From bdae99c0847556dd8103f172fc1836eb83ae4c4a Mon Sep 17 00:00:00 2001 From: Vincent Breitmoser Date: Fri, 12 Feb 2016 17:08:09 +0100 Subject: mime: try to decode with given charset or utf-8 while file is read --- .../keychain/operations/InputDataOperation.java | 86 ++++++++++++++++++++-- .../operations/results/OperationResult.java | 4 + 2 files changed, 84 insertions(+), 6 deletions(-) (limited to 'OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations') diff --git a/OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations/InputDataOperation.java b/OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations/InputDataOperation.java index bb8d6ad73..80f9d6368 100644 --- a/OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations/InputDataOperation.java +++ b/OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations/InputDataOperation.java @@ -23,6 +23,13 @@ import java.io.FileNotFoundException; import java.io.IOException; import java.io.InputStream; import java.io.OutputStream; +import java.nio.ByteBuffer; +import java.nio.CharBuffer; +import java.nio.charset.Charset; +import java.nio.charset.CharsetDecoder; +import java.nio.charset.CoderResult; +import java.nio.charset.CodingErrorAction; +import java.nio.charset.UnsupportedCharsetException; import java.util.ArrayList; import android.content.ClipDescription; @@ -67,10 +74,15 @@ import org.sufficientlysecure.keychain.service.input.CryptoInputParcel; */ public class InputDataOperation extends BaseOperation { - final private byte[] buf = new byte[256]; + private final byte[] buf = new byte[256]; + private final ByteBuffer bufWrap; + private final CharBuffer dummyOutput; public InputDataOperation(Context context, ProviderHelper providerHelper, Progressable progressable) { super(context, providerHelper, progressable); + + bufWrap = ByteBuffer.wrap(buf); + dummyOutput = CharBuffer.allocate(256); } Uri mSignedDataUri; @@ -326,20 +338,82 @@ public class InputDataOperation extends BaseOperation { throw new IOException("Error getting file for writing!"); } + boolean isPossibleTextMimeType = ClipDescription.compareMimeTypes(mimeType, "application/octet-stream") + || ClipDescription.compareMimeTypes(mimeType, "application/x-download") + || ClipDescription.compareMimeTypes(mimeType, "text/*"); + + // If this data looks like text, we pipe the incoming data into a charset + // decoder, to see if the data is legal for the assumed charset. + String charset; + boolean charsetIsFaulty; + boolean charsetIsGuessed; + CharsetDecoder charsetDecoder = null; + if (isPossibleTextMimeType) { + charset = bd.getCharset(); + // the charset defaults to us-ascii, but we want to default to utf-8 + if (charset == null || "us-ascii".equals(charset)) { + charset = "utf-8"; + charsetIsGuessed = true; + } else { + charsetIsGuessed = false; + } + + try { + charsetDecoder = Charset.forName(charset).newDecoder(); + charsetDecoder.onMalformedInput(CodingErrorAction.REPORT); + charsetDecoder.onUnmappableCharacter(CodingErrorAction.REPORT); + charsetDecoder.reset(); + charsetIsFaulty = false; + } catch (UnsupportedCharsetException e) { + charsetIsFaulty = true; + } + } else { + charsetIsFaulty = true; + charsetIsGuessed = false; + charset = null; + } + int totalLength = 0; do { totalLength += len; out.write(buf, 0, len); + + if (isPossibleTextMimeType && !charsetIsFaulty) { + bufWrap.rewind(); + bufWrap.limit(len); + dummyOutput.rewind(); + CoderResult result = charsetDecoder.decode(bufWrap, dummyOutput, false); + if (result.isError()) { + charsetIsFaulty = true; + } + } } while ((len = is.read(buf)) > 0); - log.add(LogType.MSG_DATA_MIME_LENGTH, 3, Long.toString(totalLength)); + if (!charsetIsFaulty) { + bufWrap.rewind(); + bufWrap.limit(0); + dummyOutput.rewind(); + CoderResult result = charsetDecoder.decode(bufWrap, dummyOutput, true); + if (result.isError()) { + charsetIsFaulty = true; + } + } - String charset = bd.getCharset(); - // the charset defaults to us-ascii, but we want to default to utf-8 - if ("us-ascii".equals(charset)) { - charset = "utf-8"; + if (isPossibleTextMimeType) { + if (charsetIsFaulty && charsetIsGuessed) { + log.add(LogType.MSG_DATA_MIME_CHARSET_UNKNOWN, 3, charset); + charset = null; + } else if (charsetIsFaulty) { + log.add(LogType.MSG_DATA_MIME_CHARSET_FAULTY, 3, charset); + } else if (charsetIsGuessed) { + log.add(LogType.MSG_DATA_MIME_CHARSET_GUESS, 3, charset); + } else { + log.add(LogType.MSG_DATA_MIME_CHARSET, 3, charset); + } } + log.add(LogType.MSG_DATA_MIME_LENGTH, 3, Long.toString(totalLength)); + OpenPgpMetadata metadata = new OpenPgpMetadata(mFilename, mimeType, 0L, totalLength, charset); out.close(); diff --git a/OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations/results/OperationResult.java b/OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations/results/OperationResult.java index f9c2db8e8..ec2fddbd0 100644 --- a/OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations/results/OperationResult.java +++ b/OpenKeychain/src/main/java/org/sufficientlysecure/keychain/operations/results/OperationResult.java @@ -860,6 +860,10 @@ public abstract class OperationResult implements Parcelable { MSG_DATA_MIME_FROM_EXTENSION (LogLevel.DEBUG, R.string.msg_data_mime_from_extension), MSG_DATA_MIME_FILENAME (LogLevel.DEBUG, R.string.msg_data_mime_filename), MSG_DATA_MIME_LENGTH (LogLevel.DEBUG, R.string.msg_data_mime_length), + MSG_DATA_MIME_CHARSET (LogLevel.DEBUG, R.string.msg_data_mime_charset), + MSG_DATA_MIME_CHARSET_FAULTY (LogLevel.WARN, R.string.msg_data_mime_charset_faulty), + MSG_DATA_MIME_CHARSET_GUESS (LogLevel.DEBUG, R.string.msg_data_mime_charset_guess), + MSG_DATA_MIME_CHARSET_UNKNOWN (LogLevel.DEBUG, R.string.msg_data_mime_charset_unknown), MSG_DATA_MIME (LogLevel.DEBUG, R.string.msg_data_mime), MSG_DATA_MIME_OK (LogLevel.INFO, R.string.msg_data_mime_ok), MSG_DATA_MIME_NONE (LogLevel.DEBUG, R.string.msg_data_mime_none), -- cgit v1.2.3