Changeset 996

Timestamp:

03/27/08 16:46:28 (5 months ago)

Author:

normalperson

Message:

http11_parser: accept '"' (double-quote), '<', and '>' characters in URLs

Some broken web browsers don't properly escape ", <, and > characters
in URLs, however these URLs to occasionally legitimate and sometimes
show up.

This patch was submitted by Eden Li here:

http://rubyforge.org/pipermail/mongrel-users/2006-October/001845.html

This patch was accepted by Zed Shaw here:

http://rubyforge.org/pipermail/mongrel-users/2006-October/001847.html

Files:

• trunk/ext/http11/http11_parser.c (modified) (12 diffs)
• trunk/ext/http11/http11_parser_common.rl (modified) (1 diff)
• trunk/test/test_http11.rb (modified) (1 diff)

trunk/ext/http11/http11_parser.c

@@ -485 +485 @@
-        switch( (*p) ) {
-                case 32: goto tr30;
+                case 35: goto st0;
-                case 37: goto tr31;
-                case 60: goto st0;
-                case 62: goto st0;
-                case 127: goto st0;
-        }
-
-
-
-
+
-                goto st0;
-        goto tr29;
@@ -504 +500 @@
-                goto _out21;
-case 21:
-7
+3
-        switch( (*p) ) {
-                case 32: goto tr30;
+                case 35: goto st0;
-                case 37: goto st22;
-                case 60: goto st0;
-                case 62: goto st0;
-                case 127: goto st0;
-        }
-
-
-
-
+
-                goto st0;
-        goto st21;
@@ -526 +518 @@
-                goto _out22;
-case 22:
-9
+1
-        if ( (*p) < 65 ) {
-                if ( 48 <= (*p) && (*p) <= 57 )
@@ -557 +549 @@
-                goto _out24;
-case 24:
-60
+52
-        switch( (*p) ) {
-                case 43: goto st24;
@@ -582 +574 @@
-                goto _out25;
-case 25:
-85
+77
-        switch( (*p) ) {
-                case 32: goto tr8;
-                case 34: goto st0;
-                case 35: goto tr9;
-                case 37: goto st26;
-                case 60: goto st0;
-                case 62: goto st0;
-                case 127: goto st0;
-        }
@@ -629 +618 @@
-                goto _out28;
-case 28:
-32
+21
-        switch( (*p) ) {
-                case 32: goto tr40;
-                case 34: goto st0;
-                case 35: goto tr41;
-                case 37: goto st29;
-                case 59: goto tr43;
-                case 60: goto st0;
-                case 62: goto st0;
-                case 63: goto tr44;
-                case 127: goto st0;
@@ -681 +667 @@
-                goto _out31;
-case 31:
-84
+70
-        switch( (*p) ) {
-                case 32: goto tr8;
-                case 34: goto st0;
-                case 35: goto tr9;
-                case 37: goto st32;
-                case 60: goto st0;
-                case 62: goto st0;
-                case 63: goto st34;
-                case 127: goto st0;
@@ -732 +715 @@
-                goto _out34;
-case 34:
-35
+18
-        switch( (*p) ) {
-                case 32: goto tr51;
-                case 34: goto st0;
-                case 35: goto tr52;
-                case 37: goto tr53;
-                case 60: goto st0;
-                case 62: goto st0;
-                case 127: goto st0;
-        }
@@ -753 +733 @@
-                goto _out35;
-case 35:
-5
+3
-        switch( (*p) ) {
-                case 32: goto tr55;
-                case 34: goto st0;
-                case 35: goto tr56;
-                case 37: goto st36;
-                case 60: goto st0;
-                case 62: goto st0;
-                case 127: goto st0;
-        }
@@ -774 +751 @@
-                goto _out36;
-case 36:
-77
+54
-        if ( (*p) < 65 ) {
-                if ( 48 <= (*p) && (*p) <= 57 )
@@ -1204 +1181 @@
-    /* final \r\n combo encountered so stop right here */
-    
-207
+184
-#line 136 "http11_parser.rl"
-    parser->nread++;
@@ -1217 +1194 @@
-
-  
-220
+197
-#line 147 "http11_parser.rl"
-

trunk/ext/http11/http11_parser_common.rl

@@ -12 +12 @@
-  extra = ("!" | "*" | "'" | "(" | ")" | ",");
-  reserved = (";" | "/" | "?" | ":" | "@" | "&" | "=" | "+");
-
+
+
-  national = any -- (alpha | digit | reserved | extra | safe | unsafe);
-  unreserved = (alpha | digit | safe | extra | national);
-  escape = ("%" xdigit xdigit);
-
+ | sorta_safe
-  pchar = (uchar | ":" | "@" | "&" | "=" | "+");
-  tspecials = ("(" | ")" | "<" | ">" | "@" | "," | ";" | ":" | "\\" | "\"" | "/" | "[" | "]" | "?" | "=" | "{" | "}" | " " | "\t");

trunk/test/test_http11.rb

@@ -53 +53 @@
-    #assert !parser.error?
-  end
+
+  def test_parse_ie6_urls
+    %w(/some/random/path"
+       /some/random/path>
+       /some/random/path<
+       /we/love/you/ie6?q=<"">
+       /url?<="&>="
+       /mal"formed"?
+    ).each do |path|
+      parser = HttpParser.new
+      req = {}
+      sorta_safe = %(GET #{path} HTTP/1.1\r\n\r\n)
+      nread = parser.execute(req, sorta_safe, 0)
+      assert_equal sorta_safe.length, nread
+      assert parser.finished?
+      assert !parser.error?
+    end
+  end
-  
-  def test_parse_error