2020-10-09 15:43:02 +00:00
|
|
|
package eu.faircode.email;
|
|
|
|
|
|
|
|
/*
|
|
|
|
This file is part of FairEmail.
|
|
|
|
|
|
|
|
FairEmail is free software: you can redistribute it and/or modify
|
|
|
|
it under the terms of the GNU General Public License as published by
|
|
|
|
the Free Software Foundation, either version 3 of the License, or
|
|
|
|
(at your option) any later version.
|
|
|
|
|
|
|
|
FairEmail is distributed in the hope that it will be useful,
|
|
|
|
but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
GNU General Public License for more details.
|
|
|
|
|
|
|
|
You should have received a copy of the GNU General Public License
|
|
|
|
along with FairEmail. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
|
|
|
|
Copyright 2018-2020 by Marcel Bokhorst (M66B)
|
|
|
|
*/
|
|
|
|
|
|
|
|
import java.nio.charset.Charset;
|
|
|
|
import java.nio.charset.StandardCharsets;
|
|
|
|
|
|
|
|
class CharsetHelper {
|
2020-10-15 06:40:20 +00:00
|
|
|
private static final int MAX_SAMPLE_SIZE = 8192;
|
2020-10-14 20:36:27 +00:00
|
|
|
|
2020-10-14 18:54:28 +00:00
|
|
|
static {
|
|
|
|
System.loadLibrary("compact_enc_det");
|
|
|
|
}
|
|
|
|
|
|
|
|
private static native String jni_detect(byte[] chars);
|
|
|
|
|
2020-10-09 15:43:02 +00:00
|
|
|
static boolean isUTF8(String text) {
|
|
|
|
// Get extended ASCII characters
|
|
|
|
byte[] octets = text.getBytes(StandardCharsets.ISO_8859_1);
|
|
|
|
|
|
|
|
int bytes;
|
|
|
|
for (int i = 0; i < octets.length; i++) {
|
|
|
|
if ((octets[i] & 0b10000000) == 0b00000000)
|
|
|
|
bytes = 1;
|
|
|
|
else if ((octets[i] & 0b11100000) == 0b11000000)
|
|
|
|
bytes = 2;
|
|
|
|
else if ((octets[i] & 0b11110000) == 0b11100000)
|
|
|
|
bytes = 3;
|
|
|
|
else if ((octets[i] & 0b11111000) == 0b11110000)
|
|
|
|
bytes = 4;
|
|
|
|
else if ((octets[i] & 0b11111100) == 0b11111000)
|
|
|
|
bytes = 5;
|
|
|
|
else if ((octets[i] & 0b11111110) == 0b11111100)
|
|
|
|
bytes = 6;
|
|
|
|
else
|
|
|
|
return false;
|
|
|
|
|
|
|
|
if (i + bytes > octets.length)
|
|
|
|
return false;
|
|
|
|
|
|
|
|
while (--bytes > 0)
|
|
|
|
if ((octets[++i] & 0b11000000) != 0b10000000)
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
|
|
|
return true;
|
|
|
|
}
|
|
|
|
|
2020-10-10 06:57:03 +00:00
|
|
|
static Charset detect(String text) {
|
|
|
|
try {
|
2020-10-10 11:20:24 +00:00
|
|
|
byte[] octets = text.getBytes(StandardCharsets.ISO_8859_1);
|
2020-10-14 20:36:27 +00:00
|
|
|
|
|
|
|
byte[] sample;
|
2020-10-15 06:40:20 +00:00
|
|
|
if (octets.length < MAX_SAMPLE_SIZE)
|
2020-10-14 20:36:27 +00:00
|
|
|
sample = octets;
|
|
|
|
else {
|
2020-10-15 06:40:20 +00:00
|
|
|
sample = new byte[MAX_SAMPLE_SIZE];
|
|
|
|
System.arraycopy(octets, 0, sample, 0, MAX_SAMPLE_SIZE);
|
2020-10-14 20:36:27 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
Log.i("compact_enc_det sample=" + sample.length);
|
|
|
|
String detected = jni_detect(sample);
|
2020-10-18 07:02:08 +00:00
|
|
|
if ("US-ASCII".equals(detected) ||
|
|
|
|
"ISO-8859-1".equals(detected) ||
|
|
|
|
"UTF-8".equals(detected))
|
2020-10-15 14:34:16 +00:00
|
|
|
Log.w("compact_enc_det result=" + detected);
|
|
|
|
else
|
2020-10-20 12:17:09 +00:00
|
|
|
// ISO-2022-JP, ISO-8859-2, windows-1250, windows-1252, windows-1257
|
2020-10-15 14:34:16 +00:00
|
|
|
Log.e("compact_enc_det result=" + detected);
|
2020-10-14 20:36:27 +00:00
|
|
|
|
2020-10-10 06:57:03 +00:00
|
|
|
return Charset.forName(detected);
|
|
|
|
} catch (Throwable ex) {
|
|
|
|
Log.w(ex);
|
|
|
|
return null;
|
|
|
|
}
|
|
|
|
}
|
2020-10-09 15:43:02 +00:00
|
|
|
}
|