正文提取就是去除掉html代码里的的内容。这段代码增加了可选择保留某些内容。
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 = "abc
efg
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 ///
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 ///
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 ///
57 /// 初始化类
58 ///
59 ///
60 /// 要分析的html代码
61 ///
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 ///
71 /// 设置要保存那些标签不要被过滤掉
72 ///
73 ///
74 ///
75 ///
76 public void KeepTag (string[] tags)
77 {
78 keepTag = tags;
79 }
80
81 ///
82 ///
83 ///
84 ///
85 /// 输出处理后的文本
86 ///
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 }
124124 }
125 } else if (!inTag && needContent) {
126 result.Append (word);
127 }
128
129 }
130 return result.ToString ();
131 }
132 ///
133 /// 判断是否要保存这个标签
134 ///
135 ///
136 /// A
139 /// A
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
