Sakurai/pdf.js
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Re-factor heuristics to recognize unknown glyphs for |toUnicode|

Browse Source
This commit is contained in:
Jonas Jenwald 2014-07-28 18:34:41 +02:00
parent 3fc746cf7c
commit b918df3547
1 changed files with 21 additions and 17 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
src/core/fonts.js
View File

@ -4310,26 +4310,30 @@ var Font = (function FontClosure() {
var glyphName = encoding[charcode];
// b) Look up the character name in the Adobe Glyph List (see the
// Bibliography) to obtain the corresponding Unicode value.
if (glyphName === '' || !(glyphName in GlyphsUnicode)) {
if (glyphName === '') {
continue;
} else if (GlyphsUnicode[glyphName] === undefined) {
// (undocumented) c) Few heuristics to recognize unknown glyphs
// NOTE: Adobe Reader does not do this step, but OSX Preview does
var code;
// Gxx glyph
if (glyphName.length === 3 &&
glyphName[0] === 'G' &&
(code = parseInt(glyphName.substr(1), 16))) {
toUnicode[charcode] = String.fromCharCode(code);
var code = 0;
switch (glyphName[0]) {
case 'G': // Gxx glyph
if (glyphName.length === 3) {
code = parseInt(glyphName.substr(1), 16);
}
break;
case 'g': // g00xx glyph
if (glyphName.length === 5) {
code = parseInt(glyphName.substr(1), 16);
}
break;
case 'C': // Cddd glyph
if (glyphName.length >= 3) {
code = +glyphName.substr(1);
}
break;
}
// g00xx glyph
if (glyphName.length === 5 &&
glyphName[0] === 'g' &&
(code = parseInt(glyphName.substr(1), 16))) {
toUnicode[charcode] = String.fromCharCode(code);
}
// Cddd glyph
if (glyphName.length >= 3 &&
glyphName[0] === 'C' &&
(code = +glyphName.substr(1))) {
if (code) {
toUnicode[charcode] = String.fromCharCode(code);
}
continue;