Commit b18e4184 authored by Rob Pike's avatar Rob Pike

document unicode, such as it is

R=rsc
DELTA=18  (9 added, 0 deleted, 9 changed)
OCL=25817
CL=25832
parent 6b8ac0a9
...@@ -10,14 +10,19 @@ ...@@ -10,14 +10,19 @@
// link in only the tables that are used by the program, // link in only the tables that are used by the program,
// etc. // etc.
// This package provides data and functions to test some properties of Unicode code points.
// It is rudimentary but will improve.
package unicode package unicode
// The representation of a range of Unicode code points. The range runs from Lo to Hi
// inclusive and has the specified stride.
type Range struct { type Range struct {
lo int; Lo int;
hi int; Hi int;
stride int; Stride int;
} }
// Upper is the set of Unicode upper case letters.
var Upper = []Range{ var Upper = []Range{
Range{0x0041, 0x005a, 1}, Range{0x0041, 0x005a, 1},
Range{0x00c0, 0x00d6, 1}, Range{0x00c0, 0x00d6, 1},
...@@ -150,6 +155,7 @@ var Upper = []Range{ ...@@ -150,6 +155,7 @@ var Upper = []Range{
Range{0x1d7ca, 0x1d7ca, 1}, Range{0x1d7ca, 0x1d7ca, 1},
} }
// Letter is the set of Unicode letters.
var Letter = []Range { var Letter = []Range {
Range{0x0041, 0x005a, 1}, Range{0x0041, 0x005a, 1},
Range{0x0061, 0x007a, 1}, Range{0x0061, 0x007a, 1},
...@@ -525,18 +531,19 @@ var Letter = []Range { ...@@ -525,18 +531,19 @@ var Letter = []Range {
Range{0x2f800, 0x2fa1d, 1}, Range{0x2f800, 0x2fa1d, 1},
} }
// Is tests whether rune is in the specified table of ranges.
func Is(ranges []Range, rune int) bool { func Is(ranges []Range, rune int) bool {
// common case: rune is ASCII or Latin-1 // common case: rune is ASCII or Latin-1
if rune < 0x100 { if rune < 0x100 {
for i := 0; i < len(ranges); i++ { for i := 0; i < len(ranges); i++ {
r := ranges[i]; r := ranges[i];
if rune > r.hi { if rune > r.Hi {
continue; continue;
} }
if rune < r.lo { if rune < r.Lo {
return false; return false;
} }
return (rune - r.lo) % r.stride == 0; return (rune - r.Lo) % r.Stride == 0;
} }
return false; return false;
} }
...@@ -547,10 +554,10 @@ func Is(ranges []Range, rune int) bool { ...@@ -547,10 +554,10 @@ func Is(ranges []Range, rune int) bool {
for lo < hi { for lo < hi {
m := lo + (hi - lo)/2; m := lo + (hi - lo)/2;
r := ranges[m]; r := ranges[m];
if r.lo <= rune && rune <= r.hi { if r.Lo <= rune && rune <= r.Hi {
return (rune - r.lo) % r.stride == 0; return (rune - r.Lo) % r.Stride == 0;
} }
if rune < r.lo { if rune < r.Lo {
hi = m; hi = m;
} else { } else {
lo = m+1; lo = m+1;
...@@ -559,10 +566,12 @@ func Is(ranges []Range, rune int) bool { ...@@ -559,10 +566,12 @@ func Is(ranges []Range, rune int) bool {
return false; return false;
} }
// IsLetter reports whether the rune is an upper case letter.
func IsUpper(rune int) bool { func IsUpper(rune int) bool {
return Is(Upper, rune); return Is(Upper, rune);
} }
// IsLetter reports whether the rune is a letter.
func IsLetter(rune int) bool { func IsLetter(rune int) bool {
return Is(Letter, rune); return Is(Letter, rune);
} }
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment