正文提取就是去除掉html代码里的a8093152e673feb7aba1828c43532094的内容。这段代码增加了可选择保留某些a8093152e673feb7aba1828c43532094内容。
1 using system;
2 using system.text;
3 namespace htmlstrip
4 {
5 class mainclass
6 {
7 public static void main (string[] args)
8 {
9 string str = "<p>abc</p><span>efg</span><br /><script>888</script><!--<pa>ww</pa-->oo";
10 //system.io.streamreader rd=new system.io.streamreader ("/home/lx/test.html");
11 //str=rd.readtoend ();
12 htmlparser t = new htmlparser (str); //
13 t.keeptag (new string[] { "br" }); //设置br标签不过虑
14 console.write (t.text ());
15 }
16
17
18
19 }
20 class htmlparser
21 {
22 private string[] htmlcode; //把html转为数组形式用于分析
23 private stringbuilder result = new stringbuilder (); //输出的结果
24 private int seek; //分析文本时候的指针位置
25 private string[] keeptag; //用于保存要保留的尖括号内容
26 private bool _intag; //标记现在的指针是不是在尖括号内
27 private bool needcontent = true; //是否要提取正文
28 private string tagname; //当前尖括号的名字
29 private string[] specialtag = new string[] { "script", "style", "!--" }; //特殊的尖括号内容，一般这些标签的正文是不要的
30
31 /// <summary>
32 /// 当指针进入尖括号内，就会触发这个属性。这里主要逻辑是提取尖括号里的标签名字
33 /// </summary>
34 public bool intag {
35 get { return _intag; }
36 set {
37 _intag = value;
38 if (!value)
39 return;
40 bool ok = true;
41 tagname = "";
42 while (ok) {
43 string word = read ();
44 if (word != " " && word != ">") {
45 tagname += word;
46 } else if (word == " " && tagname.length > 0) {
47 ok = false;
48 } else if (word == ">") {
49 ok = false;
50 intag = false;
51 seek -= 1;
52 }
53 }
54 }
55 }
56 /// <summary>
57 /// 初始化类
58 /// </summary>
59 /// <param name="html">
60 /// 要分析的html代码
61 /// </param>
62 public htmlparser (string html)
63 {
64 htmlcode = new string[html.length];
65 for (int i = 0; i < html.length; i++) {
66 htmlcode[i] = html[i].tostring ();
67 }
68 keeptag (new string[] { });
69 }
70 /// <summary>
71 /// 设置要保存那些标签不要被过滤掉
72 /// </summary>
73 /// <param name="tags">
74 ///
75 /// </param>
76 public void keeptag (string[] tags)
77 {
78 keeptag = tags;
79 }
80
81 /// <summary>
82 ///
83 /// </summary>
84 /// <returns>
85 /// 输出处理后的文本
86 /// </returns>
87 public string text ()
88 {
89 int starttag = 0;
90 int endtag = 0;
91 while (seek < htmlcode.length) {
92 string word = read ();
93 if (word.tolower () == "<") {
94 starttag = seek;
95 intag = true;
96 } else if (word.tolower () == ">") {
97 endtag = seek;
98 intag = false;
99 if (iskeeptag (tagname.replace ("/", ""))) {
100 for (int i = starttag - 1; i < endtag; i++) {
101 result.append (htmlcode[i].tostring ());
102 }
103 } else if (tagname.startswith ("!--")) {
104 bool ok = true;
105 while (ok) {
106 if (read () == "-") {
107 if (read () == "-") {
108 if (read () == ">") {
109 ok = false;
110 } else {
111 seek -= 1;
112 }
113 }
114 }
115 }
116 } else {
117 foreach (string str in specialtag) {
118 if (tagname == str) {
119 needcontent = false;
120 break;
121 } else
122 needcontent = true;
123 }
124 }
125 } else if (!intag && needcontent) {
126 result.append (word);
127 }
128
129 }
130 return result.tostring ();
131 }
132 /// <summary>
133 /// 判断是否要保存这个标签
134 /// </summary>
135 /// <param name="tag">
136 /// a <see cref="system.string"/>
137 /// </param>
138 /// <returns>
139 /// a <see cref="system.boolean"/>
140 /// </returns>
141 private bool iskeeptag (string tag)
142 {
143 foreach (string ta in keeptag) {
144 if (tag.tolower () == ta.tolower ()) {
145 return true;
146 }
147 }
148 return false;
149 }
150 private string read ()
151 {
152 return htmlcode[seek++];
153 }
154
155 }
156 }
157
以上就是如何提取html的正文以及保留某些<>内容？的详细内容。