diff options
Diffstat (limited to 'ext/http11')
-rw-r--r-- | ext/http11/ext_help.h | 1 | ||||
-rw-r--r-- | ext/http11/http11.c | 165 | ||||
-rw-r--r-- | ext/http11/http11_parser.c | 492 | ||||
-rw-r--r-- | ext/http11/http11_parser.java.rl | 1 | ||||
-rw-r--r-- | ext/http11/http11_parser.rl | 30 | ||||
-rw-r--r-- | ext/http11/http11_parser_common.rl | 2 |
6 files changed, 420 insertions, 271 deletions
diff --git a/ext/http11/ext_help.h b/ext/http11/ext_help.h index 8b4d754..08c0e1e 100644 --- a/ext/http11/ext_help.h +++ b/ext/http11/ext_help.h @@ -4,6 +4,7 @@ #define RAISE_NOT_NULL(T) if(T == NULL) rb_raise(rb_eArgError, "NULL found for " # T " when shouldn't be."); #define DATA_GET(from,type,name) Data_Get_Struct(from,type,name); RAISE_NOT_NULL(name); #define REQUIRE_TYPE(V, T) if(TYPE(V) != T) rb_raise(rb_eTypeError, "Wrong argument type for " # V " required " # T); +#define ARRAY_SIZE(x) (sizeof(x)/sizeof(x[0])) #ifdef DEBUG #define TRACE() fprintf(stderr, "> %s:%d:%s\n", __FILE__, __LINE__, __FUNCTION__) diff --git a/ext/http11/http11.c b/ext/http11/http11.c index 2982467..a228019 100644 --- a/ext/http11/http11.c +++ b/ext/http11/http11.c @@ -7,7 +7,13 @@ #include <assert.h> #include <string.h> #include "http11_parser.h" -#include <ctype.h> + +#ifndef RSTRING_PTR +#define RSTRING_PTR(s) (RSTRING(s)->ptr) +#endif +#ifndef RSTRING_LEN +#define RSTRING_LEN(s) (RSTRING(s)->len) +#endif static VALUE mMongrel; static VALUE cHttpParser; @@ -15,8 +21,9 @@ static VALUE eHttpParserError; #define id_handler_map rb_intern("@handler_map") #define id_http_body rb_intern("@http_body") +#define HTTP_PREFIX "HTTP_" +#define HTTP_PREFIX_LEN (sizeof(HTTP_PREFIX) - 1) -static VALUE global_http_prefix; static VALUE global_request_method; static VALUE global_request_uri; static VALUE global_fragment; @@ -59,10 +66,119 @@ DEF_MAX_LENGTH(REQUEST_PATH, 1024); DEF_MAX_LENGTH(QUERY_STRING, (1024 * 10)); DEF_MAX_LENGTH(HEADER, (1024 * (80 + 32))); +struct common_field { + const signed long len; + const char *name; + VALUE value; +}; + +/* + * A list of common HTTP headers we expect to receive. + * This allows us to avoid repeatedly creating identical string + * objects to be used with rb_hash_aset(). + */ +static struct common_field common_http_fields[] = { +# define f(N) { (sizeof(N) - 1), N, Qnil } + f("ACCEPT"), + f("ACCEPT_CHARSET"), + f("ACCEPT_ENCODING"), + f("ACCEPT_LANGUAGE"), + f("ALLOW"), + f("AUTHORIZATION"), + f("CACHE_CONTROL"), + f("CONNECTION"), + f("CONTENT_ENCODING"), + f("CONTENT_LENGTH"), + f("CONTENT_TYPE"), + f("COOKIE"), + f("DATE"), + f("EXPECT"), + f("FROM"), + f("HOST"), + f("IF_MATCH"), + f("IF_MODIFIED_SINCE"), + f("IF_NONE_MATCH"), + f("IF_RANGE"), + f("IF_UNMODIFIED_SINCE"), + f("KEEP_ALIVE"), /* Firefox sends this */ + f("MAX_FORWARDS"), + f("PRAGMA"), + f("PROXY_AUTHORIZATION"), + f("RANGE"), + f("REFERER"), + f("TE"), + f("TRAILER"), + f("TRANSFER_ENCODING"), + f("UPGRADE"), + f("USER_AGENT"), + f("VIA"), + f("X_FORWARDED_FOR"), /* common for proxies */ + f("X_REAL_IP"), /* common for proxies */ + f("WARNING") +# undef f +}; + +/* + * qsort(3) and bsearch(3) improve average performance slightly, but may + * not be worth it for lack of portability to certain platforms... + */ +#if defined(HAVE_QSORT_BSEARCH) +/* sort by length, then by name if there's a tie */ +static int common_field_cmp(const void *a, const void *b) +{ + struct common_field *cfa = (struct common_field *)a; + struct common_field *cfb = (struct common_field *)b; + signed long diff = cfa->len - cfb->len; + return diff ? diff : memcmp(cfa->name, cfb->name, cfa->len); +} +#endif /* HAVE_QSORT_BSEARCH */ + +static void init_common_fields(void) +{ + int i; + struct common_field *cf = common_http_fields; + char tmp[256]; /* MAX_FIELD_NAME_LENGTH */ + memcpy(tmp, HTTP_PREFIX, HTTP_PREFIX_LEN); + + for(i = 0; i < ARRAY_SIZE(common_http_fields); cf++, i++) { + memcpy(tmp + HTTP_PREFIX_LEN, cf->name, cf->len + 1); + cf->value = rb_obj_freeze(rb_str_new(tmp, HTTP_PREFIX_LEN + cf->len)); + rb_global_variable(&cf->value); + } + +#if defined(HAVE_QSORT_BSEARCH) + qsort(common_http_fields, + ARRAY_SIZE(common_http_fields), + sizeof(struct common_field), + common_field_cmp); +#endif /* HAVE_QSORT_BSEARCH */ +} + +static VALUE find_common_field_value(const char *field, size_t flen) +{ +#if defined(HAVE_QSORT_BSEARCH) + struct common_field key; + struct common_field *found; + key.name = field; + key.len = (signed long)flen; + found = (struct common_field *)bsearch(&key, common_http_fields, + ARRAY_SIZE(common_http_fields), + sizeof(struct common_field), + common_field_cmp); + return found ? found->value : Qnil; +#else /* !HAVE_QSORT_BSEARCH */ + int i; + struct common_field *cf = common_http_fields; + for(i = 0; i < ARRAY_SIZE(common_http_fields); i++, cf++) { + if (cf->len == flen && !memcmp(cf->name, field, flen)) + return cf->value; + } + return Qnil; +#endif /* !HAVE_QSORT_BSEARCH */ +} void http_field(void *data, const char *field, size_t flen, const char *value, size_t vlen) { - char *ch, *end; VALUE req = (VALUE)data; VALUE v = Qnil; VALUE f = Qnil; @@ -71,15 +187,25 @@ void http_field(void *data, const char *field, size_t flen, const char *value, s VALIDATE_MAX_LENGTH(vlen, FIELD_VALUE); v = rb_str_new(value, vlen); - f = rb_str_dup(global_http_prefix); - f = rb_str_buf_cat(f, field, flen); - for(ch = RSTRING(f)->ptr, end = ch + RSTRING(f)->len; ch < end; ch++) { - if(*ch == '-') { - *ch = '_'; - } else { - *ch = toupper(*ch); - } + f = find_common_field_value(field, flen); + + if (f == Qnil) { + /* + * We got a strange header that we don't have a memoized value for. + * Fallback to creating a new string to use as a hash key. + * + * using rb_str_new(NULL, len) here is faster than rb_str_buf_new(len) + * in my testing, because: there's no minimum allocation length (and + * no check for it, either), RSTRING_LEN(f) does not need to be + * written twice, and and RSTRING_PTR(f) will already be + * null-terminated for us. + */ + f = rb_str_new(NULL, HTTP_PREFIX_LEN + flen); + memcpy(RSTRING_PTR(f), HTTP_PREFIX, HTTP_PREFIX_LEN); + memcpy(RSTRING_PTR(f) + HTTP_PREFIX_LEN, field, flen); + assert(*(RSTRING_PTR(f) + RSTRING_LEN(f)) == '\0'); /* paranoia */ + /* fprintf(stderr, "UNKNOWN HEADER <%s>\n", RSTRING_PTR(f)); */ } rb_hash_aset(req, f, v); @@ -168,13 +294,12 @@ void header_done(void *data, const char *at, size_t length) rb_hash_aset(req, global_gateway_interface, global_gateway_interface_value); if((temp = rb_hash_aref(req, global_http_host)) != Qnil) { - /* ruby better close strings off with a '\0' dammit */ - colon = strchr(RSTRING(temp)->ptr, ':'); + colon = memchr(RSTRING_PTR(temp), ':', RSTRING_LEN(temp)); if(colon != NULL) { - rb_hash_aset(req, global_server_name, rb_str_substr(temp, 0, colon - RSTRING(temp)->ptr)); + rb_hash_aset(req, global_server_name, rb_str_substr(temp, 0, colon - RSTRING_PTR(temp))); rb_hash_aset(req, global_server_port, - rb_str_substr(temp, colon - RSTRING(temp)->ptr+1, - RSTRING(temp)->len)); + rb_str_substr(temp, colon - RSTRING_PTR(temp)+1, + RSTRING_LEN(temp))); } else { rb_hash_aset(req, global_server_name, temp); rb_hash_aset(req, global_server_port, global_port_80); @@ -295,8 +420,8 @@ VALUE HttpParser_execute(VALUE self, VALUE req_hash, VALUE data, VALUE start) DATA_GET(self, http_parser, http); from = FIX2INT(start); - dptr = RSTRING(data)->ptr; - dlen = RSTRING(data)->len; + dptr = RSTRING_PTR(data); + dlen = RSTRING_LEN(data); if(from >= dlen) { rb_raise(eHttpParserError, "Requested start is after data buffer end."); @@ -366,7 +491,6 @@ void Init_http11() mMongrel = rb_define_module("Mongrel"); - DEF_GLOBAL(http_prefix, "HTTP_"); DEF_GLOBAL(request_method, "REQUEST_METHOD"); DEF_GLOBAL(request_uri, "REQUEST_URI"); DEF_GLOBAL(fragment, "FRAGMENT"); @@ -384,7 +508,7 @@ void Init_http11() DEF_GLOBAL(server_protocol, "SERVER_PROTOCOL"); DEF_GLOBAL(server_protocol_value, "HTTP/1.1"); DEF_GLOBAL(http_host, "HTTP_HOST"); - DEF_GLOBAL(mongrel_version, "Mongrel 1.2.0"); /* XXX Why is this defined here? */ + DEF_GLOBAL(mongrel_version, "Mongrel 1.2"); /* XXX Why is this defined here? */ DEF_GLOBAL(server_software, "SERVER_SOFTWARE"); DEF_GLOBAL(port_80, "80"); @@ -399,4 +523,5 @@ void Init_http11() rb_define_method(cHttpParser, "error?", HttpParser_has_error,0); rb_define_method(cHttpParser, "finished?", HttpParser_is_finished,0); rb_define_method(cHttpParser, "nread", HttpParser_nread,0); + init_common_fields(); } diff --git a/ext/http11/http11_parser.c b/ext/http11/http11_parser.c index 1712d29..c2143de 100644 --- a/ext/http11/http11_parser.c +++ b/ext/http11/http11_parser.c @@ -10,34 +10,46 @@ #include <ctype.h> #include <string.h> +/* + * capitalizes all lower-case ASCII characters, + * converts dashes to underscores. + */ +static void snake_upcase_char(char *c) +{ + if (*c >= 'a' && *c <= 'z') + *c &= ~0x20; + else if (*c == '-') + *c = '_'; +} + #define LEN(AT, FPC) (FPC - buffer - parser->AT) #define MARK(M,FPC) (parser->M = (FPC) - buffer) #define PTR_TO(F) (buffer + parser->F) /** Machine **/ -#line 74 "http11_parser.rl" +#line 87 "http11_parser.rl" /** Data **/ -#line 25 "http11_parser.c" +#line 37 "http11_parser.c" static const int http_parser_start = 1; static const int http_parser_first_final = 57; static const int http_parser_error = 0; static const int http_parser_en_main = 1; -#line 78 "http11_parser.rl" +#line 91 "http11_parser.rl" int http_parser_init(http_parser *parser) { int cs = 0; -#line 37 "http11_parser.c" +#line 49 "http11_parser.c" { cs = http_parser_start; } -#line 82 "http11_parser.rl" +#line 95 "http11_parser.rl" parser->cs = cs; parser->body_start = 0; parser->content_len = 0; @@ -60,15 +72,15 @@ size_t http_parser_execute(http_parser *parser, const char *buffer, size_t len, p = buffer+off; pe = buffer+len; - assert(*pe == '\0' && "pointer does not end on NUL"); + /* assert(*pe == '\0' && "pointer does not end on NUL"); */ assert(pe - p == len - off && "pointers aren't same distance"); -#line 69 "http11_parser.c" +#line 81 "http11_parser.c" { if ( p == pe ) - goto _out; + goto _test_eof; switch ( cs ) { case 1: @@ -86,16 +98,17 @@ case 1: goto tr0; goto st0; st0: - goto _out0; +cs = 0; + goto _out; tr0: -#line 22 "http11_parser.rl" +#line 34 "http11_parser.rl" {MARK(mark, p); } goto st2; st2: if ( ++p == pe ) - goto _out2; + goto _test_eof2; case 2: -#line 99 "http11_parser.c" +#line 112 "http11_parser.c" switch( (*p) ) { case 32: goto tr2; case 36: goto st38; @@ -111,7 +124,7 @@ case 2: goto st38; goto st0; tr2: -#line 36 "http11_parser.rl" +#line 49 "http11_parser.rl" { if(parser->request_method != NULL) parser->request_method(parser->data, PTR_TO(mark), LEN(mark, p)); @@ -119,9 +132,9 @@ tr2: goto st3; st3: if ( ++p == pe ) - goto _out3; + goto _test_eof3; case 3: -#line 125 "http11_parser.c" +#line 138 "http11_parser.c" switch( (*p) ) { case 42: goto tr4; case 43: goto tr5; @@ -138,66 +151,75 @@ case 3: goto tr5; goto st0; tr4: -#line 22 "http11_parser.rl" +#line 34 "http11_parser.rl" {MARK(mark, p); } goto st4; st4: if ( ++p == pe ) - goto _out4; + goto _test_eof4; case 4: -#line 149 "http11_parser.c" +#line 162 "http11_parser.c" switch( (*p) ) { case 32: goto tr8; case 35: goto tr9; } goto st0; tr8: -#line 40 "http11_parser.rl" +#line 53 "http11_parser.rl" { if(parser->request_uri != NULL) parser->request_uri(parser->data, PTR_TO(mark), LEN(mark, p)); } goto st5; -tr30: -#line 44 "http11_parser.rl" +tr31: +#line 34 "http11_parser.rl" + {MARK(mark, p); } +#line 57 "http11_parser.rl" + { + if(parser->fragment != NULL) + parser->fragment(parser->data, PTR_TO(mark), LEN(mark, p)); + } + goto st5; +tr34: +#line 57 "http11_parser.rl" { if(parser->fragment != NULL) parser->fragment(parser->data, PTR_TO(mark), LEN(mark, p)); } goto st5; -tr40: -#line 60 "http11_parser.rl" +tr42: +#line 73 "http11_parser.rl" { if(parser->request_path != NULL) parser->request_path(parser->data, PTR_TO(mark), LEN(mark,p)); } -#line 40 "http11_parser.rl" +#line 53 "http11_parser.rl" { if(parser->request_uri != NULL) parser->request_uri(parser->data, PTR_TO(mark), LEN(mark, p)); } goto st5; -tr51: -#line 49 "http11_parser.rl" +tr53: +#line 62 "http11_parser.rl" {MARK(query_start, p); } -#line 50 "http11_parser.rl" +#line 63 "http11_parser.rl" { if(parser->query_string != NULL) parser->query_string(parser->data, PTR_TO(query_start), LEN(query_start, p)); } -#line 40 "http11_parser.rl" +#line 53 "http11_parser.rl" { if(parser->request_uri != NULL) parser->request_uri(parser->data, PTR_TO(mark), LEN(mark, p)); } goto st5; -tr55: -#line 50 "http11_parser.rl" +tr57: +#line 63 "http11_parser.rl" { if(parser->query_string != NULL) parser->query_string(parser->data, PTR_TO(query_start), LEN(query_start, p)); } -#line 40 "http11_parser.rl" +#line 53 "http11_parser.rl" { if(parser->request_uri != NULL) parser->request_uri(parser->data, PTR_TO(mark), LEN(mark, p)); @@ -205,55 +227,55 @@ tr55: goto st5; st5: if ( ++p == pe ) - goto _out5; + goto _test_eof5; case 5: -#line 211 "http11_parser.c" +#line 233 "http11_parser.c" if ( (*p) == 72 ) goto tr10; goto st0; tr10: -#line 22 "http11_parser.rl" +#line 34 "http11_parser.rl" {MARK(mark, p); } goto st6; st6: if ( ++p == pe ) - goto _out6; + goto _test_eof6; case 6: -#line 223 "http11_parser.c" +#line 245 "http11_parser.c" if ( (*p) == 84 ) goto st7; goto st0; st7: if ( ++p == pe ) - goto _out7; + goto _test_eof7; case 7: if ( (*p) == 84 ) goto st8; goto st0; st8: if ( ++p == pe ) - goto _out8; + goto _test_eof8; case 8: if ( (*p) == 80 ) goto st9; goto st0; st9: if ( ++p == pe ) - goto _out9; + goto _test_eof9; case 9: if ( (*p) == 47 ) goto st10; goto st0; st10: if ( ++p == pe ) - goto _out10; + goto _test_eof10; case 10: if ( 48 <= (*p) && (*p) <= 57 ) goto st11; goto st0; st11: if ( ++p == pe ) - goto _out11; + goto _test_eof11; case 11: if ( (*p) == 46 ) goto st12; @@ -262,14 +284,14 @@ case 11: goto st0; st12: if ( ++p == pe ) - goto _out12; + goto _test_eof12; case 12: if ( 48 <= (*p) && (*p) <= 57 ) goto st13; goto st0; st13: if ( ++p == pe ) - goto _out13; + goto _test_eof13; case 13: if ( (*p) == 13 ) goto tr18; @@ -277,14 +299,24 @@ case 13: goto st13; goto st0; tr18: -#line 55 "http11_parser.rl" +#line 68 "http11_parser.rl" { if(parser->http_version != NULL) parser->http_version(parser->data, PTR_TO(mark), LEN(mark, p)); } goto st14; tr26: -#line 31 "http11_parser.rl" +#line 43 "http11_parser.rl" + { MARK(mark, p); } +#line 44 "http11_parser.rl" + { + if(parser->http_field != NULL) { + parser->http_field(parser->data, PTR_TO(field_start), parser->field_len, PTR_TO(mark), LEN(mark, p)); + } + } + goto st14; +tr29: +#line 44 "http11_parser.rl" { if(parser->http_field != NULL) { parser->http_field(parser->data, PTR_TO(field_start), parser->field_len, PTR_TO(mark), LEN(mark, p)); @@ -293,15 +325,15 @@ tr26: goto st14; st14: if ( ++p == pe ) - goto _out14; + goto _test_eof14; case 14: -#line 299 "http11_parser.c" +#line 331 "http11_parser.c" if ( (*p) == 10 ) goto st15; goto st0; st15: if ( ++p == pe ) - goto _out15; + goto _test_eof15; case 15: switch( (*p) ) { case 13: goto st16; @@ -329,131 +361,137 @@ case 15: goto st0; st16: if ( ++p == pe ) - goto _out16; + goto _test_eof16; case 16: if ( (*p) == 10 ) goto tr22; goto st0; tr22: -#line 65 "http11_parser.rl" +#line 78 "http11_parser.rl" { parser->body_start = p - buffer + 1; if(parser->header_done != NULL) parser->header_done(parser->data, p + 1, pe - p - 1); - goto _out57; + {p++; cs = 57; goto _out;} } goto st57; st57: if ( ++p == pe ) - goto _out57; + goto _test_eof57; case 57: -#line 351 "http11_parser.c" +#line 383 "http11_parser.c" goto st0; tr21: -#line 25 "http11_parser.rl" +#line 37 "http11_parser.rl" { MARK(field_start, p); } +#line 38 "http11_parser.rl" + { snake_upcase_char((char *)p); } + goto st17; +tr23: +#line 38 "http11_parser.rl" + { snake_upcase_char((char *)p); } goto st17; st17: if ( ++p == pe ) - goto _out17; + goto _test_eof17; case 17: -#line 361 "http11_parser.c" +#line 399 "http11_parser.c" switch( (*p) ) { - case 33: goto st17; + case 33: goto tr23; case 58: goto tr24; - case 124: goto st17; - case 126: goto st17; + case 124: goto tr23; + case 126: goto tr23; } if ( (*p) < 45 ) { if ( (*p) > 39 ) { if ( 42 <= (*p) && (*p) <= 43 ) - goto st17; + goto tr23; } else if ( (*p) >= 35 ) - goto st17; + goto tr23; } else if ( (*p) > 46 ) { if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) - goto st17; + goto tr23; } else if ( (*p) > 90 ) { if ( 94 <= (*p) && (*p) <= 122 ) - goto st17; + goto tr23; } else - goto st17; + goto tr23; } else - goto st17; + goto tr23; goto st0; tr24: -#line 26 "http11_parser.rl" +#line 39 "http11_parser.rl" { parser->field_len = LEN(field_start, p); } goto st18; tr27: -#line 30 "http11_parser.rl" +#line 43 "http11_parser.rl" { MARK(mark, p); } goto st18; st18: if ( ++p == pe ) - goto _out18; + goto _test_eof18; case 18: -#line 400 "http11_parser.c" +#line 438 "http11_parser.c" switch( (*p) ) { case 13: goto tr26; case 32: goto tr27; } goto tr25; tr25: -#line 30 "http11_parser.rl" +#line 43 "http11_parser.rl" { MARK(mark, p); } goto st19; st19: if ( ++p == pe ) - goto _out19; + goto _test_eof19; case 19: -#line 414 "http11_parser.c" +#line 452 "http11_parser.c" if ( (*p) == 13 ) - goto tr26; + goto tr29; goto st19; tr9: -#line 40 "http11_parser.rl" +#line 53 "http11_parser.rl" { if(parser->request_uri != NULL) parser->request_uri(parser->data, PTR_TO(mark), LEN(mark, p)); } goto st20; -tr41: -#line 60 "http11_parser.rl" +tr43: +#line 73 "http11_parser.rl" { if(parser->request_path != NULL) parser->request_path(parser->data, PTR_TO(mark), LEN(mark,p)); } -#line 40 "http11_parser.rl" +#line 53 "http11_parser.rl" { if(parser->request_uri != NULL) parser->request_uri(parser->data, PTR_TO(mark), LEN(mark, p)); } goto st20; -tr52: -#line 49 "http11_parser.rl" +tr54: +#line 62 "http11_parser.rl" {MARK(query_start, p); } -#line 50 "http11_parser.rl" +#line 63 "http11_parser.rl" { if(parser->query_string != NULL) parser->query_string(parser->data, PTR_TO(query_start), LEN(query_start, p)); } -#line 40 "http11_parser.rl" +#line 53 "http11_parser.rl" { if(parser->request_uri != NULL) parser->request_uri(parser->data, PTR_TO(mark), LEN(mark, p)); } goto st20; -tr56: -#line 50 "http11_parser.rl" +tr58: +#line 63 "http11_parser.rl" { if(parser->query_string != NULL) parser->query_string(parser->data, PTR_TO(query_start), LEN(query_start, p)); } -#line 40 "http11_parser.rl" +#line 53 "http11_parser.rl" { if(parser->request_uri != NULL) parser->request_uri(parser->data, PTR_TO(mark), LEN(mark, p)); @@ -461,12 +499,12 @@ tr56: goto st20; st20: if ( ++p == pe ) - goto _out20; + goto _test_eof20; case 20: -#line 467 "http11_parser.c" +#line 505 "http11_parser.c" switch( (*p) ) { - case 32: goto tr30; - case 37: goto tr31; + case 32: goto tr31; + case 37: goto tr32; case 60: goto st0; case 62: goto st0; case 127: goto st0; @@ -476,18 +514,18 @@ case 20: goto st0; } else if ( (*p) >= 0 ) goto st0; - goto tr29; -tr29: -#line 22 "http11_parser.rl" + goto tr30; +tr30: +#line 34 "http11_parser.rl" {MARK(mark, p); } goto st21; st21: if ( ++p == pe ) - goto _out21; + goto _test_eof21; case 21: -#line 489 "http11_parser.c" +#line 527 "http11_parser.c" switch( (*p) ) { - case 32: goto tr30; + case 32: goto tr34; case 37: goto st22; case 60: goto st0; case 62: goto st0; @@ -499,15 +537,15 @@ case 21: } else if ( (*p) >= 0 ) goto st0; goto st21; -tr31: -#line 22 "http11_parser.rl" +tr32: +#line 34 "http11_parser.rl" {MARK(mark, p); } goto st22; st22: if ( ++p == pe ) - goto _out22; + goto _test_eof22; case 22: -#line 511 "http11_parser.c" +#line 549 "http11_parser.c" if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) goto st23; @@ -519,7 +557,7 @@ case 22: goto st0; st23: if ( ++p == pe ) - goto _out23; + goto _test_eof23; case 23: if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) @@ -531,14 +569,14 @@ case 23: goto st21; goto st0; tr5: -#line 22 "http11_parser.rl" +#line 34 "http11_parser.rl" {MARK(mark, p); } goto st24; st24: if ( ++p == pe ) - goto _out24; + goto _test_eof24; case 24: -#line 542 "http11_parser.c" +#line 580 "http11_parser.c" switch( (*p) ) { case 43: goto st24; case 58: goto st25; @@ -556,14 +594,14 @@ case 24: goto st24; goto st0; tr7: -#line 22 "http11_parser.rl" +#line 34 "http11_parser.rl" {MARK(mark, p); } goto st25; st25: if ( ++p == pe ) - goto _out25; + goto _test_eof25; case 25: -#line 567 "http11_parser.c" +#line 605 "http11_parser.c" switch( (*p) ) { case 32: goto tr8; case 34: goto st0; @@ -578,7 +616,7 @@ case 25: goto st25; st26: if ( ++p == pe ) - goto _out26; + goto _test_eof26; case 26: if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) @@ -591,7 +629,7 @@ case 26: goto st0; st27: if ( ++p == pe ) - goto _out27; + goto _test_eof27; case 27: if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) @@ -603,23 +641,23 @@ case 27: goto st25; goto st0; tr6: -#line 22 "http11_parser.rl" +#line 34 "http11_parser.rl" {MARK(mark, p); } goto st28; st28: if ( ++p == pe ) - goto _out28; + goto _test_eof28; case 28: -#line 614 "http11_parser.c" +#line 652 "http11_parser.c" switch( (*p) ) { - case 32: goto tr40; + case 32: goto tr42; case 34: goto st0; - case 35: goto tr41; + case 35: goto tr43; case 37: goto st29; - case 59: goto tr43; + case 59: goto tr45; case 60: goto st0; case 62: goto st0; - case 63: goto tr44; + case 63: goto tr46; case 127: goto st0; } if ( 0 <= (*p) && (*p) <= 31 ) @@ -627,7 +665,7 @@ case 28: goto st28; st29: if ( ++p == pe ) - goto _out29; + goto _test_eof29; case 29: if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) @@ -640,7 +678,7 @@ case 29: goto st0; st30: if ( ++p == pe ) - goto _out30; + goto _test_eof30; case 30: if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) @@ -651,8 +689,8 @@ case 30: } else goto st28; goto st0; -tr43: -#line 60 "http11_parser.rl" +tr45: +#line 73 "http11_parser.rl" { if(parser->request_path != NULL) parser->request_path(parser->data, PTR_TO(mark), LEN(mark,p)); @@ -660,9 +698,9 @@ tr43: goto st31; st31: if ( ++p == pe ) - goto _out31; + goto _test_eof31; case 31: -#line 666 "http11_parser.c" +#line 704 "http11_parser.c" switch( (*p) ) { case 32: goto tr8; case 34: goto st0; @@ -678,7 +716,7 @@ case 31: goto st31; st32: if ( ++p == pe ) - goto _out32; + goto _test_eof32; case 32: if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) @@ -691,7 +729,7 @@ case 32: goto st0; st33: if ( ++p == pe ) - goto _out33; + goto _test_eof33; case 33: if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) @@ -702,8 +740,8 @@ case 33: } else goto st31; goto st0; -tr44: -#line 60 "http11_parser.rl" +tr46: +#line 73 "http11_parser.rl" { if(parser->request_path != NULL) parser->request_path(parser->data, PTR_TO(mark), LEN(mark,p)); @@ -711,34 +749,34 @@ tr44: goto st34; st34: if ( ++p == pe ) - goto _out34; + goto _test_eof34; case 34: -#line 717 "http11_parser.c" +#line 755 "http11_parser.c" switch( (*p) ) { - case 32: goto tr51; + case 32: goto tr53; case 34: goto st0; - case 35: goto tr52; - case 37: goto tr53; + case 35: goto tr54; + case 37: goto tr55; case 60: goto st0; case 62: goto st0; case 127: goto st0; } if ( 0 <= (*p) && (*p) <= 31 ) goto st0; - goto tr50; -tr50: -#line 49 "http11_parser.rl" + goto tr52; +tr52: +#line 62 "http11_parser.rl" {MARK(query_start, p); } goto st35; st35: if ( ++p == pe ) - goto _out35; + goto _test_eof35; case 35: -#line 738 "http11_parser.c" +#line 776 "http11_parser.c" switch( (*p) ) { - case 32: goto tr55; + case 32: goto tr57; case 34: goto st0; - case 35: goto tr56; + case 35: goto tr58; case 37: goto st36; case 60: goto st0; case 62: goto st0; @@ -747,15 +785,15 @@ case 35: if ( 0 <= (*p) && (*p) <= 31 ) goto st0; goto st35; -tr53: -#line 49 "http11_parser.rl" +tr55: +#line 62 "http11_parser.rl" {MARK(query_start, p); } goto st36; st36: if ( ++p == pe ) - goto _out36; + goto _test_eof36; case 36: -#line 759 "http11_parser.c" +#line 797 "http11_parser.c" if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) goto st37; @@ -767,7 +805,7 @@ case 36: goto st0; st37: if ( ++p == pe ) - goto _out37; + goto _test_eof37; case 37: if ( (*p) < 65 ) { if ( 48 <= (*p) && (*p) <= 57 ) @@ -780,7 +818,7 @@ case 37: goto st0; st38: if ( ++p == pe ) - goto _out38; + goto _test_eof38; case 38: switch( (*p) ) { case 32: goto tr2; @@ -798,7 +836,7 @@ case 38: goto st0; st39: if ( ++p == pe ) - goto _out39; + goto _test_eof39; case 39: switch( (*p) ) { case 32: goto tr2; @@ -816,7 +854,7 @@ case 39: goto st0; st40: if ( ++p == pe ) - goto _out40; + goto _test_eof40; case 40: switch( (*p) ) { case 32: goto tr2; @@ -834,7 +872,7 @@ case 40: goto st0; st41: if ( ++p == pe ) - goto _out41; + goto _test_eof41; case 41: switch( (*p) ) { case 32: goto tr2; @@ -852,7 +890,7 @@ case 41: goto st0; st42: if ( ++p == pe ) - goto _out42; + goto _test_eof42; case 42: switch( (*p) ) { case 32: goto tr2; @@ -870,7 +908,7 @@ case 42: goto st0; st43: if ( ++p == pe ) - goto _out43; + goto _test_eof43; case 43: switch( (*p) ) { case 32: goto tr2; @@ -888,7 +926,7 @@ case 43: goto st0; st44: if ( ++p == pe ) - goto _out44; + goto _test_eof44; case 44: switch( (*p) ) { case 32: goto tr2; @@ -906,7 +944,7 @@ case 44: goto st0; st45: if ( ++p == pe ) - goto _out45; + goto _test_eof45; case 45: switch( (*p) ) { case 32: goto tr2; @@ -924,7 +962,7 @@ case 45: goto st0; st46: if ( ++p == pe ) - goto _out46; + goto _test_eof46; case 46: switch( (*p) ) { case 32: goto tr2; @@ -942,7 +980,7 @@ case 46: goto st0; st47: if ( ++p == pe ) - goto _out47; + goto _test_eof47; case 47: switch( (*p) ) { case 32: goto tr2; @@ -960,7 +998,7 @@ case 47: goto st0; st48: if ( ++p == pe ) - goto _out48; + goto _test_eof48; case 48: switch( (*p) ) { case 32: goto tr2; @@ -978,7 +1016,7 @@ case 48: goto st0; st49: if ( ++p == pe ) - goto _out49; + goto _test_eof49; case 49: switch( (*p) ) { case 32: goto tr2; @@ -996,7 +1034,7 @@ case 49: goto st0; st50: if ( ++p == pe ) - goto _out50; + goto _test_eof50; case 50: switch( (*p) ) { case 32: goto tr2; @@ -1014,7 +1052,7 @@ case 50: goto st0; st51: if ( ++p == pe ) - goto _out51; + goto _test_eof51; case 51: switch( (*p) ) { case 32: goto tr2; @@ -1032,7 +1070,7 @@ case 51: goto st0; st52: if ( ++p == pe ) - goto _out52; + goto _test_eof52; case 52: switch( (*p) ) { case 32: goto tr2; @@ -1050,7 +1088,7 @@ case 52: goto st0; st53: if ( ++p == pe ) - goto _out53; + goto _test_eof53; case 53: switch( (*p) ) { case 32: goto tr2; @@ -1068,7 +1106,7 @@ case 53: goto st0; st54: if ( ++p == pe ) - goto _out54; + goto _test_eof54; case 54: switch( (*p) ) { case 32: goto tr2; @@ -1086,7 +1124,7 @@ case 54: goto st0; st55: if ( ++p == pe ) - goto _out55; + goto _test_eof55; case 55: switch( (*p) ) { case 32: goto tr2; @@ -1104,73 +1142,73 @@ case 55: goto st0; st56: if ( ++p == pe ) - goto _out56; + goto _test_eof56; case 56: if ( (*p) == 32 ) goto tr2; goto st0; } - _out0: cs = 0; goto _out; - _out2: cs = 2; goto _out; - _out3: cs = 3; goto _out; - _out4: cs = 4; goto _out; - _out5: cs = 5; goto _out; - _out6: cs = 6; goto _out; - _out7: cs = 7; goto _out; - _out8: cs = 8; goto _out; - _out9: cs = 9; goto _out; - _out10: cs = 10; goto _out; - _out11: cs = 11; goto _out; - _out12: cs = 12; goto _out; - _out13: cs = 13; goto _out; - _out14: cs = 14; goto _out; - _out15: cs = 15; goto _out; - _out16: cs = 16; goto _out; - _out57: cs = 57; goto _out; - _out17: cs = 17; goto _out; - _out18: cs = 18; goto _out; - _out19: cs = 19; goto _out; - _out20: cs = 20; goto _out; - _out21: cs = 21; goto _out; - _out22: cs = 22; goto _out; - _out23: cs = 23; goto _out; - _out24: cs = 24; goto _out; - _out25: cs = 25; goto _out; - _out26: cs = 26; goto _out; - _out27: cs = 27; goto _out; - _out28: cs = 28; goto _out; - _out29: cs = 29; goto _out; - _out30: cs = 30; goto _out; - _out31: cs = 31; goto _out; - _out32: cs = 32; goto _out; - _out33: cs = 33; goto _out; - _out34: cs = 34; goto _out; - _out35: cs = 35; goto _out; - _out36: cs = 36; goto _out; - _out37: cs = 37; goto _out; - _out38: cs = 38; goto _out; - _out39: cs = 39; goto _out; - _out40: cs = 40; goto _out; - _out41: cs = 41; goto _out; - _out42: cs = 42; goto _out; - _out43: cs = 43; goto _out; - _out44: cs = 44; goto _out; - _out45: cs = 45; goto _out; - _out46: cs = 46; goto _out; - _out47: cs = 47; goto _out; - _out48: cs = 48; goto _out; - _out49: cs = 49; goto _out; - _out50: cs = 50; goto _out; - _out51: cs = 51; goto _out; - _out52: cs = 52; goto _out; - _out53: cs = 53; goto _out; - _out54: cs = 54; goto _out; - _out55: cs = 55; goto _out; - _out56: cs = 56; goto _out; + _test_eof2: cs = 2; goto _test_eof; + _test_eof3: cs = 3; goto _test_eof; + _test_eof4: cs = 4; goto _test_eof; + _test_eof5: cs = 5; goto _test_eof; + _test_eof6: cs = 6; goto _test_eof; + _test_eof7: cs = 7; goto _test_eof; + _test_eof8: cs = 8; goto _test_eof; + _test_eof9: cs = 9; goto _test_eof; + _test_eof10: cs = 10; goto _test_eof; + _test_eof11: cs = 11; goto _test_eof; + _test_eof12: cs = 12; goto _test_eof; + _test_eof13: cs = 13; goto _test_eof; + _test_eof14: cs = 14; goto _test_eof; + _test_eof15: cs = 15; goto _test_eof; + _test_eof16: cs = 16; goto _test_eof; + _test_eof57: cs = 57; goto _test_eof; + _test_eof17: cs = 17; goto _test_eof; + _test_eof18: cs = 18; goto _test_eof; + _test_eof19: cs = 19; goto _test_eof; + _test_eof20: cs = 20; goto _test_eof; + _test_eof21: cs = 21; goto _test_eof; + _test_eof22: cs = 22; goto _test_eof; + _test_eof23: cs = 23; goto _test_eof; + _test_eof24: cs = 24; goto _test_eof; + _test_eof25: cs = 25; goto _test_eof; + _test_eof26: cs = 26; goto _test_eof; + _test_eof27: cs = 27; goto _test_eof; + _test_eof28: cs = 28; goto _test_eof; + _test_eof29: cs = 29; goto _test_eof; + _test_eof30: cs = 30; goto _test_eof; + _test_eof31: cs = 31; goto _test_eof; + _test_eof32: cs = 32; goto _test_eof; + _test_eof33: cs = 33; goto _test_eof; + _test_eof34: cs = 34; goto _test_eof; + _test_eof35: cs = 35; goto _test_eof; + _test_eof36: cs = 36; goto _test_eof; + _test_eof37: cs = 37; goto _test_eof; + _test_eof38: cs = 38; goto _test_eof; + _test_eof39: cs = 39; goto _test_eof; + _test_eof40: cs = 40; goto _test_eof; + _test_eof41: cs = 41; goto _test_eof; + _test_eof42: cs = 42; goto _test_eof; + _test_eof43: cs = 43; goto _test_eof; + _test_eof44: cs = 44; goto _test_eof; + _test_eof45: cs = 45; goto _test_eof; + _test_eof46: cs = 46; goto _test_eof; + _test_eof47: cs = 47; goto _test_eof; + _test_eof48: cs = 48; goto _test_eof; + _test_eof49: cs = 49; goto _test_eof; + _test_eof50: cs = 50; goto _test_eof; + _test_eof51: cs = 51; goto _test_eof; + _test_eof52: cs = 52; goto _test_eof; + _test_eof53: cs = 53; goto _test_eof; + _test_eof54: cs = 54; goto _test_eof; + _test_eof55: cs = 55; goto _test_eof; + _test_eof56: cs = 56; goto _test_eof; + _test_eof: {} _out: {} } -#line 109 "http11_parser.rl" +#line 122 "http11_parser.rl" parser->cs = cs; parser->nread += p - (buffer + off); @@ -1182,27 +1220,11 @@ case 56: assert(parser->field_len <= len && "field has length longer than whole buffer"); assert(parser->field_start < len && "field starts after buffer end"); - if(parser->body_start) { - /* final \r\n combo encountered so stop right here */ - -#line 1189 "http11_parser.c" -#line 123 "http11_parser.rl" - parser->nread++; - } - return(parser->nread); } int http_parser_finish(http_parser *parser) { - int cs = parser->cs; - - -#line 1202 "http11_parser.c" -#line 134 "http11_parser.rl" - - parser->cs = cs; - if (http_parser_has_error(parser) ) { return -1; } else if (http_parser_is_finished(parser) ) { @@ -1217,5 +1239,5 @@ int http_parser_has_error(http_parser *parser) { } int http_parser_is_finished(http_parser *parser) { - return parser->cs == http_parser_first_final; + return parser->cs >= http_parser_first_final; } diff --git a/ext/http11/http11_parser.java.rl b/ext/http11/http11_parser.java.rl index 71f8d3c..c9be22e 100644 --- a/ext/http11/http11_parser.java.rl +++ b/ext/http11/http11_parser.java.rl @@ -13,6 +13,7 @@ public class Http11Parser { action mark {parser.mark = fpc; } action start_field { parser.field_start = fpc; } + action snake_upcase_field { /* FIXME stub */ } action write_field { parser.field_len = fpc-parser.field_start; } diff --git a/ext/http11/http11_parser.rl b/ext/http11/http11_parser.rl index fffe57a..5e551a3 100644 --- a/ext/http11/http11_parser.rl +++ b/ext/http11/http11_parser.rl @@ -9,6 +9,18 @@ #include <ctype.h> #include <string.h> +/* + * capitalizes all lower-case ASCII characters, + * converts dashes to underscores. + */ +static void snake_upcase_char(char *c) +{ + if (*c >= 'a' && *c <= 'z') + *c &= ~0x20; + else if (*c == '-') + *c = '_'; +} + #define LEN(AT, FPC) (FPC - buffer - parser->AT) #define MARK(M,FPC) (parser->M = (FPC) - buffer) #define PTR_TO(F) (buffer + parser->F) @@ -23,6 +35,7 @@ action start_field { MARK(field_start, fpc); } + action snake_upcase_field { snake_upcase_char((char *)fpc); } action write_field { parser->field_len = LEN(field_start, fpc); } @@ -101,10 +114,9 @@ size_t http_parser_execute(http_parser *parser, const char *buffer, size_t len, p = buffer+off; pe = buffer+len; - assert(*pe == '\0' && "pointer does not end on NUL"); + /* assert(*pe == '\0' && "pointer does not end on NUL"); */ assert(pe - p == len - off && "pointers aren't same distance"); - %% write exec; parser->cs = cs; @@ -117,23 +129,11 @@ size_t http_parser_execute(http_parser *parser, const char *buffer, size_t len, assert(parser->field_len <= len && "field has length longer than whole buffer"); assert(parser->field_start < len && "field starts after buffer end"); - if(parser->body_start) { - /* final \r\n combo encountered so stop right here */ - %%write eof; - parser->nread++; - } - return(parser->nread); } int http_parser_finish(http_parser *parser) { - int cs = parser->cs; - - %%write eof; - - parser->cs = cs; - if (http_parser_has_error(parser) ) { return -1; } else if (http_parser_is_finished(parser) ) { @@ -148,5 +148,5 @@ int http_parser_has_error(http_parser *parser) { } int http_parser_is_finished(http_parser *parser) { - return parser->cs == http_parser_first_final; + return parser->cs >= http_parser_first_final; } diff --git a/ext/http11/http11_parser_common.rl b/ext/http11/http11_parser_common.rl index a70d4da..53c805f 100644 --- a/ext/http11/http11_parser_common.rl +++ b/ext/http11/http11_parser_common.rl @@ -41,7 +41,7 @@ HTTP_Version = ( "HTTP/" http_number ) >mark %http_version ; Request_Line = ( Method " " Request_URI ("#" Fragment){0,1} " " HTTP_Version CRLF ) ; - field_name = ( token -- ":" )+ >start_field %write_field; + field_name = ( token -- ":" )+ >start_field $snake_upcase_field %write_field; field_value = any* >start_value %write_value; |