http_parser.rb 0.5.3 → 0.6.0.beta.1

data/http_parser.rb.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name = "http_parser.rb"
-  s.version = "0.5.3"
+  s.version = "0.6.0.beta.1"
   s.summary = "Simple callback-based HTTP request/response parser"
   s.description = "Ruby bindings to http://github.com/ry/http-parser and http://github.com/a2800276/http-parser.java"
 
@@ -16,6 +16,8 @@ Gem::Specification.new do |s|
   s.add_development_dependency 'rake-compiler', '>= 0.7.9'
   s.add_development_dependency 'rspec', '>= 2.0.1'
   s.add_development_dependency 'json', '>= 1.4.6'
+  s.add_development_dependency 'benchmark_suite'
+  s.add_development_dependency 'ffi'
 
   if RUBY_PLATFORM =~ /java/
     s.add_development_dependency 'jruby-openssl'

data/spec/parser_spec.rb

@@ -24,9 +24,6 @@ describe HTTP::Parser do
     @parser.status_code.should be_nil
 
     @parser.request_url.should be_nil
-    @parser.request_path.should be_nil
-    @parser.query_string.should be_nil
-    @parser.fragment.should be_nil
 
     @parser.header_value_type.should == :mixed
   end
@@ -78,9 +75,6 @@ describe HTTP::Parser do
     @parser.status_code.should be_nil
 
     @parser.request_url.should == '/test?ok=1'
-    @parser.request_path.should == '/test'
-    @parser.query_string.should == 'ok=1'
-    @parser.fragment.should be_empty
 
     @parser.headers.should == @headers
     @parser.headers['User-Agent'].should == 'curl/7.18.0'
@@ -119,9 +113,6 @@ describe HTTP::Parser do
     @parser.http_version.should == [1,0]
 
     @parser.request_url.should == '/'
-    @parser.request_path.should == '/'
-    @parser.query_string.should == ''
-    @parser.fragment.should == ''
 
     @parser.reset!.should be_true
 
@@ -130,9 +121,6 @@ describe HTTP::Parser do
     @parser.status_code.should be_nil
 
     @parser.request_url.should be_nil
-    @parser.request_path.should be_nil
-    @parser.query_string.should be_nil
-    @parser.fragment.should be_nil
   end
 
   it "should optionally reset parser state on no-body responses" do
@@ -282,16 +270,55 @@ describe HTTP::Parser do
     @parser.upgrade_data.should == ''
   end
 
+  it 'should stop parsing headers when instructed' do
+    request = "GET /websocket HTTP/1.1\r\n" +
+      "host: localhost\r\n" +
+      "connection: Upgrade\r\n" +
+      "upgrade: websocket\r\n" +
+      "sec-websocket-key: SD6/hpYbKjQ6Sown7pBbWQ==\r\n" +
+      "sec-websocket-version: 13\r\n" +
+      "\r\n"
+
+    @parser.on_headers_complete = proc { |e| :stop }
+    offset = (@parser << request)
+    @parser.upgrade?.should be_true
+    @parser.upgrade_data.should == ''
+    offset.should == request.length
+  end
+
+  it "should execute on_body on requests with no content-length" do
+   @parser.reset!.should be_true
+
+   @head, @complete, @body = 0, 0, 0
+   @parser.on_headers_complete = proc {|h| @head += 1 }
+   @parser.on_message_complete = proc { @complete += 1 }
+   @parser.on_body = proc {|b| @body += 1 }
+
+   head_response = "HTTP/1.1 200 OK\r\n\r\nstuff"
+
+   @parser << head_response
+   @parser << ''
+   @head.should == 1
+   @complete.should == 1
+   @body.should == 1
+  end
+
+
   %w[ request response ].each do |type|
     JSON.parse(File.read(File.expand_path("../support/#{type}s.json", __FILE__))).each do |test|
       test['headers'] ||= {}
+      next unless HTTP::Parser.strict? == test['strict']
 
       it "should parse #{type}: #{test['name']}" do
         @parser << test['raw']
 
-        @parser.keep_alive?.should == test['should_keep_alive']
-        @parser.upgrade?.should == (test['upgrade']==1)
         @parser.http_method.should == test['method']
+        @parser.keep_alive?.should == test['should_keep_alive']
+
+        if test.has_key?('upgrade') and test['upgrade'] != 0
+          @parser.upgrade?.should be_true
+          @parser.upgrade_data.should == test['upgrade']
+        end
 
         fields = %w[
           http_major
@@ -301,9 +328,6 @@ describe HTTP::Parser do
         if test['type'] == 'HTTP_REQUEST'
           fields += %w[
             request_url
-            request_path
-            query_string
-            fragment
           ]
         else
           fields += %w[

data/spec/support/requests.json

@@ -18,7 +18,8 @@
       "Host": "0.0.0.0=5000",
       "Accept": "*/*"
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "firefox get",
@@ -44,7 +45,8 @@
       "Keep-Alive": "300",
       "Connection": "keep-alive"
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "dumbfuck",
@@ -63,7 +65,8 @@
     "headers": {
       "aaaaaaaaaaaaa": "++++++++++"
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "fragment in url",
@@ -79,7 +82,8 @@
     "request_path": "/forums/1/topics/2375",
     "request_url": "/forums/1/topics/2375?page=1#posts-17408",
     "num_headers": 0,
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "get no headers no body",
@@ -95,7 +99,8 @@
     "request_path": "/get_no_headers_no_body/world",
     "request_url": "/get_no_headers_no_body/world",
     "num_headers": 0,
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "get one header no body",
@@ -114,7 +119,8 @@
     "headers": {
       "Accept": "*/*"
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "get funky content length body hello",
@@ -133,7 +139,8 @@
     "headers": {
       "conTENT-Length": "5"
     },
-    "body": "HELLO"
+    "body": "HELLO",
+    "strict": true
   },
   {
     "name": "post identity body world",
@@ -154,7 +161,8 @@
       "Transfer-Encoding": "identity",
       "Content-Length": "5"
     },
-    "body": "World"
+    "body": "World",
+    "strict": true
   },
   {
     "name": "post - chunked body: all your base are belong to us",
@@ -173,7 +181,8 @@
     "headers": {
       "Transfer-Encoding": "chunked"
     },
-    "body": "all your base are belong to us"
+    "body": "all your base are belong to us",
+    "strict": true
   },
   {
     "name": "two chunks ; triple zero ending",
@@ -192,7 +201,8 @@
     "headers": {
       "Transfer-Encoding": "chunked"
     },
-    "body": "hello world"
+    "body": "hello world",
+    "strict": true
   },
   {
     "name": "chunked with trailing headers. blech.",
@@ -213,7 +223,8 @@
       "Vary": "*",
       "Content-Type": "text/plain"
     },
-    "body": "hello world"
+    "body": "hello world",
+    "strict": true
   },
   {
     "name": "with bullshit after the length",
@@ -232,7 +243,8 @@
     "headers": {
       "Transfer-Encoding": "chunked"
     },
-    "body": "hello world"
+    "body": "hello world",
+    "strict": true
   },
   {
     "name": "with quotes",
@@ -251,7 +263,8 @@
     "headers": {
 
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "apachebench get",
@@ -272,7 +285,8 @@
       "User-Agent": "ApacheBench/2.3",
       "Accept": "*/*"
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "query url with question mark",
@@ -291,7 +305,8 @@
     "headers": {
 
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "newline prefix get",
@@ -310,12 +325,13 @@
     "headers": {
 
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "upgrade request",
     "type": "HTTP_REQUEST",
-    "raw": "GET /demo HTTP/1.1\r\nHost: example.com\r\nConnection: Upgrade\r\nSec-WebSocket-Key2: 12998 5 Y3 1  .P00\r\nSec-WebSocket-Protocol: sample\r\nUpgrade: WebSocket\r\nSec-WebSocket-Key1: 4 @1  46546xW%0l 1 5\r\nOrigin: http://example.com\r\n\r\n",
+    "raw": "GET /demo HTTP/1.1\r\nHost: example.com\r\nConnection: Upgrade\r\nSec-WebSocket-Key2: 12998 5 Y3 1  .P00\r\nSec-WebSocket-Protocol: sample\r\nUpgrade: WebSocket\r\nSec-WebSocket-Key1: 4 @1  46546xW%0l 1 5\r\nOrigin: http://example.com\r\n\r\nHot diggity dogg",
     "should_keep_alive": true,
     "message_complete_on_eof": false,
     "http_major": 1,
@@ -326,7 +342,7 @@
     "request_path": "/demo",
     "request_url": "/demo",
     "num_headers": 7,
-    "upgrade": 1,
+    "upgrade": "Hot diggity dogg",
     "headers": {
       "Host": "example.com",
       "Connection": "Upgrade",
@@ -336,12 +352,13 @@
       "Sec-WebSocket-Key1": "4 @1  46546xW%0l 1 5",
       "Origin": "http://example.com"
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "connect request",
     "type": "HTTP_REQUEST",
-    "raw": "CONNECT home0.netscape.com:443 HTTP/1.0\r\nUser-agent: Mozilla/1.1N\r\nProxy-authorization: basic aGVsbG86d29ybGQ=\r\n\r\n",
+    "raw": "CONNECT 0-home0.netscape.com:443 HTTP/1.0\r\nUser-agent: Mozilla/1.1N\r\nProxy-authorization: basic aGVsbG86d29ybGQ=\r\n\r\nsome data\r\nand yet even more data",
     "should_keep_alive": false,
     "message_complete_on_eof": false,
     "http_major": 1,
@@ -350,14 +367,15 @@
     "query_string": "",
     "fragment": "",
     "request_path": "",
-    "request_url": "home0.netscape.com:443",
+    "request_url": "0-home0.netscape.com:443",
     "num_headers": 2,
-    "upgrade": 1,
+    "upgrade": "some data\r\nand yet even more data",
     "headers": {
       "User-agent": "Mozilla/1.1N",
       "Proxy-authorization": "basic aGVsbG86d29ybGQ="
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "report request",
@@ -376,7 +394,8 @@
     "headers": {
 
     },
-    "body": ""
+    "body": "",
+    "strict": true
   },
   {
     "name": "request with no http version",
@@ -394,6 +413,199 @@
     "num_headers": 0,
     "headers": {
 
+    },
+    "body": "",
+    "strict": true
+  },
+  {
+    "name": "m-search request",
+    "type": "HTTP_REQUEST",
+    "raw": "M-SEARCH * HTTP/1.1\r\nHOST: 239.255.255.250:1900\r\nMAN: \"ssdp:discover\"\r\nST: \"ssdp:all\"\r\n\r\n",
+    "should_keep_alive": true,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 1,
+    "method": "M-SEARCH",
+    "query_string": "",
+    "fragment": "",
+    "request_path": "*",
+    "request_url": "*",
+    "num_headers": 3,
+    "headers": {
+      "HOST": "239.255.255.250:1900",
+      "MAN": "\"ssdp:discover\"",
+      "ST": "\"ssdp:all\""
+    },
+    "body": "",
+    "strict": true
+  },
+  {
+    "name": "line folding in header value",
+    "type": "HTTP_REQUEST",
+    "raw": "GET / HTTP/1.1\r\nLine1:   abc\r\n\tdef\r\n ghi\r\n\t\tjkl\r\n  mno \r\n\t \tqrs\r\nLine2: \t line2\t\r\n\r\n",
+    "should_keep_alive": true,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 1,
+    "method": "GET",
+    "query_string": "",
+    "fragment": "",
+    "request_path": "/",
+    "request_url": "/",
+    "num_headers": 2,
+    "headers": {
+      "Line1": "abcdefghijklmno qrs",
+      "Line2": "line2\t"
+    },
+    "body": "",
+    "strict": true
+  },
+  {
+    "name": "host terminated by a query string",
+    "type": "HTTP_REQUEST",
+    "raw": "GET http://hypnotoad.org?hail=all HTTP/1.1\r\n\r\n",
+    "should_keep_alive": true,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 1,
+    "method": "GET",
+    "query_string": "hail=all",
+    "fragment": "",
+    "request_path": "",
+    "request_url": "http://hypnotoad.org?hail=all",
+    "num_headers": 0,
+    "headers": {
+
+    },
+    "body": "",
+    "strict": true
+  },
+  {
+    "name": "host:port terminated by a query string",
+    "type": "HTTP_REQUEST",
+    "raw": "GET http://hypnotoad.org:1234?hail=all HTTP/1.1\r\n\r\n",
+    "should_keep_alive": true,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 1,
+    "method": "GET",
+    "query_string": "hail=all",
+    "fragment": "",
+    "request_path": "",
+    "request_url": "http://hypnotoad.org:1234?hail=all",
+    "port": 1234,
+    "num_headers": 0,
+    "headers": {
+
+    },
+    "body": "",
+    "strict": true
+  },
+  {
+    "name": "host:port terminated by a space",
+    "type": "HTTP_REQUEST",
+    "raw": "GET http://hypnotoad.org:1234 HTTP/1.1\r\n\r\n",
+    "should_keep_alive": true,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 1,
+    "method": "GET",
+    "query_string": "",
+    "fragment": "",
+    "request_path": "",
+    "request_url": "http://hypnotoad.org:1234",
+    "port": 1234,
+    "num_headers": 0,
+    "headers": {
+
+    },
+    "body": "",
+    "strict": true
+  },
+  {
+    "name": "PATCH request",
+    "type": "HTTP_REQUEST",
+    "raw": "PATCH /file.txt HTTP/1.1\r\nHost: www.example.com\r\nContent-Type: application/example\r\nIf-Match: \"e0023aa4e\"\r\nContent-Length: 10\r\n\r\ncccccccccc",
+    "should_keep_alive": true,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 1,
+    "method": "PATCH",
+    "query_string": "",
+    "fragment": "",
+    "request_path": "/file.txt",
+    "request_url": "/file.txt",
+    "num_headers": 4,
+    "headers": {
+      "Host": "www.example.com",
+      "Content-Type": "application/example",
+      "If-Match": "\"e0023aa4e\"",
+      "Content-Length": "10"
+    },
+    "body": "cccccccccc",
+    "strict": true
+  },
+  {
+    "name": "connect caps request",
+    "type": "HTTP_REQUEST",
+    "raw": "CONNECT HOME0.NETSCAPE.COM:443 HTTP/1.0\r\nUser-agent: Mozilla/1.1N\r\nProxy-authorization: basic aGVsbG86d29ybGQ=\r\n\r\n",
+    "should_keep_alive": false,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 0,
+    "method": "CONNECT",
+    "query_string": "",
+    "fragment": "",
+    "request_path": "",
+    "request_url": "HOME0.NETSCAPE.COM:443",
+    "num_headers": 2,
+    "upgrade": "",
+    "headers": {
+      "User-agent": "Mozilla/1.1N",
+      "Proxy-authorization": "basic aGVsbG86d29ybGQ="
+    },
+    "body": "",
+    "strict": true
+  },
+  {
+    "name": "utf-8 path request",
+    "type": "HTTP_REQUEST",
+    "strict": false,
+    "raw": "GET /δ¶/δt/pope?q=1#narf HTTP/1.1\r\nHost: github.com\r\n\r\n",
+    "should_keep_alive": true,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 1,
+    "method": "GET",
+    "query_string": "q=1",
+    "fragment": "narf",
+    "request_path": "/δ¶/δt/pope",
+    "request_url": "/δ¶/δt/pope?q=1#narf",
+    "num_headers": 1,
+    "headers": {
+      "Host": "github.com"
+    },
+    "body": ""
+  },
+  {
+    "name": "hostname underscore",
+    "type": "HTTP_REQUEST",
+    "strict": false,
+    "raw": "CONNECT home_0.netscape.com:443 HTTP/1.0\r\nUser-agent: Mozilla/1.1N\r\nProxy-authorization: basic aGVsbG86d29ybGQ=\r\n\r\n",
+    "should_keep_alive": false,
+    "message_complete_on_eof": false,
+    "http_major": 1,
+    "http_minor": 0,
+    "method": "CONNECT",
+    "query_string": "",
+    "fragment": "",
+    "request_path": "",
+    "request_url": "home_0.netscape.com:443",
+    "num_headers": 2,
+    "upgrade": "",
+    "headers": {
+      "User-agent": "Mozilla/1.1N",
+      "Proxy-authorization": "basic aGVsbG86d29ybGQ="
     },
     "body": ""
   }