optimize zip_findlast_entry and add benchmarks (#81)

.gitignore

@@ -5,6 +5,7 @@
 /.vscode/
 
 /test/Manifest.toml
+/benchmark/Manifest.toml
 /Manifest.toml
 fixture.tar.gz
 fixture

benchmark/Project.toml

@@ -0,0 +1,6 @@
+[deps]
+BenchmarkTools = "6e4b80f9-dd63-53aa-95a3-0cdb28fa8baf"
+Downloads = "f43a241f-c20a-4ad4-852c-f6b1247861c6"
+PkgBenchmark = "32113eaa-f34f-5b0d-bd6c-c81e245fc73d"
+Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
+ZipArchives = "49080126-0e18-4c2a-b176-c102e4b3760c"

benchmark/README.md

@@ -0,0 +1,11 @@
+# ZipArchives.jl benchmarks
+
+This directory contains benchmarks for ZipArchives. To run all the
+benchmarks, launch `julia --project=benchmark` and enter:
+
+``` julia
+using PkgBenchmark
+import ZipArchives
+
+benchmarkpkg(ZipArchives)
+```

benchmark/benchmarks.jl

@@ -0,0 +1,24 @@
+using BenchmarkTools
+using Random
+using ZipArchives
+
+const SUITE = BenchmarkGroup()
+rbench = SUITE["reading"] = BenchmarkGroup()
+
+sink = IOBuffer()
+names = String[]
+ZipWriter(sink) do w
+    for i in 1:2000
+        local name = String(rand(UInt8('a'):UInt8('z'), 100))
+        push!(names, name)
+        zip_writefile(w, name, rand(UInt8, 10000))
+    end
+end
+data = take!(sink)
+r = ZipReader(data)
+
+rbench["ZipReader"] = @benchmarkable ZipReader($(data))
+rbench["zip_findlast_entry nothing"] = @benchmarkable zip_findlast_entry($(r), $("abc"))
+rbench["zip_findlast_entry first"] = @benchmarkable zip_findlast_entry($(r), $(names[begin]))
+rbench["zip_findlast_entry last"] = @benchmarkable zip_findlast_entry($(r), $(names[end]))
+rbench["zip_readentry"] = @benchmarkable zip_readentry($(r), $(1000))

src/reader.jl

@@ -202,8 +202,9 @@ Return the index of the last entry with name `s` or `nothing` if not found.
 zip_findlast_entry(x::HasEntries, s::AbstractString)::Union{Nothing, Int} = zip_findlast_entry(x, String(s))
 function zip_findlast_entry(x::HasEntries, s::String)::Union{Nothing, Int}
     data = codeunits(s)
-    findlast(eachindex(x.entries)) do i
-        _name_view(x, i) == data
+    n = length(data)
+    findlast(x.entries) do e
+        n == length(e.name_range) && view(x.central_dir_buffer, e.name_range) == data
     end
 end