Updates per quinnj's review

Author: ScottPJones

base/strings/parse.jl renamed to base/parse.jl

@@ -1,8 +1,7 @@
 # This file is a part of Julia. License is MIT: http://julialang.org/license
 
+include("strings/types.jl")
 include("strings/basic.jl")
 include("strings/search.jl")
-include("strings/parse.jl")
-include("strings/strutil.jl")
-include("strings/strio.jl")
-
+include("strings/util.jl")
+include("strings/io.jl")

base/string.jl

@@ -67,6 +67,38 @@ function length(s::AbstractString)
     end
 end
 
+## String comparison functions ##
+
+function cmp(a::AbstractString, b::AbstractString)
+    if a === b
+        return 0
+    end
+    i = start(a)
+    j = start(b)
+    while !done(a,i) && !done(b,i)
+        c, i = next(a,i)
+        d, j = next(b,j)
+        if c != d
+            return c < d ? -1 : +1
+        end
+    end
+    done(a,i) && !done(b,j) ? -1 :
+    !done(a,i) && done(b,j) ? +1 : 0
+end
+
+==(a::AbstractString, b::AbstractString) = cmp(a,b) == 0
+isless(a::AbstractString, b::AbstractString) = cmp(a,b) < 0
+
+# faster comparisons for byte strings and symbols
+
+cmp(a::ByteString, b::ByteString) = lexcmp(a.data, b.data)
+cmp(a::Symbol, b::Symbol) = Int(sign(ccall(:strcmp, Int32, (Cstring, Cstring), a, b)))
+
+==(a::ByteString, b::ByteString) = endof(a) == endof(b) && cmp(a,b) == 0
+isless(a::Symbol, b::Symbol) = cmp(a,b) < 0
+
+## Generic validation functions ##
+
 isvalid(s::DirectIndexString, i::Integer) = (start(s) <= i <= endof(s))
 function isvalid(s::AbstractString, i::Integer)
     i < 1 && return false
@@ -79,6 +111,8 @@ function isvalid(s::AbstractString, i::Integer)
     end
 end
 
+## Generic indexing functions ##
+
 prevind(s::DirectIndexString, i::Integer) = i-1
 prevind(s::AbstractArray   , i::Integer) = i-1
 nextind(s::DirectIndexString, i::Integer) = i+1

base/strings/basic.jl

@@ -1,27 +1,5 @@
 # This file is a part of Julia. License is MIT: http://julialang.org/license
 
-## String comparison functions ##
-
-function cmp(a::AbstractString, b::AbstractString)
-    if a === b
-        return 0
-    end
-    i = start(a)
-    j = start(b)
-    while !done(a,i) && !done(b,i)
-        c, i = next(a,i)
-        d, j = next(b,j)
-        if c != d
-            return c < d ? -1 : +1
-        end
-    end
-    done(a,i) && !done(b,j) ? -1 :
-    !done(a,i) && done(b,j) ? +1 : 0
-end
-
-==(a::AbstractString, b::AbstractString) = cmp(a,b) == 0
-isless(a::AbstractString, b::AbstractString) = cmp(a,b) < 0
-
 # starts with and ends with predicates
 
 function startswith(a::AbstractString, b::AbstractString)
@@ -52,14 +30,6 @@ function endswith(a::AbstractString, b::AbstractString)
 end
 endswith(str::AbstractString, chars::Chars) = !isempty(str) && str[end] in chars
 
-# faster comparisons for byte strings and symbols
-
-cmp(a::ByteString, b::ByteString) = lexcmp(a.data, b.data)
-cmp(a::Symbol, b::Symbol) = Int(sign(ccall(:strcmp, Int32, (Cstring, Cstring), a, b)))
-
-==(a::ByteString, b::ByteString) = endof(a) == endof(b) && cmp(a,b) == 0
-isless(a::Symbol, b::Symbol) = cmp(a,b) < 0
-
 startswith(a::ByteString, b::ByteString) = startswith(a.data, b.data)
 startswith(a::Vector{UInt8}, b::Vector{UInt8}) =
     (length(a) >= length(b) && ccall(:strncmp, Int32, (Ptr{UInt8}, Ptr{UInt8}, UInt), a, b, length(b)) == 0)

base/strings/strio.jl renamed to base/strings/io.jl

@@ -88,12 +88,12 @@ include("iterator.jl")
 include("osutils.jl")
 
 # strings & printing
-include("stringtypes.jl")
 include("char.jl")
 include("ascii.jl")
-include("unicode.jl")
 include("iobuffer.jl")
 include("string.jl")
+include("unicode.jl")
+include("parse.jl")
 include("shell.jl")
 include("regex.jl")
 include("base64.jl")

base/stringtypes.jl renamed to base/strings/types.jl

@@ -16,12 +16,12 @@ Upon return, `tests` is a vector of fully-expanded test names, and
 function choosetests(choices = [])
     testnames = [
         "linalg", "core", "keywordargs", "numbers", "printf",
-        "char", "string", "stringtypes", "triplequote", "unicode",
+        "char", "string", "triplequote", "unicode",
         "dates", "dict", "hashing", "remote", "iobuffer", "staged",
         "arrayops", "tuple", "subarray", "reduce", "reducedim", "random",
         "abstractarray", "intfuncs", "simdloop", "blas", "sparse",
         "bitarray", "copy", "math", "fastmath", "functional",
-        "operators", "path", "ccall",
+        "operators", "path", "ccall", "parse",
         "bigint", "sorting", "statistics", "spawn", "backtrace",
         "priorityqueue", "file", "mmap", "version", "resolve",
         "pollfd", "mpfr", "broadcast", "complex", "socket",

base/strings/strutil.jl renamed to base/strings/util.jl

@@ -146,3 +146,122 @@ macro test999_str(args...); args; end
 @test parseall("a = \$\nb") == Expr(:block, Expr(:(=), :a, :$), :b)
 @test parseall(":(a = &\nb)") == Expr(:quote, Expr(:(=), :a, Expr(:&, :b)))
 @test parseall(":(a = \$\nb)") == Expr(:quote, Expr(:(=), :a, Expr(:$, :b)))
+
+# integer parsing
+@test is(parse(Int32,"0",36),Int32(0))
+@test is(parse(Int32,"1",36),Int32(1))
+@test is(parse(Int32,"9",36),Int32(9))
+@test is(parse(Int32,"A",36),Int32(10))
+@test is(parse(Int32,"a",36),Int32(10))
+@test is(parse(Int32,"B",36),Int32(11))
+@test is(parse(Int32,"b",36),Int32(11))
+@test is(parse(Int32,"F",36),Int32(15))
+@test is(parse(Int32,"f",36),Int32(15))
+@test is(parse(Int32,"Z",36),Int32(35))
+@test is(parse(Int32,"z",36),Int32(35))
+
+@test parse(Int,"0") == 0
+@test parse(Int,"-0") == 0
+@test parse(Int,"1") == 1
+@test parse(Int,"-1") == -1
+@test parse(Int,"9") == 9
+@test parse(Int,"-9") == -9
+@test parse(Int,"10") == 10
+@test parse(Int,"-10") == -10
+@test parse(Int64,"3830974272") == 3830974272
+@test parse(Int64,"-3830974272") == -3830974272
+@test parse(Int,'3') == 3
+@test parse(Int,'3', 8) == 3
+
+parsebin(s) = parse(Int,s,2)
+parseoct(s) = parse(Int,s,8)
+parsehex(s) = parse(Int,s,16)
+
+@test parsebin("0") == 0
+@test parsebin("-0") == 0
+@test parsebin("1") == 1
+@test parsebin("-1") == -1
+@test parsebin("10") == 2
+@test parsebin("-10") == -2
+@test parsebin("11") == 3
+@test parsebin("-11") == -3
+@test parsebin("1111000011110000111100001111") == 252645135
+@test parsebin("-1111000011110000111100001111") == -252645135
+
+@test parseoct("0") == 0
+@test parseoct("-0") == 0
+@test parseoct("1") == 1
+@test parseoct("-1") == -1
+@test parseoct("7") == 7
+@test parseoct("-7") == -7
+@test parseoct("10") == 8
+@test parseoct("-10") == -8
+@test parseoct("11") == 9
+@test parseoct("-11") == -9
+@test parseoct("72") == 58
+@test parseoct("-72") == -58
+@test parseoct("3172207320") == 434704080
+@test parseoct("-3172207320") == -434704080
+
+@test parsehex("0") == 0
+@test parsehex("-0") == 0
+@test parsehex("1") == 1
+@test parsehex("-1") == -1
+@test parsehex("9") == 9
+@test parsehex("-9") == -9
+@test parsehex("a") == 10
+@test parsehex("-a") == -10
+@test parsehex("f") == 15
+@test parsehex("-f") == -15
+@test parsehex("10") == 16
+@test parsehex("-10") == -16
+@test parsehex("0BADF00D") == 195948557
+@test parsehex("-0BADF00D") == -195948557
+@test parse(Int64,"BADCAB1E",16) == 3135023902
+@test parse(Int64,"-BADCAB1E",16) == -3135023902
+@test parse(Int64,"CafeBabe",16) == 3405691582
+@test parse(Int64,"-CafeBabe",16) == -3405691582
+@test parse(Int64,"DeadBeef",16) == 3735928559
+@test parse(Int64,"-DeadBeef",16) == -3735928559
+
+@test parse(Int,"2\n") == 2
+@test parse(Int,"   2 \n ") == 2
+@test parse(Int," 2 ") == 2
+@test parse(Int,"2 ") == 2
+@test parse(Int," 2") == 2
+@test parse(Int,"+2\n") == 2
+@test parse(Int,"-2") == -2
+@test_throws ArgumentError parse(Int,"   2 \n 0")
+@test_throws ArgumentError parse(Int,"2x")
+@test_throws ArgumentError parse(Int,"-")
+
+# multibyte spaces
+@test parse(Int, "3\u2003\u202F") == 3
+@test_throws ArgumentError parse(Int, "3\u2003\u202F,")
+
+@test parse(Int,'a') == 10
+@test_throws ArgumentError parse(Int,typemax(Char))
+
+@test parse(Int,"1234") == 1234
+@test parse(Int,"0x1234") == 0x1234
+@test parse(Int,"0o1234") == 0o1234
+@test parse(Int,"0b1011") == 0b1011
+@test parse(Int,"-1234") == -1234
+@test parse(Int,"-0x1234") == -Int(0x1234)
+@test parse(Int,"-0o1234") == -Int(0o1234)
+@test parse(Int,"-0b1011") == -Int(0b1011)
+
+## FIXME: #4905, do these tests for Int128/UInt128!
+for T in (Int8, Int16, Int32, Int64)
+    @test parse(T,string(typemin(T))) == typemin(T)
+    @test parse(T,string(typemax(T))) == typemax(T)
+    @test_throws OverflowError parse(T,string(big(typemin(T))-1))
+    @test_throws OverflowError parse(T,string(big(typemax(T))+1))
+end
+
+for T in (UInt8,UInt16,UInt32,UInt64)
+    @test parse(T,string(typemin(T))) == typemin(T)
+    @test parse(T,string(typemax(T))) == typemax(T)
+    @test_throws ArgumentError parse(T,string(big(typemin(T))-1))
+    @test_throws OverflowError parse(T,string(big(typemax(T))+1))
+end

base/sysimg.jl

@@ -1,7 +1,7 @@
 # This file is a part of Julia. License is MIT: http://julialang.org/license
 
 include("strings/basic.jl")
+include("strings/types.jl")
 include("strings/search.jl")
-include("strings/parse.jl")
-include("strings/strutil.jl")
-include("strings/strio.jl")
+include("strings/util.jl")
+include("strings/io.jl")