加入 Gitee
与超过 1200万 开发者一起发现、参与优秀开源项目,私有仓库也完全免费 :)
免费加入
文件
克隆/下载
http_parser.c 69.90 KB
一键复制 编辑 原始数据 按行查看 历史
linxyruffy 提交于 2016-07-08 09:59 . 创建httpclient
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721722723724725726727728729730731732733734735736737738739740741742743744745746747748749750751752753754755756757758759760761762763764765766767768769770771772773774775776777778779780781782783784785786787788789790791792793794795796797798799800801802803804805806807808809810811812813814815816817818819820821822823824825826827828829830831832833834835836837838839840841842843844845846847848849850851852853854855856857858859860861862863864865866867868869870871872873874875876877878879880881882883884885886887888889890891892893894895896897898899900901902903904905906907908909910911912913914915916917918919920921922923924925926927928929930931932933934935936937938939940941942943944945946947948949950951952953954955956957958959960961962963964965966967968969970971972973974975976977978979980981982983984985986987988989990991992993994995996997998999100010011002100310041005100610071008100910101011101210131014101510161017101810191020102110221023102410251026102710281029103010311032103310341035103610371038103910401041104210431044104510461047104810491050105110521053105410551056105710581059106010611062106310641065106610671068106910701071107210731074107510761077107810791080108110821083108410851086108710881089109010911092109310941095109610971098109911001101110211031104110511061107110811091110111111121113111411151116111711181119112011211122112311241125112611271128112911301131113211331134113511361137113811391140114111421143114411451146114711481149115011511152115311541155115611571158115911601161116211631164116511661167116811691170117111721173117411751176117711781179118011811182118311841185118611871188118911901191119211931194119511961197119811991200120112021203120412051206120712081209121012111212121312141215121612171218121912201221122212231224122512261227122812291230123112321233123412351236123712381239124012411242124312441245124612471248124912501251125212531254125512561257125812591260126112621263126412651266126712681269127012711272127312741275127612771278127912801281128212831284128512861287128812891290129112921293129412951296129712981299130013011302130313041305130613071308130913101311131213131314131513161317131813191320132113221323132413251326132713281329133013311332133313341335133613371338133913401341134213431344134513461347134813491350135113521353135413551356135713581359136013611362136313641365136613671368136913701371137213731374137513761377137813791380138113821383138413851386138713881389139013911392139313941395139613971398139914001401140214031404140514061407140814091410141114121413141414151416141714181419142014211422142314241425142614271428142914301431143214331434143514361437143814391440144114421443144414451446144714481449145014511452145314541455145614571458145914601461146214631464146514661467146814691470147114721473147414751476147714781479148014811482148314841485148614871488148914901491149214931494149514961497149814991500150115021503150415051506150715081509151015111512151315141515151615171518151915201521152215231524152515261527152815291530153115321533153415351536153715381539154015411542154315441545154615471548154915501551155215531554155515561557155815591560156115621563156415651566156715681569157015711572157315741575157615771578157915801581158215831584158515861587158815891590159115921593159415951596159715981599160016011602160316041605160616071608160916101611161216131614161516161617161816191620162116221623162416251626162716281629163016311632163316341635163616371638163916401641164216431644164516461647164816491650165116521653165416551656165716581659166016611662166316641665166616671668166916701671167216731674167516761677167816791680168116821683168416851686168716881689169016911692169316941695169616971698169917001701170217031704170517061707170817091710171117121713171417151716171717181719172017211722172317241725172617271728172917301731173217331734173517361737173817391740174117421743174417451746174717481749175017511752175317541755175617571758175917601761176217631764176517661767176817691770177117721773177417751776177717781779178017811782178317841785178617871788178917901791179217931794179517961797179817991800180118021803180418051806180718081809181018111812181318141815181618171818181918201821182218231824182518261827182818291830183118321833183418351836183718381839184018411842184318441845184618471848184918501851185218531854185518561857185818591860186118621863186418651866186718681869187018711872187318741875187618771878187918801881188218831884188518861887188818891890189118921893189418951896189718981899190019011902190319041905190619071908190919101911191219131914191519161917191819191920192119221923192419251926192719281929193019311932193319341935193619371938193919401941194219431944194519461947194819491950195119521953195419551956195719581959196019611962196319641965196619671968196919701971197219731974197519761977197819791980198119821983198419851986198719881989199019911992199319941995199619971998199920002001200220032004200520062007200820092010201120122013201420152016201720182019202020212022202320242025202620272028202920302031203220332034203520362037203820392040204120422043204420452046204720482049205020512052205320542055205620572058205920602061206220632064206520662067206820692070207120722073207420752076207720782079208020812082208320842085208620872088208920902091209220932094209520962097209820992100210121022103210421052106210721082109211021112112211321142115211621172118211921202121212221232124212521262127212821292130213121322133213421352136213721382139214021412142214321442145214621472148214921502151215221532154215521562157215821592160216121622163216421652166216721682169217021712172217321742175217621772178217921802181218221832184218521862187218821892190219121922193219421952196219721982199220022012202220322042205220622072208220922102211221222132214221522162217221822192220222122222223222422252226222722282229223022312232223322342235223622372238223922402241224222432244224522462247224822492250225122522253225422552256225722582259226022612262226322642265226622672268226922702271227222732274227522762277227822792280228122822283228422852286228722882289229022912292229322942295229622972298229923002301230223032304230523062307230823092310231123122313231423152316231723182319232023212322232323242325232623272328232923302331233223332334233523362337233823392340234123422343234423452346234723482349235023512352235323542355235623572358235923602361236223632364236523662367236823692370237123722373237423752376237723782379238023812382238323842385238623872388238923902391239223932394239523962397239823992400240124022403240424052406240724082409241024112412241324142415241624172418241924202421242224232424242524262427242824292430243124322433243424352436243724382439244024412442244324442445244624472448244924502451245224532454245524562457245824592460246124622463246424652466246724682469247024712472
/* 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)
#define CURRENT_STATE() p_state
#define UPDATE_STATE(V) p_state = (enum state) (V);
#define RETURN(V) \
do { \
parser->state = CURRENT_STATE(); \
return (V); \
} while (0);
#define REEXECUTE() \
goto reexecute; \
#ifdef __GNUC__
# define LIKELY(X) __builtin_expect(!!(X), 1)
# define UNLIKELY(X) __builtin_expect(!!(X), 0)
#else
# define LIKELY(X) (X)
# define UNLIKELY(X) (X)
#endif
/* Run the notify callback FOR, returning ER if it fails */
#define CALLBACK_NOTIFY_(FOR, ER) \
do { \
assert(HTTP_PARSER_ERRNO(parser) == HPE_OK); \
\
if (LIKELY(settings->on_##FOR)) { \
parser->state = CURRENT_STATE(); \
if (UNLIKELY(0 != settings->on_##FOR(parser))) { \
SET_ERRNO(HPE_CB_##FOR); \
} \
UPDATE_STATE(parser->state); \
\
/* We either errored above or got paused; get out */ \
if (UNLIKELY(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 (LIKELY(settings->on_##FOR)) { \
parser->state = CURRENT_STATE(); \
if (UNLIKELY(0 != \
settings->on_##FOR(parser, FOR##_mark, (LEN)))) { \
SET_ERRNO(HPE_CB_##FOR); \
} \
UPDATE_STATE(parser->state); \
\
/* We either errored above or got paused; get out */ \
if (UNLIKELY(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)
/* 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.
*/
#define COUNT_HEADER_SIZE(V) \
do { \
parser->nread += (V); \
if (UNLIKELY(parser->nread > (HTTP_MAX_HEADER_SIZE))) { \
SET_ERRNO(HPE_HEADER_OVERFLOW); \
goto error; \
} \
} 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_token_start
, h_matching_connection_keep_alive
, h_matching_connection_close
, h_matching_connection_upgrade
, h_matching_connection_token
, h_transfer_encoding_chunked
, h_connection_keep_alive
, h_connection_close
, h_connection_upgrade
};
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_v6_zone_start
, s_http_host_v6_zone
, 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) == ',')
#define STRICT_TOKEN(c) (tokens[(unsigned char)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;
enum state p_state = (enum state) parser->state;
/* We're in an error state. Don't bother doing anything. */
if (HTTP_PARSER_ERRNO(parser) != HPE_OK) {
return 0;
}
if (len == 0) {
switch (CURRENT_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 (CURRENT_STATE() == s_header_field)
header_field_mark = data;
if (CURRENT_STATE() == s_header_value)
header_value_mark = data;
switch (CURRENT_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;
default:
break;
}
for (p=data; p != data + len; p++) {
ch = *p;
if (PARSING_HEADER(CURRENT_STATE()))
COUNT_HEADER_SIZE(1);
reexecute:
switch (CURRENT_STATE()) {
case s_dead:
/* this state is used after a 'Connection: close' message
* the parser will error out if it reads another message
*/
if (LIKELY(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') {
UPDATE_STATE(s_res_or_resp_H);
CALLBACK_NOTIFY(message_begin);
} else {
parser->type = HTTP_REQUEST;
UPDATE_STATE(s_start_req);
REEXECUTE();
}
break;
}
case s_res_or_resp_H:
if (ch == 'T') {
parser->type = HTTP_RESPONSE;
UPDATE_STATE(s_res_HT);
} else {
if (UNLIKELY(ch != 'E')) {
SET_ERRNO(HPE_INVALID_CONSTANT);
goto error;
}
parser->type = HTTP_REQUEST;
parser->method = HTTP_HEAD;
parser->index = 2;
UPDATE_STATE(s_req_method);
}
break;
case s_start_res:
{
parser->flags = 0;
parser->content_length = ULLONG_MAX;
switch (ch) {
case 'H':
UPDATE_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');
UPDATE_STATE(s_res_HT);
break;
case s_res_HT:
STRICT_CHECK(ch != 'T');
UPDATE_STATE(s_res_HTT);
break;
case s_res_HTT:
STRICT_CHECK(ch != 'P');
UPDATE_STATE(s_res_HTTP);
break;
case s_res_HTTP:
STRICT_CHECK(ch != '/');
UPDATE_STATE(s_res_first_http_major);
break;
case s_res_first_http_major:
if (UNLIKELY(ch < '0' || ch > '9')) {
SET_ERRNO(HPE_INVALID_VERSION);
goto error;
}
parser->http_major = ch - '0';
UPDATE_STATE(s_res_http_major);
break;
/* major HTTP version or dot */
case s_res_http_major:
{
if (ch == '.') {
UPDATE_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 (UNLIKELY(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 (UNLIKELY(!IS_NUM(ch))) {
SET_ERRNO(HPE_INVALID_VERSION);
goto error;
}
parser->http_minor = ch - '0';
UPDATE_STATE(s_res_http_minor);
break;
/* minor HTTP version or end of request line */
case s_res_http_minor:
{
if (ch == ' ') {
UPDATE_STATE(s_res_first_status_code);
break;
}
if (UNLIKELY(!IS_NUM(ch))) {
SET_ERRNO(HPE_INVALID_VERSION);
goto error;
}
parser->http_minor *= 10;
parser->http_minor += ch - '0';
if (UNLIKELY(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';
UPDATE_STATE(s_res_status_code);
break;
}
case s_res_status_code:
{
if (!IS_NUM(ch)) {
switch (ch) {
case ' ':
UPDATE_STATE(s_res_status_start);
break;
case CR:
UPDATE_STATE(s_res_line_almost_done);
break;
case LF:
UPDATE_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 (UNLIKELY(parser->status_code > 999)) {
SET_ERRNO(HPE_INVALID_STATUS);
goto error;
}
break;
}
case s_res_status_start:
{
if (ch == CR) {
UPDATE_STATE(s_res_line_almost_done);
break;
}
if (ch == LF) {
UPDATE_STATE(s_header_field_start);
break;
}
MARK(status);
UPDATE_STATE(s_res_status);
parser->index = 0;
break;
}
case s_res_status:
if (ch == CR) {
UPDATE_STATE(s_res_line_almost_done);
CALLBACK_DATA(status);
break;
}
if (ch == LF) {
UPDATE_STATE(s_header_field_start);
CALLBACK_DATA(status);
break;
}
break;
case s_res_line_almost_done:
STRICT_CHECK(ch != LF);
UPDATE_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 (UNLIKELY(!IS_ALPHA(ch))) {
SET_ERRNO(HPE_INVALID_METHOD);
goto error;
}
parser->method = (enum http_method) 0;
parser->index = 1;
switch (ch) {
case 'A': parser->method = HTTP_ACL; break;
case 'B': parser->method = HTTP_BIND; break;
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, MKCALENDAR */ 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; /* or REBIND */ 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, UNBIND */ break;
default:
SET_ERRNO(HPE_INVALID_METHOD);
goto error;
}
UPDATE_STATE(s_req_method);
CALLBACK_NOTIFY(message_begin);
break;
}
case s_req_method:
{
const char *matcher;
if (UNLIKELY(ch == '\0')) {
SET_ERRNO(HPE_INVALID_METHOD);
goto error;
}
matcher = method_strings[parser->method];
if (ch == ' ' && matcher[parser->index] == '\0') {
UPDATE_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 if (parser->index == 3 && ch == 'A') {
parser->method = HTTP_MKCALENDAR;
} 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->method == HTTP_REPORT) {
if (parser->index == 2 && ch == 'B') {
parser->method = HTTP_REBIND;
} 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 if(ch == 'B') {
parser->method = HTTP_UNBIND;
} 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) {
UPDATE_STATE(s_req_server_start);
}
UPDATE_STATE(parse_url_char(CURRENT_STATE(), ch));
if (UNLIKELY(CURRENT_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:
UPDATE_STATE(parse_url_char(CURRENT_STATE(), ch));
if (UNLIKELY(CURRENT_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 ' ':
UPDATE_STATE(s_req_http_start);
CALLBACK_DATA(url);
break;
case CR:
case LF:
parser->http_major = 0;
parser->http_minor = 9;
UPDATE_STATE((ch == CR) ?
s_req_line_almost_done :
s_header_field_start);
CALLBACK_DATA(url);
break;
default:
UPDATE_STATE(parse_url_char(CURRENT_STATE(), ch));
if (UNLIKELY(CURRENT_STATE() == s_dead)) {
SET_ERRNO(HPE_INVALID_URL);
goto error;
}
}
break;
}
case s_req_http_start:
switch (ch) {
case 'H':
UPDATE_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');
UPDATE_STATE(s_req_http_HT);
break;
case s_req_http_HT:
STRICT_CHECK(ch != 'T');
UPDATE_STATE(s_req_http_HTT);
break;
case s_req_http_HTT:
STRICT_CHECK(ch != 'P');
UPDATE_STATE(s_req_http_HTTP);
break;
case s_req_http_HTTP:
STRICT_CHECK(ch != '/');
UPDATE_STATE(s_req_first_http_major);
break;
/* first digit of major HTTP version */
case s_req_first_http_major:
if (UNLIKELY(ch < '1' || ch > '9')) {
SET_ERRNO(HPE_INVALID_VERSION);
goto error;
}
parser->http_major = ch - '0';
UPDATE_STATE(s_req_http_major);
break;
/* major HTTP version or dot */
case s_req_http_major:
{
if (ch == '.') {
UPDATE_STATE(s_req_first_http_minor);
break;
}
if (UNLIKELY(!IS_NUM(ch))) {
SET_ERRNO(HPE_INVALID_VERSION);
goto error;
}
parser->http_major *= 10;
parser->http_major += ch - '0';
if (UNLIKELY(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 (UNLIKELY(!IS_NUM(ch))) {
SET_ERRNO(HPE_INVALID_VERSION);
goto error;
}
parser->http_minor = ch - '0';
UPDATE_STATE(s_req_http_minor);
break;
/* minor HTTP version or end of request line */
case s_req_http_minor:
{
if (ch == CR) {
UPDATE_STATE(s_req_line_almost_done);
break;
}
if (ch == LF) {
UPDATE_STATE(s_header_field_start);
break;
}
/* XXX allow spaces after digit? */
if (UNLIKELY(!IS_NUM(ch))) {
SET_ERRNO(HPE_INVALID_VERSION);
goto error;
}
parser->http_minor *= 10;
parser->http_minor += ch - '0';
if (UNLIKELY(parser->http_minor > 999)) {
SET_ERRNO(HPE_INVALID_VERSION);
goto error;
}
break;
}
/* end of request line */
case s_req_line_almost_done:
{
if (UNLIKELY(ch != LF)) {
SET_ERRNO(HPE_LF_EXPECTED);
goto error;
}
UPDATE_STATE(s_header_field_start);
break;
}
case s_header_field_start:
{
if (ch == CR) {
UPDATE_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*/
UPDATE_STATE(s_headers_almost_done);
REEXECUTE();
}
c = TOKEN(ch);
if (UNLIKELY(!c)) {
SET_ERRNO(HPE_INVALID_HEADER_TOKEN);
goto error;
}
MARK(header_field);
parser->index = 0;
UPDATE_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:
{
const char* start = p;
for (; p != data + len; p++) {
ch = *p;
c = TOKEN(ch);
if (!c)
break;
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;
}
}
COUNT_HEADER_SIZE(p - start);
if (p == data + len) {
--p;
break;
}
if (ch == ':') {
UPDATE_STATE(s_header_value_discard_ws);
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) {
UPDATE_STATE(s_header_value_discard_ws_almost_done);
break;
}
if (ch == LF) {
UPDATE_STATE(s_header_value_discard_lws);
break;
}
/* FALLTHROUGH */
case s_header_value_start:
{
MARK(header_value);
UPDATE_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 (UNLIKELY(!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 if (c == 'u') {
parser->header_state = h_matching_connection_upgrade;
} else {
parser->header_state = h_matching_connection_token;
}
break;
/* Multi-value `Connection` header */
case h_matching_connection_token_start:
break;
default:
parser->header_state = h_general;
break;
}
break;
}
case s_header_value:
{
const char* start = p;
enum header_states h_state = (enum header_states) parser->header_state;
for (; p != data + len; p++) {
ch = *p;
if (ch == CR) {
UPDATE_STATE(s_header_almost_done);
parser->header_state = h_state;
CALLBACK_DATA(header_value);
break;
}
if (ch == LF) {
UPDATE_STATE(s_header_almost_done);
COUNT_HEADER_SIZE(p - start);
parser->header_state = h_state;
CALLBACK_DATA_NOADVANCE(header_value);
REEXECUTE();
}
c = LOWER(ch);
switch (h_state) {
case h_general:
{
const char* p_cr;
const char* p_lf;
size_t limit = data + len - p;
limit = MIN(limit, HTTP_MAX_HEADER_SIZE);
p_cr = (const char*) memchr(p, CR, limit);
p_lf = (const char*) memchr(p, LF, limit);
if (p_cr != NULL) {
if (p_lf != NULL && p_cr >= p_lf)
p = p_lf;
else
p = p_cr;
} else if (UNLIKELY(p_lf != NULL)) {
p = p_lf;
} else {
p = data + len;
}
--p;
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 (UNLIKELY(!IS_NUM(ch))) {
SET_ERRNO(HPE_INVALID_CONTENT_LENGTH);
parser->header_state = h_state;
goto error;
}
t = parser->content_length;
t *= 10;
t += ch - '0';
/* Overflow? Test against a conservative limit for simplicity. */
if (UNLIKELY((ULLONG_MAX - 10) / 10 < parser->content_length)) {
SET_ERRNO(HPE_INVALID_CONTENT_LENGTH);
parser->header_state = h_state;
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]) {
h_state = h_general;
} else if (parser->index == sizeof(CHUNKED)-2) {
h_state = h_transfer_encoding_chunked;
}
break;
case h_matching_connection_token_start:
/* looking for 'Connection: keep-alive' */
if (c == 'k') {
h_state = h_matching_connection_keep_alive;
/* looking for 'Connection: close' */
} else if (c == 'c') {
h_state = h_matching_connection_close;
} else if (c == 'u') {
h_state = h_matching_connection_upgrade;
} else if (STRICT_TOKEN(c)) {
h_state = h_matching_connection_token;
} else if (c == ' ' || c == '\t') {
/* Skip lws */
} else {
h_state = h_general;
}
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]) {
h_state = h_matching_connection_token;
} else if (parser->index == sizeof(KEEP_ALIVE)-2) {
h_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]) {
h_state = h_matching_connection_token;
} else if (parser->index == sizeof(CLOSE)-2) {
h_state = h_connection_close;
}
break;
/* looking for 'Connection: upgrade' */
case h_matching_connection_upgrade:
parser->index++;
if (parser->index > sizeof(UPGRADE) - 1 ||
c != UPGRADE[parser->index]) {
h_state = h_matching_connection_token;
} else if (parser->index == sizeof(UPGRADE)-2) {
h_state = h_connection_upgrade;
}
break;
case h_matching_connection_token:
if (ch == ',') {
h_state = h_matching_connection_token_start;
parser->index = 0;
}
break;
case h_transfer_encoding_chunked:
if (ch != ' ') h_state = h_general;
break;
case h_connection_keep_alive:
case h_connection_close:
case h_connection_upgrade:
if (ch == ',') {
if (h_state == h_connection_keep_alive) {
parser->flags |= F_CONNECTION_KEEP_ALIVE;
} else if (h_state == h_connection_close) {
parser->flags |= F_CONNECTION_CLOSE;
} else if (h_state == h_connection_upgrade) {
parser->flags |= F_CONNECTION_UPGRADE;
}
h_state = h_matching_connection_token_start;
parser->index = 0;
} else if (ch != ' ') {
h_state = h_matching_connection_token;
}
break;
default:
UPDATE_STATE(s_header_value);
h_state = h_general;
break;
}
}
parser->header_state = h_state;
COUNT_HEADER_SIZE(p - start);
if (p == data + len)
--p;
break;
}
case s_header_almost_done:
{
STRICT_CHECK(ch != LF);
UPDATE_STATE(s_header_value_lws);
break;
}
case s_header_value_lws:
{
if (ch == ' ' || ch == '\t') {
UPDATE_STATE(s_header_value_start);
REEXECUTE();
}
/* 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;
case h_connection_upgrade:
parser->flags |= F_CONNECTION_UPGRADE;
break;
default:
break;
}
UPDATE_STATE(s_header_field_start);
REEXECUTE();
}
case s_header_value_discard_ws_almost_done:
{
STRICT_CHECK(ch != LF);
UPDATE_STATE(s_header_value_discard_lws);
break;
}
case s_header_value_discard_lws:
{
if (ch == ' ' || ch == '\t') {
UPDATE_STATE(s_header_value_discard_ws);
break;
} else {
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_connection_upgrade:
parser->flags |= F_CONNECTION_UPGRADE;
break;
case h_transfer_encoding_chunked:
parser->flags |= F_CHUNKED;
break;
default:
break;
}
/* header value was empty */
MARK(header_value);
UPDATE_STATE(s_header_field_start);
CALLBACK_DATA_NOADVANCE(header_value);
REEXECUTE();
}
}
case s_headers_almost_done:
{
STRICT_CHECK(ch != LF);
if (parser->flags & F_TRAILING) {
/* End of a chunked request */
UPDATE_STATE(s_message_done);
CALLBACK_NOTIFY_NOADVANCE(chunk_complete);
REEXECUTE();
}
UPDATE_STATE(s_headers_done);
/* Set this here so that on_headers_complete() callbacks can see it */
parser->upgrade =
((parser->flags & (F_UPGRADE | F_CONNECTION_UPGRADE)) ==
(F_UPGRADE | F_CONNECTION_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);
}
REEXECUTE();
}
case s_headers_done:
{
int hasBody;
STRICT_CHECK(ch != LF);
parser->nread = 0;
hasBody = parser->flags & F_CHUNKED ||
(parser->content_length > 0 && parser->content_length != ULLONG_MAX);
if (parser->upgrade && (parser->method == HTTP_CONNECT ||
(parser->flags & F_SKIPBODY) || !hasBody)) {
/* Exit, the rest of the message is in a different protocol. */
UPDATE_STATE(NEW_MESSAGE());
CALLBACK_NOTIFY(message_complete);
RETURN((p - data) + 1);
}
if (parser->flags & F_SKIPBODY) {
UPDATE_STATE(NEW_MESSAGE());
CALLBACK_NOTIFY(message_complete);
} else if (parser->flags & F_CHUNKED) {
/* chunked encoding - ignore Content-Length header */
UPDATE_STATE(s_chunk_size_start);
} else {
if (parser->content_length == 0) {
/* Content-Length header given but zero: Content-Length: 0\r\n */
UPDATE_STATE(NEW_MESSAGE());
CALLBACK_NOTIFY(message_complete);
} else if (parser->content_length != ULLONG_MAX) {
/* Content-Length header given and non-zero */
UPDATE_STATE(s_body_identity);
} else {
if (!http_message_needs_eof(parser)) {
/* Assume content-length 0 - read the next */
UPDATE_STATE(NEW_MESSAGE());
CALLBACK_NOTIFY(message_complete);
} else {
/* Read body until EOF */
UPDATE_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) {
UPDATE_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);
REEXECUTE();
}
break;
}
/* read until EOF */
case s_body_identity_eof:
MARK(body);
p = data + len - 1;
break;
case s_message_done:
UPDATE_STATE(NEW_MESSAGE());
CALLBACK_NOTIFY(message_complete);
if (parser->upgrade) {
/* Exit, the rest of the message is in a different protocol. */
RETURN((p - data) + 1);
}
break;
case s_chunk_size_start:
{
assert(parser->nread == 1);
assert(parser->flags & F_CHUNKED);
unhex_val = unhex[(unsigned char)ch];
if (UNLIKELY(unhex_val == -1)) {
SET_ERRNO(HPE_INVALID_CHUNK_SIZE);
goto error;
}
parser->content_length = unhex_val;
UPDATE_STATE(s_chunk_size);
break;
}
case s_chunk_size:
{
uint64_t t;
assert(parser->flags & F_CHUNKED);
if (ch == CR) {
UPDATE_STATE(s_chunk_size_almost_done);
break;
}
unhex_val = unhex[(unsigned char)ch];
if (unhex_val == -1) {
if (ch == ';' || ch == ' ') {
UPDATE_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 (UNLIKELY((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) {
UPDATE_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;
UPDATE_STATE(s_header_field_start);
} else {
UPDATE_STATE(s_chunk_data);
}
CALLBACK_NOTIFY(chunk_header);
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) {
UPDATE_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);
UPDATE_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;
UPDATE_STATE(s_chunk_size_start);
CALLBACK_NOTIFY(chunk_complete);
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;
}
void
http_parser_settings_init(http_parser_settings *settings)
{
memset(settings, 0, sizeof(*settings));
}
const char *
http_errno_name(enum http_errno err) {
assert(((size_t) err) < ARRAY_SIZE(http_strerror_tab));
return http_strerror_tab[err].name;
}
const char *
http_errno_description(enum http_errno err) {
assert(((size_t) err) < ARRAY_SIZE(http_strerror_tab));
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;
}
if (s == s_http_host_v6 && ch == '%') {
return s_http_host_v6_zone_start;
}
break;
case s_http_host_v6_zone:
if (ch == ']') {
return s_http_host_v6_end;
}
/* FALLTHROUGH */
case s_http_host_v6_zone_start:
/* RFC 6874 Zone ID consists of 1*( unreserved / pct-encoded) */
if (IS_ALPHANUM(ch) || ch == '%' || ch == '.' || ch == '-' || ch == '_' ||
ch == '~') {
return s_http_host_v6_zone;
}
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;
size_t buflen = 0;
const char *p;
assert(u->field_set & (1 << UF_HOST));
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_v6_zone_start:
case s_http_host_v6_zone:
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_v6_zone_start:
case s_http_host_v6_zone:
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;
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)) &&
(u->field_set & (1 << UF_HOST)) == 0) {
return 1;
}
if (u->field_set & (1 << UF_HOST)) {
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;
}
马建仓 AI 助手
尝试更多
代码解读
代码找茬
代码优化