Skip to content

CHANGELOG.md

Latest commit

 

History

History
238 lines (178 loc) · 6.63 KB

CHANGELOG.md

File metadata and controls

238 lines (178 loc) · 6.63 KB
Raw

0.6.5

  • Update OrderedCollections compat

0.6.4

  • Support for directly reading from and writing to Vector{UInt8}
  • Fix for working with opened IOStream objects
  • Proper boundscheck for internal read_bytestring
  • Fixes for working in IOStream mode

0.6.3

  • Fix edge case in datatype loading (#692)
  • Add keyword argument constructors to Filters
  • Allow negative compression levels for the ZstdFilter

0.6.2

  • Fix an edge case for Vector{Union{}}

0.6.1

  • Improved backwards compat for lz4 compression
  • Docs improvements
  • updated compat bounds for filter pkgs

0.6.0

  • Breaking: Rework compression API. Expose filter approach of hdf5 broadly similar to HDF5.jl. See docs for description. The basic compress=true api still works but backend libraries were replaced.
  • Added new filter pkgs JLD2Lz4 and JLD2Bzip2 to enable lz4 and bzip2 filters without making them dependencies by default.

0.5.17

  • update hash implementation for internal types to eliminate downstream invalidations

0.5.16

  • Improve readmmap to work on more types and misaligned data
  • Make default iotype selection dynamic. Makes precompile files relocatable to win7. (#632)

0.5.15

  • Add full control over type mapping by providing a typemap function

0.5.14

  • Add support for Memory and MemoryRef

0.5.13

  • Make UnPack.jl compat into an extension
  • bumps lower compat for julia to v1.9

0.5.12

  • Fix precompile workload for loading on latest julia

0.5.11

  • Make Upgrade work for types that sit inside Union{..}

0.5.10

  • fix regression for UInt32
  • Deprecation: Do not rely on JLD2 to load a compression library. This feature will be removed in a future release. Instead explicitly add using statements into your scripts.

0.5.9

  • fix regression for Union{Bool,Nothing} array elements (#617)
  • fix printing issue in printtoc

0.5.8

  • Stop using Base.module_keys as it is removed on nightly

0.5.7

  • Fix edge case for uninitialized Vlens

0.5.6

  • Add Aqua tests and eliminate method ambiguities

0.5.5

  • Experimental support for writing to and reading from IO objects e.g. jldopen(io, "r")

0.5.4

  • Important correctness fix when storing very many equally sized objects that may get GC'ed while storing is in progress! (#603)

0.5.3

  • Bugfix for <: Function structs

0.5.2

  • Remove left-over warning
  • Restrict warning for storing Function objects to anonymous functions.

0.5.1

  • Bugfix and added test for bug introduced in v0.5.0

0.5.0

  • Improved encoding of committed datatypes. This fixes longstanding issues but new files will not be loaded correctly with JLD2 versions prior to v0.5.0. Existing files remain readable.
  • JLD2 now warns when storing structures containing <: Function objects.

0.4.53

  • Experimental: Slicing and inplace updating of array datasets
  • updated CI workflows
  • improve pretty printing of attribute header message
  • fix storing of datatype info for h5 compat

0.4.52

  • fix attribute loading
  • new features: readmmap ismmappable and allocate_early (api experimental)
  • adds Downgrade testing
  • new feature: disable committing datatypes. (restrict to h5 numbers, strings, and arrays)
  • internal cleanup
  • new experimental feature: reconstruct all committed types as NamedTuples

0.4.51

  • remove Unicode normalization support due to excessive performance loss
  • rework of header message internals
  • Implement addition and loading of attributes to datasets and groups
  • Add preliminary Dataset structure and API
  • cleanup tests
  • remove Pkg dependency
  • fix Upgrade for Singleton types

0.4.50

  • Don't hide exception data during loading and saving (#569)

0.4.49

  • update compat bounds
  • add support Zstdcompression

0.4.48

  • fix behaviour for unnormalized strings
  • add missing method for load_attributes
  • clean up using statements

0.4.47

  • fix loading structs with more than 256 fields (#558)

0.4.46

  • remove usage of IOBuffer internals

0.4.45

  • fix loading of unknown struct with UnionAll (unknown) type parameter (#538)

0.4.44

  • fix dispatch around Union{} as type parameter

0.4.43

  • do not try to rconvert #undef array elements (#529)

0.4.42

  • improvements in time-to-first save/load on julia v1.10 and up
  • discourage from using custom serialization to Array and stop using it internally as it cannot properly preserve object identity.
  • fix inline group show
  • fix a bug in group loading for the IOStream backend
  • reduce invalidations to zero

0.4.41

  • fix ntuple type with typevar length
  • fix OpaqueData test (HDF5 compat)

0.4.40

  • fix unitialized custom-serialized objects
  • allow serializing pkg-modules by name

0.4.39

  • backward compatibility support for windows 7 by changing the default IO type to IOStream (#509) (@HongBinYu-hub)

0.4.38

  • restrict default Dict encoding to Base implementations

0.4.37

  • Update Dict encoding for latest julia

0.4.36

  • compat bound for TranscodingStreams.jl

0.4.35

  • fix roundtrip of Type{} objects #484

0.4.34

  • reclose file when opening fails

0.4.33

  • fix Upgrade for parametric types
  • new type reconstruction when matching DataType cannot be found (eval-free)
  • new parallel_read keyword for creating stand-alone file handles for multithreaded file reading (@ejmeitz)

0.4.32

  • add experimental JLD2.readas function for customized reading of custom serialized objects (#468)

0.4.31

  • fix UInt32 truncation error for absurdly large array sizes
  • move test-files to a separate repo

0.4.30

  • allow loading compressed files during precompilation #446 (@marius311)

0.4.29

  • added Upgrade feature

0.4.28

  • compatibility to julia v1.9-dev (@eschnett)

0.4.26

  • fix identity relations with custom serialization

0.4.25

  • remove leftover debug statement

0.4.24

  • read-only support for JLD.jl files
  • read-only support for many HDF5 files. Most test files of HDF5.jl are covered
  • read Opaque bit fields
  • read some other string encodings
  • read big endian numbers
  • read typical chunking formats

0.4.23

  • Support for const fields in mutable structs

0.4.22

  • Fix reconstruction of partially initialized structs

0.4.21

  • Add explicit type mapping

0.4.20

  • TTFX improvements
  • Add a comment on jldsave (@BoundaryValueProblems)

0.4.19

  • Don't inline long ntuples

0.4.18

  • Bugfix for dynamic loading of CodecZlib

0.4.17

  • Automatically transform symbol keys in strings with a warning (@theogf)

0.4.16

  • Fix @save macro
  • Improve auto-conversion of convertible types

0.4.15

  • Implement Base.keytype for JLDFile and Group