Remove some chars from tokens[] per RFC.

- Treat ' ' specially, as apparently IIS6.0 can send this in headers. Allow this character through if we're not in strict mode. - Move some test code around so that test indices don't break when HTTP_PARSER_STRICT changes. Fixes #13.
13 years ago · 48a4364fdd
parent b47c44d7a6
commit 48a4364fdd
2 changed files with 79 additions and 77 deletions
--- a/http_parser.c
+++ b/http_parser.c
@ -133,9 +133,9 @@ static const char tokens[256] = {
 /*  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,       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  ?  */
@ -155,7 +155,7 @@ static const char tokens[256] = {
 /* 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 };
+       'x',     'y',     'z',      0,      '|',      0,      '~',       0 };


 static const int8_t unhex[256] =
@ -311,15 +311,16 @@ enum header_states
 #define CR                  '\r'
 #define LF                  '\n'
 #define LOWER(c)            (unsigned char)(c | 0x20)
-#define TOKEN(c)            (tokens[(unsigned char)c])
 #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))

 #if HTTP_PARSER_STRICT
+#define TOKEN(c)            (tokens[(unsigned char)c])
 #define IS_URL_CHAR(c)      (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)                                                         \
  (normal_url_char[(unsigned char) (c)] || ((c) & 0x80))
 #define IS_HOST_CHAR(c)                                                        \
--- a/test.c
+++ b/test.c
@ -593,29 +593,35 @@ const struct message requests[] =
  ,.body= ""
  }

-#if !HTTP_PARSER_STRICT
-#define UTF8_PATH_REQ 24
-, {.name= "utf-8 path request"
+#define PATCH_REQ 24
+, {.name = "PATCH request"
  ,.type= HTTP_REQUEST
-  ,.raw= "GET /δ¶/δt/pope?q=1#narf HTTP/1.1\r\n"
-         "Host: github.com\r\n"
+  ,.raw= "PATCH /file.txt HTTP/1.1\r\n"
+         "Host: www.example.com\r\n"
+         "Content-Type: application/example\r\n"
+         "If-Match: \"e0023aa4e\"\r\n"
+         "Content-Length: 10\r\n"
         "\r\n"
+         "cccccccccc"
  ,.should_keep_alive= TRUE
  ,.message_complete_on_eof= FALSE
  ,.http_major= 1
  ,.http_minor= 1
-  ,.method= HTTP_GET
-  ,.request_url= "/δ¶/δt/pope?q=1#narf"
-  ,.num_headers= 1
-  ,.headers= { {"Host", "github.com" }
+  ,.method= HTTP_PATCH
+  ,.request_url= "/file.txt"
+  ,.num_headers= 4
+  ,.headers= { { "Host", "www.example.com" }
+             , { "Content-Type", "application/example" }
+             , { "If-Match", "\"e0023aa4e\"" }
+             , { "Content-Length", "10" }
             }
-  ,.body= ""
+  ,.body= "cccccccccc"
  }

-#define HOSTNAME_UNDERSCORE 25
-, {.name = "hostname underscore"
+#define CONNECT_CAPS_REQUEST 25
+, {.name = "connect caps request"
  ,.type= HTTP_REQUEST
-  ,.raw= "CONNECT home_0.netscape.com:443 HTTP/1.0\r\n"
+  ,.raw= "CONNECT HOME0.NETSCAPE.COM:443 HTTP/1.0\r\n"
         "User-agent: Mozilla/1.1N\r\n"
         "Proxy-authorization: basic aGVsbG86d29ybGQ=\r\n"
         "\r\n"
@ -624,7 +630,7 @@ const struct message requests[] =
  ,.http_major= 1
  ,.http_minor= 0
  ,.method= HTTP_CONNECT
-  ,.request_url= "home_0.netscape.com:443"
+  ,.request_url= "HOME0.NETSCAPE.COM:443"
  ,.num_headers= 2
  ,.upgrade=""
  ,.headers= { { "User-agent", "Mozilla/1.1N" }
@ -632,37 +638,30 @@ const struct message requests[] =
             }
  ,.body= ""
  }
-#endif  /* !HTTP_PARSER_STRICT */

-#define PATCH_REQ 26
-, {.name = "PATCH request"
+#if !HTTP_PARSER_STRICT
+#define UTF8_PATH_REQ 26
+, {.name= "utf-8 path request"
  ,.type= HTTP_REQUEST
-  ,.raw= "PATCH /file.txt HTTP/1.1\r\n"
-         "Host: www.example.com\r\n"
-         "Content-Type: application/example\r\n"
-         "If-Match: \"e0023aa4e\"\r\n"
-         "Content-Length: 10\r\n"
+  ,.raw= "GET /δ¶/δt/pope?q=1#narf HTTP/1.1\r\n"
+         "Host: github.com\r\n"
         "\r\n"
-         "cccccccccc"
  ,.should_keep_alive= TRUE
  ,.message_complete_on_eof= FALSE
  ,.http_major= 1
  ,.http_minor= 1
-  ,.method= HTTP_PATCH
-  ,.request_url= "/file.txt"
-  ,.num_headers= 4
-  ,.headers= { { "Host", "www.example.com" }
-             , { "Content-Type", "application/example" }
-             , { "If-Match", "\"e0023aa4e\"" }
-             , { "Content-Length", "10" }
+  ,.method= HTTP_GET
+  ,.request_url= "/δ¶/δt/pope?q=1#narf"
+  ,.num_headers= 1
+  ,.headers= { {"Host", "github.com" }
             }
-  ,.body= "cccccccccc"
+  ,.body= ""
  }

-#define CONNECT_CAPS_REQUEST 27
-, {.name = "connect caps request"
+#define HOSTNAME_UNDERSCORE 27
+, {.name = "hostname underscore"
  ,.type= HTTP_REQUEST
-  ,.raw= "CONNECT HOME0.NETSCAPE.COM:443 HTTP/1.0\r\n"
+  ,.raw= "CONNECT home_0.netscape.com:443 HTTP/1.0\r\n"
         "User-agent: Mozilla/1.1N\r\n"
         "Proxy-authorization: basic aGVsbG86d29ybGQ=\r\n"
         "\r\n"
@ -671,7 +670,7 @@ const struct message requests[] =
  ,.http_major= 1
  ,.http_minor= 0
  ,.method= HTTP_CONNECT
-  ,.request_url= "HOME0.NETSCAPE.COM:443"
+  ,.request_url= "home_0.netscape.com:443"
  ,.num_headers= 2
  ,.upgrade=""
  ,.headers= { { "User-agent", "Mozilla/1.1N" }
@ -679,6 +678,7 @@ const struct message requests[] =
             }
  ,.body= ""
  }
+#endif  /* !HTTP_PARSER_STRICT */

 , {.name= NULL } /* sentinel */
 };
@ -945,40 +945,7 @@ const struct message responses[] =
  ,.body= ""
  }

-#define SPACE_IN_FIELD_RES 9
-/* Should handle spaces in header fields */
-, {.name= "field space"
-  ,.type= HTTP_RESPONSE
-  ,.raw= "HTTP/1.1 200 OK\r\n"
-         "Server: Microsoft-IIS/6.0\r\n"
-         "X-Powered-By: ASP.NET\r\n"
-         "en-US Content-Type: text/xml\r\n" /* this is the problem */
-         "Content-Type: text/xml\r\n"
-         "Content-Length: 16\r\n"
-         "Date: Fri, 23 Jul 2010 18:45:38 GMT\r\n"
-         "Connection: keep-alive\r\n"
-         "\r\n"
-         "<xml>hello</xml>" /* fake body */
-  ,.should_keep_alive= TRUE
-  ,.message_complete_on_eof= FALSE
-  ,.http_major= 1
-  ,.http_minor= 1
-  ,.status_code= 200
-  ,.num_headers= 7
-  ,.headers=
-    { { "Server",  "Microsoft-IIS/6.0" }
-    , { "X-Powered-By", "ASP.NET" }
-    , { "en-US Content-Type", "text/xml" }
-    , { "Content-Type", "text/xml" }
-    , { "Content-Length", "16" }
-    , { "Date", "Fri, 23 Jul 2010 18:45:38 GMT" }
-    , { "Connection", "keep-alive" }
-    }
-  ,.body= "<xml>hello</xml>"
-  }
-
-
-#define RES_FIELD_UNDERSCORE 10
+#define RES_FIELD_UNDERSCORE 9
 /* Should handle spaces in header fields */
 , {.name= "field underscore"
  ,.type= HTTP_RESPONSE
@ -1018,7 +985,7 @@ const struct message responses[] =
  ,.body= ""
  }

-#define NON_ASCII_IN_STATUS_LINE 11
+#define NON_ASCII_IN_STATUS_LINE 10
 /* Should handle non-ASCII in status line */
 , {.name= "non-ASCII in status line"
  ,.type= HTTP_RESPONSE
@ -1041,7 +1008,7 @@ const struct message responses[] =
  ,.body= ""
  }

-#define HTTP_VERSION_0_9 12
+#define HTTP_VERSION_0_9 11
 /* Should handle HTTP/0.9 */
 , {.name= "http version 0.9"
  ,.type= HTTP_RESPONSE
@ -1058,7 +1025,7 @@ const struct message responses[] =
  ,.body= ""
  }

-#define NO_CONTENT_LENGTH_NO_TRANSFER_ENCODING_RESPONSE 13
+#define NO_CONTENT_LENGTH_NO_TRANSFER_ENCODING_RESPONSE 12
 /* The client should wait for the server's EOF. That is, when neither
 * content-length nor transfer-encoding is specified, the end of body
 * is specified by the EOF.
@ -1081,7 +1048,7 @@ const struct message responses[] =
  ,.body= "hello world"
  }

-#define NO_HEADERS_NO_BODY_204 14
+#define NO_HEADERS_NO_BODY_204 13
 , {.name= "204 no headers no body"
  ,.type= HTTP_RESPONSE
  ,.raw= "HTTP/1.1 204 No Content\r\n\r\n"
@ -1096,6 +1063,40 @@ const struct message responses[] =
  ,.body= ""
  }

+#if !HTTP_PARSER_STRICT
+#define SPACE_IN_FIELD_RES 14
+/* Should handle spaces in header fields */
+, {.name= "field space"
+  ,.type= HTTP_RESPONSE
+  ,.raw= "HTTP/1.1 200 OK\r\n"
+         "Server: Microsoft-IIS/6.0\r\n"
+         "X-Powered-By: ASP.NET\r\n"
+         "en-US Content-Type: text/xml\r\n" /* this is the problem */
+         "Content-Type: text/xml\r\n"
+         "Content-Length: 16\r\n"
+         "Date: Fri, 23 Jul 2010 18:45:38 GMT\r\n"
+         "Connection: keep-alive\r\n"
+         "\r\n"
+         "<xml>hello</xml>" /* fake body */
+  ,.should_keep_alive= TRUE
+  ,.message_complete_on_eof= FALSE
+  ,.http_major= 1
+  ,.http_minor= 1
+  ,.status_code= 200
+  ,.num_headers= 7
+  ,.headers=
+    { { "Server",  "Microsoft-IIS/6.0" }
+    , { "X-Powered-By", "ASP.NET" }
+    , { "en-US Content-Type", "text/xml" }
+    , { "Content-Type", "text/xml" }
+    , { "Content-Length", "16" }
+    , { "Date", "Fri, 23 Jul 2010 18:45:38 GMT" }
+    , { "Connection", "keep-alive" }
+    }
+  ,.body= "<xml>hello</xml>"
+  }
+#endif /* !HTTP_PARSER_STRICT */
+
 , {.name= NULL } /* sentinel */
 };