Navigation
API > API/Runtime > API/Runtime/Core > API/Runtime/Core/Serialization > API/Runtime/Core/Serialization/FArchive > API/Runtime/Core/Serialization/FArchive/SerializeCompressedNew
References
| Module | Core |
| Header | /Engine/Source/Runtime/Core/Public/Serialization/Archive.h |
| Include | #include "Serialization/Archive.h" |
| Source | /Engine/Source/Runtime/Core/Private/Serialization/Archive.cpp |
void SerializeCompressedNew
(
void * V,
int64 Length,
FName CompressionFormatToEncode,
FName CompressionFormatToDecodeOldV1Files,
ECompressionFlags Flags,
bool bTreatBufferAsFileReader,
int64 * OutPartialReadLength
)
Remarks
Serializes and compresses/ uncompresses data. This is a shared helper function for compression support.
call SerializeCompressedNew instead of SerializeCompressed Typically you should not serializing data compressed if it will be packaged or stored in the DDC. Prefer to allow the package/iostore system to do the compression for you instead.
SerializeCompressedNew can read existing data written by old SerializeCompressed calls.
Parameters
| Name | Description |
|---|---|
| V | Data pointer to serialize data from/ to |
| Length | Length of source data if we're saving, unused otherwise |
| CompressionFormatToEncode | Compression Format to use for encoding, can be changed freely without breaking compatibility |
| CompressionFormatToDecodeOldV1Files | Compression Format to decode old data with that didn't write compressor in header, cannot change, usually NAME_Zlib |
| Flags | Flags to control what method to use for [de]compression and optionally control memory vs speed when compressing |
| bTreatBufferAsFileReader | true if V is actually an FArchive, which is used when saving to read data - helps to avoid single huge allocations of source data |
| OutPartialReadLength | if not null, partial reads are allowed and the size is filled here |