Adds method for splitting data into periods

.github/workflows/CI.yml

@@ -15,9 +15,8 @@ jobs:
       fail-fast: false
       matrix:
         version:
-          - "1.6.7"  # LTS
-          - "1.6"
-          - "1"    # Latest Release
+          - "1.10.7" # LTS
+          - "1" # Latest Release
         os:
           - ubuntu-latest
         arch:
@@ -56,7 +55,7 @@ jobs:
       - uses: actions/checkout@v2
       - uses: julia-actions/setup-julia@latest
         with:
-          version: '1'
+          version: "1"
       - run: |
           git config --global user.name name
           git config --global user.email email

docs/src/split.md

@@ -124,3 +124,14 @@ using MarketData
 tail(cl)
 tail(cl, 3)
 ```
+
+## Splitting by period
+
+Splitting data by a given function, e.g. `Dates.day` into periods.
+
+```@repl
+using TimeSeries
+using MarketData
+
+split(cl, Dates.day)
+```

src/TimeSeries.jl

@@ -12,12 +12,12 @@ using Tables
 using PrettyTables: pretty_table
 
 export TimeArray, AbstractTimeSeries,
-       when, from, to, findwhen, timestamp, values, colnames, meta, head, tail,
-       lag, lead, diff, percentchange, moving, upto,
-       uniformspaced, uniformspace, dropnan,
-       basecall,
-       merge, collapse,
-       readtimearray, writetimearray
+    when, from, to, findwhen, timestamp, values, colnames, meta, head, tail,
+    lag, lead, diff, percentchange, moving, upto,
+    uniformspaced, uniformspace, dropnan,
+    basecall,
+    merge, collapse,
+    readtimearray, writetimearray
 
 # modify.jl
 export rename, rename!

src/split.jl

@@ -7,17 +7,17 @@ when(ta::TimeArray, period::Function, t::String) =
 
 # from, to ######################
 
-from(ta::TimeArray{T, N, D}, d::D) where {T, N, D} =
+from(ta::TimeArray{T,N,D}, d::D) where {T,N,D} =
     length(ta) == 0 ? ta :
-        d < timestamp(ta)[1] ? ta :
-        d > timestamp(ta)[end] ? ta[1:0] :
-        ta[searchsortedfirst(timestamp(ta), d):end]
+    d < timestamp(ta)[1] ? ta :
+    d > timestamp(ta)[end] ? ta[1:0] :
+    ta[searchsortedfirst(timestamp(ta), d):end]
 
-to(ta::TimeArray{T, N, D}, d::D) where {T, N, D} =
+to(ta::TimeArray{T,N,D}, d::D) where {T,N,D} =
     length(ta) == 0 ? ta :
-        d < timestamp(ta)[1] ? ta[1:0] :
-        d > timestamp(ta)[end] ? ta :
-        ta[1:searchsortedlast(timestamp(ta), d)]
+    d < timestamp(ta)[1] ? ta[1:0] :
+    d > timestamp(ta)[end] ? ta :
+    ta[1:searchsortedlast(timestamp(ta), d)]
 
 ###### findall ##################
 
@@ -43,7 +43,7 @@ findwhen(ta::TimeArray{Bool,1}) = timestamp(ta)[findall(values(ta))]
     end
 end
 
- @generated function tail(ta::TimeArray{T,N}, n::Int=6) where {T,N}
+@generated function tail(ta::TimeArray{T,N}, n::Int=6) where {T,N}
     new_values = (N == 1) ? :(values(ta)[start:end]) : :(values(ta)[start:end, :])
 
     quote
@@ -58,3 +58,35 @@ end
 Base.first(ta::TimeArray) = head(ta, 1)
 
 Base.last(ta::TimeArray) = tail(ta, 1)
+
+
+"""
+    split(data::TimeSeries.TimeArray, period::Function)
+
+Split `data` by `period` function, returns a vector of `TimeSeries.TimeArray`.
+
+## Arguments
+
+- `data::TimeSeries.TimeArray`: Data to split
+- `period::Function`: Function, e.g. `Dates.day` that is used to split the `data`.
+"""
+Base.split(data::TimeSeries.TimeArray, period::Function) = Iterators.map(i -> data[i], _split(TimeSeries.timestamp(data), period))
+
+function _split(ts::AbstractVector{D}, period::Function) where {D<:TimeType}
+    m = length(ts)
+    idx = UnitRange{Int}[]
+    if !isempty(ts)
+        sizehint!(idx, m)
+        t0 = period(ts[1])
+        j = 1
+        for i in 1:(m-1)
+            t1 = period(ts[i+1])
+            t0 == t1 && continue
+            push!(idx, j:i)
+            j = i + 1
+            t0 = t1
+        end
+        push!(idx, j:m)
+    end
+    return Iterators.map(i -> ts[i], idx)
+end

test/split.jl

@@ -1,121 +1,129 @@
 using Dates
 using Test
-
 using MarketData
-
 using TimeSeries
 
-
 @testset "split" begin
 
-
-@testset "find methods" begin
-    @testset "find returns correct row numbers array" begin
-        @test timestamp(cl[findall(cl .> op)])[1] == Date(2000, 1, 3)
-        @test length(findall(cl .> op))           == 244
+    @testset "find methods" begin
+        @testset "find returns correct row numbers array" begin
+            @test timestamp(cl[findall(cl .> op)])[1] == Date(2000, 1, 3)
+            @test length(findall(cl .> op)) == 244
+        end
+
+        @testset "findwhen returns correct Dates array" begin
+            @test findwhen(cl .> op)[2] == Date(2000, 1, 5)
+            @test length(findwhen(cl .> op)) == 244
+        end
+
+        @testset "findall(f::Function, ta)" begin
+            @test findall(cl .> 100) == findall(x -> x > 100, cl)
+            @test findall(cl .> 100) == findall(x -> x[4] > 100, ohlc)
+        end
     end
 
-    @testset "findwhen returns correct Dates array" begin
-       @test findwhen(cl .> op)[2]      == Date(2000, 1, 5)
-       @test length(findwhen(cl .> op)) == 244
+    @testset "split date operations" begin
+        @testset "from and to correctly subset non-zero and zero-length time arrays" begin
+            @test length(from(cl, Date(2001, 12, 28))) == 2
+            @test length(from(cl, Date(2002, 1, 1))) == 0
+            @test length(from(from(cl, Date(2002, 1, 1)), Date(2012, 1, 1))) == 0
+
+            @test length(to(cl, Date(2000, 1, 4))) == 2
+            @test length(to(cl, Date(1999, 1, 4))) == 0
+            @test length(to(to(cl, Date(1999, 1, 4)), Date(1912, 1, 1))) == 0
+        end
+
+        @testset "when method correctly subset" begin
+            @test timestamp(when(cl, day, 4))[1] == Date(2000, 1, 4)
+            @test timestamp(when(cl, dayname, "Friday"))[1] == Date(2000, 1, 7)
+            @test timestamp(when(cl, week, 5))[1] == Date(2000, 1, 31)
+            @test timestamp(when(cl, month, 5))[1] == Date(2000, 5, 1)
+            @test timestamp(when(cl, monthname, "June"))[1] == Date(2000, 6, 1)
+            @test timestamp(when(cl, year, 2001))[1] == Date(2001, 1, 2)
+            @test timestamp(when(cl, dayofweek, 1))[1] == Date(2000, 1, 3)
+            # all the days in the nth week of each month
+            @test timestamp(when(cl, dayofweekofmonth, 5))[1] == Date(2000, 1, 31)
+            @test timestamp(when(cl, dayofyear, 365))[1] == Date(2001, 12, 31)
+            @test timestamp(when(cl, quarterofyear, 4))[1] == Date(2000, 10, 2)
+            @test timestamp(when(cl, dayofquarter, 1))[1] == Date(2001, 10, 1)
+        end
     end
 
-    @testset "findall(f::Function, ta)" begin
-        @test findall(cl .> 100) == findall(x -> x > 100, cl)
-        @test findall(cl .> 100) == findall(x -> x[4] > 100, ohlc)
+    @testset "head, tail, first and last methods" begin
+        @testset "head, tail, first and last methods work with default n value on single column TimeArray" begin
+            @test length(head(cl, 6)) == 6
+            @test timestamp(head(cl)) == [Date(2000, 1, 3), Date(2000, 1, 4), Date(2000, 1, 5),
+                Date(2000, 1, 6), Date(2000, 1, 7), Date(2000, 1, 10)]
+            @test values(head(cl)) == [111.94, 102.5, 104.0, 95.0, 99.5, 97.75]
+
+            @test length(tail(cl, 6)) == 6
+            @test timestamp(tail(cl)) == [Date(2001, 12, 21), Date(2001, 12, 24), Date(2001, 12, 26),
+                Date(2001, 12, 27), Date(2001, 12, 28), Date(2001, 12, 31)]
+            @test values(tail(cl)) == [21.0, 21.36, 21.49, 22.07, 22.43, 21.9]
+
+            @test length(first(cl)) == 1
+            @test timestamp(first(cl))[1] == Date(2000, 1, 3)
+            @test values(first(cl))[1] == 111.94
+            @test meta(first(cl)) == "AAPL"
+
+            @test length(last(cl)) == 1
+            @test timestamp(last(cl))[1] == Date(2001, 12, 31)
+            @test values(last(cl))[1] == 21.9
+            @test meta(last(cl)) == "AAPL"
+        end
+
+        @testset "head, tail, first and last methods work with default n value on multi column TimeArray" begin
+            @test length(head(ohlc)) == 6
+            @test values(head(ohlc, 1)) == [104.88 112.5 101.69 111.94]
+
+            @test length(tail(ohlc)) == 6
+            @test values(tail(ohlc, 1)) == [22.51 22.66 21.83 21.9]
+
+            @test length(first(ohlc)) == 1
+            @test timestamp(first(ohlc))[1] == Date(2000, 1, 3)
+            @test values(first(ohlc)) == [104.88 112.5 101.69 111.94]
+            @test meta(first(ohlc)) == "AAPL"
+
+            @test length(last(ohlc)) == 1
+            @test timestamp(last(ohlc))[1] == Date(2001, 12, 31)
+            @test values(last(ohlc)) == [22.51 22.66 21.83 21.9]
+            @test meta(last(ohlc)) == "AAPL"
+        end
+
+        @testset "head, tail, first and last methods work with custom periods on single column TimeArray" begin
+            @test length(head(cl, 2)) == 2
+            @test length(head(cl, 500)) == length(cl)
+            @test length(tail(cl, 2)) == 2
+            @test length(tail(cl, 500)) == length(cl)
+
+            @test length(first(cl)) == 1
+            @test length(last(cl)) == 1
+        end
+
+        @testset "head, tail, first and last methods work with custom periods on multi column TimeArray" begin
+            @test length(head(ohlc, 2)) == 2
+            @test length(head(ohlc, 500)) == length(ohlc)
+            @test length(tail(ohlc, 2)) == 2
+            @test length(tail(ohlc, 500)) == length(ohlc)
+
+            @test length(first(ohlc)) == 1
+            @test length(last(ohlc)) == 1
+        end
     end
-end
-
-
-@testset "split date operations" begin
-    @testset "from and to correctly subset non-zero and zero-length time arrays" begin
-        @test length(from(cl, Date(2001,12,28))) == 2
-        @test length(from(cl, Date(2002,1,1)))   == 0
-        @test length(from(from(cl, Date(2002,1,1)), Date(2012,1,1))) == 0
 
-        @test length(to(cl, Date(2000,1,4)))     == 2
-        @test length(to(cl, Date(1999,1,4)))     == 0
-        @test length(to(to(cl, Date(1999,1,4)), Date(1912,1,1))) == 0
+    @testset "split period" begin
+        for period in [day, week, month, year]
+            for cl_ in split(cl, period)
+                @test allequal(period.(timestamp(cl_)))
+            end
+        end
+        @test length(split(cl, day)) == 500
+        @test length(split(cl, week)) == 105
+        @test length(split(cl, month)) == 24
+        @test length(split(cl, year)) == 2
+
+        # test empty timearray
+        @test length(split(to(cl, Date(2000)), week)) == 0
     end
 
-    @testset "when method correctly subset" begin
-        @test timestamp(when(cl, day, 4))[1]              == Date(2000,1,4)
-        @test timestamp(when(cl, dayname, "Friday"))[1]   == Date(2000,1,7)
-        @test timestamp(when(cl, week, 5))[1]             == Date(2000,1,31)
-        @test timestamp(when(cl, month, 5))[1]            == Date(2000,5,1)
-        @test timestamp(when(cl, monthname, "June"))[1]   == Date(2000,6,1)
-        @test timestamp(when(cl, year, 2001))[1]          == Date(2001,1,2)
-        @test timestamp(when(cl, dayofweek, 1))[1]        == Date(2000,1,3)
-        # all the days in the nth week of each month
-        @test timestamp(when(cl, dayofweekofmonth, 5))[1] == Date(2000,1,31)
-        @test timestamp(when(cl, dayofyear, 365))[1]      == Date(2001,12,31)
-        @test timestamp(when(cl, quarterofyear, 4))[1]    == Date(2000,10,2)
-        @test timestamp(when(cl, dayofquarter, 1))[1]     == Date(2001,10,1)
-    end
-end
-
-
-@testset "head, tail, first and last methods" begin
-    @testset "head, tail, first and last methods work with default n value on single column TimeArray" begin
-        @test length(head(cl,6))  == 6
-        @test timestamp(head(cl)) == [Date(2000,1,3), Date(2000,1,4), Date(2000,1,5),
-                                      Date(2000,1,6), Date(2000,1,7), Date(2000,1,10)]
-        @test values(head(cl))    == [111.94, 102.5, 104.0, 95.0, 99.5, 97.75]
-
-        @test length(tail(cl,6))  == 6
-        @test timestamp(tail(cl)) == [Date(2001,12,21), Date(2001,12,24), Date(2001,12,26),
-                                      Date(2001,12,27), Date(2001,12,28), Date(2001,12,31)]
-        @test values(tail(cl))    == [21.0, 21.36, 21.49, 22.07, 22.43, 21.9]
-
-        @test length(first(cl))       == 1
-        @test timestamp(first(cl))[1] == Date(2000,1,3)
-        @test values(first(cl))[1]    == 111.94
-        @test meta(first(cl))         == "AAPL"
-
-        @test length(last(cl))       == 1
-        @test timestamp(last(cl))[1] == Date(2001,12,31)
-        @test values(last(cl))[1]    == 21.9
-        @test meta(last(cl))         == "AAPL"
-    end
-
-    @testset "head, tail, first and last methods work with default n value on multi column TimeArray" begin
-        @test length(head(ohlc))         == 6
-        @test values(head(ohlc, 1))      == [104.88 112.5 101.69 111.94]
-
-        @test length(tail(ohlc))         == 6
-        @test values(tail(ohlc, 1))      == [22.51 22.66 21.83 21.9]
-
-        @test length(first(ohlc))       == 1
-        @test timestamp(first(ohlc))[1] == Date(2000,1,3)
-        @test values(first(ohlc))       == [104.88 112.5 101.69 111.94]
-        @test meta(first(ohlc))         == "AAPL"
-
-        @test length(last(ohlc))       == 1
-        @test timestamp(last(ohlc))[1] == Date(2001,12,31)
-        @test values(last(ohlc))       == [22.51 22.66 21.83 21.9]
-        @test meta(last(ohlc))         == "AAPL"
-    end
-
-    @testset "head, tail, first and last methods work with custom periods on single column TimeArray" begin
-        @test length(head(cl, 2))   == 2
-        @test length(head(cl, 500)) == length(cl)
-        @test length(tail(cl, 2))   == 2
-        @test length(tail(cl, 500)) == length(cl)
-
-        @test length(first(cl))     == 1
-        @test length(last(cl))      == 1
-    end
-
-    @testset "head, tail, first and last methods work with custom periods on multi column TimeArray" begin
-        @test length(head(ohlc, 2))   == 2
-        @test length(head(ohlc, 500)) == length(ohlc)
-        @test length(tail(ohlc, 2))   == 2
-        @test length(tail(ohlc, 500)) == length(ohlc)
-
-        @test length(first(ohlc))     == 1
-        @test length(last(ohlc))      == 1
-    end
-end
-
-
 end  # @testset "split"