PFRecordWrite class
This class is used to write data to a file contained inside a DBPF.
Use PFRecordRead::
Base classes
- class PFRecordBase
- Base implementation for records in a .packagefile (DatabasePackedFile).
Public static variables
Constructors, destructors, conversion operators
- 
              PFRecordWrite(void* pData,
              size_t nSize, bool bUsePointer, ResourceKey& name, DatabasePackedFile* pParentDBPF) 
- 
              PFRecordWrite(size_t nChunkOffset, size_ t nSize, bool bUsePointer, ResourceKey& name, DatabasePackedFile* pParentDBPF) 
- ~PFRecordWrite() virtual
Public functions
- auto AddRef() -> int override
- auto Release() -> int override
- auto GetKey() -> const ResourceKey& override
- Gets the ResourceKey name that identifies this record.
- auto SetKey(const ResourceKey& key) -> void override
- Sets the ResourceKey name that identifies this record.
- 
              auto GetStream() -> IO::IStream* override 
- Returns the IStream used to access this record data.
- auto GetDatabase() -> Database* override
- Returns the database that this record belongs to.
- auto RecordOpen() -> bool override
- Opens this record, allowing its data to be used.
- auto RecordClose() -> bool override
- Closes this record, forbidding any use of its data.
- auto DoPostClose() -> int override
- 
              auto GetType() const -> uint32_t override 
- Returns the type of the stream, which is different for each Stream subclass.
- 
              auto GetAccessFlags() const -> IO::AccessFlags override 
- Returns one of enum AccessFlags.
- 
              auto GetState() const -> IO::FileError override 
- Returns the error state of the stream.
- 
              auto GetSize() const -> IO::size_type override 
- Returns the size of the stream, which is not the same as the size of bytes remaining to be read from the stream.
- 
              auto SetSize(IO::size_type size) -> bool override 
- Sets the size of the stream, if possible.
- 
              auto GetPosition(IO::PositionType positionType = IO:: PositionType:: Begin) const -> int override 
- Gets the current read/write position within the stream.
- 
              auto SetPosition(int distance,
              IO::PositionType positionType = IO:: PositionType:: Begin) -> bool override 
- Sets the read/write position of the stream.
- auto GetAvailable() const -> int override
- Returns the number of bytes available for reading.
- 
              auto Read(void* pData,
              size_t nSize) -> int override 
- Reads bytes from the stream given by the input count 'nSize'.
- auto Flush() -> bool override
- Flush any non-empty stream write buffers.
- 
              auto Write(const void* pData,
              size_t nSize) -> int override 
- Writes bytes to the stream.
Protected variables
Function documentation
              void Resource::
            Sets the ResourceKey name that identifies this record.
| Parameters | |
|---|---|
| key | |
              IO::
            Returns the IStream used to access this record data.
| Returns | 
|---|
              IO::
            Returns one of enum AccessFlags.
This function also tells you if the stream is open, as a return value of zero means the stream is not open. It is not allowed that a stream
be open with no type of access.
              IO::
            Returns the error state of the stream.
Returns FileError::Success if OK, else an error code. This function is similar to 'errno' or a 'get last error' facility.
              IO::
            Returns the size of the stream, which is not the same as the size of bytes remaining to be read from the stream.
Returns (size_type)-2 if the stream is of indeterminate size. Returns (size_type)-1 (a.k.a. kSizeTypeError) upon error.
              bool Resource::
            Sets the size of the stream, if possible.
| Parameters | |
|---|---|
| size | |
It is debatable whether this function should be present in IStream or only in suclasses of StreamBase which are writable. For consistency with GetSize, we put the function here. But also consider that a SetSize function is not necessarily a data writing function, depending on the stream implementation.
              int Resource::
            Gets the current read/write position within the stream.
| Parameters | |
|---|---|
| positionType | |
The read and write positions of a stream must be the same value; you cannot have a read position that is different from a write position. However, a Stream subclass can provide such functionality if needed. Returns -1 upon error.
              bool Resource::
            Sets the read/write position of the stream.
| Parameters | |
|---|---|
| distance | |
| positionType | |
If the specified position is beyond the size of a fixed stream, the position is set to the end of the stream. A writable stream subclass may provide a policy whereby setting the position beyond the end of the stream results in an increase in the stream size.
              int Resource::
            Returns the number of bytes available for reading.
Returns (size_type)-1 (a.k.a. kSizeTypeError) upon error. This function is non-blocking; it should return immediately.
              int Resource::
            Reads bytes from the stream given by the input count 'nSize'.
| Parameters | |
|---|---|
| pData | |
| nSize | |
If less then nSize bytes are available, then those bytes will be read. Returns the number of bytes read. A return value of zero means that there were no bytes to be read or no bytes were requested to be read. A return value of zero means the end of file was reached. A return value > 0 but < 'nSize' is possible, and it does not necessarily mean that the end of the file was reached. Returns (size_type)-1 (a.k.a. kSizeTypeError) if there was an error. You can use this return value or IStream::GetState to determine the error. Input size values equal to (size_type)-1 (a.k.a. kSizeTypeError) are not valid input. Upon error, the stream pointer is at the position it was upon the error occurrence.
              bool Resource::
            Flush any non-empty stream write buffers.
If the return value is false, GetState will give the error code. This function implements the flushing as per the underlying file system. The behavior of the Flush function varies with the underlying platform.
A common use of Flush is write a file to disk immediately in order to prevent the file from being corrupted if the application crashes before the file is closed. However, on desktop platforms such as Windows this strategy is unnecesary, as the Windows OS file flush doesn't write the file to disk as might be expected. This actually is not a problem, because the Windows OS manages files outside the process and if your process crashes the OS will take care of safely closing the files. Only if the machine power is lost or if certain kinds of kernel-level crashes occur may you lose file data.