Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Faster creation of OrderedDict, revisited #221

Closed
wants to merge 19 commits into from
Closed
Show file tree
Hide file tree
Changes from 14 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions REQUIRE
Original file line number Diff line number Diff line change
@@ -1 +1,2 @@
julia 0.4
Compat 0.9.4
2 changes: 2 additions & 0 deletions src/DataStructures.jl
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,8 @@ __precompile__()

module DataStructures

using Compat

import Base: <, <=, ==, length, isempty, start, next, done,
show, dump, empty!, getindex, setindex!, get, get!,
in, haskey, keys, merge, copy, cat,
Expand Down
91 changes: 55 additions & 36 deletions src/ordered_dict.jl
Original file line number Diff line number Diff line change
Expand Up @@ -8,8 +8,12 @@ import Base: haskey, get, get!, getkey, delete!, push!, pop!, empty!,
hash, eltype, KeyIterator, ValueIterator, convert, copy,
merge

if VERSION >= v"0.5.0"
import Base: HasLength, HasShape, SizeUnknown, iteratorsize
end

"""
OrderedDict
OrderedDict

`OrderedDict`s are simply dictionaries whose entries have a particular order. The order
refers to insertion order, which allows deterministic iteration over the dictionary or set.
Expand All @@ -21,25 +25,28 @@ type OrderedDict{K,V} <: Associative{K,V}
ndel::Int
dirty::Bool

function OrderedDict()
new(zeros(Int32,16), Array(K,0), Array(V,0), 0, false)
end
OrderedDict() = new(zeros(Int32,16), Array(K,0), Array(V,0), 0, false)

function OrderedDict(kv)
h = OrderedDict{K,V}()
for (k,v) in kv
h[k] = v
end
return h
end
OrderedDict(p::Pair) = setindex!(OrderedDict{K,V}(), p.second, p.first)
function OrderedDict(ps::Pair...)
h = OrderedDict{K,V}()
sizehint!(h, length(ps))
for p in ps
h[p.first] = p.second
(n, slotsz) = get_n_slotsz(kv)
h = new(zeros(Int32,slotsz), Array(K,n), Array(V,n), 0, false)
if n > 0
i = 0
for (k,v) in kv
i = i + 1
index = ht_keyindex2(h, k)
@inbounds h.keys[i] = k
@inbounds h.vals[i] = v
@inbounds h.slots[-index] = i
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Closer, but there needs to be a little more defensive programming here. Since kv is an arbitrary set of key-value pairs, it's possible that a key is repeated, so you'll need to check the return value of index and respond accordingly. You should double check against setindex! to make sure there aren't any other corner cases.

end
else # Unknown length
for (k,v) in kv
h[k] = v
end
end
return h
end

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Minor nitpick: spaces at end of line.

function OrderedDict(d::OrderedDict{K,V})
if d.ndel > 0
rehash!(d)
Expand All @@ -52,24 +59,10 @@ OrderedDict() = OrderedDict{Any,Any}()
OrderedDict(kv::Tuple{}) = OrderedDict()
copy(d::OrderedDict) = OrderedDict(d)


# TODO: this can probably be simplified using `eltype` as a THT (Tim Holy trait)
# OrderedDict{K,V}(kv::Tuple{Vararg{Tuple{K,V}}}) = OrderedDict{K,V}(kv)
# OrderedDict{K }(kv::Tuple{Vararg{Tuple{K,Any}}}) = OrderedDict{K,Any}(kv)
# OrderedDict{V }(kv::Tuple{Vararg{Tuple{Any,V}}}) = OrderedDict{Any,V}(kv)
OrderedDict{K,V}(kv::Tuple{Vararg{Pair{K,V}}}) = OrderedDict{K,V}(kv)
OrderedDict{K}(kv::Tuple{Vararg{Pair{K}}}) = OrderedDict{K,Any}(kv)
OrderedDict{V}(kv::Tuple{Vararg{Pair{TypeVar(:K),V}}}) = OrderedDict{Any,V}(kv)
OrderedDict(kv::Tuple{Vararg{Pair}}) = OrderedDict{Any,Any}(kv)

OrderedDict{K,V}(kv::AbstractArray{Tuple{K,V}}) = OrderedDict{K,V}(kv)
OrderedDict{K,V}(kv::AbstractArray{Pair{K,V}}) = OrderedDict{K,V}(kv)
OrderedDict{K,V}(kv::Associative{K,V}) = OrderedDict{K,V}(kv)

OrderedDict{K,V}(ps::Pair{K,V}...) = OrderedDict{K,V}(ps)
OrderedDict{K}(ps::Pair{K}...,) = OrderedDict{K,Any}(ps)
OrderedDict{V}(ps::Pair{TypeVar(:K),V}...,) = OrderedDict{Any,V}(ps)
OrderedDict(ps::Pair...) = OrderedDict{Any,Any}(ps)

function OrderedDict(kv)
try
Expand All @@ -84,9 +77,35 @@ function OrderedDict(kv)
end
end

dict_with_eltype{K,V}(kv, ::Type{Tuple{K,V}}) = OrderedDict{K,V}(kv)
dict_with_eltype{K,V}(kv, ::Type{Pair{K,V}}) = OrderedDict{K,V}(kv)
dict_with_eltype(kv, t) = OrderedDict{Any,Any}(kv)
@static if VERSION >= v"0.5.0"
get_n_slotsz(kv) = get_n_slotsz(kv, iteratorsize(kv))
get_n_slotsz(kv, isz::SizeUnknown) = (0, 16)
function get_n_slotsz(kv, isz::Union{HasLength, HasShape})
n = length(kv)
slotsz = max(16, (n*3)>>1)
return n, slotsz
end
else
get_n_slotsz(kv) = (0, 16)
end

OrderedDict{K,V}(kv::Tuple{Vararg{Pair{K, V}}}) = OrderedDict{K, V}(kv)
OrderedDict{K}(kv::Tuple{Vararg{Pair{K}}}) = OrderedDict{K, Any}(kv)
OrderedDict{V}(kv::Tuple{Vararg{Pair{TypeVar(:K), V}}}) = OrderedDict{Any, V}(kv)
OrderedDict(kv::Tuple{Vararg{Pair}}) = OrderedDict{Any, Any}(kv)

OrderedDict{K,V}(kv::AbstractArray{Tuple{K, V}}) = OrderedDict{K, V}(kv)
OrderedDict{K,V}(kv::AbstractArray{Pair{K, V}}) = OrderedDict{K, V}(kv)
OrderedDict{K,V}(kv::Associative{K, V}) = OrderedDict{K, V}(kv)

OrderedDict{K,V}(ps::Pair{K, V}...) = OrderedDict{K, V}(ps)
OrderedDict{K}(ps::Pair{K}..., ) = OrderedDict{K, Any}(ps)
OrderedDict{V}(ps::Pair{TypeVar(:K), V}..., ) = OrderedDict{Any, V}(ps)
OrderedDict(ps::Pair...) = OrderedDict{Any, Any}(ps)

dict_with_eltype{K,V}(kv, ::Type{Tuple{K, V}}) = OrderedDict{K, V}(kv)
dict_with_eltype{K,V}(kv, ::Type{Pair{K, V}}) = OrderedDict{K, V}(kv)
dict_with_eltype(kv,t) = OrderedDict{Any, Any}(kv)

similar{K,V}(d::OrderedDict{K,V}) = OrderedDict{K,V}()

Expand Down Expand Up @@ -292,7 +311,7 @@ function setindex!{K,V}(h::OrderedDict{K,V}, v0, key0)
if !isequal(key,key0)
throw(ArgumentError("$key0 is not a valid key for type $K"))
end
v = convert(V, v0)
v = convert(V, v0)

index = ht_keyindex2(h, key)

Expand All @@ -316,7 +335,7 @@ function get!{K,V}(h::OrderedDict{K,V}, key0, default)

index > 0 && return h.vals[index]

v = convert(V, default)
v = convert(V, default)
_setindex!(h, v, key, -index)
return v
end
Expand All @@ -332,7 +351,7 @@ function get!{K,V}(default::Base.Callable, h::OrderedDict{K,V}, key0)
index > 0 && return h.vals[index]

h.dirty = false
v = convert(V, default())
v = convert(V, default())
if h.dirty
index = ht_keyindex2(h, key)
end
Expand Down
10 changes: 6 additions & 4 deletions test/test_ordered_dict.jl
Original file line number Diff line number Diff line change
Expand Up @@ -3,11 +3,13 @@
@test isa(OrderedDict(), OrderedDict{Any,Any})
@test isa(OrderedDict([(1,2.0)]), OrderedDict{Int,Float64})
@test isa(OrderedDict([("a",1),("b",2)]), OrderedDict{Compat.ASCIIString,Int})
@test isa(OrderedDict(Pair(1, 1.0)), OrderedDict{Int,Float64})
@test isa(OrderedDict(Pair(1, 1.0), Pair(2, 2.0)), OrderedDict{Int,Float64})
@test isa(OrderedDict(Pair(1, 1.0), Pair(2, 2.0), Pair(3, 3.0)), OrderedDict{Int,Float64})
@test isa(OrderedDict(1 => 1.0), OrderedDict{Int,Float64})
@test isa(OrderedDict(1 => 1.0, 2 => 2.0), OrderedDict{Int,Float64})
@test isa(OrderedDict(1 => 1.0, 2 => 2.0, 3 => 3.0), OrderedDict{Int,Float64})
@test isa(OrderedDict([1 => 1.0, 2 => 2.0, 3 => 3.0]), OrderedDict{Int,Float64})

# empty dictionary

d = OrderedDict{Char, Int}()
@test length(d) == 0
@test isempty(d)
Expand All @@ -20,7 +22,7 @@ empty!(d)
@test isempty(d)

# access, modification

d = OrderedDict{Char, Int}()
for c in 'a':'z'
d[c] = c-'a'+1
end
Expand Down