Mercurial > hg4j
view src/org/tmatesoft/hg/internal/EncodingHelper.java @ 634:170b6ecc890e
Issue 46: Files with backslashes (legal on Linux) cause troubles
author | Artem Tikhomirov <tikhomirov.artem@gmail.com> |
---|---|
date | Thu, 06 Jun 2013 14:21:11 +0200 |
parents | 47b7bedf0569 |
children | fba85bc1dfb8 |
line wrap: on
line source
/* * Copyright (c) 2011-2013 TMate Software Ltd * * This program 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; version 2 of the License. * * This program 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. * * For information on how to redistribute this software under * the terms of a license other than GNU General Public License * contact TMate Software at support@hg4j.com */ package org.tmatesoft.hg.internal; import static org.tmatesoft.hg.util.LogFacility.Severity.Error; import java.nio.ByteBuffer; import java.nio.CharBuffer; import java.nio.charset.CharacterCodingException; import java.nio.charset.Charset; import java.nio.charset.CharsetDecoder; import java.nio.charset.CharsetEncoder; import org.tmatesoft.hg.core.SessionContext; /** * Keep all encoding-related issues in the single place * NOT thread-safe (encoder and decoder requires synchronized access) * @author Artem Tikhomirov * @author TMate Software Ltd. */ public class EncodingHelper { // XXX perhaps, shall not be full of statics, but rather an instance coming from e.g. HgRepository? /* * To understand what Mercurial thinks of UTF-8 and Unix byte approach to names, see * http://mercurial.808500.n3.nabble.com/Unicode-support-request-td3430704.html */ private final SessionContext sessionContext; private final CharsetEncoder encoder; private final CharsetDecoder decoder; EncodingHelper(Charset fsEncoding, SessionContext ctx) { sessionContext = ctx; decoder = fsEncoding.newDecoder(); encoder = fsEncoding.newEncoder(); } /** * Translate file names from manifest to amazing Unicode string */ public String fromManifest(byte[] data, int start, int length) { return decodeWithSystemDefaultFallback(data, start, length); } /** * @return byte representation of the string directly comparable to bytes in manifest */ public byte[] toManifest(CharSequence s) { if (s == null) { // perhaps, can return byte[0] in this case? throw new IllegalArgumentException(); } return encodeWithSystemDefaultFallback(s); } /** * Translate file names from dirstate to amazing Unicode string */ public String fromDirstate(byte[] data, int start, int length) { return decodeWithSystemDefaultFallback(data, start, length); } public byte[] toDirstate(CharSequence fname) { if (fname == null) { throw new IllegalArgumentException(); } return encodeWithSystemDefaultFallback(fname); } private String decodeWithSystemDefaultFallback(byte[] data, int start, int length) { try { return decoder.decode(ByteBuffer.wrap(data, start, length)).toString(); } catch (CharacterCodingException ex) { sessionContext.getLog().dump(getClass(), Error, ex, String.format("Use of charset %s failed, resort to system default", charset().name())); // resort to system-default return new String(data, start, length); } } private byte[] encodeWithSystemDefaultFallback(CharSequence s) { try { // synchronized(encoder) { ByteBuffer bb = encoder.encode(CharBuffer.wrap(s)); // } byte[] rv = new byte[bb.remaining()]; bb.get(rv, 0, rv.length); return rv; } catch (CharacterCodingException ex) { sessionContext.getLog().dump(getClass(), Error, ex, String.format("Use of charset %s failed, resort to system default", charset().name())); // resort to system-default return s.toString().getBytes(); } } private Charset charset() { return encoder.charset(); } public static Charset getUTF8() { return Charset.forName("UTF-8"); } }