\r
'use strict';\r
\r
-// Character types for symbols from 0000 to 00FF.\r
-var baseTypes = [\r
- 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'S', 'B', 'S', 'WS',\r
- 'B', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN',\r
- 'BN', 'BN', 'B', 'B', 'B', 'S', 'WS', 'ON', 'ON', 'ET', 'ET', 'ET', 'ON',\r
- 'ON', 'ON', 'ON', 'ON', 'ON', 'CS', 'ON', 'CS', 'ON', 'EN', 'EN', 'EN', 'EN',\r
- 'EN', 'EN', 'EN', 'EN', 'EN', 'EN', 'ON', 'ON', 'ON', 'ON', 'ON', 'ON', 'ON',\r
- 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
- 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'ON', 'ON', 'ON', 'ON',\r
- 'ON', 'ON', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
- 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'ON', 'ON',\r
- 'ON', 'ON', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'B', 'BN', 'BN', 'BN', 'BN',\r
- 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN',\r
- 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'CS', 'ON', 'ET', 'ET',\r
- 'ET', 'ET', 'ON', 'ON', 'ON', 'ON', 'L', 'ON', 'ON', 'ON', 'ON', 'ON', 'ET',\r
- 'ET', 'EN', 'EN', 'ON', 'L', 'ON', 'ON', 'ON', 'EN', 'L', 'ON', 'ON', 'ON',\r
- 'ON', 'ON', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
- 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'ON', 'L', 'L', 'L', 'L',\r
- 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
- 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'ON', 'L', 'L',\r
- 'L', 'L', 'L', 'L', 'L', 'L'\r
-];\r
-\r
-// Character types for symbols from 0600 to 06FF\r
-var arabicTypes = [\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'CS',\r
- 'AL', 'ON', 'ON', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'NSM',\r
- 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM',\r
- 'NSM', 'NSM', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AN', 'AN', 'AN',\r
- 'AN', 'AN', 'AN', 'AN', 'AN', 'AN', 'AN', 'ET', 'AN', 'AN', 'AL', 'AL', 'AL',\r
- 'NSM', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'NSM',\r
- 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM',\r
- 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'ON', 'NSM', 'NSM', 'NSM',\r
- 'NSM', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
- 'AL', 'AL', 'AL', 'AL', 'AL', 'AL'\r
-];\r
-\r
-function bidi(text, startLevel) {\r
+var bidi = (function bidiClosure() {\r
+ // Character types for symbols from 0000 to 00FF.\r
+ var baseTypes = [\r
+ 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'S', 'B', 'S', 'WS',\r
+ 'B', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN',\r
+ 'BN', 'BN', 'B', 'B', 'B', 'S', 'WS', 'ON', 'ON', 'ET', 'ET', 'ET', 'ON',\r
+ 'ON', 'ON', 'ON', 'ON', 'ON', 'CS', 'ON', 'CS', 'ON', 'EN', 'EN', 'EN',\r
+ 'EN', 'EN', 'EN', 'EN', 'EN', 'EN', 'EN', 'ON', 'ON', 'ON', 'ON', 'ON',\r
+ 'ON', 'ON', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
+ 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'ON', 'ON',\r
+ 'ON', 'ON', 'ON', 'ON', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
+ 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
+ 'L', 'ON', 'ON', 'ON', 'ON', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'B', 'BN',\r
+ 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN',\r
+ 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN', 'BN',\r
+ 'BN', 'CS', 'ON', 'ET', 'ET', 'ET', 'ET', 'ON', 'ON', 'ON', 'ON', 'L', 'ON',\r
+ 'ON', 'ON', 'ON', 'ON', 'ET', 'ET', 'EN', 'EN', 'ON', 'L', 'ON', 'ON', 'ON',\r
+ 'EN', 'L', 'ON', 'ON', 'ON', 'ON', 'ON', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
+ 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
+ 'L', 'ON', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
+ 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L',\r
+ 'L', 'L', 'L', 'ON', 'L', 'L', 'L', 'L', 'L', 'L', 'L', 'L'\r
+ ];\r
+\r
+ // Character types for symbols from 0600 to 06FF\r
+ var arabicTypes = [\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'CS', 'AL', 'ON', 'ON', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM',\r
+ 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AN', 'AN', 'AN', 'AN', 'AN', 'AN', 'AN', 'AN', 'AN',\r
+ 'AN', 'ET', 'AN', 'AN', 'AL', 'AL', 'AL', 'NSM', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM',\r
+ 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'NSM', 'ON', 'NSM',\r
+ 'NSM', 'NSM', 'NSM', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL',\r
+ 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL', 'AL'\r
+ ];\r
+\r
function isOdd(i) {\r
return (i & 1) != 0;\r
}\r
}\r
}\r
\r
- var str = text.str;\r
- var strLength = str.length;\r
- if (strLength == 0)\r
- return str;\r
+ return (function bidi(text, startLevel) {\r
+ var str = text.str;\r
+ var strLength = str.length;\r
+ if (strLength == 0)\r
+ return str;\r
+\r
+ // get types, fill arrays\r
+\r
+ var chars = new Array(strLength);\r
+ var types = new Array(strLength);\r
+ var oldtypes = new Array(strLength);\r
+ var numBidi = 0;\r
+\r
+ for (var i = 0; i < strLength; ++i) {\r
+ chars[i] = str.charAt(i);\r
+\r
+ var charCode = str.charCodeAt(i);\r
+ var charType = 'L';\r
+ if (charCode <= 0x00ff)\r
+ charType = baseTypes[charCode];\r
+ else if (0x0590 <= charCode && charCode <= 0x05f4)\r
+ charType = 'R';\r
+ else if (0x0600 <= charCode && charCode <= 0x06ff)\r
+ charType = arabicTypes[charCode & 0xff];\r
+ else if (0x0700 <= charCode && charCode <= 0x08AC)\r
+ charType = 'AL';\r
+\r
+ if (charType == 'R' || charType == 'AL' || charType == 'AN')\r
+ numBidi++;\r
+\r
+ oldtypes[i] = types[i] = charType;\r
+ }\r
\r
- // get types, fill arrays\r
+ // detect the bidi method\r
+ // if there are no rtl characters then no bidi needed\r
+ // if less than 30% chars are rtl then string is primarily ltr\r
+ // if more than 30% chars are rtl then string is primarily rtl\r
+ if (numBidi == 0) {\r
+ text.direction = 'ltr';\r
+ return str;\r
+ }\r
\r
- var chars = new Array(strLength);\r
- var types = new Array(strLength);\r
- var oldtypes = new Array(strLength);\r
- var numBidi = 0;\r
+ if (startLevel == -1) {\r
+ if ((strLength / numBidi) < 0.3) {\r
+ text.direction = 'ltr';\r
+ startLevel = 0;\r
+ } else {\r
+ text.direction = 'rtl';\r
+ startLevel = 1;\r
+ }\r
+ }\r
\r
- for (var i = 0; i < strLength; ++i) {\r
- chars[i] = str.charAt(i);\r
+ var levels = new Array(strLength);\r
\r
- var charCode = str.charCodeAt(i);\r
- var charType = 'L';\r
- if (charCode <= 0x00ff)\r
- charType = baseTypes[charCode];\r
- else if (0x0590 <= charCode && charCode <= 0x05f4)\r
- charType = 'R';\r
- else if (0x0600 <= charCode && charCode <= 0x06ff)\r
- charType = arabicTypes[charCode & 0xff];\r
- else if (0x0700 <= charCode && charCode <= 0x08AC)\r
- charType = 'AL';\r
+ for (var i = 0; i < strLength; ++i) {\r
+ levels[i] = startLevel;\r
+ }\r
\r
- if (charType == 'R' || charType == 'AL' || charType == 'AN')\r
- numBidi++;\r
+ var diffChars = new Array(strLength);\r
+ var diffLevels = new Array(strLength);\r
+ var diffTypes = new Array(strLength);\r
\r
- oldtypes[i] = types[i] = charType;\r
- }\r
+ /*\r
+ X1-X10: skip most of this, since we are NOT doing the embeddings.\r
+ */\r
\r
- // detect the bidi method\r
- // if there are no rtl characters then no bidi needed\r
- // if less than 30% chars are rtl then string is primarily ltr\r
- // if more than 30% chars are rtl then string is primarily rtl\r
- if (numBidi == 0) {\r
- text.direction = 'ltr';\r
- return str;\r
- }\r
+ var e = isOdd(startLevel) ? 'R' : 'L';\r
+ var sor = e;\r
+ var eor = sor;\r
\r
- if (startLevel == -1) {\r
- if ((strLength / numBidi) < 0.3) {\r
- text.direction = 'ltr';\r
- startLevel = 0;\r
- } else {\r
- text.direction = 'rtl';\r
- startLevel = 1;\r
+ /*\r
+ W1. Examine each non-spacing mark (NSM) in the level run, and change the\r
+ type of the NSM to the type of the previous character. If the NSM is at the\r
+ start of the level run, it will get the type of sor.\r
+ */\r
+\r
+ var lastType = sor;\r
+ for (var i = 0; i < strLength; ++i) {\r
+ if (types[i] == 'NSM')\r
+ types[i] = lastType;\r
+ else\r
+ lastType = types[i];\r
}\r
- }\r
\r
- var levels = new Array(strLength);\r
+ /*\r
+ W2. Search backwards from each instance of a European number until the\r
+ first strong type (R, L, AL, or sor) is found. If an AL is found, change\r
+ the type of the European number to Arabic number.\r
+ */\r
\r
- for (var i = 0; i < strLength; ++i) {\r
- levels[i] = startLevel;\r
- }\r
+ var lastType = sor;\r
+ for (var i = 0; i < strLength; ++i) {\r
+ var t = types[i];\r
+ if (t == 'EN')\r
+ types[i] = (lastType == 'AL') ? 'AN' : 'EN';\r
+ else if (t == 'R' || t == 'L' || t == 'AL')\r
+ lastType = t;\r
+ }\r
\r
- var diffChars = new Array(strLength);\r
- var diffLevels = new Array(strLength);\r
- var diffTypes = new Array(strLength);\r
-\r
- /*\r
- X1-X10: skip most of this, since we are NOT doing the embeddings.\r
- */\r
-\r
- var e = isOdd(startLevel) ? 'R' : 'L';\r
- var sor = e;\r
- var eor = sor;\r
-\r
- /*\r
- W1. Examine each non-spacing mark (NSM) in the level run, and change the type\r
- of the NSM to the type of the previous character. If the NSM is at the start\r
- of the level run, it will get the type of sor.\r
- */\r
-\r
- var lastType = sor;\r
- for (var i = 0; i < strLength; ++i) {\r
- if (types[i] == 'NSM')\r
- types[i] = lastType;\r
- else\r
- lastType = types[i];\r
- }\r
+ /*\r
+ W3. Change all ALs to R.\r
+ */\r
\r
- /*\r
- W2. Search backwards from each instance of a European number until the first\r
- strong type (R, L, AL, or sor) is found. If an AL is found, change the type\r
- of the European number to Arabic number.\r
- */\r
-\r
- var lastType = sor;\r
- for (var i = 0; i < strLength; ++i) {\r
- var t = types[i];\r
- if (t == 'EN')\r
- types[i] = (lastType == 'AL') ? 'AN' : 'EN';\r
- else if (t == 'R' || t == 'L' || t == 'AL')\r
- lastType = t;\r
- }\r
+ for (var i = 0; i < strLength; ++i) {\r
+ var t = types[i];\r
+ if (t == 'AL')\r
+ types[i] = 'R';\r
+ }\r
\r
- /*\r
- W3. Change all ALs to R.\r
- */\r
+ /*\r
+ W4. A single European separator between two European numbers changes to a\r
+ European number. A single common separator between two numbers of the same\r
+ type changes to that type:\r
+ */\r
\r
- for (var i = 0; i < strLength; ++i) {\r
- var t = types[i];\r
- if (t == 'AL')\r
- types[i] = 'R';\r
- }\r
+ for (var i = 1; i < strLength - 1; ++i) {\r
+ if (types[i] == 'ES' && types[i - 1] == 'EN' && types[i + 1] == 'EN')\r
+ types[i] = 'EN';\r
+ if (types[i] == 'CS' && (types[i - 1] == 'EN' || types[i - 1] == 'AN') &&\r
+ types[i + 1] == types[i - 1])\r
+ types[i] = types[i - 1];\r
+ }\r
\r
- /*\r
- W4. A single European separator between two European numbers changes to a\r
- European number. A single common separator between two numbers of the same\r
- type changes to that type:\r
- */\r
-\r
- for (var i = 1; i < strLength - 1; ++i) {\r
- if (types[i] == 'ES' && types[i - 1] == 'EN' && types[i + 1] == 'EN')\r
- types[i] = 'EN';\r
- if (types[i] == 'CS' && (types[i - 1] == 'EN' || types[i - 1] == 'AN') &&\r
- types[i + 1] == types[i - 1])\r
- types[i] = types[i - 1];\r
- }\r
+ /*\r
+ W5. A sequence of European terminators adjacent to European numbers changes\r
+ to all European numbers:\r
+ */\r
\r
- /*\r
- W5. A sequence of European terminators adjacent to European numbers changes\r
- to all European numbers:\r
- */\r
-\r
- for (var i = 0; i < strLength; ++i) {\r
- if (types[i] == 'EN') {\r
- // do before\r
- for (var j = i - 1; j >= 0; --j) {\r
- if (types[j] != 'ET')\r
- break;\r
- types[j] = 'EN';\r
- }\r
- // do after\r
- for (var j = i + 1; j < strLength; --j) {\r
- if (types[j] != 'ET')\r
- break;\r
- types[j] = 'EN';\r
+ for (var i = 0; i < strLength; ++i) {\r
+ if (types[i] == 'EN') {\r
+ // do before\r
+ for (var j = i - 1; j >= 0; --j) {\r
+ if (types[j] != 'ET')\r
+ break;\r
+ types[j] = 'EN';\r
+ }\r
+ // do after\r
+ for (var j = i + 1; j < strLength; --j) {\r
+ if (types[j] != 'ET')\r
+ break;\r
+ types[j] = 'EN';\r
+ }\r
}\r
}\r
- }\r
\r
- /*\r
- W6. Otherwise, separators and terminators change to Other Neutral:\r
- */\r
+ /*\r
+ W6. Otherwise, separators and terminators change to Other Neutral:\r
+ */\r
\r
- for (var i = 0; i < strLength; ++i) {\r
- var t = types[i];\r
- if (t == 'WS' || t == 'ES' || t == 'ET' || t == 'CS')\r
- types[i] = 'ON';\r
- }\r
+ for (var i = 0; i < strLength; ++i) {\r
+ var t = types[i];\r
+ if (t == 'WS' || t == 'ES' || t == 'ET' || t == 'CS')\r
+ types[i] = 'ON';\r
+ }\r
\r
- /*\r
- W7. Search backwards from each instance of a European number until the first\r
- strong type (R, L, or sor) is found. If an L is found, then change the type\r
- of the European number to L.\r
- */\r
-\r
- var lastType = sor;\r
- for (var i = 0; i < strLength; ++i) {\r
- var t = types[i];\r
- if (t == 'EN')\r
- types[i] = (lastType == 'L') ? 'L' : 'EN';\r
- else if (t == 'R' || t == 'L')\r
- lastType = t;\r
- }\r
+ /*\r
+ W7. Search backwards from each instance of a European number until the\r
+ first strong type (R, L, or sor) is found. If an L is found, then change\r
+ the type of the European number to L.\r
+ */\r
\r
- /*\r
- N1. A sequence of neutrals takes the direction of the surrounding strong text\r
- if the text on both sides has the same direction. European and Arabic numbers\r
- are treated as though they were R. Start-of-level-run (sor) and\r
- end-of-level-run (eor) are used at level run boundaries.\r
- */\r
-\r
- for (var i = 0; i < strLength; ++i) {\r
- if (types[i] == 'ON') {\r
- var end = findUnequal(types, i + 1, 'ON');\r
- var before = sor;\r
- if (i > 0)\r
- before = types[i - 1];\r
- var after = eor;\r
- if (end + 1 < strLength)\r
- after = types[end + 1];\r
- if (before != 'L')\r
- before = 'R';\r
- if (after != 'L')\r
- after = 'R';\r
- if (before == after)\r
- setValues(types, i, end, before);\r
- i = end - 1; // reset to end (-1 so next iteration is ok)\r
+ var lastType = sor;\r
+ for (var i = 0; i < strLength; ++i) {\r
+ var t = types[i];\r
+ if (t == 'EN')\r
+ types[i] = (lastType == 'L') ? 'L' : 'EN';\r
+ else if (t == 'R' || t == 'L')\r
+ lastType = t;\r
}\r
- }\r
-\r
- /*\r
- N2. Any remaining neutrals take the embedding direction.\r
- */\r
\r
- for (var i = 0; i < strLength; ++i) {\r
- if (types[i] == 'ON')\r
- types[i] = e;\r
- }\r
+ /*\r
+ N1. A sequence of neutrals takes the direction of the surrounding strong\r
+ text if the text on both sides has the same direction. European and Arabic\r
+ numbers are treated as though they were R. Start-of-level-run (sor) and\r
+ end-of-level-run (eor) are used at level run boundaries.\r
+ */\r
\r
- /*\r
- I1. For all characters with an even (left-to-right) embedding direction,\r
- those of type R go up one level and those of type AN or EN go up two levels.\r
- I2. For all characters with an odd (right-to-left) embedding direction, those\r
- of type L, EN or AN go up one level.\r
- */\r
-\r
- for (var i = 0; i < strLength; ++i) {\r
- var t = types[i];\r
- if (isEven(levels[i])) {\r
- if (t == 'R') {\r
- levels[i] += 1;\r
- } else if (t == 'AN' || t == 'EN') {\r
- levels[i] += 2;\r
+ for (var i = 0; i < strLength; ++i) {\r
+ if (types[i] == 'ON') {\r
+ var end = findUnequal(types, i + 1, 'ON');\r
+ var before = sor;\r
+ if (i > 0)\r
+ before = types[i - 1];\r
+ var after = eor;\r
+ if (end + 1 < strLength)\r
+ after = types[end + 1];\r
+ if (before != 'L')\r
+ before = 'R';\r
+ if (after != 'L')\r
+ after = 'R';\r
+ if (before == after)\r
+ setValues(types, i, end, before);\r
+ i = end - 1; // reset to end (-1 so next iteration is ok)\r
}\r
- } else { // isOdd, so\r
- if (t == 'L' || t == 'AN' || t == 'EN') {\r
- levels[i] += 1;\r
+ }\r
+\r
+ /*\r
+ N2. Any remaining neutrals take the embedding direction.\r
+ */\r
+\r
+ for (var i = 0; i < strLength; ++i) {\r
+ if (types[i] == 'ON')\r
+ types[i] = e;\r
+ }\r
+\r
+ /*\r
+ I1. For all characters with an even (left-to-right) embedding direction,\r
+ those of type R go up one level and those of type AN or EN go up two\r
+ levels.\r
+ I2. For all characters with an odd (right-to-left) embedding direction,\r
+ those of type L, EN or AN go up one level.\r
+ */\r
+\r
+ for (var i = 0; i < strLength; ++i) {\r
+ var t = types[i];\r
+ if (isEven(levels[i])) {\r
+ if (t == 'R') {\r
+ levels[i] += 1;\r
+ } else if (t == 'AN' || t == 'EN') {\r
+ levels[i] += 2;\r
+ }\r
+ } else { // isOdd, so\r
+ if (t == 'L' || t == 'AN' || t == 'EN') {\r
+ levels[i] += 1;\r
+ }\r
}\r
}\r
- }\r
\r
- /*\r
- L1. On each line, reset the embedding level of the following characters to\r
- the paragraph embedding level:\r
-\r
- segment separators,\r
- paragraph separators,\r
- any sequence of whitespace characters preceding a segment separator or\r
- paragraph separator, and any sequence of white space characters at the end\r
- of the line.\r
- */\r
-\r
- // don't bother as text is only single line\r
-\r
- /*\r
- L2. From the highest level found in the text to the lowest odd level on each\r
- line, reverse any contiguous sequence of characters that are at that level or\r
- higher.\r
- */\r
-\r
- // find highest level & lowest odd level\r
-\r
- var highestLevel = -1;\r
- var lowestOddLevel = 99;\r
- for (var i = 0, ii = levels.length; i < ii; ++i) {\r
- var level = levels[i];\r
- if (highestLevel < level)\r
- highestLevel = level;\r
- if (lowestOddLevel > level && isOdd(level))\r
- lowestOddLevel = level;\r
- }\r
+ /*\r
+ L1. On each line, reset the embedding level of the following characters to\r
+ the paragraph embedding level:\r
+\r
+ segment separators,\r
+ paragraph separators,\r
+ any sequence of whitespace characters preceding a segment separator or\r
+ paragraph separator, and any sequence of white space characters at the end\r
+ of the line.\r
+ */\r
+\r
+ // don't bother as text is only single line\r
\r
- // now reverse between those limits\r
+ /*\r
+ L2. From the highest level found in the text to the lowest odd level on\r
+ each line, reverse any contiguous sequence of characters that are at that\r
+ level or higher.\r
+ */\r
+\r
+ // find highest level & lowest odd level\r
\r
- for (var level = highestLevel; level >= lowestOddLevel; --level) {\r
- // find segments to reverse\r
- var start = -1;\r
+ var highestLevel = -1;\r
+ var lowestOddLevel = 99;\r
for (var i = 0, ii = levels.length; i < ii; ++i) {\r
- if (levels[i] < level) {\r
- if (start >= 0) {\r
- reverseValues(chars, start, i);\r
- start = -1;\r
+ var level = levels[i];\r
+ if (highestLevel < level)\r
+ highestLevel = level;\r
+ if (lowestOddLevel > level && isOdd(level))\r
+ lowestOddLevel = level;\r
+ }\r
+\r
+ // now reverse between those limits\r
+\r
+ for (var level = highestLevel; level >= lowestOddLevel; --level) {\r
+ // find segments to reverse\r
+ var start = -1;\r
+ for (var i = 0, ii = levels.length; i < ii; ++i) {\r
+ if (levels[i] < level) {\r
+ if (start >= 0) {\r
+ reverseValues(chars, start, i);\r
+ start = -1;\r
+ }\r
+ } else if (start < 0) {\r
+ start = i;\r
}\r
- } else if (start < 0) {\r
- start = i;\r
+ }\r
+ if (start >= 0) {\r
+ reverseValues(chars, start, levels.length);\r
}\r
}\r
- if (start >= 0) {\r
- reverseValues(chars, start, levels.length);\r
- }\r
- }\r
\r
- /*\r
- L3. Combining marks applied to a right-to-left base character will at this\r
- point precede their base character. If the rendering engine expects them to\r
- follow the base characters in the final display process, then the ordering of\r
- the marks and the base character must be reversed.\r
- */\r
+ /*\r
+ L3. Combining marks applied to a right-to-left base character will at this\r
+ point precede their base character. If the rendering engine expects them to\r
+ follow the base characters in the final display process, then the ordering\r
+ of the marks and the base character must be reversed.\r
+ */\r
\r
- // don't bother for now\r
+ // don't bother for now\r
\r
- /*\r
- L4. A character that possesses the mirrored property as specified by\r
- Section 4.7, Mirrored, must be depicted by a mirrored glyph if the resolved\r
- directionality of that character is R.\r
- */\r
+ /*\r
+ L4. A character that possesses the mirrored property as specified by\r
+ Section 4.7, Mirrored, must be depicted by a mirrored glyph if the resolved\r
+ directionality of that character is R.\r
+ */\r
\r
- // don't mirror as characters are already mirrored in the pdf\r
+ // don't mirror as characters are already mirrored in the pdf\r
\r
- // Finally, return string\r
+ // Finally, return string\r
\r
- var result = '';\r
- for (var i = 0, ii = chars.length; i < ii; ++i) {\r
- var ch = chars[i];\r
- if (ch != '<' && ch != '>')\r
- result += ch;\r
- }\r
- return result;\r
-}\r
+ var result = '';\r
+ for (var i = 0, ii = chars.length; i < ii; ++i) {\r
+ var ch = chars[i];\r
+ if (ch != '<' && ch != '>')\r
+ result += ch;\r
+ }\r
+ return result;\r
+ });\r
+})();\r
projects / pdf.js.git / commitdiff