LumeWeb
/
chardet
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

declared variable confidence before use, to make it work in 'strict mode' environment.

This commit is contained in:
Spiky 2015-11-27 16:17:28 -05:00
parent 17aebe7b28
commit b0f828d8ad
1 changed files with 77 additions and 73 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

150
encoding/unicode.js
View File

@ -1,107 +1,111 @@
'use strict';
var util = require('util'), var util = require('util'),
Match = require ('../match'); Match = require ('../match');
/** /**
* This class matches UTF-16 and UTF-32, both big- and little-endian. The * This class matches UTF-16 and UTF-32, both big- and little-endian. The
* BOM will be used if it is present. * BOM will be used if it is present.
*/ */
module.exports.UTF_16BE = function() { module.exports.UTF_16BE = function() {
this.name = function() { this.name = function() {
return 'UTF-16BE'; return 'UTF-16BE';
}; };
this.match = function(det) { this.match = function(det) {
var input = det.fRawInput; var input = det.fRawInput;
if (input.length >= 2 && ((input[0] & 0xff) == 0xfe && (input[1] & 0xff) == 0xff)) if (input.length >= 2 && ((input[0] & 0xff) == 0xfe && (input[1] & 0xff) == 0xff)) {
return new Match(det, this, confidence = 100); var confidence = 100;
return new Match(det, this, confidence);
}
// TODO: Do some statistics to check for unsigned UTF-16BE // TODO: Do some statistics to check for unsigned UTF-16BE
return null; return null;
}; };
}; };
module.exports.UTF_16LE = function() { module.exports.UTF_16LE = function() {
this.name = function() { this.name = function() {
return 'UTF-16LE'; return 'UTF-16LE';
}; };
this.match = function(det) { this.match = function(det) {
var input = det.fRawInput; var input = det.fRawInput;
if (input.length >= 2 && ((input[0] & 0xff) == 0xff && (input[1] & 0xff) == 0xfe)) { if (input.length >= 2 && ((input[0] & 0xff) == 0xff && (input[1] & 0xff) == 0xfe)) {
// An LE BOM is present. // An LE BOM is present.
if (input.length >= 4 && input[2] == 0x00 && input[3] == 0x00) if (input.length >= 4 && input[2] == 0x00 && input[3] == 0x00) {
// It is probably UTF-32 LE, not UTF-16 // It is probably UTF-32 LE, not UTF-16
return null; return null;
}
var confidence = 100;
return new Match(det, this, confidence);
}
return new Match(det, this, confidence = 100); // TODO: Do some statistics to check for unsigned UTF-16LE
return null;
} }
// TODO: Do some statistics to check for unsigned UTF-16LE
return null;
}
}; };
function UTF_32() {}; function UTF_32() {};
UTF_32.prototype.match = function(det) { UTF_32.prototype.match = function(det) {
var input = det.fRawInput, var input = det.fRawInput,
limit = (det.fRawLength / 4) * 4, limit = (det.fRawLength / 4) * 4,
numValid = 0, numValid = 0,
numInvalid = 0, numInvalid = 0,
hasBOM = false, hasBOM = false,
confidence = 0; confidence = 0;
if (limit == 0) if (limit == 0)
return null; return null;
if (this.getChar(input, 0) == 0x0000FEFF) if (this.getChar(input, 0) == 0x0000FEFF)
hasBOM = true; hasBOM = true;
for (var i = 0; i < limit; i += 4) { for (var i = 0; i < limit; i += 4) {
var ch = this.getChar(input, i); var ch = this.getChar(input, i);
if (ch < 0 || ch >= 0x10FFFF || (ch >= 0xD800 && ch <= 0xDFFF)) if (ch < 0 || ch >= 0x10FFFF || (ch >= 0xD800 && ch <= 0xDFFF))
numInvalid += 1; numInvalid += 1;
else else
numValid += 1; numValid += 1;
} }
// Cook up some sort of confidence score, based on presence of a BOM // Cook up some sort of confidence score, based on presence of a BOM
// and the existence of valid and/or invalid multi-byte sequences. // and the existence of valid and/or invalid multi-byte sequences.
if (hasBOM && numInvalid == 0) { if (hasBOM && numInvalid == 0) {
confidence = 100; confidence = 100;
} else if (hasBOM && numValid > numInvalid * 10) { } else if (hasBOM && numValid > numInvalid * 10) {
confidence = 80; confidence = 80;
} else if (numValid > 3 && numInvalid == 0) { } else if (numValid > 3 && numInvalid == 0) {
confidence = 100; confidence = 100;
} else if (numValid > 0 && numInvalid == 0) { } else if (numValid > 0 && numInvalid == 0) {
confidence = 80; confidence = 80;
} else if (numValid > numInvalid * 10) { } else if (numValid > numInvalid * 10) {
// Probably corrupt UTF-32BE data. Valid sequences aren't likely by chance. // Probably corrupt UTF-32BE data. Valid sequences aren't likely by chance.
confidence = 25; confidence = 25;
} }
// return confidence == 0 ? null : new CharsetMatch(det, this, confidence); // return confidence == 0 ? null : new CharsetMatch(det, this, confidence);
return confidence == 0 ? null : new Match(det, this, confidence); return confidence == 0 ? null : new Match(det, this, confidence);
}; };
module.exports.UTF_32BE = function() { module.exports.UTF_32BE = function() {
this.name = function() { this.name = function() {
return 'UTF-32BE'; return 'UTF-32BE';
}; };
this.getChar = function(input, index) { this.getChar = function(input, index) {
return (input[index + 0] & 0xff) << 24 | (input[index + 1] & 0xff) << 16 | return (input[index + 0] & 0xff) << 24 | (input[index + 1] & 0xff) << 16 |
(input[index + 2] & 0xff) << 8 | (input[index + 3] & 0xff); (input[index + 2] & 0xff) << 8 | (input[index + 3] & 0xff);
}; };
}; };
util.inherits(module.exports.UTF_32BE, UTF_32); util.inherits(module.exports.UTF_32BE, UTF_32);
module.exports.UTF_32LE = function() { module.exports.UTF_32LE = function() {
this.name = function() { this.name = function() {
return 'UTF-32LE'; return 'UTF-32LE';
}; };
this.getChar = function(input, index) { this.getChar = function(input, index) {
return (input[index + 3] & 0xff) << 24 | (input[index + 2] & 0xff) << 16 | return (input[index + 3] & 0xff) << 24 | (input[index + 2] & 0xff) << 16 |
(input[index + 1] & 0xff) << 8 | (input[index + 0] & 0xff); (input[index + 1] & 0xff) << 8 | (input[index + 0] & 0xff);
}; };
}; };
util.inherits(module.exports.UTF_32LE, UTF_32); util.inherits(module.exports.UTF_32LE, UTF_32);