super-test 1.0.9

data/lib/common/http/html_helper.rb

@@ -0,0 +1,362 @@
+#!/usr/bin/env ruby
+# coding: utf-8
+#
+# Author:: hanyu
+# Date:: 2009.5.7
+
+
+require 'hpricot'
+require 'common/http/fix_hpricot'
+
+# 解析纯文本、HTML、JSON
+#
+# - 打印可读日志
+module HtmlHelper
+
+    # === 功能:
+    # 正则表达式提取
+    # 
+    # === 参数解释:
+    # - regexp    提取的正则表达式
+    # - group    正则表达式匹配组号，默认为1
+    #
+    # === Example:
+    # Example #1:
+    #
+    #    get "http://www.baidu.com/"
+    #    puts find_grep %r"<title>(.*?)</title>"
+    #    # => 百度一下，你就知道
+    #
+    # Example #2:
+    #
+    #    get "http://www.baidu.com/"
+    #    puts find_grep /<title>(.*?)，(.*?)<\/title>/, 2    
+    #    # => 你就知道
+    def find_grep(regexp, group=1)
+        raise "no any http request before!" if @response == nil
+        m = regexp.match @response.body
+        raise HtmlError,"find_grep regexp[#{regexp}] not exist!" if m.nil?
+        return m[group]
+    end
+
+    # === 功能:
+    # 验证是否包含某字符串，不带正则表达式的功能
+    # 
+    # === 参数解释:
+    # - str        给定的字符串
+    #
+    # === Example:
+    # Example :
+    #    get "http://hi.baidu.com/"
+    #    p find_has? "百度空间"    #=> true
+    def find_has?(str)
+        raise "no any http request before!" if @response == nil
+        @response.body.include? str
+    end
+        
+
+    
+    # === 功能:
+    # 使用正则表达式的扫描，取出满足的集合
+    # 
+    # === 参数解释:
+    # - regexp    扫描的正则表达式
+    # - return    一个二维数组，第一维是匹配上的串，第二维是正则表达式里面的group情况
+    #
+    # === Example:
+    # Example #1: 非block
+    #
+    #    get "http://hi.baidu.com/腚腚熊/album"
+    #    p find_scan /^imgarr\[len\]=\{purl:"(.*)",psrc:"(.*)",pname:"(.*)",pnum:"(.*)"/
+    #    # => [["/%EB%EB%EB%EB%D0%DC/album/%C4%AC%C8%CF%CF%E0%B2%E1", "http://hiphotos.baidu.com/%EB%EB%EB%EB%D0%DC/abpic/item/8ddde41f55e355daa78669b4.jpg", "默认相册", "32"], ["/%EB%EB%EB%EB%D0%DC/album/%CE%D2%B5%C4%D1%D0%BE%BF%C9%FA%CA%B1%B4%FA", "http://hiphotos.baidu.com/%EB%EB%EB%EB%D0%DC/abpic/item/b7c672197f61435542a9ad2f.jpg", "我的研究生时代", "14"], ["/%EB%EB%EB%EB%D0%DC/album/3", "http://hiphotos.baidu.com/%EB%EB%EB%EB%D0%DC/abpic/item/0ded0b80781ddbcb9123d9b2.jpg", "3", "1"], ["/%EB%EB%EB%EB%D0%DC/album/4", "http://hiphotos.baidu.com/%EB%EB%EB%EB%D0%DC/abpic/item/e62608ea188ae1cfd539c991.jpg", "4", "1"], ["/%EB%EB%EB%EB%D0%DC/album/Kongde", "http://hiphotos.baidu.com/%EB%EB%EB%EB%D0%DC/abpic/item/8c3e0aef76094b36acafd5f0.jpg", "Kongde", "0"]]
+    #
+    # Example #2: block方式
+    #
+    #    get "http://hi.baidu.com/腚腚熊/album"
+    #    find_scan /^imgarr\[len\]=\{purl:"(.*)",psrc:"(.*)",pname:"(.*)",pnum:"(.*)"/ do |url,src,name,num|
+    #        p "url:#{url} src:#{src} name:#{name} num:#{num}"
+    #    end
+    #    # => url:/%EB%EB%EB%EB%D0%DC/album/%C4%AC%C8%CF%CF%E0%B2%E1 src:http://hiphotos.baidu.com/%EB%EB%EB%EB%D0%DC/abpic/item/8ddde41f55e355daa78669b4.jpg name:默认相册 num:32
+    #    # => ...
+    def find_scan(regexp,&blk)
+        raise "no any http request before!" if @response == nil
+        return @response.body.scan regexp,&blk
+    end
+    
+    # === 功能:
+    # 判断元素是否存在
+    # 
+    # === 参数解释:
+    # - xpath       定位，可以使xpath or selector
+    # - return      返回boolean
+    #
+    # === Example:
+    #    get "http://www.baidu.com/"
+    #    p find_exist? "title"    # => true
+    def find_exist?(xpath)
+        raise "no any http request before!" if @response == nil
+        @hpricot ||= Hpricot(@response.body)
+        elems = @hpricot.search(xpath)
+        if block_given? 
+            index = -1
+            elems = elems.select do |elem|
+                index = index + 1
+                yield elem,index
+            end
+        end
+        return !elems.empty?
+    end
+
+    # === 功能:
+    # 使用XPath或CSS Selector来提取html片段
+    # 
+    # === 参数解释:
+    # - xpath    定位，可以使xpath or selector
+    # - return    返回String
+    #
+    # === Example:
+    # Example #1:
+    #
+    #    get "http://www.baidu.com/"
+    #    # HTML片段如下
+    #    # <p id=km><a href=http://hi.baidu.com>空间</a>&nbsp;&nbsp;<a href=http://www.hao123.com>hao123</a>
+    #    # &nbsp;|&nbsp;<a href=/more/>更多<span style="font-family:宋体">>></span></a></p> 
+    #
+    #    puts find_html "p#km"
+    #    # <a href=http://hi.baidu.com>空间</a>&nbsp;&nbsp;<a href=http://www.hao123.com>hao123</a>
+    #    # &nbsp;|&nbsp;<a href=/more/>更多<span style="font-family:宋体">>></span></a>
+    #
+    #    puts find_html "//p[@id='km']"
+    #    # 与上同
+    def find_html(xpath)
+        raise "no any http request before!" if @response == nil
+        @hpricot ||= Hpricot(@response.body)
+        elems = @hpricot.search(xpath)
+        if block_given? 
+            index = -1
+            elems = elems.select do |elem|
+                index = index + 1
+                yield elem,index
+            end
+        end
+        raise HtmlError, "find_html xpath[#{xpath}] is not exist!" if elems.empty?
+        return elems.first.html.force_encoding "gbk"
+    end
+
+    # === 功能:
+    # 使用XPath或CSS Selector来提取text内容，不会包含HTML标签
+    # 
+    # === 参数解释:
+    # - xpath    定位，可以使xpath or selector
+    # - return    返回String
+    #
+    # === Example:
+    # Example #1:
+    #
+    #    get "http://www.baidu.com/"
+    #    # HTML片段如下
+    #    # <p id=km><a href=http://hi.baidu.com>空间</a>&nbsp;&nbsp;
+    #    # <a href=http://www.hao123.com>hao123</a>&nbsp;|&nbsp;<a href=/more/>更多<span style="font-family:宋体">
+    #    # >></span></a></p> 
+    #
+    #    puts find_text "p#km"
+    #    # => 空间&nbsp;&nbsp;hao123&nbsp;|&nbsp;更多>>
+    #
+    #    puts find_text "//p[@id='km']"
+    #    # 与上同
+    def find_text(xpath)
+        raise "no any http request before!" if @response == nil
+        @hpricot ||= Hpricot(@response.body)
+        elems = @hpricot.search(xpath)
+        if block_given? 
+            index = -1
+            elems = elems.select do |elem|
+                index = index + 1
+                yield elem,index
+            end
+        end
+        raise HtmlError, "find_html xpath[#{xpath}] is not exist!" if elems.empty?
+        return elems.first.inner_text
+    end
+
+    # === 功能:
+    # 使用XPath或CSS Selector来提取HTML元素的属性值
+    # 
+    # === 参数解释:
+    # - xpath    定位，可以使xpath or selector
+    # - attr    属性名
+    # - return    返回String
+    #
+    # === Example:
+    # Example #1:
+    #
+    #    get "http://www.baidu.com/"
+    #    # HTML片段如下
+    #    # <p id=km><a href=http://hi.baidu.com>空间</a>&nbsp;&nbsp;<a href=http://www.hao123.com>hao123</a>&nbsp;|&nbsp;<a href=/more/>更多<span style="font-family:宋体">>></span></a></p> 
+    #
+    #    puts find_attr "p#km a:first", "href"
+    #    # => http://hi.baidu.com
+    def find_attr(xpath, attr)
+        raise "no any http request before!" if @response == nil
+        @hpricot ||= Hpricot(@response.body)
+        elems = @hpricot.search(xpath)
+        if block_given? 
+            index = -1
+            elems = elems.select do |elem|
+                index = index + 1
+                yield elem,index
+            end
+        end
+        raise HtmlError, "find_attr xpath[#{xpath}] is not exist!" if elems.empty?
+        elem = elems.first
+        raise HtmlError, "find_attr xpath[#{xpath}] ok, but elem[#{elem}] has no attr[#{attr}]!" unless elem.has_attribute? attr
+        return elem.get_attribute(attr)
+    end
+
+    # === 功能:
+    # 此方法不推荐使用，请使用 find_attr
+    def find_raw_attr(xpath, attr)
+        raise "no any http request before!" if @response == nil
+        @hpricot ||= Hpricot(@response.body)
+        elems = @hpricot.search(xpath.downcase)
+        if block_given? 
+            index = -1
+            elems = elems.select do |elem|
+                index = index + 1
+                yield elem,index
+            end
+        end
+        raise HtmlError, "find_attr xpath[#{xpath}] is not exist!" if elems.empty?
+        elem = elems.first
+        raise HtmlError, "find_attr xpath[#{xpath}] ok, but elem[#{elem}] has no attr[#{attr}]!" unless elem.has_attribute? attr
+        return elem.raw_attributes[attr]
+    end
+
+
+    # === 功能:
+    # 使用XPath或CSS Selector来提取HTML元素集合的个数
+    # 
+    # === 参数解释:
+    # - xpath    定位，可以使xpath or selector
+    # - return    返回Fixnum
+    #
+    # === Example:
+    # Example #1:
+    #
+    #    get "http://www.baidu.com/"
+    #    # HTML片段如下
+    #    # <p id=km><a href=http://hi.baidu.com>空间</a>&nbsp;&nbsp;<a href=http://www.hao123.com>hao123</a>&nbsp;|&nbsp;<a href=/more/>更多<span style="font-family:宋体">>></span></a></p> 
+    #
+    #    puts find_size "p#km a"
+    #    # => 3
+    #    # 上层的意义可能是“输入框下面这一行显示3个超链接
+    def find_size(xpath)
+        raise "no any http request before!" if @response == nil
+        @hpricot ||= Hpricot(@response.body)
+        elems = @hpricot.search(xpath)
+        if block_given? 
+            index = -1
+            elems = elems.select do |elem|
+                index = index + 1
+                yield elem,index
+            end
+        end
+        return elems.size
+    end
+
+    # 使用XPath或CSS Selector来提取HTML元素集合的个数
+    # 
+    # === 参数解释:
+    # - xpath       父亲节点，可以使xpath or selector
+    # - excludes    排除的子节点名，长度可变参数
+    # - RETURN      返回非elems内的儿子节点集合
+    #
+    # === Example:
+    #      get "http://hi.baidu.com/sys/search?word=我&from=wise"
+    #      p find_children("entry > list > blog:first")
+    #          #=> ["title", "content", "spsspaceurlorilist", "url", "time", "spsspacenamelist"]
+    #      p find_children("entry > list > blog:first",%w{title content url time spsSpaceURLOriList})
+    #          #=> ["spsspacenamelist"]
+    def find_children(xpath,excludes=[])
+        raise "no any http request before!" if @response == nil
+        @hpricot ||= Hpricot(@response.body)
+        elems = @hpricot.search(xpath)
+        if block_given? 
+            index = -1
+            elems = elems.select do |elem|
+                index = index + 1
+                yield elem,index
+            end
+        end
+        raise HtmlError, "find_attr xpath[#{xpath}] is not exist!" if elems.empty?
+        elem = elems.first
+
+        rs = []
+        elem.children.each do |el|
+            if el.class == Hpricot::Elem and !excludes.include?(el.name)
+                rs << el.name
+            end
+        end
+        return rs
+    end
+
+
+    # === 功能:
+    # 高级接口，将 Hpricot 暴露出来，请参看 Hpricot 的使用手册
+    # 返回数组
+    # 
+    # === 参数解释:
+    # - xpath    定位，可以使xpath or selector
+    # - return    返回Hpricot::Elems
+    #
+    # === Example:
+    #    get "http://www.baidu.com"
+    #    hpricot_search("div").each |elem|
+    #        p elem
+    #    end
+    def hpricot_search(xpath, &blk)
+        raise "no any http request before!" if @response == nil
+        @hpricot ||= Hpricot(@response.body)
+        @hpricot.search(xpath, &blk)
+    end
+        
+
+end
+
+
+#
+class HtmlError < StandardError
+    # TODO 
+end
+
+
+if __FILE__ == $0
+    p __FILE__ 
+    p $0
+    class MyTest
+    include HtmlHelper
+
+        # 高级接口，将 Hpricot 暴露出来
+        def hpricot_search(xpath, &blk)
+            @hpricot ||= Hpricot("<tr><td>111</td><td id='q'>222</td><br/><td>333</td></tr>")
+            @hpricot.search(xpath, &blk)
+        end
+        def test
+            #post("http://passport.baidu.com/?login","username=hannyu5122&password=aaaa")
+            h = Hpricot("<tr><td>111</td><td id='q'>222</td><br/><td>333</td></tr>")
+            p h.search("tr td").text
+            p h.search("tr td#q").html
+            p h.search("tr td#q")[0]
+            p h.search("tr td#q")[0].next.html
+            p h.search("tr td")[0].get_attribute("id")
+            p h.search("tr td")[0].inner_text
+            p h.search("tr td#q")[0].html
+            p h.search("tt td")
+            p h.search("td").size
+            p h.search("tt td")[0].html
+        end
+
+    end
+    MyTest.new.test2
+end
+

data/lib/common/http/http_helper.rb

@@ -0,0 +1,469 @@
+#!/usr/bin/env ruby
+# coding: utf-8
+#
+# Author:: hanyu
+# Date:: 2009.5.7
+
+require 'net/http'
+require 'net/https'
+require 'cgi'
+require 'uri'
+require 'webrick'
+require "common/http/multipart"
+
+# 实现HTTP协议的支持，可以在action或case中调用
+#
+# - 封装 Net:HTTP 和 Net:HTTPS
+# - 模拟浏览器的Cookie策略，不完全实现，不支持Expire（没必要）
+# - 上下文环境的保持，request & response
+# - 编码转换（统一使用gbk）
+# - Referer、User-Agent等的自动填入
+# - 打印可读日志
+module HttpHelper
+
+    # === 功能:
+    # 发送POST请求
+    # 
+    # === 参数解释:
+    # - url        http请求的url，比如: http://hi.baidu.com/，前面的http和后面的/都不能省略
+    # - data    post的数据，Hash格式，键可以是字符串或符号，值可以是字符串或其他
+    # - header    http请求携带的header，比如referer、p3p等，形式与data类似
+    # - file    支持上传文件，可以传多个，Hash格式，键为参数名，值为文件的本地路径
+    #
+    # === Example:
+    # Example #1:
+    #    # 模拟登录
+    #    post "http://passport.baidu.com/?login", {
+    #        :username => 'xxx',
+    #        :password => 'yyy'
+    #    }
+    #
+    # Example #2:
+    #    # 自定义header
+    #    login user
+    #    post "http://hi.baidu.com/xxx/commit", {
+    #        :ct => 1,
+    #        :cm => 2,
+    #        ........
+    #    }, {
+    #        "Myhead" => "okoko"
+    #    }
+    #
+    # Example #3:
+    #    post "http://hi.baidu.com/upload",{
+    #        :xxx => 'yyy'
+    #    }
+    #    # 环境变量的注入
+    #    puts @request.method
+    #    puts @response.body
+    #    puts @response.header["Content-Length"]
+    #    puts @cookies
+    #
+    # Example #4:
+    #    # 上传文件
+    #    post "http://hi.baidu.com/upload",{
+    #        :xxx => 'yyy'
+    #    },{},{
+    #        "param_name" => "file_path"
+    #    }
+    #
+    def post(url,data={},header={},file={})
+        #解析成URI
+        $myparse = $myparse || URI::Parser.new(:RESERVED=>URI::REGEXP::PATTERN::RESERVED+"%")
+        uri = $myparse.parse $myparse.escape(url)
+        raise HttpError, "url[#{url}] parse fail! e.g. http://hi.baidu.com/" if uri.scheme.nil? or uri.host.nil? or uri.path.empty?
+
+        # CGI:URI.path 需要加上query部分才能构成一个正常的请求
+        path = (uri.query==nil) ? uri.path : uri.path+"?"+uri.query
+
+        # 把header符号变成字符串
+        header.each do |k,v|
+            if k.class == Symbol
+                header[k.to_s] = v.to_s
+                header.delete k
+            end
+        end
+
+        # 初始化Post请求
+        req = Net::HTTP::Post.new(path, header)
+        req.set_content_type "application/x-www-form-urlencoded" unless req.content_type
+
+        # 发送请求
+        send_http req,uri,data,file
+
+    end
+
+    # === 功能:
+    # 发送GET请求
+    # 
+    # === 参数解释:
+    # - url        http请求的url，比如: http://hi.baidu.com/，前面的http和后面的/都不能省略
+    # - data    数据将被编码并附在url的后面，Hash格式，键可以是字符串或符号，值可以是字符串或其他
+    # - header    http请求携带的header，比如referer、p3p等，形式与data类似
+    # 
+    # === Example:
+    # Example #1:
+    #    # 注意url要写全，带上协议头"http"，以及路径，若是根路径末尾要加"/"
+    #    get "http://www.baidu.com/"
+    #
+    # Example #2:
+    #    # 支持https
+    #    get "https://passport.baidu.com/"
+    #
+    # Example #3:
+    #    get "http://www.baidu.com/"
+    #    # 环境变量的注入
+    #    puts @request.method
+    #    puts @response.body
+    #    puts @response.header["Content-Length"]
+    #    puts @cookies
+    #
+    # Example #4:
+    #    # 可自定义 HTTP Header
+    #    get "http://hi.baidu.com/",{
+    #        :q => "空间",
+    #        :t => "haha"
+    #    },{
+    #        "Cookie" => "BDUSS=xxx;",
+    #        "Referer" => "google.com"
+    #    }
+    #
+    # Example #5:
+    #    # 带参数
+    #    get "https://passport.baidu.com/",{
+    #        :name => "walle",
+    #        :password => "1111"
+    #    }
+    #
+    def get(url,data={},header={})
+        #解析成URI
+        $myparse = $myparse || URI::Parser.new(:RESERVED=>URI::REGEXP::PATTERN::RESERVED+"%")
+        uri = $myparse.parse $myparse.escape(url)
+        raise HttpError, "url[#{url}] parse fail! e.g. http://hi.baidu.com/" if uri.scheme.nil? or uri.host.nil? or uri.path.empty?
+        
+        # CGI:URI.path 需要加上query部分才能构成一个正常的请求
+        path = (uri.query==nil) ? uri.path : uri.path+"?"+uri.query
+
+        # 将data转成url中的参数
+        path << "?" if data.size>0 && !path.include?("?")
+        data.each do |k,v|
+            path << "&" if !path.end_with?("&") && !path.end_with?("?")
+            path << k.to_s + "=" + CGI::escape(v.to_s)
+        end
+
+        # 把header符号变成字符串
+        header.each do |k,v|
+            if k.class == Symbol
+                header[k.to_s] = v.to_s
+                header.delete k
+            end
+        end
+
+        # 初始化Get请求
+        req = Net::HTTP::Get.new(path, header)
+
+        # 发送请求
+        send_http req,uri
+
+    end
+
+    def download(uri, path=nil)
+        begin
+            if path
+                file_name = path + uri.split('/').last
+                file_name.sub(/\/\//, '/')
+            else
+                file_name = uri.split('/').last
+            end
+
+            file_content = get uri
+            #puts file_name
+            File.open(file_name, "w+") do |file|
+                file.binmode # must be in binary mode
+                file.write file_content
+                file.rewind
+            end
+        rescue => err
+            puts err
+            return ''
+        end
+    end
+
+    # === 功能:
+    # 发送PUT请求
+    # 未实现，有需求吗？
+    # 
+    # === 参数解释:
+    # 
+    # === Example:
+    #
+    def put(url,data="",header=nil,file=nil)
+        #TODO
+    end
+
+    # === 功能:
+    # 发送DELETE请求
+    # 未实现，有需求吗？
+    # 
+    # === 参数解释:
+    # 
+    # === Example:
+    #
+    def delete(url,data=nil,header=nil)
+        #TODO
+    end
+
+
+    # === 功能:
+    # 清空当前cookie
+    # 
+    # === 参数解释:
+    #  
+    # === Example:
+    #    get "http://www.baidu.com/"
+    #    clear_cookie        # => 清空cookie
+    #    get "http://hi.baidu.com/"
+    def clear_cookie
+        @cookies = {}
+    end
+
+    # === 功能:
+    # 指定不要进行转码
+    #
+    # === 参数解释:
+    #  
+    # === Example:
+    # Example #1: block方式
+    #
+    #    dont_encode do
+    #        get "..."
+    #        post "..."
+    #    end
+    #    # 凡是块里面的请求都不会有转码，但是块外面的代码仍然可以
+    #
+    # Example #2: 非block方式
+    #
+    #    get "http://www.baidu.com/"
+    #    # 返回页面经过转码，转成GBK
+    #    dont_encode
+    #    get "http://www.baidu.com/"
+    #    # 返回页面在任何情况下都不会做任何转码
+    #    # ...
+    #    do_encode
+    #    # 恢复转码方式
+    #    get "http://www.baidu.com/"
+    #    # 返回页面经过转码，转成GBK
+    def dont_encode 
+        if block_given? 
+            @dont_encode = true
+            yield
+            @dont_encode = false
+        else
+            @dont_encode = true
+        end
+    end
+
+    # === 功能:
+    # 参见dont_encode
+    def do_encode
+        if block_given? 
+            @dont_encode = false
+            yield
+            @dont_encode = true
+        else
+            @dont_encode = false
+        end
+    end
+
+    # === 功能:
+    # 打印当前HTTP请求返回的content，可用于调试
+    #
+    # === 参数解释:
+    # 无
+    #  
+    # === Example:
+    #       get "http://www.baidu.com/"
+    #       print_body
+    #       #=> <html><head><meta http-equiv=Content-Type content="text/html;charset=gb2312"><title>百度一下，你就知道      </title>.....(省略)
+    #
+    def print_body
+        puts @response.body unless @response.nil?
+    end
+
+    # === 功能:
+    # 打印当前HTTP请求返回的所有Header，可用于调试
+    #
+    # === 参数解释:
+    # 无
+    #  
+    # === Example:
+    #       get "http://www.baidu.com/"
+    #       print_head
+    #       #=> {"date"=>["Sun, 27 Dec 2009 12:04:29 GMT"], "server"=>["BWS/1.0"], "content-length"=>["3657"], "content-type"=>["text/html;charset=gb2312"], "cache-control"=>["private"], "expires"=>["Sun, 27 Dec 2009 12:04:29 GMT"], "set-cookie"=>["BAIDUID=95C50755A927A287A68E2014CEF4CF1E:FG=1; expires=Sun, 27-Dec-39 12:04:29 GMT; path=/; domain=.baidu.com"], "p3p"=>["CP=\" OTI DSP COR IVA OUR IND COM \""]}
+    #
+    def print_head
+        puts @response.to_hash unless @response.nil?
+    end
+    alias print_header print_head
+
+    # 以下方法为私有
+    private
+    
+    # 封装发送HTTP细节
+    def send_http(req,uri,data=nil,file={})
+        #注入上下文环境中的cookie
+        @cookies = {} if @cookies.nil?
+        @cookies.each do |key,coos|
+            if uri.host.end_with? key
+                coos.each do |coo|
+                    if req.key? "cookie"
+                        req["cookie"] += "#{coo.name}=#{coo.value};"
+                    else
+                        req["cookie"] = "#{coo.name}=#{coo.value};"
+                    end
+                end
+            end
+        end
+
+        # 填入默认header
+        # req["User-Agent"] = "#{ConfigHelper::CONF['user_agent']}" if req["User-Agent"] == "Ruby"
+        # req["Referer"] = "#{@referer}" unless req.key? "referer"
+
+        # 如果上传文件，需要Multipart的支持
+        if file.size > 0
+            file.each do |k,v|
+                file[k] = File.new v
+            end
+            data_str, headers = Multipart::Post.prepare_query data.merge(file)
+            req["Content-Type"] = headers["Content-Type"]
+        else
+            #转义data
+            if data.nil?
+                data_str = nil
+            elsif data.kind_of? Hash
+                data_str = data.inject("") do |mem, (key, value)|
+                    mem << "&" if mem.size > 0
+                    mem << key.to_s + "=" + CGI::escape(value.to_s)
+                end
+            else
+                raise "data must be Hash"
+            end
+            
+        end
+        
+        #如果是HTTPS 安全连接
+        conn = Net::HTTP.new(uri.host, uri.port)
+        if uri.scheme == "https"
+            conn.use_ssl = true
+            conn.verify_mode = OpenSSL::SSL::VERIFY_NONE
+        end
+
+        take = 0
+        #加入异常捕获机器
+        begin
+            ts1 = Time.now
+            #发送HTTP请求
+            res = conn.start do |http|
+                #puts "data_str.size: #{data_str.size}" unless data_str.nil?
+                http.request(req,data_str)
+            end
+            ts2 = Time.now
+            take = ts2 - ts1
+        rescue SocketError => err
+            #域名不能解析，机器下线？
+            $log.error "#{req.method} url[#{uri}] data[#{data}] file[#{file}] time[#{take}s] FAIL! Host[#{uri.host}] is unknown! Maybe it is out of the DNS."
+            raise HttpError, "host[#{uri.host}] is unknown! Maybe it is out of the DNS."
+        rescue EOFError => err
+            #连接被中断，被transmit拒？
+            $log.error "#{req.method} url[#{uri}] data[#{data}] file[#{file}] time[#{take}s] FAIL! Connection with uri[#{uri}] is interrupted! Maybe transmit did this."
+            raise HttpError, "Connection with uri[#{uri}] is interrupted! Maybe transmit did this."
+        rescue SystemCallError => err
+            #服务被拒绝，端口未开放？
+            $log.error "#{req.method} url[#{uri}] data[#{data}] file[#{file}] time[#{take}s] FAIL! Connection with uri[#{uri}] is refused! Maybe the port[#{uri.port}] is not open."
+            raise HttpError, "Connection with uri[#{uri}] is refused! Maybe the port[#{uri.port}] is not open."
+        end
+        #将返回的cookie存入上下文环境中
+        if res.key?('set-cookie')
+            res.get_fields('set-cookie').each do |str|
+                cookie = WEBrick::Cookie.parse_set_cookie(str)
+                cookie.domain = uri.host if cookie.domain == nil
+                if @cookies.key?(cookie.domain)
+                    @cookies[cookie.domain].delete_if do |coo|
+                        coo.name == cookie.name
+                    end
+                    @cookies[cookie.domain] << cookie if cookie.value != "deleted"
+                else
+                    @cookies[cookie.domain] = [cookie]
+                end
+            end
+        end
+
+        #置上下文变量
+        @referer = uri.to_s
+        @request = req
+        @response = res
+        @hpricot = nil
+
+        # #编码转换
+        # unless res.body.nil? or
+        #     @dont_encode or
+        #     res["Content-Type"].nil? or                 # 如果没有 content-type 字段，那么不做转换
+        #     res["Content-Type"].include? "image" or     # 图片不转
+        #     res["Content-Type"].include? "video" or     # 视频不转
+        #     res["Content-Type"].include? "audio"        # 声音不转
+        #     ct = res["Content-Type"].scan(/charset=(.*?);?$/)[0]
+        #     encod = ct[0].downcase unless ct.nil?
+        #     if ! encod.nil?
+        #         if encod=="gbk" or encod=="gb2312"
+        #             res.body = res.body.force_encoding("gbk")
+        #         else
+        #             res.body = res.body.force_encoding(encod).encode "gbk"
+        #         end
+        #     else
+        #         if ConfigHelper::CONF["default_encoding"]=="gbk"
+        #             res.body = res.body.force_encoding("gbk")
+        #         else
+        #             res.body = res.body.force_encoding(ConfigHelper::CONF["default_encoding"]).encode "gbk"
+        #         end
+        #     end
+        # end
+        res.body
+    end
+
+end
+
+class HttpError < StandardError
+    # do nothing
+end
+
+
+
+if __FILE__ == $0
+    $LIB_ROOT = "#{File.dirname(__FILE__)}/.."
+    $TEST_ROOT = "#{$LIB_ROOT}/.."
+    $: << "#{$LIB_ROOT}"
+    require 'common/log_helper'
+    require 'common/config_helper'
+
+    class MyTest
+    include HttpHelper
+
+        def test1
+            @cookies = {}
+            #post("http://passport.baidu.com/?login","username=hannyu5122&password=aaaa")
+            get "http://www.baidu.com/"
+            puts @response.body
+            get "http://sz-testing-space07.sz01.baidu.com:5120/haoba?oko"
+        end
+        def test2
+            @cookies = {}
+            assert_raise HttpError do
+                get "http://db-testing-space13.db01.baidu.com:4444/developers"
+                #get "http://db-testing-space13.db01.baidu.com:8080/developers"
+                get "http://db-testing-sp.db01.baidu.com:8080/developers"
+            end
+        end
+    end
+    MyTest.new.test2
+end
+