//===- BinaryStreamReader.h - Reads objects from a binary stream *- C++ -*-===// // // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions. // See https://llvm.org/LICENSE.txt for license information. // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception // //===----------------------------------------------------------------------===// #ifndef LLVM_SUPPORT_BINARYSTREAMREADER_H #define LLVM_SUPPORT_BINARYSTREAMREADER_H #include "llvm/ADT/ArrayRef.h" #include "llvm/ADT/StringRef.h" #include "llvm/Support/Alignment.h" #include "llvm/Support/BinaryStreamArray.h" #include "llvm/Support/BinaryStreamRef.h" #include "llvm/Support/ConvertUTF.h" #include "llvm/Support/Endian.h" #include "llvm/Support/Error.h" #include namespace llvm { /// Provides read only access to a subclass of `BinaryStream`. Provides /// bounds checking and helpers for writing certain common data types such as /// null-terminated strings, integers in various flavors of endianness, etc. /// Can be subclassed to provide reading of custom datatypes, although no /// are overridable. class BinaryStreamReader { public: BinaryStreamReader() = default; explicit BinaryStreamReader(BinaryStreamRef Ref); explicit BinaryStreamReader(BinaryStream &Stream); explicit BinaryStreamReader(ArrayRef Data, llvm::endianness Endian); explicit BinaryStreamReader(StringRef Data, llvm::endianness Endian); BinaryStreamReader(const BinaryStreamReader &Other) = default; BinaryStreamReader &operator=(const BinaryStreamReader &Other) = default; virtual ~BinaryStreamReader() = default; /// Read as much as possible from the underlying string at the current offset /// without invoking a copy, and set \p Buffer to the resulting data slice. /// Updates the stream's offset to point after the newly read data. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readLongestContiguousChunk(ArrayRef &Buffer); /// Read \p Size bytes from the underlying stream at the current offset and /// and set \p Buffer to the resulting data slice. Whether a copy occurs /// depends on the implementation of the underlying stream. Updates the /// stream's offset to point after the newly read data. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readBytes(ArrayRef &Buffer, uint32_t Size); /// Read an integer of the specified endianness into \p Dest and update the /// stream's offset. The data is always copied from the stream's underlying /// buffer into \p Dest. Updates the stream's offset to point after the newly /// read data. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. template Error readInteger(T &Dest) { static_assert(std::is_integral_v, "Cannot call readInteger with non-integral value!"); ArrayRef Bytes; if (auto EC = readBytes(Bytes, sizeof(T))) return EC; Dest = llvm::support::endian::read(Bytes.data(), Stream.getEndian()); return Error::success(); } /// Similar to readInteger. template Error readEnum(T &Dest) { static_assert(std::is_enum::value, "Cannot call readEnum with non-enum value!"); std::underlying_type_t N; if (auto EC = readInteger(N)) return EC; Dest = static_cast(N); return Error::success(); } /// Read an unsigned LEB128 encoded value. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readULEB128(uint64_t &Dest); /// Read a signed LEB128 encoded value. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readSLEB128(int64_t &Dest); /// Read a null terminated string from \p Dest. Whether a copy occurs depends /// on the implementation of the underlying stream. Updates the stream's /// offset to point after the newly read data. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readCString(StringRef &Dest); /// Similar to readCString, however read a null-terminated UTF16 string /// instead. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readWideString(ArrayRef &Dest); /// Read a \p Length byte string into \p Dest. Whether a copy occurs depends /// on the implementation of the underlying stream. Updates the stream's /// offset to point after the newly read data. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readFixedString(StringRef &Dest, uint32_t Length); /// Read the entire remainder of the underlying stream into \p Ref. This is /// equivalent to calling getUnderlyingStream().slice(Offset). Updates the /// stream's offset to point to the end of the stream. Never causes a copy. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readStreamRef(BinaryStreamRef &Ref); /// Read \p Length bytes from the underlying stream into \p Ref. This is /// equivalent to calling getUnderlyingStream().slice(Offset, Length). /// Updates the stream's offset to point after the newly read object. Never /// causes a copy. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readStreamRef(BinaryStreamRef &Ref, uint32_t Length); /// Read \p Length bytes from the underlying stream into \p Ref. This is /// equivalent to calling getUnderlyingStream().slice(Offset, Length). /// Updates the stream's offset to point after the newly read object. Never /// causes a copy. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. Error readSubstream(BinarySubstreamRef &Ref, uint32_t Length); /// Get a pointer to an object of type T from the underlying stream, as if by /// memcpy, and store the result into \p Dest. It is up to the caller to /// ensure that objects of type T can be safely treated in this manner. /// Updates the stream's offset to point after the newly read object. Whether /// a copy occurs depends upon the implementation of the underlying /// stream. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. template Error readObject(const T *&Dest) { ArrayRef Buffer; if (auto EC = readBytes(Buffer, sizeof(T))) return EC; Dest = reinterpret_cast(Buffer.data()); return Error::success(); } /// Get a reference to a \p NumElements element array of objects of type T /// from the underlying stream as if by memcpy, and store the resulting array /// slice into \p array. It is up to the caller to ensure that objects of /// type T can be safely treated in this manner. Updates the stream's offset /// to point after the newly read object. Whether a copy occurs depends upon /// the implementation of the underlying stream. /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. template Error readArray(ArrayRef &Array, uint32_t NumElements) { ArrayRef Bytes; if (NumElements == 0) { Array = ArrayRef(); return Error::success(); } if (NumElements > UINT32_MAX / sizeof(T)) return make_error( stream_error_code::invalid_array_size); if (auto EC = readBytes(Bytes, NumElements * sizeof(T))) return EC; assert(isAddrAligned(Align::Of(), Bytes.data()) && "Reading at invalid alignment!"); Array = ArrayRef(reinterpret_cast(Bytes.data()), NumElements); return Error::success(); } /// Read a VarStreamArray of size \p Size bytes and store the result into /// \p Array. Updates the stream's offset to point after the newly read /// array. Never causes a copy (although iterating the elements of the /// VarStreamArray may, depending upon the implementation of the underlying /// stream). /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. template Error readArray(VarStreamArray &Array, uint32_t Size, uint32_t Skew = 0) { BinaryStreamRef S; if (auto EC = readStreamRef(S, Size)) return EC; Array.setUnderlyingStream(S, Skew); return Error::success(); } /// Read a FixedStreamArray of \p NumItems elements and store the result into /// \p Array. Updates the stream's offset to point after the newly read /// array. Never causes a copy (although iterating the elements of the /// FixedStreamArray may, depending upon the implementation of the underlying /// stream). /// /// \returns a success error code if the data was successfully read, otherwise /// returns an appropriate error code. template Error readArray(FixedStreamArray &Array, uint32_t NumItems) { if (NumItems == 0) { Array = FixedStreamArray(); return Error::success(); } if (NumItems > UINT32_MAX / sizeof(T)) return make_error( stream_error_code::invalid_array_size); BinaryStreamRef View; if (auto EC = readStreamRef(View, NumItems * sizeof(T))) return EC; Array = FixedStreamArray(View); return Error::success(); } bool empty() const { return bytesRemaining() == 0; } void setOffset(uint64_t Off) { Offset = Off; } uint64_t getOffset() const { return Offset; } uint64_t getLength() const { return Stream.getLength(); } uint64_t bytesRemaining() const { return getLength() - getOffset(); } /// Advance the stream's offset by \p Amount bytes. /// /// \returns a success error code if at least \p Amount bytes remain in the /// stream, otherwise returns an appropriate error code. Error skip(uint64_t Amount); /// Examine the next byte of the underlying stream without advancing the /// stream's offset. If the stream is empty the behavior is undefined. /// /// \returns the next byte in the stream. uint8_t peek() const; Error padToAlignment(uint32_t Align); std::pair split(uint64_t Offset) const; private: BinaryStreamRef Stream; uint64_t Offset = 0; }; } // namespace llvm #endif // LLVM_SUPPORT_BINARYSTREAMREADER_H