I am calling a shared .dll function using JNA Java. From the documentation, the function can be invoked to receive parameters using Visual C++ as below;
PMSifEncodeKcdLcl(PCHAR ff, PCHAR Dta, BOOL Dbg, PCHAR szOpId, PCHAR szOpFirst, PCHAR szOpLast);
From the doc:
ff - A single ASCII character.
Dta - Points to a null-terminated string.
Dbg - a boolean flag
szOpId - points to a null-terminated string
szOpFirst - points to a null-terminated string
szOpLast - points to a null-terminated string
The string is built from a number of Data Fields. The format for each Data Field within the string is as follows:
RS FI data
RS = Record Separator.
Indicates the start of the Data Field. A single ASCII Record Separator [RS] character (hex 1E)
FI = Field Identifier - Indicates the type of data in the field. A single ASCII character.
data = the actual data. A number of ASCII characters, dependent on the Field Identifier. Sometimes the data is variable in length. The Record Separator of the following field indicates the end of a Data Field (or for the last field, the NULL character at the end of the string).
An Answer Code is returned in field ff. Answer Data (if any) is returned in field Dta
I have cross checked the JNA documentation to confirm field mappings but still no success. After trying for days. I came up with the code below;
My Java Code:
/* JNA interface class
*/
public class JNALocksInterface {
public interface LockLibrary extends StdCallLibrary {
LockLibrary INSTANCE = (LockLibrary) Native.loadLibrary("path_to_dll", LockLibrary.class);
public void PMSifEncodeKcdLcl(byte[] ff, byte[] dta, boolean debug, String szOpid, String szOpFirst, String szOpLast);
}
}
/*My Calling Class Code*/
JNALocksInterface.LockLibrary INSTANCE = JNALocksInterface.LockLibrary.INSTANCE;
String dta = "*R101*L101*TSingle Room*NMatu*FZachary*URegular Guest*D201805021347*O201805030111";
String ff = "A";
byte[] dataBytes = new byte[dta.length() + 1];
System.arraycopy(dta.getBytes("UTF-8"), 0, dataBytes, 0, dta.length());
dataBytes[dta.length()] = 0;
byte[] dtaByteArray = new byte[dta.length() + 1];
byte[] ffByteArray = ff.getBytes("UTF-8");
for (int i = 0; i < dataBytes.length; i++) {
String s1 = String.format("%8s", Integer.toBinaryString(dataBytes[i] & 0xFF)).replace(' ', '0');
// System.out.println(s1);
if((char)dataBytes[i] == '*')
{
dtaByteArray[i] = 30;
}
else{
int val = Integer.parseInt(s1, 2);
byte b = (byte) val;
dtaByteArray[i] = b;
}
}
byte[] commandCodeFinal = new byte[1];
for (int i = 0; i < ffByteArray.length; i++) {
String s2 = String.format("%8s", Integer.toBinaryString(ffByteArray[i] & 0xFF)).replace(' ', '0');
System.out.println(s2);
int val = Integer.parseInt(s2, 2);
byte b = (byte) val;
commandCodeFinal[i] = b;
}
String userNameBytes = "test";
String userFirstNameBytes = "test";
String userLastNameBytes = "test";
INSTANCE.PMSifEncodeKcdLcl(commandCodeFinal, dtaByteArray, false, userNameBytes, userFirstNameBytes, userLastNameBytes);
I am getting a wrong response on field ff and dta as shown below.
FF Response >> :
DTA Response >> 0101IR101L101TSingle RoomNMatuFZacharyURegular GuestD201805021347O2018050
I am replacing "*" with the ascii record separator.
Can someone show me how to correctly call the function using JNA? I've searched all over but still no success.
Solved IT. Used Unicode Field separator and used JNA Memory object and it Worked!
Was also using WIndows 10 64 bit. Changed to Windows 7 32 bit and it worked!!
Replaced the code with below snippet;
String fieldSeparator = "\u001e"
String dataTest = fieldSeparator+"R101"+fieldSeparator+"TSingle Room"+fieldSeparator+"FShujaa"+fieldSeparator+"NMatoke"
+ fieldSeparator+"URegular Guest"+fieldSeparator+"D201805040842"+fieldSeparator+"O201805051245";
String dataTestPadded = org.apache.commons.lang.StringUtils.rightPad(dataTest,30,'0');
System.out.println("Padded string >> " + dataTestPadded);
String data = dataTest;
//getPayloadToSend(payLoadSample) + (char)00;
String commandCode = "A";
Memory commandCodeMemory = new Memory(commandCode.length()+1);
commandCodeMemory.setString(0, commandCode);
Memory dataMemory = new Memory(data.length()+1);
dataMemory.setString(0, data);
//dataMemory.setString(1, "0");
System.out.println("Registerring >> " + INSTANCE.PMSifRegister("42860149", "BatchClient")) ;
INSTANCE.PMSifEncodeKcdLcl(commandCodeMemory, dataMemory, false, "ZKMATU", "zACHARY", "tESTING");
System.out.println("FF Response >> " + commandCodeMemory.getString(0));
System.out.println("DTA Response >> " + dataMemory.getString(0));
INSTANCE.PMSifUnregister();
Related
I am trying to recreate the following logic I created in JAVA to swift:
public String xorMessage(String message, String key) {
try {
if (message == null || key == null) return null;
char[] keys = key.toCharArray();
char[] mesg = message.toCharArray();
int ml = mesg.length;
int kl = keys.length;
char[] newmsg = new char[ml];
for (int i = 0; i < ml; i++) {
newmsg[i] = (char)(mesg[i] ^ keys[i % kl]);
}//for i
return new String(newmsg);
} catch (Exception e) {
return null;
}
I have reached till here while coding in swift3:
import UIKit
import Foundation
let t = "22-Jun-2017 12:30 pm"
let m = "message"
print(UInt8(t))
let a :[UInt8] = Array(t.utf8)
let v = m.characters.map{String ($0) }
print(v)
func encodeWithXorByte(key: UInt8 , Input : String) -> String {
return String(bytes: Input.utf8.map{$0 ^ key}, encoding: String.Encoding.utf8) ?? ""
}
var ml :Int = Int( m.characters.count )
var kl :Int = Int (t.characters.count)
var f = [String]()
for i in 0..<ml{
let key = a[i%kl]
let input = v[i]
f.append(String(bytes: input.utf8.map{$0 ^ key} , encoding : String.Encoding.utf8)!)
// f.append(<#T##newElement: Character##Character#>)
//m[i] = input.utf8.map{$0 ^ key}
}
I am trying to obtain a string(message) which has been xor'ed with a key passed into the above function. But my code in swift is not working as it is returning character array and I want a string, if I try to cast the character array to string it does not show the unicode like \u{0001} etc in the string...
Suppose I get following output :
["_", "W", "^", "9", "\u{14}", "\t", "H"]
and then I try to convert to string, I get this:
_W^9 H
I want :
_W^9\u{14}\tH
Please help.
There are different problems. First, if your intention is to print
"unprintable" characters in a string \u{} escaped then you can use
the .debugDescription method. Example:
let s = "a\u{04}\u{08}b"
print(s) // ab
print(s.debugDescription) // "a\u{04}\u{08}b"
Next, your Swift code converts the string to UTF-8, xor's the bytes
and then converts the result back to a String. That can easily fail
if the xor'ed byte sequence is not valid UTF-8.
The Java code operates on UTF-16 code units, so the equivalent Swift
code would be
func xorMessage(message: String, key: String) -> String {
let keyChars = Array(key.utf16)
let keyLen = keyChars.count
let newMsg = message.utf16.enumerated().map { $1 ^ keyChars[$0 % keyLen] }
return String(utf16CodeUnits: newMsg, count: newMsg.count)
}
Example:
let t = "22-Jun-2017 12:30 pm"
let m = "message"
let encrypted = xorMessage(message: m, key: t)
print(encrypted.debugDescription) // "_W^9\u{14}\tH"
Finally, even that can produce unexpected results unless you restrict
the input (key and message) to ASCII characters. Example:
let m = "😀"
print(Array(m.utf16).map { String($0, radix: 16)} ) // ["d83d", "de00"]
let t = "a€"
print(Array(t.utf16).map { String($0, radix: 16)} ) // ["61", "20ac"]
let e = xorMessage(message: m, key: t)
print(Array(e.utf16).map { String($0, radix: 16)} ) // ["fffd", "feac"]
let d = xorMessage(message: e, key: t)
print(Array(d.utf16).map { String($0, radix: 16)} ) // ["ff9c", "fffd"]
print(d) // ﾜ�
print(d == m) // false
The problem is that the xor'ing produces an invalid UTF-16 sequence
(an unbalanced surrogate pair), which is then replaced by the
"replacement character" U+FFFD.
I don't know how Java handles this, but Swift strings cannot invalid
Unicode scalar values, so the only solution would be to represent
the result as an [UInt16] array instead of a String.
I want to convert every character of a String to a new binary String. Here is what I do :
public static void main(String args[]) {
String MESSAGE = "%";
String binaryResult = "";
for (char c : MESSAGE.toCharArray()){
binaryResult += Integer.toBinaryString( (int) c);
}
System.err.println(binaryResult);
}
For exemple with the input : "%", I get the following output : "100101"
My problem is that the leading "0" is deleted ...
I want to have : "0100101". Does anyone have ideas?
What you're really saying is "How can I pad my binary string representation of a character to 7 digits"?
Replace this line:
binaryResult += Integer.toBinaryString( (int) c);
With these:
String binString = Integer.toBinaryString( (int) c );
binaryResult += ("0000000" + binString).substring(binString.length());
This presumes that you only have 7-bit characters... if you need more, then add 0's to the "00000" string to match the length of string (with padded 0s) you want.
I would suggest a couple of changes to your existing code. Since, you are concatenating to an string, inside a loop, this would cause the creation of a bunch of new string objects since they are immutable. The problem may be solved by use of a StringBuilder.
public static void main(String args[]) {
String MESSAGE = "%";
StringBuilder binaryResult = new StringBuilder();
for (char c : MESSAGE.toCharArray()) {
StringBuilder curValue = new StringBuilder(Integer.toBinaryString((int)c));
// calculate padding 0 bits to fill to 8 bits
int paddingLength = 8 - curValue.length();
char[] paddingArr = new char[paddingLength];
Arrays.fill(paddingArr, '0');
// insert padding bytes to the front
curValue.insert(0, paddingArr);
// add to stringbuilder for `MESSAGE`
binaryResult.append(curValue);
}
System.err.println(binaryResult.toString());
}
I'm trying to convert Java characters to JIS X 0208 "x-JIS0208" encoding (or any compatible, like EUC-JP, but not Shift-JIS), but I want unified (merged) codepoints to be handled correctly.
For example, 高 is assigned to row 25 column 66 in this JISX0208 chart, and a look-alike character 髙, while classified as an unassigned codepoint, is merged with the former. I quote from wikipedia: "both the form [ ] (高) and the less common form with a ladder-like construction (髙) are subsumed into the same code point".
I tried this in code the code below, and whatever encoding I try, I always get either an exception or the unassigned-character-placeholder ? (either ASCII or full-width).
Is there a way, perhaps a different endoding or an entirely different way of converting, so both these characters return the same codepoint? Alternatively, is there an API to find such characters so I can merge them before converting?
static Charset charset1 = Charset.forName("x-JIS0208");
static Charset charset2 = Charset.forName("EUC-JP");
static Charset[] charsets = {charset1, charset2};
static CharBuffer in = CharBuffer.allocate(1);
public static void main(String[] args) throws Exception
{
CharsetEncoder[] encoders = new CharsetEncoder[charsets.length];
for (int i = 0; i < charsets.length; i++)
encoders[i] = charsets[i].newEncoder();
char[] testChars = {' ', 'Ａ', '？', '亜', '唖', '蔭', '高', '髙'};
for (char ch : testChars)
{
System.out.print("'" + ch + "'\t(" + Integer.toHexString(ch) + ")\t=");
for (int i = 0; i < charsets.length; i++)
{
System.out.print("\t" + interpret(encode1(encoders[i], ch)));
System.out.print("\t" + interpret(encode2(charsets[i], ch)));
}
System.out.println();
}
}
private static String interpret(int i)
{
if (i == -1)
return "excepti";
if (i < 0x80)
return "'" + (char)i + "'";
return Integer.toHexString(i);
}
private static int encode1(CharsetEncoder encoder, char ch)
{
in.rewind();
in.put(ch);
in.rewind();
try
{
ByteBuffer out = encoder.encode(in);
if (out.limit() == 1)
return out.get(0) & 0xFF;
return out.get(1) & 0xFF | (out.get(0) & 0xFF) << 8;
}
catch (CharacterCodingException e)
{
return -1;
}
}
private static int encode2(Charset charset, char ch)
{
in.rewind();
in.put(ch);
in.rewind();
ByteBuffer out = charset.encode(in);
if (out.limit() == 1)
return out.get(0) & 0xFF;
return out.get(1) & 0xFF | (out.get(0) & 0xFF) << 8;
}
The output:
' ' (3000) = 2121 2121 a1a1 a1a1
'Ａ' (ff21) = 2341 2341 a3c1 a3c1
'？' (ff1f) = 2129 2129 a1a9 a1a9
'亜' (4e9c) = 3021 3021 b0a1 b0a1
'唖' (5516) = 3022 3022 b0a2 b0a2
'蔭' (852d) = 307e 307e b0fe b0fe
'高' (9ad8) = 3962 3962 b9e2 b9e2
'髙' (9ad9) = excepti 2129 excepti '?'
Note: I'm only interested in converting single characters, lots of them, not strings or streams, so I actually prefer a different method (if exists) that doesn't allocate a ByteBuffer every conversion.
髙 is not contained in JIS X 0208, but is containd in Microsoft Windows code page 932 (MS932). This is a variant of Shift JIS encoding, and is a superset of JIS X 0208 charset.
You should use the name "Windows-31j" for MS932, like:
Charset.forName("Windows-31j");
rather than Charset.forName("x-JIS0208");.
EDIT
The mapping table for some characters like 𨦇 and 鋏 (scissors) is distributed from the government of Japan, like National Tax Agency (see JIS縮退マップ（Ver.1.0.0）) .
But these mapping tables don't contain the character 髙. I think this is because 髙 is not contained in JIS X 0208 nor JIS X 0213.
So, I think you will have to replace 髙 with 高 manually (with String#replaceAll()), or make your own custom Charset with CharsetProvider.
I only knew that in the spec "ARIB STD-B24" (for ISDB-T 1seg in JP), this character is coding with DRCS pattern data, from DRCS-1 to DRCS-15, and each set consists of 94
characters.
I need to convert this java code in force.com apex. i tried to use Crypto class to get same encryption but not getting how can i get same value for the variable "fingerprintHash" in the last in APEX . Can Anyone help me in this technical issue?
Random generator = new Random();
sequence =Long.parseLong(sequence+""+generator.nextInt(1000));
timeStamp = System.currentTimeMillis() / 1000;
try {
SecretKey key = new SecretKeySpec(transactionKey.getBytes(), "HmacMD5");
Mac mac = Mac.getInstance("HmacMD5");
mac.init(key);
String inputstring = loginID + "^" + sequence + "^" + timeStamp + "^" + amount + "^";
byte[] result = mac.doFinal(inputstring.getBytes());
StringBuffer strbuf = new StringBuffer(result.length * 2);
for (int i = 0; i < result.length; i++) {
if (((int) result[i] & 0xff) < 0x10) {
strbuf.append("0");
}
strbuf.append(Long.toString((int) result[i] & 0xff, 16));
}
fingerprintHash = strbuf.toString(); //need this result for variable x_fp_hash
The apex code I was trying is :-
String API_Login_Id='6########';
String TXn_Key='6###############';
String amount='55';
sequence = '300';
long timeStamp = System.currentTimeMillis()/1000;
String inputStr = API_Login_Id + '^' + sequence + '^' + timeStamp + '^' + amount + '^';
String algorithmName = 'hmacMD5';
Blob mac = Crypto.generateMac(algorithmName,Blob.valueOf(inputStr),Blob.valueOf( TXn_Key));
String macUrl =EncodingUtil.urlEncode(EncodingUtil.base64Encode(mac), 'UTF-8');
The problem would seem to be that you are hex encoding the output on the javaside, but base64 encoding the output on the apex side, try using EncodingUtils.convertToHex instead of EncodingUtils.base64Encode
You look like you're heading along the right lines with regards to the encryption, however you're using a time stamp as part of your input string, and so unless you're astronomically lucky you're always encoding different strings. While you're working on porting the code, remove the timestamp so that you can be sure your input strings are the same - if they're not the same then you'll never get the same result.
Once you've established that your encryption is working as desired, then you can put the timestamp back into the code safe in the knowledge that it'll be functioning the same way as the original java code.
i've a particular js function that encrypts some form inputs into base64, but I need to run it in my Java app. So my question is, how can i call that function inside a java class? Otherwise I'll have to translate it but I think will be more complicated. Here's some of js code:
function encode64(input)
{
//alert(input);
//alert(input);
input = escape(input);
var output = "";
var chr1, chr2, chr3 = "";
var enc1, enc2, enc3, enc4 = "";
var i = 0;
var nMod =( input.length) % 3;
//alert(nMod);
//alert(input.length);
do
{
chr1 = input.charCodeAt(i++);
chr2 = input.charCodeAt(i++);
chr3 = input.charCodeAt(i++);
enc1 = chr1 >> 2;
enc2 = (((chr1 << 4) | (chr2 >> 4))& 0x3f);
enc3 = (((chr2 << 2) | (chr3 >> 6))& 0x3f);
enc4 = chr3 & 0x3f;
output = output +
keyStr.charAt(enc1) +
keyStr.charAt(enc2) +
keyStr.charAt(enc3) +
keyStr.charAt(enc4);
chr1 = chr2 = chr3 = "";
enc1 = enc2 = enc3 = enc4 = "";
} while (i < input.length);
if(nMod == 1)
{
chr1 = input.charCodeAt(i++);
enc1 = ((chr1 & 192)>>2);
enc2 = ((chr1 & 3) <<4);
enc3 = "=";
enc4 = "=";
output = output + keyStr.charAt(enc1)
+ keyStr.charAt(enc2)
+ keyStr.charAt(enc3)
+ keyStr.charAt(enc4);
}
if(nMod == 2)
{
chr1 = input.charCodeAt(i++);
chr2 = input.charCodeAt(i++);
enc1 = ((chr1 & 192)>>2);
enc2 = ((chr1 & 3) << 4 )|((chr2 & 0xf0) >> 4);
enc3 = ((chr2 & 15) <<2);
enc4 = "=";
output = output + keyStr.charAt(enc1)
+ keyStr.charAt(enc2)
+ keyStr.charAt(enc3)
+ keyStr.charAt(enc4);
}
return output;
}
Thanks so much!
but I think will be more complicated.
Why would it be more complicated? You can perfectly do that in Java. If your actual problem is already the first line
input = escape(input);
then it's good to know that the Java equivalent is the URLEncoder#encode(). As to the remnant of the coding, it's ultimately straightforward. Just replace var by String or char here and there, align the methods according java.lang.String API and you'll be fine.
Edit: for some downvoting nitpickers out here: I did NOT say that URLEncoder#encode() does the Base64 encoding. It just does URL encoding the same way as Javascript's escape() function does. That was just the first line of his to-be-translated Javascript code. Please read answers, do not scan answers.
Or better use commons-codec 's Base64 class
P.S. You do not "encrypt" into Base64 - you "encode"
You could do that using a javascript engine written in Java, but I think it's better to simply translate it into java.
I have used http://iharder.sourceforge.net/current/java/base64/ and it works.
Translate the code into Java - it will be quicker. (Or search for an existing example in Java!)