http-parser updated to latest version.

src/3rdparty/http-parser/http_parser.c

@@ -381,7 +381,10 @@ enum header_states
   , h_transfer_encoding
   , h_upgrade
 
+  , h_matching_transfer_encoding_token_start
   , h_matching_transfer_encoding_chunked
+  , h_matching_transfer_encoding_token
+
   , h_matching_connection_token_start
   , h_matching_connection_keep_alive
   , h_matching_connection_close
@@ -1257,9 +1260,9 @@ reexecute:
 
           switch (parser->header_state) {
             case h_general: {
-              size_t limit = data + len - p;
+              size_t left = data + len - p;
-              limit = MIN(limit, max_header_size);
+              const char* pe = p + MIN(left, max_header_size);
-              while (p+1 < data + limit && TOKEN(p[1])) {
+              while (p+1 < pe && TOKEN(p[1])) {
                 p++;
               }
               break;
@@ -1335,6 +1338,7 @@ reexecute:
                 parser->header_state = h_general;
               } else if (parser->index == sizeof(TRANSFER_ENCODING)-2) {
                 parser->header_state = h_transfer_encoding;
+                parser->flags |= F_TRANSFER_ENCODING;
               }
               break;
 
@@ -1416,10 +1420,14 @@ reexecute:
             if ('c' == c) {
               parser->header_state = h_matching_transfer_encoding_chunked;
             } else {
-              parser->header_state = h_general;
+              parser->header_state = h_matching_transfer_encoding_token;
             }
             break;
 
+          /* Multi-value `Transfer-Encoding` header */
+          case h_matching_transfer_encoding_token_start:
+            break;
+
           case h_content_length:
             if (UNLIKELY(!IS_NUM(ch))) {
               SET_ERRNO(HPE_INVALID_CONTENT_LENGTH);
@@ -1496,28 +1504,25 @@ reexecute:
 
           switch (h_state) {
             case h_general:
-            {
+              {
-              const char* p_cr;
+                size_t left = data + len - p;
-              const char* p_lf;
+                const char* pe = p + MIN(left, max_header_size);
-              size_t limit = data + len - p;
 
-              limit = MIN(limit, max_header_size);
+                for (; p != pe; p++) {
-
+                  ch = *p;
-              p_cr = (const char*) memchr(p, CR, limit);
+                  if (ch == CR || ch == LF) {
-              p_lf = (const char*) memchr(p, LF, limit);
+                    --p;
-              if (p_cr != NULL) {
+                    break;
-                if (p_lf != NULL && p_cr >= p_lf)
+                  }
-                  p = p_lf;
+                  if (!lenient && !IS_HEADER_CHAR(ch)) {
-                else
+                    SET_ERRNO(HPE_INVALID_HEADER_TOKEN);
-                  p = p_cr;
+                    goto error;
-              } else if (UNLIKELY(p_lf != NULL)) {
+                  }
-                p = p_lf;
+                }
-              } else {
+                if (p == data + len)
-                p = data + len;
+                  --p;
+                break;
               }
-              --p;
-              break;
-            }
 
             case h_connection:
             case h_transfer_encoding:
@@ -1566,16 +1571,41 @@ reexecute:
               goto error;
 
             /* Transfer-Encoding: chunked */
+            case h_matching_transfer_encoding_token_start:
+              /* looking for 'Transfer-Encoding: chunked' */
+              if ('c' == c) {
+                h_state = h_matching_transfer_encoding_chunked;
+              } else if (STRICT_TOKEN(c)) {
+                /* TODO(indutny): similar code below does this, but why?
+                 * At the very least it seems to be inconsistent given that
+                 * h_matching_transfer_encoding_token does not check for
+                 * `STRICT_TOKEN`
+                 */
+                h_state = h_matching_transfer_encoding_token;
+              } else if (c == ' ' || c == '\t') {
+                /* Skip lws */
+              } else {
+                h_state = h_general;
+              }
+              break;
+
             case h_matching_transfer_encoding_chunked:
               parser->index++;
               if (parser->index > sizeof(CHUNKED)-1
                   || c != CHUNKED[parser->index]) {
-                h_state = h_general;
+                h_state = h_matching_transfer_encoding_token;
               } else if (parser->index == sizeof(CHUNKED)-2) {
                 h_state = h_transfer_encoding_chunked;
               }
               break;
 
+            case h_matching_transfer_encoding_token:
+              if (ch == ',') {
+                h_state = h_matching_transfer_encoding_token_start;
+                parser->index = 0;
+              }
+              break;
+
             case h_matching_connection_token_start:
               /* looking for 'Connection: keep-alive' */
               if (c == 'k') {
@@ -1634,7 +1664,7 @@ reexecute:
               break;
 
             case h_transfer_encoding_chunked:
-              if (ch != ' ') h_state = h_general;
+              if (ch != ' ') h_state = h_matching_transfer_encoding_token;
               break;
 
             case h_connection_keep_alive:
@@ -1768,12 +1798,17 @@ reexecute:
           REEXECUTE();
         }
 
-        /* Cannot use chunked encoding and a content-length header together
+        /* Cannot us transfer-encoding and a content-length header together
-           per the HTTP specification. */
+           per the HTTP specification. (RFC 7230 Section 3.3.3) */
-        if ((parser->flags & F_CHUNKED) &&
+        if ((parser->flags & F_TRANSFER_ENCODING) &&
             (parser->flags & F_CONTENTLENGTH)) {
-          SET_ERRNO(HPE_UNEXPECTED_CONTENT_LENGTH);
+          /* Allow it for lenient parsing as long as `Transfer-Encoding` is
-          goto error;
+           * not `chunked`
+           */
+          if (!lenient || (parser->flags & F_CHUNKED)) {
+            SET_ERRNO(HPE_UNEXPECTED_CONTENT_LENGTH);
+            goto error;
+          }
         }
 
         UPDATE_STATE(s_headers_done);
@@ -1848,8 +1883,31 @@ reexecute:
           UPDATE_STATE(NEW_MESSAGE());
           CALLBACK_NOTIFY(message_complete);
         } else if (parser->flags & F_CHUNKED) {
-          /* chunked encoding - ignore Content-Length header */
+          /* chunked encoding - ignore Content-Length header,
+           * prepare for a chunk */
           UPDATE_STATE(s_chunk_size_start);
+        } else if (parser->flags & F_TRANSFER_ENCODING) {
+          if (parser->type == HTTP_REQUEST && !lenient) {
+            /* RFC 7230 3.3.3 */
+
+            /* If a Transfer-Encoding header field
+             * is present in a request and the chunked transfer coding is not
+             * the final encoding, the message body length cannot be determined
+             * reliably; the server MUST respond with the 400 (Bad Request)
+             * status code and then close the connection.
+             */
+            SET_ERRNO(HPE_INVALID_TRANSFER_ENCODING);
+            RETURN(p - data); /* Error */
+          } else {
+            /* RFC 7230 3.3.3 */
+
+            /* If a Transfer-Encoding header field is present in a response and
+             * the chunked transfer coding is not the final encoding, the
+             * message body length is determined by reading the connection until
+             * it is closed by the server.
+             */
+            UPDATE_STATE(s_body_identity_eof);
+          }
         } else {
           if (parser->content_length == 0) {
             /* Content-Length header given but zero: Content-Length: 0\r\n */
@@ -2103,6 +2161,12 @@ http_message_needs_eof (const http_parser *parser)
     return 0;
   }
 
+  /* RFC 7230 3.3.3, see `s_headers_almost_done` */
+  if ((parser->flags & F_TRANSFER_ENCODING) &&
+      (parser->flags & F_CHUNKED) == 0) {
+    return 1;
+  }
+
   if ((parser->flags & F_CHUNKED) || parser->content_length != ULLONG_MAX) {
     return 0;
   }

src/3rdparty/http-parser/http_parser.h

@@ -27,7 +27,7 @@ extern "C" {
 /* Also update SONAME in the Makefile whenever you change these. */
 #define HTTP_PARSER_VERSION_MAJOR 2
 #define HTTP_PARSER_VERSION_MINOR 9
-#define HTTP_PARSER_VERSION_PATCH 0
+#define HTTP_PARSER_VERSION_PATCH 3
 
 #include <stddef.h>
 #if defined(_WIN32) && !defined(__MINGW32__) && \
@@ -225,6 +225,7 @@ enum flags
   , F_UPGRADE               = 1 << 5
   , F_SKIPBODY              = 1 << 6
   , F_CONTENTLENGTH         = 1 << 7
+  , F_TRANSFER_ENCODING     = 1 << 8
   };
 
 
@@ -271,6 +272,8 @@ enum flags
      "unexpected content-length header")                             \
   XX(INVALID_CHUNK_SIZE,                                             \
      "invalid character in chunk size header")                       \
+  XX(INVALID_TRANSFER_ENCODING,                                      \
+     "request has invalid transfer-encoding")                        \
   XX(INVALID_CONSTANT, "invalid constant string")                    \
   XX(INVALID_INTERNAL_STATE, "encountered unexpected internal state")\
   XX(STRICT, "strict mode assertion failed")                         \
@@ -293,11 +296,11 @@ enum http_errno {
 struct http_parser {
   /** PRIVATE **/
   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;
+  unsigned int flags : 16;       /* F_* values from 'flags' enum; semi-public */
 
   uint32_t nread;          /* # bytes read in various scenarios */
   uint64_t content_length; /* # bytes in body (0 if no Content-Length header) */