Skip to content

JuliaIO/JLD2.jl

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

JLD2

CI codecov.io JLD2 Downloads Aqua QA

JLD2 saves and loads Julia data structures in a format comprising a subset of HDF5, without any dependency on the HDF5 C library. JLD2 is able to read most HDF5 files created by other HDF5 implementations supporting HDF5 File Format Specification Version 3.0 (i.e. libhdf5 1.10 or later) and similarly those should be able to read the files that JLD2 produces. JLD2 provides read-only support for files created with the JLD package.

Reading and writing data

save and load functions

The save and load functions, provided by FileIO, provide a mechanism to read and write data from a JLD2 file. To use these functions, you may either write using FileIO or using JLD2. FileIO will determine the correct package automatically.

The save function accepts an AbstractDict yielding the key/value pairs, where the key is a string representing the name of the dataset and the value represents its contents:

using FileIO
save("example.jld2", Dict("hello" => "world", "foo" => :bar))

The save function can also accept the dataset names and contents as arguments:

save("example.jld2", "hello", "world", "foo", :bar)

When using the save function, the file extension must be .jld2, since the extension .jld currently belongs to the previous JLD package.

If called with a filename argument only, the load function loads all datasets from the given file into a Dict:

load("example.jld2") # -> Dict{String,Any}("hello" => "world", "foo" => :bar)

If called with a single dataset name, load returns the contents of that dataset from the file:

load("example.jld2", "hello") # -> "world"

If called with multiple dataset names, load returns the contents of the given datasets as a tuple:

load("example.jld2", "hello", "foo") # -> ("world", :bar)

A new interface: jldsave

jldsave makes use of julia's keyword argument syntax to store files, thus leveraging the parser and not having to rely on macros. The new interface can be imported with using JLD2. To use it, write

using JLD2

x = 1
y = 2
z = 42

# The simplest case:
jldsave("example.jld2"; x, y, z)
# it is equivalent to 
jldsave("example.jld2"; x=x, y=y, z=z)

# You can assign new names selectively
jldsave("example.jld2"; x, a=y, z)

# and if you want to confuse your future self and everyone else, do
jldsave("example.jld2"; z=x, x=y, y=z)

In the above examples, ; after the filename is important. Compression and non-default IO types may be set via positional arguments like:

jldopen("example.jld2", "w"; compress = true) do f
    f["large_array"] = zeros(10000)
end

File interface

It is also possible to interact with JLD2 files using a file-like interface. The jldopen function accepts a file name and an argument specifying how the file should be opened:

using JLD2

f = jldopen("example.jld2", "r")  # open read-only (default)
f = jldopen("example.jld2", "r+") # open read/write, failing if no file exists
f = jldopen("example.jld2", "w")  # open read/write, overwriting existing file
f = jldopen("example.jld2", "a+") # open read/write, preserving contents of existing file or creating a new file

Data can be written to the file using write(f, "name", data) or f["name"] = data, or read from the file using read(f, "name") or f["name"]. When you are done with the file, remember to call close(f).

Like open, jldopen also accepts a function as the first argument, permitting do-block syntax:

jldopen("example.jld2", "w") do file
    file["bigdata"] = randn(5)
end

Groups

It is possible to construct groups within a JLD2 file, which may or may not be useful for organizing your data. You can create groups explicitly:

jldopen("example.jld2", "w") do file
    mygroup = JLD2.Group(file, "mygroup")
    mygroup["mystuff"] = 42
end

or implicitly, by saving a variable with a name containing slashes as path delimiters:

jldopen("example.jld2", "w") do file
    file["mygroup/mystuff"] = 42
end
# or save("example.jld2", "mygroup/mystuff", 42)

Both of these examples yield the same group structure, which you can see at the REPL:

julia> file = jldopen("example.jld2", "r")
JLDFile /Users/simon/example.jld2 (read-only)
 β””β”€πŸ“‚ mygroup
    β””β”€πŸ”’ mystuff

Similarly, you can access groups directly:

jldopen("example.jld2", "r") do file
    @assert file["mygroup"]["mystuff"] == 42
end

or using slashes as path delimiters:

@assert load("example.jld2", "mygroup/mystuff") == 42

When loading files with nested groups these will be unrolled into paths by default but yield nested dictionaries but with the nested keyword argument.

load("example.jld2") # -> Dict("mygroup/mystuff" => 42)
load("example.jld2"; nested=true) # -> Dict("mygroup" => Dict("mystuff" => 42))

Custom Serialization

The API is simple enough, to enable custom serialization for your type A you define a new type e.g. ASerialization that contains the fields you want to store and define JLD2.writeas(::Type{A}) = ASerialization. Internally JLD2 will call Base.convert when writing and loading, so you need to make sure to extend that for your type.

struct A
    x::Int
end

struct ASerialization
    x::Vector{Int}
end

JLD2.writeas(::Type{A}) = ASerialization
Base.convert(::Type{ASerialization}, a::A) = ASerialization([a.x])
Base.convert(::Type{A}, a::ASerialization) = A(only(a.x))

If you do not want to overload Base.convert then you can also define

JLD2.wconvert(::Type{ASerialization}, a::A) = ASerialization([a.x])
JLD2.rconvert(::Type{A}, a::ASerialization) = A(only(a.x))

instead. This may be particularly relevant when types are involved that are not your own.

struct B
    x::Float64
end

JLD2.writeas(::Type{B}) = Float64
JLD2.wconvert(::Type{Float64}, b::B) = b.x
JLD2.rconvert(::Type{B}, x::Float64) = B(x)

arr = [B(rand()) for i=1:10]

jldsave("test.jld2"; arr)

In this example JLD2 converts the array of B structs to a plain Vector{Float64} prior to storing to disk.

Unpack.jl API

When additionally loading the UnPack.jl package, its @unpack and @pack! macros can be used to quickly save and load data from the file-like interface. Example:

using UnPack
file = jldopen("example.jld2", "w")
x, y = rand(2)

@pack! file = x, y # equivalent to file["x"] = x; file["y"] = y
@unpack x, y = file # equivalent to x = file["x"]; y = file["y"]

The group file_group = Group(file, "mygroup") can be accessed with the same file-like interface as the "full" struct.