package jsqlite;
/**
* String encoder/decoder for SQLite.
*
* This module was kindly donated by Eric van der Maarel of Nedap N.V.
*
* This encoder was implemented based on an original idea from an anonymous
* author in the source code of the SQLite distribution.
* I feel obliged to provide a quote from the original C-source code:
*
* "The author disclaims copyright to this source code. In place of
* a legal notice, here is a blessing:
*
* May you do good and not evil.
* May you find forgiveness for yourself and forgive others.
* May you share freely, never taking more than you give."
*
*/
public class StringEncoder {
/**
* Encodes the given byte array into a string that can be used by
* the SQLite database. The database cannot handle null (0x00) and
* the character '\'' (0x27). The encoding consists of escaping
* these characters with a reserved character (0x01). The escaping
* is applied after determining and applying a shift that minimizes
* the number of escapes required.
* With this encoding the data of original size n is increased to a
* maximum of 1+(n*257)/254.
* For sufficiently large n the overhead is thus less than 1.2%.
* @param a the byte array to be encoded. A null reference is handled as
* an empty array.
* @return the encoded bytes as a string. When an empty array is
* provided a string of length 1 is returned, the value of
* which is bogus.
* When decoded with this class' <code>decode</code> method
* a string of size 1 will return an empty byte array.
*/
public static String encode(byte[] a) {
// check input
if (a == null || a.length == 0) {
// bogus shift, no data
return "x";
}
// determine count
int[] cnt = new int[256];
for (int i = 0 ; i < a.length; i++) {
cnt[a[i] & 0xff]++;
}
// determine shift for minimum number of escapes
int shift = 1;
int nEscapes = a.length;
for (int i = 1; i < 256; i++) {
if (i == '\'') {
continue;
}
int sum = cnt[i] + cnt[(i + 1) & 0xff] + cnt[(i + '\'') & 0xff];
if (sum < nEscapes) {
nEscapes = sum;
shift = i;
if (nEscapes == 0) {
// cannot become smaller
break;
}
}
}
// construct encoded output
int outLen = a.length + nEscapes + 1;
StringBuffer out = new StringBuffer(outLen);
out.append((char)shift);
for (int i = 0; i < a.length; i++) {
// apply shift
char c = (char)((a[i] - shift)&0xff);
// insert escapes
if (c == 0) { // forbidden
out.append((char)1);
out.append((char)1);
} else if (c == 1) { // escape character
out.append((char)1);
out.append((char)2);
} else if (c == '\'') { // forbidden
out.append((char)1);
out.append((char)3);
} else {
out.append(c);
}
}
return out.toString();
}
/**
* Decodes the given string that is assumed to be a valid encoding
* of a byte array. Typically the given string is generated by
* this class' <code>encode</code> method.
* @param s the given string encoding.
* @return the byte array obtained from the decoding.
* @throws IllegalArgumentException when the string given is not
* a valid encoded string for this encoder.
*/
public static byte[] decode(String s) {
char[] a = s.toCharArray();
if (a.length > 2 && a[0] == 'X' &&
a[1] == '\'' && a[a.length-1] == '\'') {
// SQLite3 BLOB syntax
byte[] result = new byte[(a.length-3)/2];
for (int i = 2, k = 0; i < a.length - 1; i += 2, k++) {
byte tmp;
switch (a[i]) {
case '0': tmp = 0; break;
case '1': tmp = 1; break;
case '2': tmp = 2; break;
case '3': tmp = 3; break;
case '4': tmp = 4; break;
case '5': tmp = 5; break;
case '6': tmp = 6; break;
case '7': tmp = 7; break;
case '8': tmp = 8; break;
case '9': tmp = 9; break;
case 'A':
case 'a': tmp = 10; break;
case 'B':
case 'b': tmp = 11; break;
case 'C':
case 'c': tmp = 12; break;
case 'D':
case 'd': tmp = 13; break;
case 'E':
case 'e': tmp = 14; break;
case 'F':
case 'f': tmp = 15; break;
default: tmp = 0; break;
}
result[k] = (byte) (tmp << 4);
switch (a[i+1]) {
case '0': tmp = 0; break;
case '1': tmp = 1; break;
case '2': tmp = 2; break;
case '3': tmp = 3; break;
case '4': tmp = 4; break;
case '5': tmp = 5; break;
case '6': tmp = 6; break;
case '7': tmp = 7; break;
case '8': tmp = 8; break;
case '9': tmp = 9; break;
case 'A':
case 'a': tmp = 10; break;
case 'B':
case 'b': tmp = 11; break;
case 'C':
case 'c': tmp = 12; break;
case 'D':
case 'd': tmp = 13; break;
case 'E':
case 'e': tmp = 14; break;
case 'F':
case 'f': tmp = 15; break;
default: tmp = 0; break;
}
result[k] |= tmp;
}
return result;
}
// first element is the shift
byte[] result = new byte[a.length-1];
int i = 0;
int shift = s.charAt(i++);
int j = 0;
while (i < s.length()) {
int c;
if ((c = s.charAt(i++)) == 1) { // escape character found
if ((c = s.charAt(i++)) == 1) {
c = 0;
} else if (c == 2) {
c = 1;
} else if (c == 3) {
c = '\'';
} else {
throw new IllegalArgumentException(
"invalid string passed to decoder: " + j);
}
}
// do shift
result[j++] = (byte)((c + shift) & 0xff);
}
int outLen = j;
// provide array of correct length
if (result.length != outLen) {
result = byteCopy(result, 0, outLen, new byte[outLen]);
}
return result;
}
/**
* Copies count elements from source, starting at element with
* index offset, to the given target.
* @param source the source.
* @param offset the offset.
* @param count the number of elements to be copied.
* @param target the target to be returned.
* @return the target being copied to.
*/
private static byte[] byteCopy(byte[] source, int offset,
int count, byte[] target) {
for (int i = offset, j = 0; i < offset + count; i++, j++) {
target[j] = source[i];
}
return target;
}
static final char[] xdigits = {
'0', '1', '2', '3', '4', '5', '6', '7',
'8', '9', 'A', 'B', 'C', 'D', 'E', 'F'
};
/**
* Encodes the given byte array into SQLite3 blob notation, ie X'..'
* @param a the byte array to be encoded. A null reference is handled as
* an empty array.
* @return the encoded bytes as a string.
*/
public static String encodeX(byte[] a) {
// check input
if (a == null || a.length == 0) {
return "X''";
}
char[] out = new char[a.length * 2 + 3];
int i = 2;
for (int j = 0; j < a.length; j++) {
out[i++] = xdigits[(a[j] >> 4) & 0x0F];
out[i++] = xdigits[a[j] & 0x0F];
}
out[0] = 'X';
out[1] = '\'';
out[i] = '\'';
return new String(out);
}
}