Merge upstream version 2.9.3

Makefile

@@ -24,7 +24,7 @@ BINEXT ?=
 SOLIBNAME = libhttp_parser
 SOMAJOR = 2
 SOMINOR = 9
-SOREV   = 2
+SOREV   = 3
 ifeq (darwin,$(PLATFORM))
 SOEXT ?= dylib
 SONAME ?= $(SOLIBNAME).$(SOMAJOR).$(SOMINOR).$(SOEXT)
@@ -133,14 +133,14 @@ tags: http_parser.c http_parser.h test.c
 install: library
 	$(INSTALL) -D  http_parser.h $(DESTDIR)$(INCLUDEDIR)/http_parser.h
 	$(INSTALL) -D $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(LIBNAME)
-	ln -s $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(SONAME)
-	ln -s $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(SOLIBNAME).$(SOEXT)
+	ln -sf $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(SONAME)
+	ln -sf $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(SOLIBNAME).$(SOEXT)
 
 install-strip: library
 	$(INSTALL) -D  http_parser.h $(DESTDIR)$(INCLUDEDIR)/http_parser.h
 	$(INSTALL) -D -s $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(LIBNAME)
-	ln -s $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(SONAME)
-	ln -s $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(SOLIBNAME).$(SOEXT)
+	ln -sf $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(SONAME)
+	ln -sf $(LIBNAME) $(DESTDIR)$(LIBDIR)/$(SOLIBNAME).$(SOEXT)
 
 uninstall:
 	rm $(DESTDIR)$(INCLUDEDIR)/http_parser.h

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
@@ -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);
@@ -1563,16 +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') {
@@ -1631,7 +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:
@@ -1765,12 +1798,17 @@ reexecute:
           REEXECUTE();
         }
 
-        /* Cannot use chunked encoding and a content-length header together
-           per the HTTP specification. */
-        if ((parser->flags & F_CHUNKED) &&
+        /* Cannot us transfer-encoding and a content-length header together
+           per the HTTP specification. (RFC 7230 Section 3.3.3) */
+        if ((parser->flags & F_TRANSFER_ENCODING) &&
             (parser->flags & F_CONTENTLENGTH)) {
-          SET_ERRNO(HPE_UNEXPECTED_CONTENT_LENGTH);
-          goto error;
+          /* Allow it for lenient parsing as long as `Transfer-Encoding` is
+           * not `chunked`
+           */
+          if (!lenient || (parser->flags & F_CHUNKED)) {
+            SET_ERRNO(HPE_UNEXPECTED_CONTENT_LENGTH);
+            goto error;
+          }
         }
 
         UPDATE_STATE(s_headers_done);
@@ -1845,8 +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 */
@@ -2100,6 +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;
   }

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 2
+#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) */

test.c

@@ -262,7 +262,6 @@ const struct message requests[] =
   ,.type= HTTP_REQUEST
   ,.raw= "POST /post_identity_body_world?q=search#hey HTTP/1.1\r\n"
          "Accept: */*\r\n"
-         "Transfer-Encoding: identity\r\n"
          "Content-Length: 5\r\n"
          "\r\n"
          "World"
@@ -275,10 +274,9 @@ const struct message requests[] =
   ,.fragment= "hey"
   ,.request_path= "/post_identity_body_world"
   ,.request_url= "/post_identity_body_world?q=search#hey"
-  ,.num_headers= 3
+  ,.num_headers= 2
   ,.headers=
     { { "Accept", "*/*" }
-    , { "Transfer-Encoding", "identity" }
     , { "Content-Length", "5" }
     }
   ,.body= "World"
@@ -1193,6 +1191,61 @@ const struct message requests[] =
   ,.headers= { { "Host", "example.com" } }
   ,.body= ""
   }
+
+#define POST_MULTI_TE_LAST_CHUNKED 43
+, {.name= "post - multi coding transfer-encoding chunked body"
+  ,.type= HTTP_REQUEST
+  ,.raw= "POST / HTTP/1.1\r\n"
+         "Transfer-Encoding: deflate, chunked\r\n"
+         "\r\n"
+         "1e\r\nall your base are belong to us\r\n"
+         "0\r\n"
+         "\r\n"
+  ,.should_keep_alive= TRUE
+  ,.message_complete_on_eof= FALSE
+  ,.http_major= 1
+  ,.http_minor= 1
+  ,.method= HTTP_POST
+  ,.query_string= ""
+  ,.fragment= ""
+  ,.request_path= "/"
+  ,.request_url= "/"
+  ,.num_headers= 1
+  ,.headers=
+    { { "Transfer-Encoding" , "deflate, chunked" }
+    }
+  ,.body= "all your base are belong to us"
+  ,.num_chunks_complete= 2
+  ,.chunk_lengths= { 0x1e }
+  }
+
+#define POST_MULTI_LINE_TE_LAST_CHUNKED 43
+, {.name= "post - multi coding transfer-encoding chunked body"
+  ,.type= HTTP_REQUEST
+  ,.raw= "POST / HTTP/1.1\r\n"
+         "Transfer-Encoding: deflate,\r\n"
+         " chunked\r\n"
+         "\r\n"
+         "1e\r\nall your base are belong to us\r\n"
+         "0\r\n"
+         "\r\n"
+  ,.should_keep_alive= TRUE
+  ,.message_complete_on_eof= FALSE
+  ,.http_major= 1
+  ,.http_minor= 1
+  ,.method= HTTP_POST
+  ,.query_string= ""
+  ,.fragment= ""
+  ,.request_path= "/"
+  ,.request_url= "/"
+  ,.num_headers= 1
+  ,.headers=
+    { { "Transfer-Encoding" , "deflate, chunked" }
+    }
+  ,.body= "all your base are belong to us"
+  ,.num_chunks_complete= 2
+  ,.chunk_lengths= { 0x1e }
+  }
 };
 
 /* * R E S P O N S E S * */
@@ -1970,6 +2023,28 @@ const struct message responses[] =
   ,.num_chunks_complete= 3
   ,.chunk_lengths= { 2, 2 }
   }
+#define HTTP_200_MULTI_TE_NOT_LAST_CHUNKED 28
+, {.name= "HTTP 200 response with `chunked` being *not last* Transfer-Encoding"
+  ,.type= HTTP_RESPONSE
+  ,.raw= "HTTP/1.1 200 OK\r\n"
+         "Transfer-Encoding: chunked, identity\r\n"
+         "\r\n"
+         "2\r\n"
+         "OK\r\n"
+         "0\r\n"
+         "\r\n"
+  ,.should_keep_alive= FALSE
+  ,.message_complete_on_eof= TRUE
+  ,.http_major= 1
+  ,.http_minor= 1
+  ,.status_code= 200
+  ,.response_status= "OK"
+  ,.num_headers= 1
+  ,.headers= { { "Transfer-Encoding", "chunked, identity" }
+             }
+  ,.body= "2\r\nOK\r\n0\r\n\r\n"
+  ,.num_chunks_complete= 0
+  }
 };
 
 /* strnlen() is a POSIX.2008 addition. Can't rely on it being available so
@@ -3663,7 +3738,7 @@ test_chunked_content_length_error (int req)
   parsed = http_parser_execute(&parser, &settings_null, buf, strlen(buf));
   assert(parsed == strlen(buf));
 
-  buf = "Transfer-Encoding: chunked\r\nContent-Length: 1\r\n\r\n";
+  buf = "Transfer-Encoding: anything\r\nContent-Length: 1\r\n\r\n";
   size_t buflen = strlen(buf);
 
   parsed = http_parser_execute(&parser, &settings_null, buf, buflen);
@@ -4332,6 +4407,12 @@ main (void)
               "fooba",
               HPE_OK);
 
+  // Unknown Transfer-Encoding in request
+  test_simple("GET / HTTP/1.1\r\n"
+              "Transfer-Encoding: unknown\r\n"
+              "\r\n",
+              HPE_INVALID_TRANSFER_ENCODING);
+
   static const char *all_methods[] = {
     "DELETE",
     "GET",