Use htmlparser to parse HTML (C)

To parse HTML, extract HTML data, or modify HTML data, htmlparser is a good choice.

Htmlparser can be used to parse HTML data on the local and network:

Parser = new Parser (New winista. Text. htmlparser. http. httpprotocol (New uri ("uristring ")));

Parser = new Parser (New winista. Text. htmlparser. Lex. lexer ("htmlstring "));

System. Io. Stream stream = new system. Io. filestream ("filepath", system. Io. filemode. Open );

Parser = new Parser (New winista. text. htmlparser. lex. lexer (New winista. text. htmlparser. lex. page (stream, "charset"); you can also analyze the data of certain nodes and use nodeclassfilter to specify the node type to be analyzed:

Nodefilter filter = new nodeclassfilter (typeof (winista. Text. htmlparser. tags. Div); Use the parse method of the parser instance to obtain the node Array

Nodelist = parser. parse (null );

Nodelist = parser. parse (filter); the following section of HTML is analyzed:

<Div class = "divcss" id = "div_1">

<Div name = "Div" class = "divcss" id = "div_2"> div_2 </div>

<Table name = "table" id = "table_1">

<Tr>

<TD> htmlparser </TD>

<TD> <Div id = "div_3"> <font color = "red"> htmlparser </font> </div> </TD>

</Tr>

</Table>

</Div>

Txtresult is used to display the data processed by analysis, and txtsource is the text box for reading HTML data.

// Record the start position of each node to avoid repeated Processing

Ilist <int> Start = new list <int> ();

Protected void button#click (Object sender, eventargs E)

{

This.txt result. Text = string. empty;

Lexer = new lexer (this.txt source. Text );

Parser = new Parser (lexer );

Nodefilter filter = new nodeclassfilter (typeof (winista. Text. htmlparser. tags. Div ));

Nodelist = parser. parse (null );

If (nodelist. Count = 0)

Txtresult. Text = "No compliant node ";

Else

{

For (INT I = 0; I <nodelist. Count; I ++)

{

Paserdata (nodelist [I]);

}

}

}

Private ITAG gettag (inode node)

{

If (node = NULL)

Return NULL;

Return node is ITAG? Node as ITAG: NULL;

}

Private void paserdata (inode node)

{

ITAG tag = gettag (node );

If (tag! = NULL &&! Tag. isendtag ()&&! Start. Contains (tag. startposition ))

{

Object OID = tag. getattribute ("ID ");

Object oname = tag. getattribute ("name ");

Object oclass = tag. getattribute ("class ");

This.txt result. text + = tag. tagname + ": \ r \ NID:" + OID + "name:" + oname + "class:" + oclass + "startposition:" + tag. startposition. tostring () + "\ r \ n ";

Start. Add (tag. startposition );

}

// Subnode

If (node. Children! = NULL & node. Children. Count> 0)

{

Paserdata (node. firstchild );

}

// Sibling Node

Inode siblingnode = node. nextsibling;

While (siblingnode! = NULL)

{

Paserdata (siblingnode );

Siblingnode = siblingnode. nextsibling;

}

} The data displayed by txtresult is:

Div:

ID: div_1 name: Class: divcss startposition: 0

Div:

ID: div_2 name: div class: divcss startposition: 34

Table:

ID: table_1 name: Table class: startposition: 90

TR:

ID: Name: Class: startposition: 127

TD:

ID: Name: Class: startposition: 136

TD:

ID: Name: Class: startposition: 160

Div:

ID: div_3 name: Class: startposition: 164

Font:

ID: Name: Class: startposition: 180

Htmlparser analyzes the specified data and makes some modifications to the data to be analyzed: To specify attributes without the name and class attributes:

Object OID = tag. getattribute ("ID ");

Object oname = tag. getattribute ("name ");

Object oclass = tag. getattribute ("class ");

If (oname = NULL)

{

Oname = "name ";

Tag. setattribute ("name", oname. tostring ());

}

If (oclass = NULL)

{

Oclass = "class ";

Tag. setattribute ("name", oclass. tostring ());

}

This.txt result. Text + = tag. tagname + ": \ r \ NID:" + OID + "name:" + oname

+ "Class:" + oclass + "startposition:" + tag. startposition. tostring () + "\ r \ n ";

Start. Add (tag. startposition); the data displayed by txtresult is:

Div:

ID: div_1 name: Name class: divcss startposition: 0

Div:

ID: div_2 name: div class: divcss startposition: 34

Table:

ID: table_1 name: Table class: Class startposition: 90

TR:

ID: Name class: Class startposition: 127

TD:

ID: Name class: Class startposition: 136

TD:

ID: Name class: Class startposition: 160

Div:

ID: div_3 name: Name class: Class startpoint: 164

Font:

ID: Name class: Class startposition: 180

Htmlparser achieves our goal. Now we are adding a subnode to the node whose node is Div and whose ID is div_3:

Object OID = tag. getattribute ("ID ");

Object oname = tag. getattribute ("name ");

Object oclass = tag. getattribute ("class ");

If (tag. tagname = "Div" & tag. getattribute ("ID") = "div_3 ")

{

Inode newnode = new textnode ("Add a new node ");

Tag. Children. Add (newnode );

}

This.txt result. Text + = tag. tagname + ": \ r \ NID:" + OID + "name:" + oname

+ "Class:" + oclass + "startposition:" + tag. startposition. tostring () + "\ r \ n"; Output nodelist [0]. tohtml ():

<Div class = "divcss" id = "div_1">

<Div name = "Div" class = "divcss" id = "div_2"> div_2 </div>

<Table name = "table" id = "table_1">

<Tr>

<TD> htmlparser </TD>

<TD> <Div id = "div_3"> <font color = "red"> htmlparser </font> Add a new node </div> </TD>

</Tr>

</Table>

</Div> the DIV node with ID div_3 is followed by the data to be added.