#region Copyright notice and license
// Protocol Buffers - Google's data interchange format
// Copyright 2008 Google Inc. All rights reserved.
// http://github.com/jskeet/dotnet-protobufs/
// Original C++/Java/Python code:
// http://code.google.com/p/protobuf/
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are
// met:
//
// * Redistributions of source code must retain the above copyright
// notice, this list of conditions and the following disclaimer.
// * Redistributions in binary form must reproduce the above
// copyright notice, this list of conditions and the following disclaimer
// in the documentation and/or other materials provided with the
// distribution.
// * Neither the name of Google Inc. nor the names of its
// contributors may be used to endorse or promote products derived from
// this software without specific prior written permission.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
// "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
// LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
// A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
// OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
// SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
// LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
// DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
// THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
// (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
// OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#endregion
using System;
using System.Collections.Generic;
using System.IO;
using System.Text;
using Google.ProtocolBuffers.Descriptors;
namespace Google.ProtocolBuffers {
///
/// Readings and decodes protocol message fields.
///
///
/// This class contains two kinds of methods: methods that read specific
/// protocol message constructs and field types (e.g. ReadTag and
/// ReadInt32) and methods that read low-level values (e.g.
/// ReadRawVarint32 and ReadRawBytes). If you are reading encoded protocol
/// messages, you should use the former methods, but if you are reading some
/// other format of your own design, use the latter. The names of the former
/// methods are taken from the protocol buffer type names, not .NET types.
/// (Hence ReadFloat instead of ReadSingle, and ReadBool instead of ReadBoolean.)
///
/// TODO(jonskeet): Consider whether recursion and size limits shouldn't be readonly,
/// set at construction time.
///
public sealed class CodedInputStream {
private readonly byte[] buffer;
private int bufferSize;
private int bufferSizeAfterLimit = 0;
private int bufferPos = 0;
private readonly Stream input;
private uint lastTag = 0;
internal const int DefaultRecursionLimit = 64;
internal const int DefaultSizeLimit = 64 << 20; // 64MB
internal const int BufferSize = 4096;
///
/// The total number of bytes read before the current buffer. The
/// total bytes read up to the current position can be computed as
/// totalBytesRetired + bufferPos.
///
private int totalBytesRetired = 0;
///
/// The absolute position of the end of the current message.
///
private int currentLimit = int.MaxValue;
///
///
///
private int recursionDepth = 0;
private int recursionLimit = DefaultRecursionLimit;
///
///
///
private int sizeLimit = DefaultSizeLimit;
#region Construction
///
/// Creates a new CodedInputStream reading data from the given
/// stream.
///
public static CodedInputStream CreateInstance(Stream input) {
return new CodedInputStream(input);
}
///
/// Creates a new CodedInputStream reading data from the given
/// byte array.
///
public static CodedInputStream CreateInstance(byte[] buf) {
return new CodedInputStream(buf);
}
private CodedInputStream(byte[] buffer) {
this.buffer = buffer;
this.bufferSize = buffer.Length;
this.input = null;
}
private CodedInputStream(Stream input) {
this.buffer = new byte[BufferSize];
this.bufferSize = 0;
this.input = input;
}
#endregion
#region Validation
///
/// Verifies that the last call to ReadTag() returned the given tag value.
/// This is used to verify that a nested group ended with the correct
/// end tag.
///
/// The last
/// tag read was not the one specified
[CLSCompliant(false)]
public void CheckLastTagWas(uint value) {
if (lastTag != value) {
throw InvalidProtocolBufferException.InvalidEndTag();
}
}
#endregion
#region Reading of tags etc
///
/// Attempt to read a field tag, returning 0 if we have reached the end
/// of the input data. Protocol message parsers use this to read tags,
/// since a protocol message may legally end wherever a tag occurs, and
/// zero is not a valid tag number.
///
[CLSCompliant(false)]
public uint ReadTag() {
if (IsAtEnd) {
lastTag = 0;
return 0;
}
lastTag = ReadRawVarint32();
if (lastTag == 0) {
// If we actually read zero, that's not a valid tag.
throw InvalidProtocolBufferException.InvalidTag();
}
return lastTag;
}
///
/// Read a double field from the stream.
///
public double ReadDouble() {
#if SILVERLIGHT2 || COMPACT_FRAMEWORK_35
byte[] bytes = ReadRawBytes(8);
return BitConverter.ToDouble(bytes, 0);
#else
return BitConverter.Int64BitsToDouble((long) ReadRawLittleEndian64());
#endif
}
///
/// Read a float field from the stream.
///
public float ReadFloat() {
// TODO(jonskeet): Test this on different endiannesses
uint raw = ReadRawLittleEndian32();
byte[] rawBytes = BitConverter.GetBytes(raw);
return BitConverter.ToSingle(rawBytes, 0);
}
///
/// Read a uint64 field from the stream.
///
[CLSCompliant(false)]
public ulong ReadUInt64() {
return ReadRawVarint64();
}
///
/// Read an int64 field from the stream.
///
public long ReadInt64() {
return (long) ReadRawVarint64();
}
///
/// Read an int32 field from the stream.
///
public int ReadInt32() {
return (int) ReadRawVarint32();
}
///
/// Read a fixed64 field from the stream.
///
[CLSCompliant(false)]
public ulong ReadFixed64() {
return ReadRawLittleEndian64();
}
///
/// Read a fixed32 field from the stream.
///
[CLSCompliant(false)]
public uint ReadFixed32() {
return ReadRawLittleEndian32();
}
///
/// Read a bool field from the stream.
///
public bool ReadBool() {
return ReadRawVarint32() != 0;
}
///
/// Reads a string field from the stream.
///
public String ReadString() {
int size = (int) ReadRawVarint32();
// No need to read any data for an empty string.
if (size == 0) {
return "";
}
if (size <= bufferSize - bufferPos) {
// Fast path: We already have the bytes in a contiguous buffer, so
// just copy directly from it.
String result = Encoding.UTF8.GetString(buffer, bufferPos, size);
bufferPos += size;
return result;
}
// Slow path: Build a byte array first then copy it.
return Encoding.UTF8.GetString(ReadRawBytes(size), 0, size);
}
///
/// Reads a group field value from the stream.
///
public void ReadGroup(int fieldNumber, IBuilder builder,
ExtensionRegistry extensionRegistry) {
if (recursionDepth >= recursionLimit) {
throw InvalidProtocolBufferException.RecursionLimitExceeded();
}
++recursionDepth;
builder.WeakMergeFrom(this, extensionRegistry);
CheckLastTagWas(WireFormat.MakeTag(fieldNumber, WireFormat.WireType.EndGroup));
--recursionDepth;
}
///
/// Reads a group field value from the stream and merges it into the given
/// UnknownFieldSet.
///
public void ReadUnknownGroup(int fieldNumber, UnknownFieldSet.Builder builder) {
if (recursionDepth >= recursionLimit) {
throw InvalidProtocolBufferException.RecursionLimitExceeded();
}
++recursionDepth;
builder.MergeFrom(this);
CheckLastTagWas(WireFormat.MakeTag(fieldNumber, WireFormat.WireType.EndGroup));
--recursionDepth;
}
///
/// Reads an embedded message field value from the stream.
///
public void ReadMessage(IBuilder builder, ExtensionRegistry extensionRegistry) {
int length = (int) ReadRawVarint32();
if (recursionDepth >= recursionLimit) {
throw InvalidProtocolBufferException.RecursionLimitExceeded();
}
int oldLimit = PushLimit(length);
++recursionDepth;
builder.WeakMergeFrom(this, extensionRegistry);
CheckLastTagWas(0);
--recursionDepth;
PopLimit(oldLimit);
}
///
/// Reads a bytes field value from the stream.
///
public ByteString ReadBytes() {
int size = (int) ReadRawVarint32();
if (size < bufferSize - bufferPos && size > 0) {
// Fast path: We already have the bytes in a contiguous buffer, so
// just copy directly from it.
ByteString result = ByteString.CopyFrom(buffer, bufferPos, size);
bufferPos += size;
return result;
} else {
// Slow path: Build a byte array first then copy it.
return ByteString.CopyFrom(ReadRawBytes(size));
}
}
///
/// Reads a uint32 field value from the stream.
///
[CLSCompliant(false)]
public uint ReadUInt32() {
return ReadRawVarint32();
}
///
/// Reads an enum field value from the stream. The caller is responsible
/// for converting the numeric value to an actual enum.
///
public int ReadEnum() {
return (int) ReadRawVarint32();
}
///
/// Reads an sfixed32 field value from the stream.
///
public int ReadSFixed32() {
return (int) ReadRawLittleEndian32();
}
///
/// Reads an sfixed64 field value from the stream.
///
public long ReadSFixed64() {
return (long) ReadRawLittleEndian64();
}
///
/// Reads an sint32 field value from the stream.
///
public int ReadSInt32() {
return DecodeZigZag32(ReadRawVarint32());
}
///
/// Reads an sint64 field value from the stream.
///
public long ReadSInt64() {
return DecodeZigZag64(ReadRawVarint64());
}
///
/// Reads a field of any primitive type. Enums, groups and embedded
/// messages are not handled by this method.
///
public object ReadPrimitiveField(FieldType fieldType) {
switch (fieldType) {
case FieldType.Double: return ReadDouble();
case FieldType.Float: return ReadFloat();
case FieldType.Int64: return ReadInt64();
case FieldType.UInt64: return ReadUInt64();
case FieldType.Int32: return ReadInt32();
case FieldType.Fixed64: return ReadFixed64();
case FieldType.Fixed32: return ReadFixed32();
case FieldType.Bool: return ReadBool();
case FieldType.String: return ReadString();
case FieldType.Bytes: return ReadBytes();
case FieldType.UInt32: return ReadUInt32();
case FieldType.SFixed32: return ReadSFixed32();
case FieldType.SFixed64: return ReadSFixed64();
case FieldType.SInt32: return ReadSInt32();
case FieldType.SInt64: return ReadSInt64();
case FieldType.Group:
throw new ArgumentException("ReadPrimitiveField() cannot handle nested groups.");
case FieldType.Message:
throw new ArgumentException("ReadPrimitiveField() cannot handle embedded messages.");
// We don't handle enums because we don't know what to do if the
// value is not recognized.
case FieldType.Enum:
throw new ArgumentException("ReadPrimitiveField() cannot handle enums.");
default:
throw new ArgumentOutOfRangeException("Invalid field type " + fieldType);
}
}
#endregion
#region Underlying reading primitives
///
/// Same code as ReadRawVarint32, but read each byte individually, checking for
/// buffer overflow.
///
private uint SlowReadRawVarint32() {
int tmp = ReadRawByte();
if (tmp < 128) {
return (uint)tmp;
}
int result = tmp & 0x7f;
if ((tmp = ReadRawByte()) < 128) {
result |= tmp << 7;
} else {
result |= (tmp & 0x7f) << 7;
if ((tmp = ReadRawByte()) < 128) {
result |= tmp << 14;
} else {
result |= (tmp & 0x7f) << 14;
if ((tmp = ReadRawByte()) < 128) {
result |= tmp << 21;
} else {
result |= (tmp & 0x7f) << 21;
result |= (tmp = ReadRawByte()) << 28;
if (tmp >= 128) {
// Discard upper 32 bits.
for (int i = 0; i < 5; i++) {
if (ReadRawByte() < 128) return (uint)result;
}
throw InvalidProtocolBufferException.MalformedVarint();
}
}
}
}
return (uint)result;
}
///
/// Read a raw Varint from the stream. If larger than 32 bits, discard the upper bits.
/// This method is optimised for the case where we've got lots of data in the buffer.
/// That means we can check the size just once, then just read directly from the buffer
/// without constant rechecking of the buffer length.
///
[CLSCompliant(false)]
public uint ReadRawVarint32() {
if (bufferPos + 5 > bufferSize) {
return SlowReadRawVarint32();
}
int tmp = buffer[bufferPos++];
if (tmp < 128) {
return (uint)tmp;
}
int result = tmp & 0x7f;
if ((tmp = buffer[bufferPos++]) < 128) {
result |= tmp << 7;
} else {
result |= (tmp & 0x7f) << 7;
if ((tmp = buffer[bufferPos++]) < 128) {
result |= tmp << 14;
} else {
result |= (tmp & 0x7f) << 14;
if ((tmp = buffer[bufferPos++]) < 128) {
result |= tmp << 21;
} else {
result |= (tmp & 0x7f) << 21;
result |= (tmp = buffer[bufferPos++]) << 28;
if (tmp >= 128) {
// Discard upper 32 bits.
// Note that this has to use ReadRawByte() as we only ensure we've
// got at least 5 bytes at the start of the method. This lets us
// use the fast path in more cases, and we rarely hit this section of code.
for (int i = 0; i < 5; i++) {
if (ReadRawByte() < 128) return (uint)result;
}
throw InvalidProtocolBufferException.MalformedVarint();
}
}
}
}
return (uint)result;
}
///
/// Reads a varint from the input one byte at a time, so that it does not
/// read any bytes after the end of the varint. If you simply wrapped the
/// stream in a CodedInputStream and used ReadRawVarint32(Stream)}
/// then you would probably end up reading past the end of the varint since
/// CodedInputStream buffers its input.
///
///
///
internal static uint ReadRawVarint32(Stream input) {
int result = 0;
int offset = 0;
for (; offset < 32; offset += 7) {
int b = input.ReadByte();
if (b == -1) {
throw InvalidProtocolBufferException.TruncatedMessage();
}
result |= (b & 0x7f) << offset;
if ((b & 0x80) == 0) {
return (uint) result;
}
}
// Keep reading up to 64 bits.
for (; offset < 64; offset += 7) {
int b = input.ReadByte();
if (b == -1) {
throw InvalidProtocolBufferException.TruncatedMessage();
}
if ((b & 0x80) == 0) {
return (uint) result;
}
}
throw InvalidProtocolBufferException.MalformedVarint();
}
///
/// Read a raw varint from the stream.
///
[CLSCompliant(false)]
public ulong ReadRawVarint64() {
int shift = 0;
ulong result = 0;
while (shift < 64) {
byte b = ReadRawByte();
result |= (ulong)(b & 0x7F) << shift;
if ((b & 0x80) == 0) {
return result;
}
shift += 7;
}
throw InvalidProtocolBufferException.MalformedVarint();
}
///
/// Read a 32-bit little-endian integer from the stream.
///
[CLSCompliant(false)]
public uint ReadRawLittleEndian32() {
uint b1 = ReadRawByte();
uint b2 = ReadRawByte();
uint b3 = ReadRawByte();
uint b4 = ReadRawByte();
return b1 | (b2 << 8) | (b3 << 16) | (b4 << 24);
}
///
/// Read a 64-bit little-endian integer from the stream.
///
[CLSCompliant(false)]
public ulong ReadRawLittleEndian64() {
ulong b1 = ReadRawByte();
ulong b2 = ReadRawByte();
ulong b3 = ReadRawByte();
ulong b4 = ReadRawByte();
ulong b5 = ReadRawByte();
ulong b6 = ReadRawByte();
ulong b7 = ReadRawByte();
ulong b8 = ReadRawByte();
return b1 | (b2 << 8) | (b3 << 16) | (b4 << 24)
| (b5 << 32) | (b6 << 40) | (b7 << 48) | (b8 << 56);
}
#endregion
///
/// Decode a 32-bit value with ZigZag encoding.
///
///
/// ZigZag encodes signed integers into values that can be efficiently
/// encoded with varint. (Otherwise, negative values must be
/// sign-extended to 64 bits to be varint encoded, thus always taking
/// 10 bytes on the wire.)
///
[CLSCompliant(false)]
public static int DecodeZigZag32(uint n) {
return (int)(n >> 1) ^ -(int)(n & 1);
}
///
/// Decode a 32-bit value with ZigZag encoding.
///
///
/// ZigZag encodes signed integers into values that can be efficiently
/// encoded with varint. (Otherwise, negative values must be
/// sign-extended to 64 bits to be varint encoded, thus always taking
/// 10 bytes on the wire.)
///
[CLSCompliant(false)]
public static long DecodeZigZag64(ulong n) {
return (long)(n >> 1) ^ -(long)(n & 1);
}
///
/// Set the maximum message recursion depth.
///
///
/// In order to prevent malicious
/// messages from causing stack overflows, CodedInputStream limits
/// how deeply messages may be nested. The default limit is 64.
///
public int SetRecursionLimit(int limit) {
if (limit < 0) {
throw new ArgumentOutOfRangeException("Recursion limit cannot be negative: " + limit);
}
int oldLimit = recursionLimit;
recursionLimit = limit;
return oldLimit;
}
///
/// Set the maximum message size.
///
///
/// In order to prevent malicious messages from exhausting memory or
/// causing integer overflows, CodedInputStream limits how large a message may be.
/// The default limit is 64MB. You should set this limit as small
/// as you can without harming your app's functionality. Note that
/// size limits only apply when reading from an InputStream, not
/// when constructed around a raw byte array (nor with ByteString.NewCodedInput).
/// If you want to read several messages from a single CodedInputStream, you
/// can call ResetSizeCounter() after each message to avoid hitting the
/// size limit.
///
public int SetSizeLimit(int limit) {
if (limit < 0) {
throw new ArgumentOutOfRangeException("Size limit cannot be negative: " + limit);
}
int oldLimit = sizeLimit;
sizeLimit = limit;
return oldLimit;
}
#region Internal reading and buffer management
///
/// Resets the current size counter to zero (see SetSizeLimit).
///
public void ResetSizeCounter() {
totalBytesRetired = 0;
}
///
/// Sets currentLimit to (current position) + byteLimit. This is called
/// when descending into a length-delimited embedded message. The previous
/// limit is returned.
///
/// The old limit.
public int PushLimit(int byteLimit) {
if (byteLimit < 0) {
throw InvalidProtocolBufferException.NegativeSize();
}
byteLimit += totalBytesRetired + bufferPos;
int oldLimit = currentLimit;
if (byteLimit > oldLimit) {
throw InvalidProtocolBufferException.TruncatedMessage();
}
currentLimit = byteLimit;
RecomputeBufferSizeAfterLimit();
return oldLimit;
}
private void RecomputeBufferSizeAfterLimit() {
bufferSize += bufferSizeAfterLimit;
int bufferEnd = totalBytesRetired + bufferSize;
if (bufferEnd > currentLimit) {
// Limit is in current buffer.
bufferSizeAfterLimit = bufferEnd - currentLimit;
bufferSize -= bufferSizeAfterLimit;
} else {
bufferSizeAfterLimit = 0;
}
}
///
/// Discards the current limit, returning the previous limit.
///
public void PopLimit(int oldLimit) {
currentLimit = oldLimit;
RecomputeBufferSizeAfterLimit();
}
///
/// Returns whether or not all the data before the limit has been read.
///
///
public bool ReachedLimit {
get {
if (currentLimit == int.MaxValue) {
return false;
}
int currentAbsolutePosition = totalBytesRetired + bufferPos;
return currentAbsolutePosition >= currentLimit;
}
}
///
/// Returns true if the stream has reached the end of the input. This is the
/// case if either the end of the underlying input source has been reached or
/// the stream has reached a limit created using PushLimit.
///
public bool IsAtEnd {
get {
return bufferPos == bufferSize && !RefillBuffer(false);
}
}
///
/// Called when buffer is empty to read more bytes from the
/// input. If is true, RefillBuffer() gurantees that
/// either there will be at least one byte in the buffer when it returns
/// or it will throw an exception. If is false,
/// RefillBuffer() returns false if no more bytes were available.
///
///
///
private bool RefillBuffer(bool mustSucceed) {
if (bufferPos < bufferSize) {
throw new InvalidOperationException("RefillBuffer() called when buffer wasn't empty.");
}
if (totalBytesRetired + bufferSize == currentLimit) {
// Oops, we hit a limit.
if (mustSucceed) {
throw InvalidProtocolBufferException.TruncatedMessage();
} else {
return false;
}
}
totalBytesRetired += bufferSize;
bufferPos = 0;
bufferSize = (input == null) ? 0 : input.Read(buffer, 0, buffer.Length);
if (bufferSize < 0) {
throw new InvalidOperationException("Stream.Read returned a negative count");
}
if (bufferSize == 0) {
if (mustSucceed) {
throw InvalidProtocolBufferException.TruncatedMessage();
} else {
return false;
}
} else {
RecomputeBufferSizeAfterLimit();
int totalBytesRead =
totalBytesRetired + bufferSize + bufferSizeAfterLimit;
if (totalBytesRead > sizeLimit || totalBytesRead < 0) {
throw InvalidProtocolBufferException.SizeLimitExceeded();
}
return true;
}
}
///
/// Read one byte from the input.
///
///
/// the end of the stream or the current limit was reached
///
public byte ReadRawByte() {
if (bufferPos == bufferSize) {
RefillBuffer(true);
}
return buffer[bufferPos++];
}
///
/// Read a fixed size of bytes from the input.
///
///
/// the end of the stream or the current limit was reached
///
public byte[] ReadRawBytes(int size) {
if (size < 0) {
throw InvalidProtocolBufferException.NegativeSize();
}
if (totalBytesRetired + bufferPos + size > currentLimit) {
// Read to the end of the stream anyway.
SkipRawBytes(currentLimit - totalBytesRetired - bufferPos);
// Then fail.
throw InvalidProtocolBufferException.TruncatedMessage();
}
if (size <= bufferSize - bufferPos) {
// We have all the bytes we need already.
byte[] bytes = new byte[size];
Array.Copy(buffer, bufferPos, bytes, 0, size);
bufferPos += size;
return bytes;
} else if (size < BufferSize) {
// Reading more bytes than are in the buffer, but not an excessive number
// of bytes. We can safely allocate the resulting array ahead of time.
// First copy what we have.
byte[] bytes = new byte[size];
int pos = bufferSize - bufferPos;
Array.Copy(buffer, bufferPos, bytes, 0, pos);
bufferPos = bufferSize;
// We want to use RefillBuffer() and then copy from the buffer into our
// byte array rather than reading directly into our byte array because
// the input may be unbuffered.
RefillBuffer(true);
while (size - pos > bufferSize) {
Array.Copy(buffer, 0, bytes, pos, bufferSize);
pos += bufferSize;
bufferPos = bufferSize;
RefillBuffer(true);
}
Array.Copy(buffer, 0, bytes, pos, size - pos);
bufferPos = size - pos;
return bytes;
} else {
// The size is very large. For security reasons, we can't allocate the
// entire byte array yet. The size comes directly from the input, so a
// maliciously-crafted message could provide a bogus very large size in
// order to trick the app into allocating a lot of memory. We avoid this
// by allocating and reading only a small chunk at a time, so that the
// malicious message must actually *be* extremely large to cause
// problems. Meanwhile, we limit the allowed size of a message elsewhere.
// Remember the buffer markers since we'll have to copy the bytes out of
// it later.
int originalBufferPos = bufferPos;
int originalBufferSize = bufferSize;
// Mark the current buffer consumed.
totalBytesRetired += bufferSize;
bufferPos = 0;
bufferSize = 0;
// Read all the rest of the bytes we need.
int sizeLeft = size - (originalBufferSize - originalBufferPos);
List chunks = new List();
while (sizeLeft > 0) {
byte[] chunk = new byte[Math.Min(sizeLeft, BufferSize)];
int pos = 0;
while (pos < chunk.Length) {
int n = (input == null) ? -1 : input.Read(chunk, pos, chunk.Length - pos);
if (n <= 0) {
throw InvalidProtocolBufferException.TruncatedMessage();
}
totalBytesRetired += n;
pos += n;
}
sizeLeft -= chunk.Length;
chunks.Add(chunk);
}
// OK, got everything. Now concatenate it all into one buffer.
byte[] bytes = new byte[size];
// Start by copying the leftover bytes from this.buffer.
int newPos = originalBufferSize - originalBufferPos;
Array.Copy(buffer, originalBufferPos, bytes, 0, newPos);
// And now all the chunks.
foreach (byte[] chunk in chunks) {
Array.Copy(chunk, 0, bytes, newPos, chunk.Length);
newPos += chunk.Length;
}
// Done.
return bytes;
}
}
///
/// Reads and discards a single field, given its tag value.
///
/// false if the tag is an end-group tag, in which case
/// nothing is skipped. Otherwise, returns true.
[CLSCompliant(false)]
public bool SkipField(uint tag) {
switch (WireFormat.GetTagWireType(tag)) {
case WireFormat.WireType.Varint:
ReadInt32();
return true;
case WireFormat.WireType.Fixed64:
ReadRawLittleEndian64();
return true;
case WireFormat.WireType.LengthDelimited:
SkipRawBytes((int) ReadRawVarint32());
return true;
case WireFormat.WireType.StartGroup:
SkipMessage();
CheckLastTagWas(
WireFormat.MakeTag(WireFormat.GetTagFieldNumber(tag),
WireFormat.WireType.EndGroup));
return true;
case WireFormat.WireType.EndGroup:
return false;
case WireFormat.WireType.Fixed32:
ReadRawLittleEndian32();
return true;
default:
throw InvalidProtocolBufferException.InvalidWireType();
}
}
///
/// Reads and discards an entire message. This will read either until EOF
/// or until an endgroup tag, whichever comes first.
///
public void SkipMessage() {
while (true) {
uint tag = ReadTag();
if (tag == 0 || !SkipField(tag)) {
return;
}
}
}
///
/// Reads and discards bytes.
///
/// the end of the stream
/// or the current limit was reached
public void SkipRawBytes(int size) {
if (size < 0) {
throw InvalidProtocolBufferException.NegativeSize();
}
if (totalBytesRetired + bufferPos + size > currentLimit) {
// Read to the end of the stream anyway.
SkipRawBytes(currentLimit - totalBytesRetired - bufferPos);
// Then fail.
throw InvalidProtocolBufferException.TruncatedMessage();
}
if (size <= bufferSize - bufferPos) {
// We have all the bytes we need already.
bufferPos += size;
} else {
// Skipping more bytes than are in the buffer. First skip what we have.
int pos = bufferSize - bufferPos;
totalBytesRetired += pos;
bufferPos = 0;
bufferSize = 0;
// Then skip directly from the InputStream for the rest.
if (pos < size) {
if (input == null) {
throw InvalidProtocolBufferException.TruncatedMessage();
}
SkipImpl(size - pos);
totalBytesRetired += size - pos;
}
}
}
///
/// Abstraction of skipping to cope with streams which can't really skip.
///
private void SkipImpl(int amountToSkip) {
if (input.CanSeek) {
long previousPosition = input.Position;
input.Position += amountToSkip;
if (input.Position != previousPosition + amountToSkip) {
throw InvalidProtocolBufferException.TruncatedMessage();
}
} else {
byte[] skipBuffer = new byte[1024];
while (amountToSkip > 0) {
int bytesRead = input.Read(skipBuffer, 0, skipBuffer.Length);
if (bytesRead <= 0) {
throw InvalidProtocolBufferException.TruncatedMessage();
}
amountToSkip -= bytesRead;
}
}
}
#endregion
}
}