kiwix/kiwix-js
1
0
Fork 0
mirror of https://github.com/kiwix/kiwix-js.git synced 2025-09-24 04:54:51 -04:00
Code Issues Packages Projects Releases Wiki Activity

Fix UTF-8 encoding of article titles

Browse Source
This commit is contained in:
mossroy 2012-12-25 14:13:25 +01:00
parent 68a2f6a5f3
commit 2701f47a87
1 changed files with 28 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

36
evopedia-html5/WebContent/evopedia.js
View File

@ -43,6 +43,31 @@ function readIntegerFrom4Bytes(byteArray,firstIndex) {
return byteArray[firstIndex] + byteArray[firstIndex+1]*256 + byteArray[firstIndex+2]*65536 + byteArray[firstIndex+3]*16777216; return byteArray[firstIndex] + byteArray[firstIndex+1]*256 + byteArray[firstIndex+2]*65536 + byteArray[firstIndex+3]*16777216;
} }
/**
* Converts a UTF-8 byte array to JavaScript's 16-bit Unicode.
* @param {Array.<number>} bytes UTF-8 byte array.
* @return {string} 16-bit Unicode string.
* Copied from http://closure-library.googlecode.com/svn/docs/closure_goog_crypt.js.source.html (Apache License 2.0)
*/
function utf8ByteArrayToString(bytes,startIndex,endIndex) {
var out = [], pos = startIndex, c = 0;
while (pos < bytes.length && pos < endIndex) {
var c1 = bytes[pos++];
if (c1 < 128) {
out[c++] = String.fromCharCode(c1);
} else if (c1 > 191 && c1 < 224) {
var c2 = bytes[pos++];
out[c++] = String.fromCharCode((c1 & 31) << 6 | c2 & 63);
} else {
var c2 = bytes[pos++];
var c3 = bytes[pos++];
out[c++] = String.fromCharCode(
(c1 & 15) << 12 | (c2 & 63) << 6 | c3 & 63);
}
}
return out.join('');
};
function readAllTitlesFromIndex(titleFile) { function readAllTitlesFromIndex(titleFile) {
if (titleFile) { if (titleFile) {
var reader = new FileReader(); var reader = new FileReader();
@ -88,18 +113,13 @@ function readAllTitlesFromIndex(titleFile) {
while (newLineIndex<byteArray.length && byteArray[newLineIndex]!=128) { while (newLineIndex<byteArray.length && byteArray[newLineIndex]!=128) {
newLineIndex++; newLineIndex++;
} }
/*
for (var j=i+15;j<newLineIndex;j++) { for (var j=i+15;j<newLineIndex;j++) {
title += String.fromCharCode(byteArray[j]); title += String.fromCharCode(byteArray[j]);
} }
// TODO : Read the title properly with UTF-8 encoding
/*
var buf = new ArrayBuffer();
var bufView = new Uint16Array(buf);
for (var j=0;j<newLineIndex-i-15;j++) {
bufView[j]=byteArray[j+i+15];
}
title = String.fromCharCode(bufView);
*/ */
// TODO : Read the title properly with UTF-8 encoding
title = utf8ByteArrayToString(byteArray,i+15,newLineIndex);
comboTitleList.options[titleNumber] = new Option (title, filenumber+"|"+blockstart+"|"+blockoffset+"|"+length); comboTitleList.options[titleNumber] = new Option (title, filenumber+"|"+blockstart+"|"+blockoffset+"|"+length);
titleNumber++; titleNumber++;