clean up

Author: sl-solution

docs/src/man/gallery.md

@@ -135,7 +135,7 @@ julia> byrow(ds, fill!, :, with = byrow(ds, mean, :));
 > since in the original post there are some specifications for checking equality of values we define a customised equality function
 
 ```julia
-julia> a1 = Dataset(z1=[1,missing,3,4,5],x1=string.(3:7),z2=[2,missing,4,5,6],x2=[3,5,4,7,5])
+julia> a1 = Dataset(z1 = [1,missing,3,4,5], x1 = string.(3:7), z2 = [2,missing,4,5,6], x2 = [3,5,4,7,5])
 5×4 Dataset
  Row │ z1        x1        z2        x2       
      │ identity  identity  identity  identity
@@ -149,10 +149,10 @@ julia> a1 = Dataset(z1=[1,missing,3,4,5],x1=string.(3:7),z2=[2,missing,4,5,6],x2
 
 julia> eq(x, y) = isequal(x, y)
 julia> eq(x::String, y) = isequal(parse(Int,x), y)
-julia> eq(x, y::String) = isequal(x,parse(Int, y))
-julia> eq(x::String, y::String) = isequal(parse(Int, x),parse(Int, y))
+julia> eq(x, y::String) = isequal(x, parse(Int, y))
+julia> eq(x::String, y::String) = isequal(parse(Int, x), parse(Int, y))
 julia> eq(::Missing, ::Missing) = false
-julia> a1.result = reduce((x,y)-> x .|= byrow(a1, in, r"2", item = y, eq = eq), names(a1, r"1"), init = zeros(Bool, nrow(a1)))
+julia> a1.result = reduce((x,y) -> x .|= byrow(a1, in, r"2", item = y, eq = eq), names(a1, r"1"), init = zeros(Bool, nrow(a1)))
 5-element Vector{Bool}:
  1
  0

docs/src/man/tutorial.md

@@ -4,7 +4,7 @@ An extended version of this tutorial is available [here](https://github.com/sl-s
 
 ## Getting and loading the data
 
-We use the `DLMReader` package to load a csv file into `InMemoryDatasets`, `DLMReader` is a high-performance and versatile package for reading delimited files into `InMemoryDatasets` (there are many alternative packages in Julia ecosystem, users can use any of them to read data into `InMemoryDatasets`):
+We recommend the `DLMReader` package to load a csv file into `InMemoryDatasets`, `DLMReader` is a high-performance and versatile package for reading delimited files into `InMemoryDatasets` (there are many alternative packages in Julia ecosystem, users can use any of them to read data into `InMemoryDatasets`):
 
 
 ```julia

src/byrow/util.jl

@@ -122,10 +122,11 @@ end
 
 #TODO this doesn't work - since in writeshortest() a must be a vector{UInt8} but we pass a view of a matrix in DLMReader
 function write_vals!(a, pos, x::Union{Float16, Float32, Float64})
-    needed_space = Base.Ryu.neededdigits(typeof(x))
-    available_space = length(a)-pos+1
-    # needed_space > available_space && throw(ArgumentError("not enough space in buffer to write value into it"))
+    # needed_space = Base.Ryu.neededdigits(typeof(x))
+    # available_space = length(a)-pos+1
+    # # needed_space > available_space && throw(ArgumentError("not enough space in buffer to write value into it"))
     _writeshortest(a, pos, x)
+
 end
 
 function write_vals!(a, pos, x::Bool)

src/precompile/warmup.jl

@@ -34,7 +34,6 @@ function warmup()
     combine(gatherby(ds,1), Ref([1,2,3,7,8]) .=> [median, sort])
     combine(gatherby(ds,1), Ref([1,2,3,7,8]) .=> [sum, mean, length, maximum, minimum, var, std])
     combine(gatherby(ds,1), r"x1$" .=> [sum, mean, length, maximum, minimum, var, std])
-    IMD._gather_groups_hugeds_multicols(ds, 1:6, Val(Int32), threads = true)
 
     ds2 = ds[1:2, [1,3,7]]
     innerjoin(ds, ds2, on = [:x1, :x3, :x7])