http_parser.rb 0.5.2 → 0.7.0

data/ext/ruby_http_parser/vendor/http-parser/http_parser.gyp

@@ -0,0 +1,111 @@
+# This file is used with the GYP meta build system.
+# http://code.google.com/p/gyp/
+# To build try this:
+#   svn co http://gyp.googlecode.com/svn/trunk gyp
+#   ./gyp/gyp -f make --depth=`pwd` http_parser.gyp 
+#   ./out/Debug/test 
+{
+  'target_defaults': {
+    'default_configuration': 'Debug',
+    'configurations': {
+      # TODO: hoist these out and put them somewhere common, because
+      #       RuntimeLibrary MUST MATCH across the entire project
+      'Debug': {
+        'defines': [ 'DEBUG', '_DEBUG' ],
+        'cflags': [ '-Wall', '-Wextra', '-O0', '-g', '-ftrapv' ],
+        'msvs_settings': {
+          'VCCLCompilerTool': {
+            'RuntimeLibrary': 1, # static debug
+          },
+        },
+      },
+      'Release': {
+        'defines': [ 'NDEBUG' ],
+        'cflags': [ '-Wall', '-Wextra', '-O3' ],
+        'msvs_settings': {
+          'VCCLCompilerTool': {
+            'RuntimeLibrary': 0, # static release
+          },
+        },
+      }
+    },
+    'msvs_settings': {
+      'VCCLCompilerTool': {
+      },
+      'VCLibrarianTool': {
+      },
+      'VCLinkerTool': {
+        'GenerateDebugInformation': 'true',
+      },
+    },
+    'conditions': [
+      ['OS == "win"', {
+        'defines': [
+          'WIN32'
+        ],
+      }]
+    ],
+  },
+
+  'targets': [
+    {
+      'target_name': 'http_parser',
+      'type': 'static_library',
+      'include_dirs': [ '.' ],
+      'direct_dependent_settings': {
+        'defines': [ 'HTTP_PARSER_STRICT=0' ],
+        'include_dirs': [ '.' ],
+      },
+      'defines': [ 'HTTP_PARSER_STRICT=0' ],
+      'sources': [ './http_parser.c', ],
+      'conditions': [
+        ['OS=="win"', {
+          'msvs_settings': {
+            'VCCLCompilerTool': {
+              # Compile as C++. http_parser.c is actually C99, but C++ is
+              # close enough in this case.
+              'CompileAs': 2,
+            },
+          },
+        }]
+      ],
+    },
+
+    {
+      'target_name': 'http_parser_strict',
+      'type': 'static_library',
+      'include_dirs': [ '.' ],
+      'direct_dependent_settings': {
+        'defines': [ 'HTTP_PARSER_STRICT=1' ],
+        'include_dirs': [ '.' ],
+      },
+      'defines': [ 'HTTP_PARSER_STRICT=1' ],
+      'sources': [ './http_parser.c', ],
+      'conditions': [
+        ['OS=="win"', {
+          'msvs_settings': {
+            'VCCLCompilerTool': {
+              # Compile as C++. http_parser.c is actually C99, but C++ is
+              # close enough in this case.
+              'CompileAs': 2,
+            },
+          },
+        }]
+      ],
+    },
+
+    {
+      'target_name': 'test-nonstrict',
+      'type': 'executable',
+      'dependencies': [ 'http_parser' ],
+      'sources': [ 'test.c' ]
+    },
+
+    {
+      'target_name': 'test-strict',
+      'type': 'executable',
+      'dependencies': [ 'http_parser_strict' ],
+      'sources': [ 'test.c' ]
+    }
+  ]
+}

data/ext/ruby_http_parser/vendor/http-parser/http_parser.h

@@ -24,11 +24,15 @@
 extern "C" {
 #endif
 
-#define HTTP_PARSER_VERSION_MAJOR 1
-#define HTTP_PARSER_VERSION_MINOR 0
+/* Also update SONAME in the Makefile whenever you change these. */
+#define HTTP_PARSER_VERSION_MAJOR 2
+#define HTTP_PARSER_VERSION_MINOR 8
+#define HTTP_PARSER_VERSION_PATCH 1
 
-#include <sys/types.h>
-#if defined(_WIN32) && !defined(__MINGW32__)
+#include <stddef.h>
+#if defined(_WIN32) && !defined(__MINGW32__) && \
+  (!defined(_MSC_VER) || _MSC_VER<1600) && !defined(__WINE__)
+#include <BaseTsd.h>
 typedef __int8 int8_t;
 typedef unsigned __int8 uint8_t;
 typedef __int16 int16_t;
@@ -37,9 +41,6 @@ typedef __int32 int32_t;
 typedef unsigned __int32 uint32_t;
 typedef __int64 int64_t;
 typedef unsigned __int64 uint64_t;
-
-typedef unsigned int size_t;
-typedef int ssize_t;
 #else
 #include <stdint.h>
 #endif
@@ -49,14 +50,18 @@ typedef int ssize_t;
  */
 #ifndef HTTP_PARSER_STRICT
 # define HTTP_PARSER_STRICT 1
-#else
-# define HTTP_PARSER_STRICT 0
 #endif
 
-
-/* Maximium header size allowed */
-#define HTTP_MAX_HEADER_SIZE (80*1024)
-
+/* Maximium header size allowed. If the macro is not defined
+ * before including this header then the default is used. To
+ * change the maximum header size, define the macro in the build
+ * environment (e.g. -DHTTP_MAX_HEADER_SIZE=<value>). To remove
+ * the effective limit on the size of the header, define the macro
+ * to a very large number (e.g. -DHTTP_MAX_HEADER_SIZE=0x7fffffff)
+ */
+#ifndef HTTP_MAX_HEADER_SIZE
+# define HTTP_MAX_HEADER_SIZE (80*1024)
+#endif
 
 typedef struct http_parser http_parser;
 typedef struct http_parser_settings http_parser_settings;
@@ -71,72 +76,245 @@ typedef struct http_parser_settings http_parser_settings;
  * HEAD request which may contain 'Content-Length' or 'Transfer-Encoding:
  * chunked' headers that indicate the presence of a body.
  *
- * http_data_cb does not return data chunks. It will be call arbitrarally
- * many times for each string. E.G. you might get 10 callbacks for "on_path"
+ * Returning `2` from on_headers_complete will tell parser that it should not
+ * expect neither a body nor any futher responses on this connection. This is
+ * useful for handling responses to a CONNECT request which may not contain
+ * `Upgrade` or `Connection: upgrade` headers.
+ *
+ * http_data_cb does not return data chunks. It will be called arbitrarily
+ * many times for each string. E.G. you might get 10 callbacks for "on_url"
  * each providing just a few characters more data.
  */
 typedef int (*http_data_cb) (http_parser*, const char *at, size_t length);
 typedef int (*http_cb) (http_parser*);
 
 
+/* Status Codes */
+#define HTTP_STATUS_MAP(XX)                                                 \
+  XX(100, CONTINUE,                        Continue)                        \
+  XX(101, SWITCHING_PROTOCOLS,             Switching Protocols)             \
+  XX(102, PROCESSING,                      Processing)                      \
+  XX(200, OK,                              OK)                              \
+  XX(201, CREATED,                         Created)                         \
+  XX(202, ACCEPTED,                        Accepted)                        \
+  XX(203, NON_AUTHORITATIVE_INFORMATION,   Non-Authoritative Information)   \
+  XX(204, NO_CONTENT,                      No Content)                      \
+  XX(205, RESET_CONTENT,                   Reset Content)                   \
+  XX(206, PARTIAL_CONTENT,                 Partial Content)                 \
+  XX(207, MULTI_STATUS,                    Multi-Status)                    \
+  XX(208, ALREADY_REPORTED,                Already Reported)                \
+  XX(226, IM_USED,                         IM Used)                         \
+  XX(300, MULTIPLE_CHOICES,                Multiple Choices)                \
+  XX(301, MOVED_PERMANENTLY,               Moved Permanently)               \
+  XX(302, FOUND,                           Found)                           \
+  XX(303, SEE_OTHER,                       See Other)                       \
+  XX(304, NOT_MODIFIED,                    Not Modified)                    \
+  XX(305, USE_PROXY,                       Use Proxy)                       \
+  XX(307, TEMPORARY_REDIRECT,              Temporary Redirect)              \
+  XX(308, PERMANENT_REDIRECT,              Permanent Redirect)              \
+  XX(400, BAD_REQUEST,                     Bad Request)                     \
+  XX(401, UNAUTHORIZED,                    Unauthorized)                    \
+  XX(402, PAYMENT_REQUIRED,                Payment Required)                \
+  XX(403, FORBIDDEN,                       Forbidden)                       \
+  XX(404, NOT_FOUND,                       Not Found)                       \
+  XX(405, METHOD_NOT_ALLOWED,              Method Not Allowed)              \
+  XX(406, NOT_ACCEPTABLE,                  Not Acceptable)                  \
+  XX(407, PROXY_AUTHENTICATION_REQUIRED,   Proxy Authentication Required)   \
+  XX(408, REQUEST_TIMEOUT,                 Request Timeout)                 \
+  XX(409, CONFLICT,                        Conflict)                        \
+  XX(410, GONE,                            Gone)                            \
+  XX(411, LENGTH_REQUIRED,                 Length Required)                 \
+  XX(412, PRECONDITION_FAILED,             Precondition Failed)             \
+  XX(413, PAYLOAD_TOO_LARGE,               Payload Too Large)               \
+  XX(414, URI_TOO_LONG,                    URI Too Long)                    \
+  XX(415, UNSUPPORTED_MEDIA_TYPE,          Unsupported Media Type)          \
+  XX(416, RANGE_NOT_SATISFIABLE,           Range Not Satisfiable)           \
+  XX(417, EXPECTATION_FAILED,              Expectation Failed)              \
+  XX(421, MISDIRECTED_REQUEST,             Misdirected Request)             \
+  XX(422, UNPROCESSABLE_ENTITY,            Unprocessable Entity)            \
+  XX(423, LOCKED,                          Locked)                          \
+  XX(424, FAILED_DEPENDENCY,               Failed Dependency)               \
+  XX(426, UPGRADE_REQUIRED,                Upgrade Required)                \
+  XX(428, PRECONDITION_REQUIRED,           Precondition Required)           \
+  XX(429, TOO_MANY_REQUESTS,               Too Many Requests)               \
+  XX(431, REQUEST_HEADER_FIELDS_TOO_LARGE, Request Header Fields Too Large) \
+  XX(451, UNAVAILABLE_FOR_LEGAL_REASONS,   Unavailable For Legal Reasons)   \
+  XX(500, INTERNAL_SERVER_ERROR,           Internal Server Error)           \
+  XX(501, NOT_IMPLEMENTED,                 Not Implemented)                 \
+  XX(502, BAD_GATEWAY,                     Bad Gateway)                     \
+  XX(503, SERVICE_UNAVAILABLE,             Service Unavailable)             \
+  XX(504, GATEWAY_TIMEOUT,                 Gateway Timeout)                 \
+  XX(505, HTTP_VERSION_NOT_SUPPORTED,      HTTP Version Not Supported)      \
+  XX(506, VARIANT_ALSO_NEGOTIATES,         Variant Also Negotiates)         \
+  XX(507, INSUFFICIENT_STORAGE,            Insufficient Storage)            \
+  XX(508, LOOP_DETECTED,                   Loop Detected)                   \
+  XX(510, NOT_EXTENDED,                    Not Extended)                    \
+  XX(511, NETWORK_AUTHENTICATION_REQUIRED, Network Authentication Required) \
+
+enum http_status
+  {
+#define XX(num, name, string) HTTP_STATUS_##name = num,
+  HTTP_STATUS_MAP(XX)
+#undef XX
+  };
+
+
 /* Request Methods */
+#define HTTP_METHOD_MAP(XX)         \
+  XX(0,  DELETE,      DELETE)       \
+  XX(1,  GET,         GET)          \
+  XX(2,  HEAD,        HEAD)         \
+  XX(3,  POST,        POST)         \
+  XX(4,  PUT,         PUT)          \
+  /* pathological */                \
+  XX(5,  CONNECT,     CONNECT)      \
+  XX(6,  OPTIONS,     OPTIONS)      \
+  XX(7,  TRACE,       TRACE)        \
+  /* WebDAV */                      \
+  XX(8,  COPY,        COPY)         \
+  XX(9,  LOCK,        LOCK)         \
+  XX(10, MKCOL,       MKCOL)        \
+  XX(11, MOVE,        MOVE)         \
+  XX(12, PROPFIND,    PROPFIND)     \
+  XX(13, PROPPATCH,   PROPPATCH)    \
+  XX(14, SEARCH,      SEARCH)       \
+  XX(15, UNLOCK,      UNLOCK)       \
+  XX(16, BIND,        BIND)         \
+  XX(17, REBIND,      REBIND)       \
+  XX(18, UNBIND,      UNBIND)       \
+  XX(19, ACL,         ACL)          \
+  /* subversion */                  \
+  XX(20, REPORT,      REPORT)       \
+  XX(21, MKACTIVITY,  MKACTIVITY)   \
+  XX(22, CHECKOUT,    CHECKOUT)     \
+  XX(23, MERGE,       MERGE)        \
+  /* upnp */                        \
+  XX(24, MSEARCH,     M-SEARCH)     \
+  XX(25, NOTIFY,      NOTIFY)       \
+  XX(26, SUBSCRIBE,   SUBSCRIBE)    \
+  XX(27, UNSUBSCRIBE, UNSUBSCRIBE)  \
+  /* RFC-5789 */                    \
+  XX(28, PATCH,       PATCH)        \
+  XX(29, PURGE,       PURGE)        \
+  /* CalDAV */                      \
+  XX(30, MKCALENDAR,  MKCALENDAR)   \
+  /* RFC-2068, section 19.6.1.2 */  \
+  XX(31, LINK,        LINK)         \
+  XX(32, UNLINK,      UNLINK)       \
+  /* icecast */                     \
+  XX(33, SOURCE,      SOURCE)       \
+
 enum http_method
-  { HTTP_DELETE    = 0
-  , HTTP_GET
-  , HTTP_HEAD
-  , HTTP_POST
-  , HTTP_PUT
-  /* pathological */
-  , HTTP_CONNECT
-  , HTTP_OPTIONS
-  , HTTP_TRACE
-  /* webdav */
-  , HTTP_COPY
-  , HTTP_LOCK
-  , HTTP_MKCOL
-  , HTTP_MOVE
-  , HTTP_PROPFIND
-  , HTTP_PROPPATCH
-  , HTTP_UNLOCK
-  /* subversion */
-  , HTTP_REPORT
-  , HTTP_MKACTIVITY
-  , HTTP_CHECKOUT
-  , HTTP_MERGE
-  /* upnp */
-  , HTTP_MSEARCH
-  , HTTP_NOTIFY
-  , HTTP_SUBSCRIBE
-  , HTTP_UNSUBSCRIBE
+  {
+#define XX(num, name, string) HTTP_##name = num,
+  HTTP_METHOD_MAP(XX)
+#undef XX
   };
 
 
 enum http_parser_type { HTTP_REQUEST, HTTP_RESPONSE, HTTP_BOTH };
 
 
+/* Flag values for http_parser.flags field */
+enum flags
+  { F_CHUNKED               = 1 << 0
+  , F_CONNECTION_KEEP_ALIVE = 1 << 1
+  , F_CONNECTION_CLOSE      = 1 << 2
+  , F_CONNECTION_UPGRADE    = 1 << 3
+  , F_TRAILING              = 1 << 4
+  , F_UPGRADE               = 1 << 5
+  , F_SKIPBODY              = 1 << 6
+  , F_CONTENTLENGTH         = 1 << 7
+  };
+
+
+/* Map for errno-related constants
+ *
+ * The provided argument should be a macro that takes 2 arguments.
+ */
+#define HTTP_ERRNO_MAP(XX)                                           \
+  /* No error */                                                     \
+  XX(OK, "success")                                                  \
+                                                                     \
+  /* Callback-related errors */                                      \
+  XX(CB_message_begin, "the on_message_begin callback failed")       \
+  XX(CB_url, "the on_url callback failed")                           \
+  XX(CB_header_field, "the on_header_field callback failed")         \
+  XX(CB_header_value, "the on_header_value callback failed")         \
+  XX(CB_headers_complete, "the on_headers_complete callback failed") \
+  XX(CB_body, "the on_body callback failed")                         \
+  XX(CB_message_complete, "the on_message_complete callback failed") \
+  XX(CB_status, "the on_status callback failed")                     \
+  XX(CB_chunk_header, "the on_chunk_header callback failed")         \
+  XX(CB_chunk_complete, "the on_chunk_complete callback failed")     \
+                                                                     \
+  /* Parsing-related errors */                                       \
+  XX(INVALID_EOF_STATE, "stream ended at an unexpected time")        \
+  XX(HEADER_OVERFLOW,                                                \
+     "too many header bytes seen; overflow detected")                \
+  XX(CLOSED_CONNECTION,                                              \
+     "data received after completed connection: close message")      \
+  XX(INVALID_VERSION, "invalid HTTP version")                        \
+  XX(INVALID_STATUS, "invalid HTTP status code")                     \
+  XX(INVALID_METHOD, "invalid HTTP method")                          \
+  XX(INVALID_URL, "invalid URL")                                     \
+  XX(INVALID_HOST, "invalid host")                                   \
+  XX(INVALID_PORT, "invalid port")                                   \
+  XX(INVALID_PATH, "invalid path")                                   \
+  XX(INVALID_QUERY_STRING, "invalid query string")                   \
+  XX(INVALID_FRAGMENT, "invalid fragment")                           \
+  XX(LF_EXPECTED, "LF character expected")                           \
+  XX(INVALID_HEADER_TOKEN, "invalid character in header")            \
+  XX(INVALID_CONTENT_LENGTH,                                         \
+     "invalid character in content-length header")                   \
+  XX(UNEXPECTED_CONTENT_LENGTH,                                      \
+     "unexpected content-length header")                             \
+  XX(INVALID_CHUNK_SIZE,                                             \
+     "invalid character in chunk size header")                       \
+  XX(INVALID_CONSTANT, "invalid constant string")                    \
+  XX(INVALID_INTERNAL_STATE, "encountered unexpected internal state")\
+  XX(STRICT, "strict mode assertion failed")                         \
+  XX(PAUSED, "parser is paused")                                     \
+  XX(UNKNOWN, "an unknown error occurred")
+
+
+/* Define HPE_* values for each errno value above */
+#define HTTP_ERRNO_GEN(n, s) HPE_##n,
+enum http_errno {
+  HTTP_ERRNO_MAP(HTTP_ERRNO_GEN)
+};
+#undef HTTP_ERRNO_GEN
+
+
+/* Get an http_errno value from an http_parser */
+#define HTTP_PARSER_ERRNO(p)            ((enum http_errno) (p)->http_errno)
+
+
 struct http_parser {
   /** PRIVATE **/
-  unsigned char type : 2;
-  unsigned char flags : 6;
-  unsigned char state;
-  unsigned char header_state;
-  unsigned char index;
+  unsigned int type : 2;         /* enum http_parser_type */
+  unsigned int flags : 8;        /* F_* values from 'flags' enum; semi-public */
+  unsigned int state : 7;        /* enum state from http_parser.c */
+  unsigned int header_state : 7; /* enum header_state from http_parser.c */
+  unsigned int index : 7;        /* index into current matcher */
+  unsigned int lenient_http_headers : 1;
 
-  uint32_t nread;
-  int64_t content_length;
+  uint32_t nread;          /* # bytes read in various scenarios */
+  uint64_t content_length; /* # bytes in body (0 if no Content-Length header) */
 
   /** READ-ONLY **/
   unsigned short http_major;
   unsigned short http_minor;
-  unsigned short status_code; /* responses only */
-  unsigned char method;    /* requests only */
+  unsigned int status_code : 16; /* responses only */
+  unsigned int method : 8;       /* requests only */
+  unsigned int http_errno : 7;
 
   /* 1 = Upgrade header was present and the parser has exited because of that.
    * 0 = No upgrade header present.
    * Should be checked when http_parser_execute() returns in addition to
    * error checking.
    */
-  char upgrade;
+  unsigned int upgrade : 1;
 
   /** PUBLIC **/
   void *data; /* A pointer to get hook to the "connection" or "socket" object */
@@ -145,21 +323,73 @@ struct http_parser {
 
 struct http_parser_settings {
   http_cb      on_message_begin;
-  http_data_cb on_path;
-  http_data_cb on_query_string;
   http_data_cb on_url;
-  http_data_cb on_fragment;
+  http_data_cb on_status;
   http_data_cb on_header_field;
   http_data_cb on_header_value;
   http_cb      on_headers_complete;
   http_data_cb on_body;
   http_cb      on_message_complete;
+  /* When on_chunk_header is called, the current chunk length is stored
+   * in parser->content_length.
+   */
+  http_cb      on_chunk_header;
+  http_cb      on_chunk_complete;
+};
+
+
+enum http_parser_url_fields
+  { UF_SCHEMA           = 0
+  , UF_HOST             = 1
+  , UF_PORT             = 2
+  , UF_PATH             = 3
+  , UF_QUERY            = 4
+  , UF_FRAGMENT         = 5
+  , UF_USERINFO         = 6
+  , UF_MAX              = 7
+  };
+
+
+/* Result structure for http_parser_parse_url().
+ *
+ * Callers should index into field_data[] with UF_* values iff field_set
+ * has the relevant (1 << UF_*) bit set. As a courtesy to clients (and
+ * because we probably have padding left over), we convert any port to
+ * a uint16_t.
+ */
+struct http_parser_url {
+  uint16_t field_set;           /* Bitmask of (1 << UF_*) values */
+  uint16_t port;                /* Converted UF_PORT string */
+
+  struct {
+    uint16_t off;               /* Offset into buffer in which field starts */
+    uint16_t len;               /* Length of run in buffer */
+  } field_data[UF_MAX];
 };
 
 
+/* Returns the library version. Bits 16-23 contain the major version number,
+ * bits 8-15 the minor version number and bits 0-7 the patch level.
+ * Usage example:
+ *
+ *   unsigned long version = http_parser_version();
+ *   unsigned major = (version >> 16) & 255;
+ *   unsigned minor = (version >> 8) & 255;
+ *   unsigned patch = version & 255;
+ *   printf("http_parser v%u.%u.%u\n", major, minor, patch);
+ */
+unsigned long http_parser_version(void);
+
 void http_parser_init(http_parser *parser, enum http_parser_type type);
 
 
+/* Initialize http_parser_settings members to 0
+ */
+void http_parser_settings_init(http_parser_settings *settings);
+
+
+/* Executes the parser. Returns number of parsed bytes. Sets
+ * `parser->http_errno` on error. */
 size_t http_parser_execute(http_parser *parser,
                            const http_parser_settings *settings,
                            const char *data,
@@ -167,15 +397,35 @@ size_t http_parser_execute(http_parser *parser,
 
 
 /* If http_should_keep_alive() in the on_headers_complete or
- * on_message_complete callback returns true, then this will be should be
+ * on_message_complete callback returns 0, then this should be
  * the last message on the connection.
  * If you are the server, respond with the "Connection: close" header.
  * If you are the client, close the connection.
  */
-int http_should_keep_alive(http_parser *parser);
+int http_should_keep_alive(const http_parser *parser);
 
 /* Returns a string version of the HTTP method. */
-const char *http_method_str(enum http_method);
+const char *http_method_str(enum http_method m);
+
+/* Return a string name of the given error */
+const char *http_errno_name(enum http_errno err);
+
+/* Return a string description of the given error */
+const char *http_errno_description(enum http_errno err);
+
+/* Initialize all http_parser_url members to 0 */
+void http_parser_url_init(struct http_parser_url *u);
+
+/* Parse a URL; return nonzero on failure */
+int http_parser_parse_url(const char *buf, size_t buflen,
+                          int is_connect,
+                          struct http_parser_url *u);
+
+/* Pause or un-pause the parser; a nonzero value pauses */
+void http_parser_pause(http_parser *parser, int paused);
+
+/* Checks if this is the final chunk of the body. */
+int http_body_is_final(const http_parser *parser);
 
 #ifdef __cplusplus
 }