZE=0x7fffffff)
+ */
+#ifndef HTTP_MAX_HEADER_SIZE
+# define HTTP_MAX_HEADER_SIZE (80*1024)
+#endif
+
+typedef struct http_parser http_parser;
+typedef struct http_parser_settings http_parser_settings;
+
+
+/* Callbacks should return non-zero to indicate an error. The parser will
+ * then halt execution.
+ *
+ * The one exception is on_headers_complete. In a HTTP_RESPONSE parser
+ * returning '1' from on_headers_complete will tell the parser that it
+ * should not expect a body. This is used when receiving a response to a
+ * HEAD request which may contain 'Content-Length' or 'Transfer-Encoding:
+ * chunked' headers that indicate the presence of a body.
+ *
+ * http_data_cb does not return data chunks. It will be call arbitrarally
+ * many times for each string. E.G. you might get 10 callbacks for "on_url"
+ * each providing just a few characters more data.
+ */
+typedef int (*http_data_cb) (http_parser*, const char *at, size_t length);
+typedef int (*http_cb) (http_parser*);
+
+
+/* Request Methods */
+#define HTTP_METHOD_MAP(XX) \
+ XX(0, DELETE, DELETE) \
+ XX(1, GET, GET) \
+ XX(2, HEAD, HEAD) \
+ XX(3, POST, POST) \
+ XX(4, PUT, PUT) \
+ /* pathological */ \
+ XX(5, CONNECT, CONNECT) \
+ XX(6, OPTIONS, OPTIONS) \
+ XX(7, TRACE, TRACE) \
+ /* webdav */ \
+ XX(8, COPY, COPY) \
+ XX(9, LOCK, LOCK) \
+ XX(10, MKCOL, MKCOL) \
+ XX(11, MOVE, MOVE) \
+ XX(12, PROPFIND, PROPFIND) \
+ XX(13, PROPPATCH, PROPPATCH) \
+ XX(14, SEARCH, SEARCH) \
+ XX(15, UNLOCK, UNLOCK) \
+ /* subversion */ \
+ XX(16, REPORT, REPORT) \
+ XX(17, MKACTIVITY, MKACTIVITY) \
+ XX(18, CHECKOUT, CHECKOUT) \
+ XX(19, MERGE, MERGE) \
+ /* upnp */ \
+ XX(20, MSEARCH, M-SEARCH) \
+ XX(21, NOTIFY, NOTIFY) \
+ XX(22, SUBSCRIBE, SUBSCRIBE) \
+ XX(23, UNSUBSCRIBE, UNSUBSCRIBE) \
+ /* RFC-5789 */ \
+ XX(24, PATCH, PATCH) \
+ XX(25, PURGE, PURGE) \
+
+enum http_method
+ {
+#define XX(num, name, string) HTTP_##name = num,
+ HTTP_METHOD_MAP(XX)
+#undef XX
+ };
+
+
+enum http_parser_type { HTTP_REQUEST, HTTP_RESPONSE, HTTP_BOTH };
+
+
+/* Flag values for http_parser.flags field */
+enum flags
+ { F_CHUNKED = 1 << 0
+ , F_CONNECTION_KEEP_ALIVE = 1 << 1
+ , F_CONNECTION_CLOSE = 1 << 2
+ , F_TRAILING = 1 << 3
+ , F_UPGRADE = 1 << 4
+ , F_SKIPBODY = 1 << 5
+ };
+
+
+/* Map for errno-related constants
+ *
+ * The provided argument should be a macro that takes 2 arguments.
+ */
+#define HTTP_ERRNO_MAP(XX) \
+ /* No error */ \
+ XX(OK, "success") \
+ \
+ /* Callback-related errors */ \
+ XX(CB_message_begin, "the on_message_begin callback failed") \
+ XX(CB_url, "the on_url callback failed") \
+ XX(CB_header_field, "the on_header_field callback failed") \
+ XX(CB_header_value, "the on_header_value callback failed") \
+ XX(CB_headers_complete, "the on_headers_complete callback failed") \
+ XX(CB_body, "the on_body callback failed") \
+ XX(CB_message_complete, "the on_message_complete callback failed") \
+ XX(CB_status, "the on_status callback failed") \
+ \
+ /* Parsing-related errors */ \
+ XX(INVALID_EOF_STATE, "stream ended at an unexpected time") \
+ XX(HEADER_OVERFLOW, \
+ "too many header bytes seen; overflow detected") \
+ XX(CLOSED_CONNECTION, \
+ "data received after completed connection: close message") \
+ XX(INVALID_VERSION, "invalid HTTP version") \
+ XX(INVALID_STATUS, "invalid HTTP status code") \
+ XX(INVALID_METHOD, "invalid HTTP method") \
+ XX(INVALID_URL, "invalid URL") \
+ XX(INVALID_HOST, "invalid host") \
+ XX(INVALID_PORT, "invalid port") \
+ XX(INVALID_PATH, "invalid path") \
+ XX(INVALID_QUERY_STRING, "invalid query string") \
+ XX(INVALID_FRAGMENT, "invalid fragment") \
+ XX(LF_EXPECTED, "LF character expected") \
+ XX(INVALID_HEADER_TOKEN, "invalid character in header") \
+ XX(INVALID_CONTENT_LENGTH, \
+ "invalid character in content-length header") \
+ XX(INVALID_CHUNK_SIZE, \
+ "invalid character in chunk size header") \
+ XX(INVALID_CONSTANT, "invalid constant string") \
+ XX(INVALID_INTERNAL_STATE, "encountered unexpected internal state")\
+ XX(STRICT, "strict mode assertion failed") \
+ XX(PAUSED, "parser is paused") \
+ XX(UNKNOWN, "an unknown error occurred")
+
+
+/* Define HPE_* values for each errno value above */
+#define HTTP_ERRNO_GEN(n, s) HPE_##n,
+enum http_errno {
+ HTTP_ERRNO_MAP(HTTP_ERRNO_GEN)
+};
+#undef HTTP_ERRNO_GEN
+
+
+/* Get an http_errno value from an http_parser */
+#define HTTP_PARSER_ERRNO(p) ((enum http_errno) (p)->http_errno)
+
+
+struct http_parser {
+ /** PRIVATE **/
+ unsigned int type : 2; /* enum http_parser_type */
+ unsigned int flags : 6; /* F_* values from 'flags' enum; semi-public */
+ unsigned int state : 8; /* enum state from http_parser.c */
+ unsigned int header_state : 8; /* enum header_state from http_parser.c */
+ unsigned int index : 8; /* index into current matcher */
+
+ uint32_t nread; /* # bytes read in various scenarios */
+ uint64_t content_length; /* # bytes in body (0 if no Content-Length header) */
+
+ /** READ-ONLY **/
+ unsigned short http_major;
+ unsigned short http_minor;
+ unsigned int status_code : 16; /* responses only */
+ unsigned int method : 8; /* requests only */
+ unsigned int http_errno : 7;
+
+ /* 1 = Upgrade header was present and the parser has exited because of that.
+ * 0 = No upgrade header present.
+ * Should be checked when http_parser_execute() returns in addition to
+ * error checking.
+ */
+ unsigned int upgrade : 1;
+
+ /** PUBLIC **/
+ void *data; /* A pointer to get hook to the "connection" or "socket" object */
+};
+
+
+struct http_parser_settings {
+ http_cb on_message_begin;
+ http_data_cb on_url;
+ http_data_cb on_status;
+ http_data_cb on_header_field;
+ http_data_cb on_header_value;
+ http_cb on_headers_complete;
+ http_data_cb on_body;
+ http_cb on_message_complete;
+};
+
+
+enum http_parser_url_fields
+ { UF_SCHEMA = 0
+ , UF_HOST = 1
+ , UF_PORT = 2
+ , UF_PATH = 3
+ , UF_QUERY = 4
+ , UF_FRAGMENT = 5
+ , UF_USERINFO = 6
+ , UF_MAX = 7
+ };
+
+
+/* Result structure for http_parser_parse_url().
+ *
+ * Callers should index into field_data[] with UF_* values iff field_set
+ * has the relevant (1 << UF_*) bit set. As a courtesy to clients (and
+ * because we probably have padding left over), we convert any port to
+ * a uint16_t.
+ */
+struct http_parser_url {
+ uint16_t field_set; /* Bitmask of (1 << UF_*) values */
+ uint16_t port; /* Converted UF_PORT string */
+
+ struct {
+ uint16_t off; /* Offset into buffer in which field starts */
+ uint16_t len; /* Length of run in buffer */
+ } field_data[UF_MAX];
+};
+
+
+/* Returns the library version. Bits 16-23 contain the major version number,
+ * bits 8-15 the minor version number and bits 0-7 the patch level.
+ * Usage example:
+ *
+ * unsigned long version = http_parser_version();
+ * unsigned major = (version >> 16) & 255;
+ * unsigned minor = (version >> 8) & 255;
+ * unsigned patch = version & 255;
+ * printf("http_parser v%u.%u.%u\n", major, minor, version);
+ */
+unsigned long http_parser_version(void);
+
+void http_parser_init(http_parser *parser, enum http_parser_type type);
+
+
+size_t http_parser_execute(http_parser *parser,
+ const http_parser_settings *settings,
+ const char *data,
+ size_t len);
+
+
+/* If http_should_keep_alive() in the on_headers_complete or
+ * on_message_complete callback returns 0, then this should be
+ * the last message on the connection.
+ * If you are the server, respond with the "Connection: close" header.
+ * If you are the client, close the connection.
+ */
+int http_should_keep_alive(const http_parser *parser);
+
+/* Returns a string version of the HTTP method. */
+const char *http_method_str(enum http_method m);
+
+/* Return a string name of the given error */
+const char *http_errno_name(enum http_errno err);
+
+/* Return a string description of the given error */
+const char *http_errno_description(enum http_errno err);
+
+/* Parse a URL; return nonzero on failure */
+int http_parser_parse_url(const char *buf, size_t buflen,
+ int is_connect,
+ struct http_parser_url *u);
+
+/* Pause or un-pause the parser; a nonzero value pauses */
+void http_parser_pause(http_parser *parser, int paused);
+
+/* Checks if this is the final chunk of the body. */
+int http_body_is_final(const http_parser *parser);
+
+#ifdef __cplusplus
+}
+#endif
+#endif
diff --git a/src/sds.c b/src/sds.c
new file mode 100644
index 0000000..5306f1f
--- /dev/null
+++ b/src/sds.c
@@ -0,0 +1,907 @@
+/* SDS (Simple Dynamic Strings), A C dynamic strings library.
+ *
+ * Copyright (c) 2006-2014, Salvatore Sanfilippo <antirez at gmail dot com>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * * Redistributions of source code must retain the above copyright notice,
+ * this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of Redis nor the names of its contributors may be used
+ * to endorse or promote products derived from this software without
+ * specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <ctype.h>
+#include <assert.h>
+
+#include "sds.h"
+
+/* Create a new sds string with the content specified by the 'init' pointer
+ * and 'initlen'.
+ * If NULL is used for 'init' the string is initialized with zero bytes.
+ *
+ * The string is always null-termined (all the sds strings are, always) so
+ * even if you create an sds string with:
+ *
+ * mystring = sdsnewlen("abc",3");
+ *
+ * You can print the string with printf() as there is an implicit \0 at the
+ * end of the string. However the string is binary safe and can contain
+ * \0 characters in the middle, as the length is stored in the sds header. */
+sds sdsnewlen(const void *init, size_t initlen) {
+ struct sdshdr *sh;
+
+ if (init) {
+ sh = malloc(sizeof *sh+initlen+1);
+ } else {
+ sh = calloc(sizeof *sh+initlen+1,1);
+ }
+ if (sh == NULL) return NULL;
+ sh->len = initlen;
+ sh->free = 0;
+ if (initlen && init)
+ memcpy(sh->buf, init, initlen);
+ sh->buf[initlen] = '\0';
+ return (char*)sh->buf;
+}
+
+/* Create an empty (zero length) sds string. Even in this case the string
+ * always has an implicit null term. */
+sds sdsempty(void) {
+ return sdsnewlen("",0);
+}
+
+/* Create a new sds string starting from a null termined C string. */
+sds sdsnew(const char *init) {
+ size_t initlen = (init == NULL) ? 0 : strlen(init);
+ return sdsnewlen(init, initlen);
+}
+
+/* Duplicate an sds string. */
+sds sdsdup(const sds s) {
+ return sdsnewlen(s, sdslen(s));
+}
+
+/* Free an sds string. No operation is performed if 's' is NULL. */
+void sdsfree(sds s) {
+ if (s == NULL) return;
+ free(s-sizeof(struct sdshdr));
+}
+
+/* Set the sds string length to the length as obtained with strlen(), so
+ * considering as content only up to the first null term character.
+ *
+ * This function is useful when the sds string is hacked manually in some
+ * way, like in the following example:
+ *
+ * s = sdsnew("foobar");
+ * s[2] = '\0';
+ * sdsupdatelen(s);
+ * printf("%d\n", sdslen(s));
+ *
+ * The output will be "2", but if we comment out the call to sdsupdatelen()
+ * the output will be "6" as the string was modified but the logical length
+ * remains 6 bytes. */
+void sdsupdatelen(sds s) {
+ struct sdshdr *sh = (void*) (s-sizeof *sh);;
+ int reallen = strlen(s);
+ sh->free += (sh->len-reallen);
+ sh->len = reallen;
+}
+
+/* Modify an sds string on-place to make it empty (zero length).
+ * However all the existing buffer is not discarded but set as free space
+ * so that next append operations will not require allocations up to the
+ * number of bytes previously available. */
+void sdsclear(sds s) {
+ struct sdshdr *sh = (void*) (s-sizeof *sh);;
+ sh->free += sh->len;
+ sh->len = 0;
+ sh->buf[0] = '\0';
+}
+
+/* Enlarge the free space at the end of the sds string so that the caller
+ * is sure that after calling this function can overwrite up to addlen
+ * bytes after the end of the string, plus one more byte for nul term.
+ *
+ * Note: this does not change the *length* of the sds string as returned
+ * by sdslen(), but only the free buffer space we have. */
+sds sdsMakeRoomFor(sds s, size_t addlen) {
+ struct sdshdr *sh, *newsh;
+ size_t free = sdsavail(s);
+ size_t len, newlen;
+
+ if (free >= addlen) return s;
+ len = sdslen(s);
+ sh = (void*) (s-sizeof *sh);;
+ newlen = (len+addlen);
+ if (newlen < SDS_MAX_PREALLOC)
+ newlen *= 2;
+ else
+ newlen += SDS_MAX_PREALLOC;
+ newsh = realloc(sh, sizeof *newsh+newlen+1);
+ if (newsh == NULL) return NULL;
+
+ newsh->free = newlen - len;
+ return newsh->buf;
+}
+
+/* Reallocate the sds string so that it has no free space at the end. The
+ * contained string remains not altered, but next concatenation operations
+ * will require a reallocation.
+ *
+ * After the call, the passed sds string is no longer valid and all the
+ * references must be substituted with the new pointer returned by the call. */
+sds sdsRemoveFreeSpace(sds s) {
+ struct sdshdr *sh;
+
+ sh = (void*) (s-sizeof *sh);;
+ sh = realloc(sh, sizeof *sh+sh->len+1);
+ sh->free = 0;
+ return sh->buf;
+}
+
+/* Return the total size of the allocation of the specifed sds string,
+ * including:
+ * 1) The sds header before the pointer.
+ * 2) The string.
+ * 3) The free buffer at the end if any.
+ * 4) The implicit null term.
+ */
+size_t sdsAllocSize(sds s) {
+ struct sdshdr *sh = (void*) (s-sizeof *sh);;
+
+ return sizeof(*sh)+sh->len+sh->free+1;
+}
+
+/* Increment the sds length and decrements the left free space at the
+ * end of the string according to 'incr'. Also set the null term
+ * in the new end of the string.
+ *
+ * This function is used in order to fix the string length after the
+ * user calls sdsMakeRoomFor(), writes something after the end of
+ * the current string, and finally needs to set the new length.
+ *
+ * Note: it is possible to use a negative increment in order to
+ * right-trim the string.
+ *
+ * Usage example:
+ *
+ * Using sdsIncrLen() and sdsMakeRoomFor() it is possible to mount the
+ * following schema, to cat bytes coming from the kernel to the end of an
+ * sds string without copying into an intermediate buffer:
+ *
+ * oldlen = sdslen(s);
+ * s = sdsMakeRoomFor(s, BUFFER_SIZE);
+ * nread = read(fd, s+oldlen, BUFFER_SIZE);
+ * ... check for nread <= 0 and handle it ...
+ * sdsIncrLen(s, nread);
+ */
+void sdsIncrLen(sds s, int incr) {
+ struct sdshdr *sh = (void*) (s-sizeof *sh);;
+
+ assert(sh->free >= incr);
+ sh->len += incr;
+ sh->free -= incr;
+ assert(sh->free >= 0);
+ s[sh->len] = '\0';
+}
+
+/* Grow the sds to have the specified length. Bytes that were not part of
+ * the original length of the sds will be set to zero.
+ *
+ * if the specified length is smaller than the current length, no operation
+ * is performed. */
+sds sdsgrowzero(sds s, size_t len) {
+ struct sdshdr *sh = (void*) (s-sizeof *sh);
+ size_t totlen, curlen = sh->len;
+
+ if (len <= curlen) return s;
+ s = sdsMakeRoomFor(s,len-curlen);
+ if (s == NULL) return NULL;
+
+ /* Make sure added region doesn't contain garbage */
+ sh = (void*)(s-sizeof *sh);
+ memset(s+curlen,0,(len-curlen+1)); /* also set trailing \0 byte */
+ totlen = sh->len+sh->free;
+ sh->len = len;
+ sh->free = totlen-sh->len;
+ return s;
+}
+
+/* Append the specified binary-safe string pointed by 't' of 'len' bytes to the
+ * end of the specified sds string 's'.
+ *
+ * After the call, the passed sds string is no longer valid and all the
+ * references must be substituted with the new pointer returned by the call. */
+sds sdscatlen(sds s, const void *t, size_t len) {
+ struct sdshdr *sh;
+ size_t curlen = sdslen(s);
+
+ s = sdsMakeRoomFor(s,len);
+ if (s == NULL) return NULL;
+ sh = (void*) (s-sizeof *sh);;
+ memcpy(s+curlen, t, len);
+ sh->len = curlen+len;
+ sh->free = sh->free-len;
+ s[curlen+len] = '\0';
+ return s;
+}
+
+/* Append the specified null termianted C string to the sds string 's'.
+ *
+ * After the call, the passed sds string is no longer valid and all the
+ * references must be substituted with the new pointer returned by the call. */
+sds sdscat(sds s, const char *t) {
+ return sdscatlen(s, t, strlen(t));
+}
+
+/* Append the specified sds 't' to the existing sds 's'.
+ *
+ * After the call, the modified sds string is no longer valid and all the
+ * references must be substituted with the new pointer returned by the call. */
+sds sdscatsds(sds s, const sds t) {
+ return sdscatlen(s, t, sdslen(t));
+}
+
+/* Destructively modify the sds string 's' to hold the specified binary
+ * safe string pointed by 't' of length 'len' bytes. */
+sds sdscpylen(sds s, const char *t, size_t len) {
+ struct sdshdr *sh = (void*) (s-sizeof *sh);;
+ size_t totlen = sh->free+sh->len;
+
+ if (totlen < len) {
+ s = sdsMakeRoomFor(s,len-sh->len);
+ if (s == NULL) return NULL;
+ sh = (void*) (s-sizeof *sh);;
+ totlen = sh->free+sh->len;
+ }
+ memcpy(s, t, len);
+ s[len] = '\0';
+ sh->len = len;
+ sh->free = totlen-len;
+ return s;
+}
+
+/* Like sdscpylen() but 't' must be a null-termined string so that the length
+ * of the string is obtained with strlen(). */
+sds sdscpy(sds s, const char *t) {
+ return sdscpylen(s, t, strlen(t));
+}
+
+/* Like sdscatpritf() but gets va_list instead of being variadic. */
+sds sdscatvprintf(sds s, const char *fmt, va_list ap) {
+ va_list cpy;
+ char *buf, *t;
+ size_t buflen = 16;
+
+ while(1) {
+ buf = malloc(buflen);
+ if (buf == NULL) return NULL;
+ buf[buflen-2] = '\0';
+ va_copy(cpy,ap);
+ vsnprintf(buf, buflen, fmt, cpy);
+ if (buf[buflen-2] != '\0') {
+ free(buf);
+ buflen *= 2;
+ continue;
+ }
+ break;
+ }
+ t = sdscat(s, buf);
+ free(buf);
+ return t;
+}
+
+/* Append to the sds string 's' a string obtained using printf-alike format
+ * specifier.
+ *
+ * After the call, the modified sds string is no longer valid and all the
+ * references must be substituted with the new pointer returned by the call.
+ *
+ * Example:
+ *
+ * s = sdsempty("Sum is: ");
+ * s = sdscatprintf(s,"%d+%d = %d",a,b,a+b).
+ *
+ * Often you need to create a string from scratch with the printf-alike
+ * format. When this is the need, just use sdsempty() as the target string:
+ *
+ * s = sdscatprintf(sdsempty(), "... your format ...", args);
+ */
+sds sdscatprintf(sds s, const char *fmt, ...) {
+ va_list ap;
+ char *t;
+ va_start(ap, fmt);
+ t = sdscatvprintf(s,fmt,ap);
+ va_end(ap);
+ return t;
+}
+
+/* Remove the part of the string from left and from right composed just of
+ * contiguous characters found in 'cset', that is a null terminted C string.
+ *
+ * After the call, the modified sds string is no longer valid and all the
+ * references must be substituted with the new pointer returned by the call.
+ *
+ * Example:
+ *
+ * s = sdsnew("AA...AA.a.aa.aHelloWorld :::");
+ * s = sdstrim(s,"A. :");
+ * printf("%s\n", s);
+ *
+ * Output will be just "Hello World".
+ */
+void sdstrim(sds s, const char *cset) {
+ struct sdshdr *sh = (void*) (s-sizeof *sh);;
+ char *start, *end, *sp, *ep;
+ size_t len;
+
+ sp = start = s;
+ ep = end = s+sdslen(s)-1;
+ while(sp <= end && strchr(cset, *sp)) sp++;
+ while(ep > start && strchr(cset, *ep)) ep--;
+ len = (sp > ep) ? 0 : ((ep-sp)+1);
+ if (sh->buf != sp) memmove(sh->buf, sp, len);
+ sh->buf[len] = '\0';
+ sh->free = sh->free+(sh->len-len);
+ sh->len = len;
+}
+
+/* Turn the string into a smaller (or equal) string containing only the
+ * substring specified by the 'start' and 'end' indexes.
+ *
+ * start and end can be negative, where -1 means the last character of the
+ * string, -2 the penultimate character, and so forth.
+ *
+ * The interval is inclusive, so the start and end characters will be part
+ * of the resulting string.
+ *
+ * The string is modified in-place.
+ *
+ * Example:
+ *
+ * s = sdsnew("Hello World");
+ * sdsrange(s,1,-1); => "ello World"
+ */
+void sdsrange(sds s, int start, int end) {
+ struct sdshdr *sh = (void*) (s-sizeof *sh);;
+ size_t newlen, len = sdslen(s);
+
+ if (len == 0) return;
+ if (start < 0) {
+ start = len+start;
+ if (start < 0) start = 0;
+ }
+ if (end < 0) {
+ end = len+end;
+ if (end < 0) end = 0;
+ }
+ newlen = (start > end) ? 0 : (end-start)+1;
+ if (newlen != 0) {
+ if (start >= (signed)len) {
+ newlen = 0;
+ } else if (end >= (signed)len) {
+ end = len-1;
+ newlen = (start > end) ? 0 : (end-start)+1;
+ }
+ } else {
+ start = 0;
+ }
+ if (start && newlen) memmove(sh->buf, sh->buf+start, newlen);
+ sh->buf[newlen] = 0;
+ sh->free = sh->free+(sh->len-newlen);
+ sh->len = newlen;
+}
+
+/* Apply tolower() to every character of the sds string 's'. */
+void sdstolower(sds s) {
+ int len = sdslen(s), j;
+
+ for (j = 0; j < len; j++) s[j] = tolower(s[j]);
+}
+
+/* Apply toupper() to every character of the sds string 's'. */
+void sdstoupper(sds s) {
+ int len = sdslen(s), j;
+
+ for (j = 0; j < len; j++) s[j] = toupper(s[j]);
+}
+
+/* Compare two sds strings s1 and s2 with memcmp().
+ *
+ * Return value:
+ *
+ * 1 if s1 > s2.
+ * -1 if s1 < s2.
+ * 0 if s1 and s2 are exactly the same binary string.
+ *
+ * If two strings share exactly the same prefix, but one of the two has
+ * additional characters, the longer string is considered to be greater than
+ * the smaller one. */
+int sdscmp(const sds s1, const sds s2) {
+ size_t l1, l2, minlen;
+ int cmp;
+
+ l1 = sdslen(s1);
+ l2 = sdslen(s2);
+ minlen = (l1 < l2) ? l1 : l2;
+ cmp = memcmp(s1,s2,minlen);
+ if (cmp == 0) return l1-l2;
+ return cmp;
+}
+
+/* Split 's' with separator in 'sep'. An array
+ * of sds strings is returned. *count will be set
+ * by reference to the number of tokens returned.
+ *
+ * On out of memory, zero length string, zero length
+ * separator, NULL is returned.
+ *
+ * Note that 'sep' is able to split a string using
+ * a multi-character separator. For example
+ * sdssplit("foo_-_bar","_-_"); will return two
+ * elements "foo" and "bar".
+ *
+ * This version of the function is binary-safe but
+ * requires length arguments. sdssplit() is just the
+ * same function but for zero-terminated strings.
+ */
+sds *sdssplitlen(const char *s, int len, const char *sep, int seplen, int *count) {
+ int elements = 0, slots = 5, start = 0, j;
+ sds *tokens;
+
+ if (seplen < 1 || len < 0) return NULL;
+
+ tokens = malloc(sizeof(sds)*slots);
+ if (tokens == NULL) return NULL;
+
+ if (len == 0) {
+ *count = 0;
+ return tokens;
+ }
+ for (j = 0; j < (len-(seplen-1)); j++) {
+ /* make sure there is room for the next element and the final one */
+ if (slots < elements+2) {
+ sds *newtokens;
+
+ slots *= 2;
+ newtokens = realloc(tokens,sizeof(sds)*slots);
+ if (newtokens == NULL) goto cleanup;
+ tokens = newtokens;
+ }
+ /* search the separator */
+ if ((seplen == 1 && *(s+j) == sep[0]) || (memcmp(s+j,sep,seplen) == 0)) {
+ tokens[elements] = sdsnewlen(s+start,j-start);
+ if (tokens[elements] == NULL) goto cleanup;
+ elements++;
+ start = j+seplen;
+ j = j+seplen-1; /* skip the separator */
+ }
+ }
+ /* Add the final element. We are sure there is room in the tokens array. */
+ tokens[elements] = sdsnewlen(s+start,len-start);
+ if (tokens[elements] == NULL) goto cleanup;
+ elements++;
+ *count = elements;
+ return tokens;
+
+cleanup:
+ {
+ int i;
+ for (i = 0; i < elements; i++) sdsfree(tokens[i]);
+ free(tokens);
+ *count = 0;
+ return NULL;
+ }
+}
+
+/* Free the result returned by sdssplitlen(), or do nothing if 'tokens' is NULL. */
+void sdsfreesplitres(sds *tokens, int count) {
+ if (!tokens) return;
+ while(count--)
+ sdsfree(tokens[count]);
+ free(tokens);
+}
+
+/* Create an sds string from a long long value. It is much faster than:
+ *
+ * sdscatprintf(sdsempty(),"%lld\n", value);
+ */
+sds sdsfromlonglong(long long value) {
+ char buf[32], *p;
+ unsigned long long v;
+
+ v = (value < 0) ? -value : value;
+ p = buf+31; /* point to the last character */
+ do {
+ *p-- = '0'+(v%10);
+ v /= 10;
+ } while(v);
+ if (value < 0) *p-- = '-';
+ p++;
+ return sdsnewlen(p,32-(p-buf));
+}
+
+/* Append to the sds string "s" an escaped string representation where
+ * all the non-printable characters (tested with isprint()) are turned into
+ * escapes in the form "\n\r\a...." or "\x<hex-number>".
+ *
+ * After the call, the modified sds string is no longer valid and all the
+ * references must be substituted with the new pointer returned by the call. */
+sds sdscatrepr(sds s, const char *p, size_t len) {
+ s = sdscatlen(s,"\"",1);
+ while(len--) {
+ switch(*p) {
+ case '\\':
+ case '"':
+ s = sdscatprintf(s,"\\%c",*p);
+ break;
+ case '\n': s = sdscatlen(s,"\\n",2); break;
+ case '\r': s = sdscatlen(s,"\\r",2); break;
+ case '\t': s = sdscatlen(s,"\\t",2); break;
+ case '\a': s = sdscatlen(s,"\\a",2); break;
+ case '\b': s = sdscatlen(s,"\\b",2); break;
+ default:
+ if (isprint(*p))
+ s = sdscatprintf(s,"%c",*p);
+ else
+ s = sdscatprintf(s,"\\x%02x",(unsigned char)*p);
+ break;
+ }
+ p++;
+ }
+ return sdscatlen(s,"\"",1);
+}
+
+/* Helper function for sdssplitargs() that returns non zero if 'c'
+ * is a valid hex digit. */
+int is_hex_digit(char c) {
+ return (c >= '0' && c <= '9') || (c >= 'a' && c <= 'f') ||
+ (c >= 'A' && c <= 'F');
+}
+
+/* Helper function for sdssplitargs() that converts a hex digit into an
+ * integer from 0 to 15 */
+int hex_digit_to_int(char c) {
+ switch(c) {
+ case '0': return 0;
+ case '1': return 1;
+ case '2': return 2;
+ case '3': return 3;
+ case '4': return 4;
+ case '5': return 5;
+ case '6': return 6;
+ case '7': return 7;
+ case '8': return 8;
+ case '9': return 9;
+ case 'a': case 'A': return 10;
+ case 'b': case 'B': return 11;
+ case 'c': case 'C': return 12;
+ case 'd': case 'D': return 13;
+ case 'e': case 'E': return 14;
+ case 'f': case 'F': return 15;
+ default: return 0;
+ }
+}
+
+/* Split a line into arguments, where every argument can be in the
+ * following programming-language REPL-alike form:
+ *
+ * foo bar "newline are supported\n" and "\xff\x00otherstuff"
+ *
+ * The number of arguments is stored into *argc, and an array
+ * of sds is returned.
+ *
+ * The caller should free the resulting array of sds strings with
+ * sdsfreesplitres().
+ *
+ * Note that sdscatrepr() is able to convert back a string into
+ * a quoted string in the same format sdssplitargs() is able to parse.
+ *
+ * The function returns the allocated tokens on success, even when the
+ * input string is empty, or NULL if the input contains unbalanced
+ * quotes or closed quotes followed by non space characters
+ * as in: "foo"bar or "foo'
+ */
+sds *sdssplitargs(const char *line, int *argc) {
+ const char *p = line;
+ char *current = NULL;
+ char **vector = NULL;
+
+ *argc = 0;
+ while(1) {
+ /* skip blanks */
+ while(*p && isspace(*p)) p++;
+ if (*p) {
+ /* get a token */
+ int inq=0; /* set to 1 if we are in "quotes" */
+ int insq=0; /* set to 1 if we are in 'single quotes' */
+ int done=0;
+
+ if (current == NULL) current = sdsempty();
+ while(!done) {
+ if (inq) {
+ if (*p == '\\' && *(p+1) == 'x' &&
+ is_hex_digit(*(p+2)) &&
+ is_hex_digit(*(p+3)))
+ {
+ unsigned char byte;
+
+ byte = (hex_digit_to_int(*(p+2))*16)+
+ hex_digit_to_int(*(p+3));
+ current = sdscatlen(current,(char*)&byte,1);
+ p += 3;
+ } else if (*p == '\\' && *(p+1)) {
+ char c;
+
+ p++;
+ switch(*p) {
+ case 'n': c = '\n'; break;
+ case 'r': c = '\r'; break;
+ case 't': c = '\t'; break;
+ case 'b': c = '\b'; break;
+ case 'a': c = '\a'; break;
+ default: c = *p; break;
+ }
+ current = sdscatlen(current,&c,1);
+ } else if (*p == '"') {
+ /* closing quote must be followed by a space or
+ * nothing at all. */
+ if (*(p+1) && !isspace(*(p+1))) goto err;
+ done=1;
+ } else if (!*p) {
+ /* unterminated quotes */
+ goto err;
+ } else {
+ current = sdscatlen(current,p,1);
+ }
+ } else if (insq) {
+ if (*p == '\\' && *(p+1) == '\'') {
+ p++;
+ current = sdscatlen(current,"'",1);
+ } else if (*p == '\'') {
+ /* closing quote must be followed by a space or
+ * nothing at all. */
+ if (*(p+1) && !isspace(*(p+1))) goto err;
+ done=1;
+ } else if (!*p) {
+ /* unterminated quotes */
+ goto err;
+ } else {
+ current = sdscatlen(current,p,1);
+ }
+ } else {
+ switch(*p) {
+ case ' ':
+ case '\n':
+ case '\r':
+ case '\t':
+ case '\0':
+ done=1;
+ break;
+ case '"':
+ inq=1;
+ break;
+ case '\'':
+ insq=1;
+ break;
+ default:
+ current = sdscatlen(current,p,1);
+ break;
+ }
+ }
+ if (*p) p++;
+ }
+ /* add the token to the vector */
+ vector = realloc(vector,((*argc)+1)*sizeof(char*));
+ vector[*argc] = current;
+ (*argc)++;
+ current = NULL;
+ } else {
+ /* Even on empty input string return something not NULL. */
+ if (vector == NULL) vector = malloc(sizeof(void*));
+ return vector;
+ }
+ }
+
+err:
+ while((*argc)--)
+ sdsfree(vector[*argc]);
+ free(vector);
+ if (current) sdsfree(current);
+ *argc = 0;
+ return NULL;
+}
+
+/* Modify the string substituting all the occurrences of the set of
+ * characters specified in the 'from' string to the corresponding character
+ * in the 'to' array.
+ *
+ * For instance: sdsmapchars(mystring, "ho", "01", 2)
+ * will have the effect of turning the string "hello" into "0ell1".
+ *
+ * The function returns the sds string pointer, that is always the same
+ * as the input pointer since no resize is needed. */
+sds sdsmapchars(sds s, const char *from, const char *to, size_t setlen) {
+ size_t j, i, l = sdslen(s);
+
+ for (j = 0; j < l; j++) {
+ for (i = 0; i < setlen; i++) {
+ if (s[j] == from[i]) {
+ s[j] = to[i];
+ break;
+ }
+ }
+ }
+ return s;
+}
+
+/* Join an array of C strings using the specified separator (also a C string).
+ * Returns the result as an sds string. */
+sds sdsjoin(char **argv, int argc, char *sep, size_t seplen) {
+ sds join = sdsempty();
+ int j;
+
+ for (j = 0; j < argc; j++) {
+ join = sdscat(join, argv[j]);
+ if (j != argc-1) join = sdscatlen(join,sep,seplen);
+ }
+ return join;
+}
+
+/* Like sdsjoin, but joins an array of SDS strings. */
+sds sdsjoinsds(sds *argv, int argc, const char *sep, size_t seplen) {
+ sds join = sdsempty();
+ int j;
+
+ for (j = 0; j < argc; j++) {
+ join = sdscatsds(join, argv[j]);
+ if (j != argc-1) join = sdscatlen(join,sep,seplen);
+ }
+ return join;
+}
+
+#ifdef SDS_TEST_MAIN
+#include <stdio.h>
+#include "testhelp.h"
+
+int main(void) {
+ {
+ struct sdshdr *sh;
+ sds x = sdsnew("foo"), y;
+
+ test_cond("Create a string and obtain the length",
+ sdslen(x) == 3 && memcmp(x,"foo\0",4) == 0)
+
+ sdsfree(x);
+ x = sdsnewlen("foo",2);
+ test_cond("Create a string with specified length",
+ sdslen(x) == 2 && memcmp(x,"fo\0",3) == 0)
+
+ x = sdscat(x,"bar");
+ test_cond("Strings concatenation",
+ sdslen(x) == 5 && memcmp(x,"fobar\0",6) == 0);
+
+ x = sdscpy(x,"a");
+ test_cond("sdscpy() against an originally longer string",
+ sdslen(x) == 1 && memcmp(x,"a\0",2) == 0)
+
+ x = sdscpy(x,"xyzxxxxxxxxxxyyyyyyyyyykkkkkkkkkk");
+ test_cond("sdscpy() against an originally shorter string",
+ sdslen(x) == 33 &&
+ memcmp(x,"xyzxxxxxxxxxxyyyyyyyyyykkkkkkkkkk\0",33) == 0)
+
+ sdsfree(x);
+ x = sdscatprintf(sdsempty(),"%d",123);
+ test_cond("sdscatprintf() seems working in the base case",
+ sdslen(x) == 3 && memcmp(x,"123\0",4) ==0)
+
+ sdsfree(x);
+ x = sdsnew("xxciaoyyy");
+ sdstrim(x,"xy");
+ test_cond("sdstrim() correctly trims characters",
+ sdslen(x) == 4 && memcmp(x,"ciao\0",5) == 0)
+
+ y = sdsdup(x);
+ sdsrange(y,1,1);
+ test_cond("sdsrange(...,1,1)",
+ sdslen(y) == 1 && memcmp(y,"i\0",2) == 0)
+
+ sdsfree(y);
+ y = sdsdup(x);
+ sdsrange(y,1,-1);
+ test_cond("sdsrange(...,1,-1)",
+ sdslen(y) == 3 && memcmp(y,"iao\0",4) == 0)
+
+ sdsfree(y);
+ y = sdsdup(x);
+ sdsrange(y,-2,-1);
+ test_cond("sdsrange(...,-2,-1)",
+ sdslen(y) == 2 && memcmp(y,"ao\0",3) == 0)
+
+ sdsfree(y);
+ y = sdsdup(x);
+ sdsrange(y,2,1);
+ test_cond("sdsrange(...,2,1)",
+ sdslen(y) == 0 && memcmp(y,"\0",1) == 0)
+
+ sdsfree(y);
+ y = sdsdup(x);
+ sdsrange(y,1,100);
+ test_cond("sdsrange(...,1,100)",
+ sdslen(y) == 3 && memcmp(y,"iao\0",4) == 0)
+
+ sdsfree(y);
+ y = sdsdup(x);
+ sdsrange(y,100,100);
+ test_cond("sdsrange(...,100,100)",
+ sdslen(y) == 0 && memcmp(y,"\0",1) == 0)
+
+ sdsfree(y);
+ sdsfree(x);
+ x = sdsnew("foo");
+ y = sdsnew("foa");
+ test_cond("sdscmp(foo,foa)", sdscmp(x,y) > 0)
+
+ sdsfree(y);
+ sdsfree(x);
+ x = sdsnew("bar");
+ y = sdsnew("bar");
+ test_cond("sdscmp(bar,bar)", sdscmp(x,y) == 0)
+
+ sdsfree(y);
+ sdsfree(x);
+ x = sdsnew("aar");
+ y = sdsnew("bar");
+ test_cond("sdscmp(bar,bar)", sdscmp(x,y) < 0)
+
+ sdsfree(y);
+ sdsfree(x);
+ x = sdsnewlen("\a\n\0foo\r",7);
+ y = sdscatrepr(sdsempty(),x,sdslen(x));
+ test_cond("sdscatrepr(...data...)",
+ memcmp(y,"\"\\a\\n\\x00foo\\r\"",15) == 0)
+
+ {
+ int oldfree;
+
+ sdsfree(x);
+ x = sdsnew("0");
+ sh = (void*) (x-(sizeof(struct sdshdr)));
+ test_cond("sdsnew() free/len buffers", sh->len == 1 && sh->free == 0);
+ x = sdsMakeRoomFor(x,1);
+ sh = (void*) (x-(sizeof(struct sdshdr)));
+ test_cond("sdsMakeRoomFor()", sh->len == 1 && sh->free > 0);
+ oldfree = sh->free;
+ x[1] = '1';
+ sdsIncrLen(x,1);
+ test_cond("sdsIncrLen() -- content", x[0] == '0' && x[1] == '1');
+ test_cond("sdsIncrLen() -- len", sh->len == 2);
+ test_cond("sdsIncrLen() -- free", sh->free == oldfree-1);
+ }
+ }
+ test_report()
+ return 0;
+}
+#endif
diff --git a/src/sds.h b/src/sds.h
new file mode 100644
index 0000000..ab6fc9c
--- /dev/null
+++ b/src/sds.h
@@ -0,0 +1,101 @@
+/* SDS (Simple Dynamic Strings), A C dynamic strings library.
+ *
+ * Copyright (c) 2006-2014, Salvatore Sanfilippo <antirez at gmail dot com>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * * Redistributions of source code must retain the above copyright notice,
+ * this list of conditions and the following disclaimer.
+ * * Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * * Neither the name of Redis nor the names of its contributors may be used
+ * to endorse or promote products derived from this software without
+ * specific prior written permission.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef __SDS_H
+#define __SDS_H
+
+#define SDS_MAX_PREALLOC (1024*1024)
+
+#include <sys/types.h>
+#include <stdarg.h>
+
+typedef char *sds;
+
+struct sdshdr {
+ int len;
+ int free;
+ char buf[];
+};
+
+static inline size_t sdslen(const sds s) {
+ struct sdshdr *sh = (void*)(s-sizeof *sh);
+ return sh->len;
+}
+
+static inline size_t sdsavail(const sds s) {
+ struct sdshdr *sh = (void*)(s-sizeof *sh);
+ return sh->free;
+}
+
+sds sdsnewlen(const void *init, size_t initlen);
+sds sdsnew(const char *init);
+sds sdsempty(void);
+size_t sdslen(const sds s);
+sds sdsdup(const sds s);
+void sdsfree(sds s);
+size_t sdsavail(const sds s);
+sds sdsgrowzero(sds s, size_t len);
+sds sdscatlen(sds s, const void *t, size_t len);
+sds sdscat(sds s, const char *t);
+sds sdscatsds(sds s, const sds t);
+sds sdscpylen(sds s, const char *t, size_t len);
+sds sdscpy(sds s, const char *t);
+
+sds sdscatvprintf(sds s, const char *fmt, va_list ap);
+#ifdef __GNUC__
+sds sdscatprintf(sds s, const char *fmt, ...)
+ __attribute__((format(printf, 2, 3)));
+#else
+sds sdscatprintf(sds s, const char *fmt, ...);
+#endif
+
+void sdstrim(sds s, const char *cset);
+void sdsrange(sds s, int start, int end);
+void sdsupdatelen(sds s);
+void sdsclear(sds s);
+int sdscmp(const sds s1, const sds s2);
+sds *sdssplitlen(const char *s, int len, const char *sep, int seplen, int *count);
+void sdsfreesplitres(sds *tokens, int count);
+void sdstolower(sds s);
+void sdstoupper(sds s);
+sds sdsfromlonglong(long long value);
+sds sdscatrepr(sds s, const char *p, size_t len);
+sds *sdssplitargs(const char *line, int *argc);
+sds sdsmapchars(sds s, const char *from, const char *to, size_t setlen);
+sds sdsjoin(char **argv, int argc, char *sep, size_t seplen);
+sds sdsjoinsds(sds *argv, int argc, const char *sep, size_t seplen);
+
+/* Low level functions exposed to the user API */
+sds sdsMakeRoomFor(sds s, size_t addlen);
+void sdsIncrLen(sds s, int incr);
+sds sdsRemoveFreeSpace(sds s);
+size_t sdsAllocSize(sds s);
+
+#endif
diff --git a/src/uvh.c b/src/uvh.c
new file mode 100644
index 0000000..f5d85f5
--- /dev/null
+++ b/src/uvh.c
@@ -0,0 +1,473 @@
+#include "uvh.h"
+#include "sds.h"
+#include "http_parser.h"
+
+#ifndef offsetof
+#ifdef __GNUC__
+#define offsetof(type, member) __builtin_offsetof (type, member)
+#endif
+#endif
+
+#ifndef container_of
+#ifdef __GNUC__
+#define container_of(ptr, type, member) ({ \
+ const typeof( ((type *)0)->member ) *__mptr = (ptr); \
+ (type *)( (char *)__mptr - offsetof(type,member) );})
+#endif
+#endif
+
+#define LOG(LEVEL, FMT, args...) \
+ fprintf(stderr, FMT "\n" , ##args)
+
+#define LOG_DEBUG(FMT, args...) LOG(0, "DEBUG " FMT , ##args)
+#define LOG_WARNING(FMT, args...) LOG(0, "WARNING " FMT , ##args)
+#define LOG_ERROR(FMT, args...) LOG(0, "ERROR " FMT , ##args)
+#define LOG_INFO(FMT, args...) LOG(0, "INFO " FMT , ##args)
+
+struct http_status_code_def
+{
+ int code;
+ const char *str;
+};
+
+static struct http_status_code_def http_status_code_defs[] =
+{
+#define XX(CODE, NAME, STR) { CODE, STR },
+ HTTP_STATUS_CODE_MAP(XX)
+#undef XX
+ { -1, NULL }
+};
+
+static void on_connection(uv_stream_t *stream, int status);
+static uv_buf_t alloc_cb(uv_handle_t *, size_t size);
+static void read_cb(uv_stream_t *stream, ssize_t nread, uv_buf_t buf);
+
+static int on_message_begin(http_parser *parser);
+static int on_url(http_parser *parser, const char *at, size_t len);
+static int on_status(http_parser *parser, const char *at, size_t len);
+static int on_header_field(http_parser *parser, const char *at, size_t len);
+static int on_header_value(http_parser *parser, const char *at, size_t len);
+static int on_headers_complete(http_parser *parser);
+static int on_body(http_parser *parser, const char *at, size_t len);
+static int on_message_complete(http_parser *parser);
+
+struct uvh_server_private
+{
+ struct uvh_server server;
+ struct sockaddr_storage addr;
+ socklen_t addr_len;
+ uv_loop_t *loop;
+ struct http_parser_settings http_parser_settings;
+ uv_tcp_t stream;
+};
+
+struct uvh_request_private
+{
+ struct uvh_request req;
+ struct http_parser parser;
+ uv_tcp_t stream;
+ sds header_name;
+ sds header_value;
+ int header_state;
+};
+
+struct uvh_server *uvh_server_init(uv_loop_t *loop, void *data,
+ uvh_request_handler_cb request_handler)
+{
+ struct uvh_server_private *server;
+ int rc;
+
+ server = calloc(1, sizeof(*server));
+
+ if (!server)
+ goto error;
+
+ server->loop = loop;
+ server->server.data = data;
+ server->server.request_handler = request_handler;
+
+ rc = uv_tcp_init(loop, &server->stream);
+
+ if (rc < 0)
+ goto error;
+
+ server->stream.data = server;
+
+ server->http_parser_settings.on_message_begin = on_message_begin;
+ server->http_parser_settings.on_url = on_url;
+ server->http_parser_settings.on_status = on_status;
+ server->http_parser_settings.on_header_field = on_header_field;
+ server->http_parser_settings.on_header_value = on_header_value;
+ server->http_parser_settings.on_headers_complete = on_headers_complete;
+ server->http_parser_settings.on_body = on_body;
+ server->http_parser_settings.on_message_complete = on_message_complete;
+
+ return &server->server;
+
+error:
+
+ if (server)
+ {
+ free(server);
+ }
+
+ return NULL;
+}
+
+int uvh_server_listen(struct uvh_server *server, const char *address,
+ short port)
+{
+ struct uvh_server_private *serverp = container_of(server, struct uvh_server_private, server);
+ struct sockaddr_in addr = uv_ip4_addr(address, port);
+
+ memcpy(&serverp->addr, &addr, sizeof(addr));
+ serverp->addr_len = sizeof(addr);
+
+ uv_tcp_bind(&serverp->stream, addr);
+
+ int r = uv_listen((uv_stream_t *) &serverp->stream, 128,
+ on_connection);
+
+ if (r)
+ return r;
+
+ return 0;
+}
+
+static void on_connection(uv_stream_t *stream, int status)
+{
+ struct uvh_server_private *priv = container_of((uv_tcp_t *) stream, struct uvh_server_private, stream);
+
+ LOG_DEBUG("%s", __FUNCTION__);
+
+ if (status == -1)
+ {
+ LOG_WARNING("on_connection: status = -1");
+ return;
+ }
+
+ struct uvh_request_private *req = calloc(1, sizeof(*req));
+ req->req.server = &priv->server;
+ req->header_state = 0;
+
+ http_parser_init(&req->parser, HTTP_REQUEST);
+ req->parser.data = req;
+
+ if (uv_tcp_init(priv->loop, &req->stream))
+ {
+ LOG_WARNING("failed to initialize uv_tcp_t");
+ goto error;
+ }
+
+ req->stream.data = req;
+
+ if (uv_accept(stream, (uv_stream_t *) &req->stream) == 0)
+ {
+ uv_read_start((uv_stream_t *) &req->stream, alloc_cb,
+ read_cb);
+ return;
+ }
+ else
+ {
+ uv_close((uv_handle_t *) &req->stream, NULL);
+ LOG_WARNING("failed to accept");
+ }
+
+error:
+
+ if (req)
+ {
+ free(req);
+ }
+}
+
+static uv_buf_t alloc_cb(uv_handle_t *handle, size_t size)
+{
+ (void) handle;
+ return uv_buf_init(calloc(1, size), size);
+}
+
+static void read_cb(uv_stream_t *stream, ssize_t nread, uv_buf_t buf)
+{
+ struct uvh_request_private *req = (struct uvh_request_private *) stream->data;
+ struct uvh_server_private *serverp = container_of(req->req.server, struct uvh_server_private, server);
+
+ LOG_DEBUG("read_cb: nread: %zd, buf.len: %zd", nread, buf.len);
+
+ if (nread < 0)
+ {
+ uv_err_t err = uv_last_error(stream->loop);
+
+ if (buf.base)
+ free(buf.base);
+
+ if (err.code == UV_EOF)
+ {
+ LOG_DEBUG("EOF");
+ http_parser_execute(&req->parser,
+ &serverp->http_parser_settings, NULL, 0);
+ }
+
+ uv_close((uv_handle_t *) stream, NULL);
+
+ return;
+ }
+
+ if (nread == 0)
+ {
+ free(buf.base);
+ return;
+ }
+
+ int nparsed = http_parser_execute(&req->parser,
+ &serverp->http_parser_settings,
+ buf.base, nread);
+
+ LOG_DEBUG("nparsed:%d", nparsed);
+
+ if (nparsed != nread)
+ {
+ LOG_ERROR("http parse error, closing connection");
+ uv_close((uv_handle_t *) stream, NULL);
+ }
+
+ free(buf.base);
+}
+
+static int on_message_begin(http_parser *parser)
+{
+ struct uvh_request_private *priv = (struct uvh_request_private *) parser->data;
+ priv->req.content = (const char *) sdsempty();
+ return 0;
+}
+
+static int on_url(http_parser *parser, const char *at, size_t len)
+{
+ struct uvh_request_private *priv = (struct uvh_request_private *) parser->data;
+
+ LOG_DEBUG("on_url: <%.*s>", (int) len, at);
+
+ if (!priv->req.url.full)
+ {
+ priv->req.url.full = sdsnewlen(at, len);
+ }
+ else
+ {
+ priv->req.url.full = sdscatlen((sds) priv->req.url.full,
+ at, len);
+ }
+
+ return 0;
+}
+
+static int on_status(http_parser *parser, const char *at, size_t len)
+{
+ LOG_DEBUG("on_status: <%.*s>", (int) len, at);
+ return 0;
+}
+
+static int on_header_field(http_parser *parser, const char *at, size_t len)
+{
+ struct uvh_request_private *priv = (struct uvh_request_private *) parser->data;
+ struct uvh_request *req = &priv->req;
+
+ if (priv->header_state == 0)
+ {
+ priv->header_name = sdsnewlen(at, len);
+ }
+ else if (priv->header_state == 1)
+ {
+ priv->header_name = sdscatlen(priv->header_name, at, len);
+ }
+ else if (priv->header_state == 2)
+ {
+ req->headers[req->header_count].name = priv->header_name;
+ req->headers[req->header_count].value = priv->header_value;
+ req->header_count += 1;
+
+ priv->header_name = sdsnewlen(at, len);
+ priv->header_value = NULL;
+ }
+
+ priv->header_state = 1;
+
+ return 0;
+}
+
+static int on_header_value(http_parser *parser, const char *at, size_t len)
+{
+ struct uvh_request_private *priv = (struct uvh_request_private *) parser->data;
+
+ if (priv->header_state == 1)
+ {
+ priv->header_value = sdsnewlen(at, len);
+ }
+ else if (priv->header_state == 2)
+ {
+ priv->header_value = sdscatlen(priv->header_value, at, len);
+ }
+
+ priv->header_state = 2;
+
+ return 0;
+}
+
+static int on_headers_complete(http_parser *parser)
+{
+ struct uvh_request_private *priv = (struct uvh_request_private *) parser->data;
+ struct uvh_request *req = &priv->req;
+ struct http_parser_url url;
+ const char *full = req->url.full;
+
+ if (priv->header_state == 2)
+ {
+ req->headers[req->header_count].name = priv->header_name;
+ req->headers[req->header_count].value = priv->header_value;
+ req->header_count += 1;
+ }
+
+ LOG_DEBUG("on_headers_complete");
+
+ http_parser_parse_url(req->url.full, sdslen((sds) req->url.full),
+ 1, &url);
+
+#define UF_OFFSET(X) url.field_data[X].off
+#define UF_LEN(X) url.field_data[X].len
+#define UF_SET(X) (url.field_set & (1 << (X)))
+#define UF_CHECK_AND_SET(X, DST) \
+ if (UF_SET(X)) \
+ (DST) = sdsnewlen(full + UF_OFFSET(X), UF_LEN(X))
+
+ UF_CHECK_AND_SET(UF_SCHEMA, req->url.schema);
+ UF_CHECK_AND_SET(UF_HOST, req->url.host);
+ UF_CHECK_AND_SET(UF_PORT, req->url.port);
+ UF_CHECK_AND_SET(UF_PATH, req->url.path);
+ UF_CHECK_AND_SET(UF_QUERY, req->url.query);
+ UF_CHECK_AND_SET(UF_FRAGMENT, req->url.fragment);
+ UF_CHECK_AND_SET(UF_USERINFO, req->url.userinfo);
+
+#undef UF_CHECK_AND_SET
+#undef UF_SET
+#undef UF_LEN
+#undef UF_OFFSET
+
+ return 0;
+}
+
+static int on_body(http_parser *parser, const char *at, size_t len)
+{
+ struct uvh_request_private *priv = (struct uvh_request_private *) parser->data;
+ priv->req.content = (const char *) sdscatlen((sds) priv->req.content,
+ at, len);
+ return 0;
+}
+
+static int on_message_complete(http_parser *parser)
+{
+ struct uvh_request_private *priv = (struct uvh_request_private *) parser->data;
+
+ LOG_DEBUG("on_message_complete");
+
+ if (priv->req.content)
+ priv->req.content_length = sdslen((sds) priv->req.content);
+ else
+ priv->req.content_length = 0;
+
+ priv->req.method = sdsnew(http_method_str(parser->method));
+
+ priv->req.version = sdsempty();
+ priv->req.version = sdscatprintf((sds) priv->req.version,
+ "HTTP/%d.%d", parser->http_major, parser->http_minor);
+
+ if (priv->req.server->request_handler)
+ priv->req.server->request_handler(&priv->req);
+
+ return 0;
+}
+
+struct uvh_write_request
+{
+ uv_buf_t buf;
+ uv_write_t wreq;
+ struct uvh_request_private *req;
+};
+
+static void after_request_write(uv_write_t *req, int status)
+{
+ struct uvh_write_request *wreq = container_of(req, struct uvh_write_request, wreq);
+ sdsfree((sds) wreq->buf.base);
+ free(wreq);
+}
+
+static void uvh_request_write_sds(struct uvh_request *req, sds data)
+{
+ struct uvh_request_private *p = container_of(req, struct uvh_request_private, req);
+ struct uvh_write_request *wreq = calloc(1, sizeof(*wreq));
+
+ wreq->buf.base = (char *) data;
+ wreq->buf.len = sdslen(data);
+
+ wreq->req = p;
+
+ uv_write(&wreq->wreq, (uv_stream_t *) &p->stream, &wreq->buf, 1,
+ &after_request_write);
+}
+
+void uvh_request_write(struct uvh_request *req,
+ const char *data, size_t len)
+{
+ uvh_request_write_sds(req, sdsnewlen(data, len));
+}
+
+void uvh_request_writef(struct uvh_request *req, const char *fmt, ...)
+{
+ va_list ap;
+ sds result;
+
+ va_start(ap, fmt);
+ result = sdscatvprintf(sdsempty(), fmt, ap);
+ va_end(ap);
+
+ uvh_request_write_sds(req, result);
+}
+
+void uvh_request_write_status(struct uvh_request *req, int status)
+{
+ uvh_request_writef(req, "%s %d %s\r\n", req->version, status,
+ http_status_code_str(status));
+}
+
+void uvh_request_write_header(struct uvh_request *req,
+ const char *name, const char *value)
+{
+ uvh_request_writef(req, "%s: %s\r\n", name, value);
+}
+
+const char *http_status_code_str(int code)
+{
+ struct http_status_code_def *def = http_status_code_defs;
+ while (def->code != -1)
+ {
+ if (def->code == code)
+ {
+ return def->str;
+ }
+ ++def;
+ }
+
+ return NULL;
+}
+
+const char *uvh_request_get_header(struct uvh_request *req,
+ const char *name)
+{
+ int i;
+
+ for (i = 0; i < req->header_count; ++i)
+ {
+ if (strcasecmp(name, req->headers[i].name) == 0)
+ return req->headers[i].value;
+ }
+
+ return NULL;
+}
diff --git a/src/uvh.h b/src/uvh.h
new file mode 100644
index 0000000..b0bb60b
--- /dev/null
+++ b/src/uvh.h
@@ -0,0 +1,129 @@
+#ifndef UVH_H
+#define UVH_H
+
+#include <stdlib.h>
+#include <string.h>
+#include <stdio.h>
+
+#include <uv.h>
+
+#define UVH_MAX_HEADERS 50
+
+#define HTTP_STATUS_CODE_MAP(XX) \
+ XX(100, CONTINUE, "Continue") \
+ XX(101, SWITCHING_PROTOCOLS, "Switching Protocols") \
+ XX(200, OK, "OK") \
+ XX(201, CREATED, "Created") \
+ XX(202, ACCEPTED, "Accepted") \
+ XX(203, NON_AUTHORITATIVE_INFORMATION, "Non-Authoritative Information") \
+ XX(204, NO_CONTENT, "No Content") \
+ XX(205, RESET_CONTENT, "Reset Content") \
+ XX(206, PARTIAL_CONTENT, "Partial Content") \
+ XX(300, MULTIPLE_CHOICES, "Multiple Choices") \
+ XX(301, MOVED_PERMANENTLY, "Moved Permanently") \
+ XX(302, FOUND, "Found") \
+ XX(303, SEE_OTHER, "See Other") \
+ XX(304, NOT_MODIFIED, "Not Modified") \
+ XX(305, USE_PROXY, "Use Proxy") \
+ XX(307, TEMPORARY_REDIRECT, "Temporary Redirect") \
+ XX(400, BAD_REQUEST, "Bad Request") \
+ XX(401, UNAUTHORIZED, "Unauthorized") \
+ XX(402, PAYMENT_REQUIRED, "Payment Required") \
+ XX(403, FORBIDDEN, "Forbidden") \
+ XX(404, NOT_FOUND, "Not Found") \
+ XX(405, METHOD_NOT_ALLOWED, "Method Not Allowed") \
+ XX(406, NOT_ACCEPTABLE, "Not Acceptable") \
+ XX(407, PROXY_AUTHENTICATION_REQUIRED, "Proxy Authentication Required") \
+ XX(408, REQUEST_TIMEOUT, "Request Timeout") \
+ XX(409, CONFLICT, "Conflict") \
+ XX(410, GONE, "Gone") \
+ XX(411, LENGTH_REQUIRED, "Length Required") \
+ XX(412, PRECONDITION_FAILED, "Precondition Failed") \
+ XX(413, REQUEST_ENTITY_TOO_LARGE, "Request Entity Too Large") \
+ XX(414, REQUEST_URI_TOO_LONG, "Request-URI Too Long") \
+ XX(415, UNSUPPORTED_MEDIA_TYPE, "Unsupported Media Type") \
+ XX(416, REQUESTED_RANGE_NOT_SATISFIABLE, "Requested Range Not Satisfiable") \
+ XX(417, EXPECTATION_FAILED, "Expectation Failed") \
+ XX(418, IM_A_TEAPOT, "I'm a teapot") /* ;-) */ \
+ XX(500, INTERNAL_SERVER_ERROR, "Internal Server Error") \
+ XX(501, NOT_IMPLEMENTED, "Not Implemented") \
+ XX(502, BAD_GATEWAY, "Bad Gateway") \
+ XX(503, SERVICE_UNAVAILABLE, "Service Unavailable") \
+ XX(504, GATEWAY_TIMEOUT, "Gateway Timeout") \
+ XX(505, HTTP_VERSION_NOT_SUPPORTED, "HTTP Version Not Supported")
+
+enum
+{
+#define XX(CODE, NAME, STR) HTTP_##NAME = CODE,
+ HTTP_STATUS_CODE_MAP(XX)
+#undef XX
+};
+
+struct uvh_request;
+
+typedef int (*uvh_request_handler_cb)(struct uvh_request *);
+
+struct uvh_server
+{
+ void *data;
+ uvh_request_handler_cb request_handler;
+};
+
+struct uvh_request
+{
+ struct uvh_server *server;
+
+ struct
+ {
+ const char *name;
+ const char *value;
+ } headers[UVH_MAX_HEADERS];
+
+ int header_count;
+
+ const char *method;
+ const char *version;
+
+ struct
+ {
+ const char *full;
+ const char *schema;
+ const char *host;
+ const char *port;
+ const char *path;
+ const char *query;
+ const char *fragment;
+ const char *userinfo;
+ } url;
+
+ const char *content;
+ int content_length;
+};
+
+struct uvh_server *uvh_server_init(uv_loop_t *loop, void *data,
+ uvh_request_handler_cb request_handler);
+
+int uvh_server_listen(struct uvh_server *server, const char *address,
+ short port);
+
+void uvh_request_write(struct uvh_request *req, const char *data,
+ size_t len);
+
+#ifdef __GNUC__
+void uvh_request_writef(struct uvh_request *req, const char *fmt, ...)
+ __attribute__((format(printf, 2, 3)));
+#else
+void uvh_request_writef(struct uvh_request *req, const char *fmt, ...);
+#endif
+
+void uvh_request_write_status(struct uvh_request *req, int status);
+
+void uvh_request_write_header(struct uvh_request *req,
+ const char *name, const char *value);
+
+const char *http_status_code_str(int code);
+
+const char *uvh_request_get_header(struct uvh_request *req,
+ const char *name);
+
+#endif /* UVH_H */
generated by cgit v1.2.3 (git 2.25.1) at 2025-12-13 02:59:07 +0000