1
0
Fork 0
mirror of https://github.com/mozilla/pdf.js.git synced 2025-04-26 10:08:06 +02:00

Merge the stream handling changes with the Font code

This commit is contained in:
Vivien Nicolas 2011-06-13 18:59:46 +02:00
parent 2c4a0aa269
commit 2dc7bda2bc
2 changed files with 22 additions and 35 deletions

View file

@ -20,7 +20,6 @@ var kMaxGlyphsCount = 1024;
*/
var Fonts = {};
/**
* 'Font' is the class the outside world should use, it encapsulate all the font
* decoding logics whatever type it is (assuming the font type is supported).
@ -44,10 +43,6 @@ var Font = function(aFontName, aFontFile, aFontType) {
var start = Date.now();
switch (aFontType) {
case "Type1":
// All Type1 font program should begin with the comment %!
if (aFontFile.getByte() != 0x25 || aFontFile.getByte() != 0x21)
error("Invalid file header");
var cff = new CFF(aFontName, aFontFile);
this.mimetype = "font/otf";
@ -569,7 +564,7 @@ var Type1Parser = function(aAsciiStream, aBinaryStream) {
var decryptedString = [];
var value = "";
var count = aStream.length;
var count = aStream.length - aStream.start;
for (var i = 0; i < count; i++) {
value = aStream.getByte();
decryptedString[i] = String.fromCharCode(value ^ (r >> 8));
@ -949,7 +944,6 @@ var Type1Parser = function(aAsciiStream, aBinaryStream) {
// and start interpreting it in order to decode it
var file = operandStack.pop();
var eexecString = decrypt(aBinaryStream, kEexecEncryptionKey, 4).join("");
dump(eexecString);
lexer = new Lexer(new StringStream(eexecString));
break;
@ -989,13 +983,12 @@ var Type1Parser = function(aAsciiStream, aBinaryStream) {
var file = operandStack.pop();
// Add '1' because of the space separator, this is dirty
var stream = lexer.stream.makeSubStream(lexer.stream.pos + 1, size);
var stream = lexer.stream.makeSubStream(lexer.stream.start + lexer.stream.pos + 1, size);
lexer.stream.skip(size + 1);
var charString = decrypt(stream, kCharStringsEncryptionKey, 4).join("");
var charStream = new StringStream(charString);
var decodedCharString = decodeCharString(charStream);
dump("decodedCharString: " + decodedCharString);
operandStack.push(decodedCharString);
// boolean indicating if the operation is a success or not
@ -1144,22 +1137,21 @@ var Type1Parser = function(aAsciiStream, aBinaryStream) {
};
var fontCount = 0;
var CFF = function(aFontName, aFontFile) {
if (!fontCount || true) {
fontCount++;
var start = Date.now();
var start = Date.now();
var ASCIIStream = aFontFile.makeSubStream(0, aFontFile.dict.get("Length1"), aFontFile.dict);
var binaryStream = aFontFile.makeSubStream(aFontFile.dict.get("Length1"), aFontFile.dict.get("Length2"), aFontFile.dict);
var length1 = aFontFile.dict.get("Length1");
var length2 = aFontFile.dict.get("Length2");
var pos = aFontFile.pos;
var ASCIIStream = aFontFile.makeSubStream(pos, length1, aFontFile.dict);
var binaryStream = aFontFile.makeSubStream(pos + length1, length2, aFontFile.dict);
this.parser = new Type1Parser(ASCIIStream, binaryStream);
var fontName = this.parser.parse();
this.font = PSFonts.get(fontName);
this.data = this.convertToCFF(this.font);
var end = Date.now();
//log("Time to parse font is:" + (end - start));
}
this.parser = new Type1Parser(ASCIIStream, binaryStream);
var fontName = this.parser.parse();
this.font = PSFonts.get(fontName);
this.data = this.convertToCFF(this.font);
var end = Date.now();
//log("Time to parse font is:" + (end - start));
};
CFF.prototype = {