flow like the river

BACK_BACK/node_modules/grapheme-breaker/test/GraphemeBreaker.coffee

@@ -0,0 +1,123 @@
+expect = require('chai').expect
+fs = require 'fs'
+punycode = require 'punycode'
+GraphemeBreaker = require '../src/GraphemeBreaker'
+
+describe 'GraphemeBreaker', ->    
+  it 'basic test', ->
+    broken = GraphemeBreaker.break('Z͑ͫ̓ͪ̂ͫ̽͏̴̙̤̞͉͚̯̞̠͍A̴̵̜̰͔ͫ͗͢L̠ͨͧͩ͘G̴̻͈͍͔̹̑͗̎̅͛́Ǫ̵̹̻̝̳͂̌̌͘!͖̬̰̙̗̿̋ͥͥ̂ͣ̐́́͜͞')
+    expect(broken).to.deep.equal(['Z͑ͫ̓ͪ̂ͫ̽͏̴̙̤̞͉͚̯̞̠͍', 'A̴̵̜̰͔ͫ͗͢', 'L̠ͨͧͩ͘', 'G̴̻͈͍͔̹̑͗̎̅͛́', 'Ǫ̵̹̻̝̳͂̌̌͘', '!͖̬̰̙̗̿̋ͥͥ̂ͣ̐́́͜͞'])
+    
+    
+  it 'nextBreak', ->
+    str = 'Z͑ͫ̓ͪ̂ͫ̽͏̴̙̤̞͉͚̯̞̠͍A̴̵̜̰͔ͫ͗͢L̠ͨͧͩ͘G̴̻͈͍͔̹̑͗̎̅͛́Ǫ̵̹̻̝̳͂̌̌͘!͖̬̰̙̗̿̋ͥͥ̂ͣ̐́́͜͞'
+    index = 0
+        
+    res = []
+    while (brk = GraphemeBreaker.nextBreak(str, index)) < str.length
+      res.push str.slice(index, brk)
+      index = brk
+      
+    res.push str.slice(index)
+    expect(res).to.deep.equal(['Z͑ͫ̓ͪ̂ͫ̽͏̴̙̤̞͉͚̯̞̠͍', 'A̴̵̜̰͔ͫ͗͢', 'L̠ͨͧͩ͘', 'G̴̻͈͍͔̹̑͗̎̅͛́', 'Ǫ̵̹̻̝̳͂̌̌͘', '!͖̬̰̙̗̿̋ͥͥ̂ͣ̐́́͜͞'])
+    
+  it 'nextBreak intermediate indexes', ->
+    str = 'Z͑ͫ̓ͪ̂ͫ̽͏̴̙̤̞͉͚̯̞̠͍A̴̵̜̰͔ͫ͗͢L̠ͨͧͩ͘G̴̻͈͍͔̹̑͗̎̅͛́Ǫ̵̹̻̝̳͂̌̌͘!͖̬̰̙̗̿̋ͥͥ̂ͣ̐́́͜͞'
+    breaks = {}
+    
+    for i in [-1...str.length] by 1
+      brk = GraphemeBreaker.nextBreak(str, i)
+      breaks[brk] = brk
+      
+    expect(Object.keys(breaks).map (b) -> breaks[b]).to.deep.equal([ 0, 19, 28, 34, 47, 58, 75 ])
+    
+  it 'previousBreak', ->
+    str = 'Z͑ͫ̓ͪ̂ͫ̽͏̴̙̤̞͉͚̯̞̠͍A̴̵̜̰͔ͫ͗͢L̠ͨͧͩ͘G̴̻͈͍͔̹̑͗̎̅͛́Ǫ̵̹̻̝̳͂̌̌͘!͖̬̰̙̗̿̋ͥͥ̂ͣ̐́́͜͞'
+    index = str.length
+        
+    res = []
+    while (brk = GraphemeBreaker.previousBreak(str, index)) > 0
+      res.push str.slice(brk, index)
+      index = brk
+      
+    res.push str.slice(0, index)
+    expect(res).to.deep.equal(['Z͑ͫ̓ͪ̂ͫ̽͏̴̙̤̞͉͚̯̞̠͍', 'A̴̵̜̰͔ͫ͗͢', 'L̠ͨͧͩ͘', 'G̴̻͈͍͔̹̑͗̎̅͛́', 'Ǫ̵̹̻̝̳͂̌̌͘', '!͖̬̰̙̗̿̋ͥͥ̂ͣ̐́́͜͞'].reverse())
+    
+  it 'previousBreak intermediate indexes', ->
+    str = 'Z͑ͫ̓ͪ̂ͫ̽͏̴̙̤̞͉͚̯̞̠͍A̴̵̜̰͔ͫ͗͢L̠ͨͧͩ͘G̴̻͈͍͔̹̑͗̎̅͛́Ǫ̵̹̻̝̳͂̌̌͘!͖̬̰̙̗̿̋ͥͥ̂ͣ̐́́͜͞'
+    breaks = {}
+    
+    for i in [str.length+1..0] by -1
+      brk = GraphemeBreaker.previousBreak(str, i)
+      breaks[brk] = brk
+      
+    expect(Object.keys(breaks).map (b) -> breaks[b]).to.deep.equal([ 0, 19, 28, 34, 47, 58, 75 ])
+    
+  it 'previousBreak handles astral characters (e.g. emoji)', ->
+    str = '😜🇺🇸👍'
+    
+    res = []
+    index = str.length
+    while (brk = GraphemeBreaker.previousBreak(str, index)) > 0
+      res.push str.slice(brk, index)
+      index = brk
+      
+    res.push str.slice(0, index)
+    expect(res).to.deep.equal([ '👍', '🇺🇸', '😜' ])
+    
+  it 'nextBreak handles astral characters (e.g. emoji)', ->
+    str = '😜🇺🇸👍'
+    
+    res = []
+    index = 0
+    while (brk = GraphemeBreaker.nextBreak(str, index)) < str.length
+      res.push str.slice(index, brk)
+      index = brk
+      
+    res.push str.slice(index)
+    expect(res).to.deep.equal([ '😜', '🇺🇸', '👍' ])
+  
+  it 'should pass all tests in GraphemeBreakTest.txt', ->
+    data = fs.readFileSync __dirname + '/GraphemeBreakTest.txt', 'utf8'
+    lines = data.split('\n')
+            
+    for line in lines
+      continue if not line or /^#/.test(line)
+  
+      [cols, comment] = line.split('#')
+      codePoints = cols.split(/\s*[×÷]\s*/).filter(Boolean).map (c) -> parseInt(c, 16)
+      str = punycode.ucs2.encode codePoints
+      
+      expected = cols.split(/\s*÷\s*/).filter(Boolean).map (c) ->
+        codes = c.split(/\s*×\s*/)
+        codes = codes.map (c) -> parseInt(c, 16)
+        punycode.ucs2.encode codes
+            
+      comment = comment.trim()
+      expect(GraphemeBreaker.break(str)).to.deep.equal expected, comment
+      expect(GraphemeBreaker.countBreaks(str)).to.equal expected.length, comment
+      
+  it 'should pass all tests in GraphemeBreakTest.txt in reverse', ->
+    data = fs.readFileSync __dirname + '/GraphemeBreakTest.txt', 'utf8'
+    lines = data.split('\n')
+            
+    for line in lines
+      continue if not line or /^#/.test(line)
+  
+      [cols, comment] = line.split('#')
+      codePoints = cols.split(/\s*[×÷]\s*/).filter(Boolean).map (c) -> parseInt(c, 16)
+      str = punycode.ucs2.encode codePoints
+      
+      expected = cols.split(/\s*÷\s*/).filter(Boolean).map (c) ->
+        codes = c.split(/\s*×\s*/)
+        codes = codes.map (c) -> parseInt(c, 16)
+        punycode.ucs2.encode codes
+        
+      res = []
+      index = str.length
+      while (brk = GraphemeBreaker.previousBreak(str, index)) > 0
+        res.push str.slice(brk, index)
+        index = brk
+      
+      res.push str.slice(0, index)
+      expect(res).to.deep.equal expected.reverse(), comment.trim()