source: test-collections/trunk/filename-encodings/bin/script/DirList2.java@ 23830

import java.io.*;
import java.lang.*;
import java.net.*;

public class DirList 
{

    public static String raw_filename_to_url_encoded(String fileName) 
    {
    String urlEncoded = "";
    try {
        byte[] bytes = fileName.getBytes();
        
        for(int i = 0; i < bytes.length; i++) {
        // mask each byte (by applying & 0xFF) to make the signed
        // byte (in the range -128 to 127) unsigned (in the range
        // 0 to 255).

        int byteVal = (int)(bytes[i] & 0xFF);
        
        if(byteVal > 127) {
            urlEncoded += String.format("%%%02X", (int)byteVal);
        } else {
            urlEncoded += String.format("%c",(char)byteVal);
        }
        }
    }
    catch (Exception e) {
        e.printStackTrace();
    }

    return urlEncoded;
    }

    // For unicode codepoints see:
    // http://unicode.org/Public/MAPPINGS/ISO8859/8859-1.TXT for ISO8859-1 (Latin-1)
    // where 0xE2 maps to codepoint 0x00E2 and is defined as "Latin small letter a with circumflex"
    // http://unicode.org/Public/MAPPINGS/ISO8859/8859-7.TXT for ISO8859-7 (Greek)
    // where 0xE2 maps to codepoint 0x03B2 and is defined as "Greek small letter beta"
    
    public static String iso_8859_1_filename_to_url_encoded(String raw_bytes_filename) 
    {
    String urlEncoded = "";

    try {
        // By this point we have a UTF-8 encoded string that captures
        // what the ISO-8859-1 (Latin-1) character is that corresponded to the
        // 8-bit numeric value for that character in the filename
        // on the file system
        
        // For example:
        //   File system char:             <lower-case beta char in Latin-7> = %E2
        //   Equivalent Latin 1 char:      <lower-case a with circumflex>    = %E2
        //   Mapped to UTF-8:              <lower-case a with circumflex>    = <C3><A2>

        // Our task is to take the string the contains <C3><A2> and ensure that
        // we "see" it as <E2>

        byte [] raw_bytes = raw_bytes_filename.getBytes("ISO-8859-1");
        String unicode_filename = new String(raw_bytes,"UTF-8");
        
        for(int i = 0; i < unicode_filename.length(); i++) {
        char charVal = unicode_filename.charAt(i);
        if((int)charVal > 127) {
            urlEncoded += String.format("%%%02X", (int)charVal);
        } else {
            urlEncoded += String.format("%c",(char)charVal);
        }
        }
    }
    catch (Exception e) {
        e.printStackTrace();
    }

    return urlEncoded;
    }


    public static void main(String[] args) 
    {
    if(args.length < 1) {
        System.out.println("Usage: DirList <directory>");
        System.exit(-1);
    }

    
    File folder = new File(args[0]);
    File[] listOfFiles = folder.listFiles();
    
    for (int i = 0; i < listOfFiles.length; i++) {
        if (listOfFiles[i].isFile()) {
        File file = listOfFiles[i];
        URI  filename_uri = file.toURI();
        try {
            // The trick:
            //  1. toASCIIString() will %xx encode values > 127
            //  2. Decode the result to "ISO-8859-1" 
            //  3. URL encode the bytes to string

            // Step 2 forces the string to be 8-bit values.  It
            // doesn't matter if the starting raw filename was *not*
            // in the ISO-8859-1 encoding, the effect is to ensure
            // we have an 8-bit byte string that (numerically)
            // captures the right value.  These numerical values are
            // then used to determine how to URL encode it

            String filename_ascii = filename_uri.toASCIIString();
            String filename_raw_bytes = URLDecoder.decode(filename_ascii,"ISO-8859-1");
            String filename_url_encoded = iso_8859_1_filename_to_url_encoded(filename_raw_bytes);

            filename_url_encoded = filename_url_encoded.replaceAll(" ","%20");

            System.out.println("File " + filename_url_encoded);

            try {
            File test_file = new File(new URI(filename_url_encoded.replaceAll(" ","%20")));
            if (test_file.exists()) {
                System.out.println("   ... and I can see it!!!");
            }
            }
            catch (Exception e) {
            e.printStackTrace();
            }


        }
        catch (Exception e) {
            e.printStackTrace();
        }
        } else if (listOfFiles[i].isDirectory()) {
        File sub_folder = listOfFiles[i];
        System.out.println("Directory " + sub_folder.getName());
        } else {
        System.out.println("*** Not file or dir. Can't see file: " + listOfFiles[i]);
        }
    }
    }

}