Merge pull request #11947 from JuliaLang/jq/countlines

Improve countlines() performance

base/datafmt.jl

@@ -13,28 +13,15 @@ export countlines, readdlm, readcsv, writedlm, writecsv
 const invalid_dlm = Char(0xfffffffe)
 const offs_chunk_size = 5000
 
-countlines(nameorfile) = countlines(nameorfile, '\n')
-function countlines(filename::AbstractString, eol::Char)
-    open(filename) do io
-        countlines(io, eol)
-    end
-end
-function countlines(io::IO, eol::Char)
-    if !isascii(eol)
-        throw(ArgumentError("only ASCII line terminators are supported"))
-    end
+countlines(f::AbstractString,eol::Char='\n') = open(io->countlines(io,eol),f)::Int
+function countlines(io::IO, eol::Char='\n')
+    isascii(eol) || throw(ArgumentError("only ASCII line terminators are supported"))
     a = Array(UInt8, 8192)
     nl = 0
-    preceded_by_eol = true
     while !eof(io)
         nb = readbytes!(io, a)
-        for i=1:nb
-            if Char(a[i]) == eol
-                preceded_by_eol = true
-            elseif preceded_by_eol
-                preceded_by_eol = false
-                nl+=1
-            end
+        @simd for i=1:nb
+            @inbounds nl += a[i] == eol
         end
     end
     nl

doc/stdlib/io-network.rst

@@ -243,7 +243,7 @@ General I/O
 
 .. function:: countlines(io,[eol::Char])
 
-   Read io until the end of the stream/file and count the number of non-empty lines. To specify a file pass the filename as the first
+   Read ``io`` until the end of the stream/file and count the number of lines. To specify a file pass the filename as the first
    argument. EOL markers other than '\\n' are supported by passing them as the second argument.
 
 .. function:: PipeBuffer()

test/readdlm.jl

@@ -1,5 +1,20 @@
 # This file is a part of Julia. License is MIT: http://julialang.org/license
 
+# countlines
+@test countlines(IOBuffer("\n")) == 1
+@test countlines(IOBuffer("\n"),'\r') == 0
+@test countlines(IOBuffer("\n\n\n\n\n\n\n\n\n\n")) == 10
+@test countlines(IOBuffer("\n \n \n \n \n \n \n \n \n \n")) == 10
+@test countlines(IOBuffer("\r\n \r\n \r\n \r\n \r\n")) == 5
+file = tempname()
+open(file,"w") do f
+    write(f,"Spiffy header\nspectacular first row\neven better 2nd row\nalmost done\n")
+end
+@test countlines(file) == 4
+@test countlines(file,'\r') == 0
+@test countlines(file,'\n') == 4
+rm(file)
+
 isequaldlm(m1, m2, t) = isequal(m1, m2) && (eltype(m1) == eltype(m2) == t)
 
 @test isequaldlm(readdlm(IOBuffer("1\t2\n3\t4\n5\t6\n")), [1. 2; 3 4; 5 6], Float64)