How to modify the shapefile attribute in geotools

In geoserver ChineseCommunityThe discussion address is: http://opengeo.cn/bbs/read.php? Tid = 1701 & page = E & #

After you use geotools to modify the shapefile, you can find that all attribute names in the DBF file are garbled, but the attribute values are not garbled. No garbled characters have been found before the modification.
In additionCodeThe encoding method has been set as follows:

 
Shapefiledatastore shape =NewShapefiledatastore (URL );
Shape. setstringcharset (charset. forname ("GBK "));
 

The modified code is as follows:

     /**  
* Modify the shapefile.
*  @ Param  Datastore
*  @ Param  Featureid of the data to be modified by fidstr
*  @ Return 
  */ 
 Public   Static   Boolean Updateshapefile (shapefiledatastore datastore, string fidstr)
{
Defaulttransaction transaction = Null ;
Simplefeaturestore store = Null ;
 Try {
Datastore. setstringcharset (charset. forname ("GBK "));
String [] featurenames = datastore. gettypenames ();
String featurename = featurenames [0];
 //  Create default transaction object  
 Transaction = New Defaulttransaction ();

 //  It also indicates the element name used by the data source. Generally, the shapefile name and the shapefile type name are the same.  
 Store = (simplefeaturestore) datastore. getfeaturesource (featurename );
 //  Associate default transactions with data sources  
 Store. settransaction (transaction );
//  Create Filter  
 Filterfactory2 FF = (filterfactory2) commonfactoryfinder. getfilterfactory2 ( Null );
Set <featureid> fids = New Hashset <featureid> ();
Featureid FID = ff. featureid (fidstr );
Fids. Add (FID );
Filter filter = (filter) ff. ID (fids );
 //  Modify the filtered feature according to the filter.  
 Store. modifyfeatures ("data table number", "test", filter );

 //  Submit  
 Transaction. Commit ();
} Catch (Ioexception e ){
E. printstacktrace ();
 Try {
 //  Rollback  
 Transaction. rollback ();
} Catch (Ioexception E1 ){
E1.printstacktrace ();
}
 Return   False ;
} Finally {
 If (Transaction! = Null ){
 //  Close  
 Transaction. Close ();
Transaction = Null ;
}
}
Return   True ;
}

Transaction. Commit ();

In addition, the content of the SHP file also contains Chinese characters, but there is no garbled problem, so start with the DBF file header.

ViewSource codeThe operations on DBF Files are found in the org \ geotools \ data \ shapefile \ DBF package. dbasefilewriter class implements write operations on DBF Files, including the property name (head) and write the attribute value (body), while the head is written in

 
PublicDbasefilewriter (dbasefileheader header, writablebytechannel out, charset)
ThrowsIoexception {
// Write the header
Header. writeheader (out );
......
}
 

The writeheader method is defined in the dbasefileheader class.

The original code is as follows:

//Write the field name
For(IntJ = 0; j <11; j ++ ){
If(Fields. fieldname. Length ()> J ){
Buffer. Put ((Byte) Fields. fieldname. charat (j ));
}Else{
Buffer. Put ((Byte) 0 );
}
}
 

This means that the attribute name is put into buffer in the form of byte. The number of bytes occupied by an attribute name cannot exceed 11. If the number is less than 11, the attribute name is supplemented with (byte) 0.
The key lies in (byte) fields. fieldname. charat (j). For Chinese characters, it occupies two bytes. Here, the conversion from Char to byte excludes the problem.
Take the following string as an example:
String STR = "asset ";
Byte [] bytes = Str. getbytes ();
For (INT I = 0; I <bytes. length; I ++)
System. Out. println (bytes );
The output result should be:
-41
-54
It occupies two bytes, which is correct.
If you follow the method in it:
Char chars = 'created ';
Byte [] bytes = {(byte) chars };
For (INT I = 0; I <bytes. length; I ++)
System. Out. println (bytes );
The output result is:
68
This is obviously incorrect.

According to the above logic, modify the original code as follows (I am only a method here, this code should be optimized again ):
Int J = 0;
For (INT counter = 0; j <11 & Counter <fields. fieldname. Length (); counter ++ ){
Char CHA = Fields. fieldname. charat (Counter );
String STR = new string (CHA + "");
Byte [] bytes = Str. getbytes (); // changeByte [] bytes = Str. getbytes ("GBK"); otherwise, Chinese characters cannot be recognized.
For (int K = 0; k <bytes. length; k ++ ){
Buffer. Put (Bytes [k]);
J ++;
}
}
If (J! = 11 ){
For (int K = 0; k <(11-j); k ++ ){
Buffer. Put (byte) 0 );
}
}

 

 

Reprinted from:

Http://opengeo.cn/bbs/read.php? Tid = 1701 & page = E & #

Http://www.cnblogs.com/lingxue3769/archive/2011/11/17/2252659.html