refactor: vendor StaticString implementation and remove StaticStrings.jl dependency (too many invalidations caused by StaticStrings.jl)

Author: Beforerr

Project.toml

@@ -12,7 +12,6 @@ LibDeflate = "9255714d-24a7-4b30-8ea3-d46a97f7e13b"
 Mmap = "a63ad114-7e13-5084-954f-fe012c677804"
 PrecompileTools = "aea7be01-6a6a-4083-8856-8a6e6704d82a"
 ResumableFunctions = "c5292f4c-5179-55e1-98c5-05642aab7184"
-StaticStrings = "4db0a0c5-418a-4e1d-8806-cb305fe13294"
 
 [weakdeps]
 CommonDataModel = "1fbeeb36-5f17-413c-809b-666fb144f157"
@@ -30,5 +29,4 @@ LibDeflate = "0.4.3"
 Mmap = "1"
 PrecompileTools = "1"
 ResumableFunctions = "1"
-StaticStrings = "0.2.6"
 julia = "1.10"

src/CommonDataFormat.jl

@@ -5,7 +5,6 @@ using Mmap
 using Dictionaries: Dictionary
 using ResumableFunctions
 using DiskArrays
-using StaticStrings: StaticString
 using Base.Threads
 using CodecZlib: GzipDecompressor, transcode
 using LibDeflate
@@ -22,6 +21,7 @@ export is_record_varying
 include("epochs.jl")
 include("enums.jl")
 include("types.jl")
+include("staticstring.jl")
 include("parsing.jl")
 include("decompress.jl")
 include("records/records.jl")

src/enums.jl

@@ -67,5 +67,5 @@ const type_map = Dict(
 
 function julia_type(cdf_type, num_elems)
     cdf_type = DataType(cdf_type)
-    return cdf_type in (CDF_CHAR, CDF_UCHAR) ? StaticString{Int(num_elems)} : type_map[cdf_type]
+    return cdf_type in (CDF_CHAR, CDF_UCHAR) ? StaticString{Int(num_elems), UInt8} : type_map[cdf_type]
 end

src/epochs.jl

@@ -149,6 +149,8 @@ function Base.show(io::IO, epoch::Epoch16)
 end
 
 Base.promote_rule(::Type{<:CDFDateTime}, ::Type{Dates.DateTime}) = Dates.DateTime
+Base.promote_rule(::Type{T}, ::Type{Dates.Date}) where {T <: CDFDateTime} = T
+# Comment out because of invalidation
 Base.convert(::Type{Dates.DateTime}, x::CDFDateTime) = Dates.DateTime(x)
 Base.bswap(x::Epoch) = Epoch(Base.bswap(x.instant))
 Base.bswap(x::Epoch16) = Epoch16(Base.bswap(x.seconds), Base.bswap(x.picoseconds))

src/parsing.jl

@@ -164,7 +164,7 @@ function validate_cdf_magic(magic_bytes)
 end
 
 _btye_swap!(data) = map!(ntoh, data, data)
-_btye_swap!(data::AbstractArray{StaticString{N}}) where {N} = data
+_btye_swap!(data::AbstractArray{<:StaticString{N}}) where {N} = data
 # function _btye_swap!(data::AbstractArray{TT2000})
 #     rd = reinterpret(Int64, data)
 #     return map!(ntoh, rd, rd)

src/staticstring.jl

@@ -0,0 +1,61 @@
+# https://github.com/mkitti/StaticStrings.jl
+# https://github.com/JuliaPy/PythonCall.jl/blob/main/src/Utils/Utils.jl
+
+struct StaticString{N, T} <: AbstractString
+    codeunits::NTuple{N, T}
+    StaticString{N, T}(codeunits::NTuple{N, T}) where {N, T} = new{N, T}(codeunits)
+end
+
+function Base.iterate(x::StaticString{N, UInt8}, i::Int = 1) where {N}
+    i > N && return
+    cs = x.codeunits
+    c = @inbounds cs[i]
+    if all(iszero, (cs[j] for j in i:N))
+        return
+    elseif (c & 0x80) == 0x00
+        return (reinterpret(Char, UInt32(c) << 24), i + 1)
+    elseif (c & 0x40) == 0x00
+        nothing
+    elseif (c & 0x20) == 0x00
+        if @inbounds (i ≤ N - 1) && ((cs[i + 1] & 0xC0) == 0x80)
+            return (
+                reinterpret(Char, (UInt32(cs[i]) << 24) | (UInt32(cs[i + 1]) << 16)),
+                i + 2,
+            )
+        end
+    elseif (c & 0x10) == 0x00
+        if @inbounds (i ≤ N - 2) && ((cs[i + 1] & 0xC0) == 0x80) && ((cs[i + 2] & 0xC0) == 0x80)
+            return (
+                reinterpret(
+                    Char,
+                    (UInt32(cs[i]) << 24) |
+                        (UInt32(cs[i + 1]) << 16) |
+                        (UInt32(cs[i + 2]) << 8),
+                ),
+                i + 3,
+            )
+        end
+    elseif (c & 0x08) == 0x00
+        if @inbounds (i ≤ N - 3) &&
+                ((cs[i + 1] & 0xC0) == 0x80) &&
+                ((cs[i + 2] & 0xC0) == 0x80) &&
+                ((cs[i + 3] & 0xC0) == 0x80)
+            return (
+                reinterpret(
+                    Char,
+                    (UInt32(cs[i]) << 24) |
+                        (UInt32(cs[i + 1]) << 16) |
+                        (UInt32(cs[i + 2]) << 8) |
+                        UInt32(cs[i + 3]),
+                ),
+                i + 4,
+            )
+        end
+    end
+    throw(StringIndexError(x, i))
+end
+
+function Base.String(x::StaticString{N, T}) where {N, T}
+    b = Base.StringVector(N)
+    return String(b .= x.codeunits)
+end

test/comprehensive_test.jl

@@ -1,15 +1,12 @@
+# Comprehensive test based on the Python CDFpp test.py
+# Tests all variables in a_cdf.cdf for expected shapes, types, values, and attributes
 using Test
 using CommonDataFormat
 import CommonDataFormat as CDF
 using Dates
 
 include("utils.jl")
 
-"""
-Comprehensive test based on the Python CDFpp test.py
-Tests all variables in a_cdf.cdf for expected shapes, types, values, and attributes
-"""
-
 # Expected variable definitions (translated from Python test.py)
 const EXPECTED_VARIABLES = Dict(
     "epoch" => (
@@ -129,6 +126,12 @@ ds = CDFDataset(file)
     @test Set(keys(ds)) == Set(keys(EXPECTED_VARIABLES))
 end
 
+@testset "StaticString" begin
+    using CommonDataFormat: StaticString
+    @test typeof(ds["var_string"][1]) == StaticString{16, UInt8}
+    @test String(ds["var_string"][1]) == "This is a string"
+end
+
 @testset "DateTime Conversions" begin
     for var in ("epoch", "epoch16", "tt2000")
         @testset "Variable: $var" begin