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

Merge pull request #4213 from nnethercote/strings

Browse Source 
Create less garbage while parsing
This commit is contained in:
Yury Delendik 2014-01-30 04:41:53 -08:00
commit acb33b3e7d
1 changed files with 58 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

90
src/core/parser.js
View File

@ -335,6 +335,13 @@ var Lexer = (function LexerClosure() {
this.stream = stream;
this.nextChar();
// While lexing, we build up many strings one char at a time. Using += for
// this can result in lots of garbage strings. It's better to build an
// array of single-char strings and then join() them together at the end.
// And reusing a single array (i.e. |this.strBuf|) over and over for this
// purpose uses less memory than using a new array for each string.
this.strBuf = [];
// The PDFs might have "glued" commands with other commands, operands or
// literals, e.g. "q1". The knownCommands is a dictionary of the valid
// commands and their prefixes. The prefixes are built the following way:
@ -389,33 +396,50 @@ var Lexer = (function LexerClosure() {
getNumber: function Lexer_getNumber() {
var floating = false;
var ch = this.currentChar;
var str = String.fromCharCode(ch);
var allDigits = ch >= 0x30 && ch <= 0x39;
var strBuf = this.strBuf;
strBuf.length = 0;
strBuf.push(String.fromCharCode(ch));
while ((ch = this.nextChar()) >= 0) {
if (ch === 0x2E && !floating) { // '.'
str += '.';
if (ch >= 0x30 && ch <= 0x39) { // '0'-'9'
strBuf.push(String.fromCharCode(ch));
} else if (ch === 0x2E && !floating) { // '.'
strBuf.push('.');
floating = true;
allDigits = false;
} else if (ch === 0x2D) { // '-'
// ignore minus signs in the middle of numbers to match
// Adobe's behavior
warn('Badly formated number');
} else if (ch >= 0x30 && ch <= 0x39) { // '0'-'9'
str += String.fromCharCode(ch);
allDigits = false;
} else if (ch === 0x45 || ch === 0x65) { // 'E', 'e'
floating = true;
allDigits = false;
} else {
// the last character doesn't belong to us
break;
}
}
var value = parseFloat(str);
if (isNaN(value))
error('Invalid floating point number: ' + value);
var value;
if (allDigits) {
value = 0;
var charCodeOfZero = 48; // '0'
for (var i = 0, ii = strBuf.length; i < ii; i++) {
value = value * 10 + (strBuf[i].charCodeAt(0) - charCodeOfZero);
}
} else {
value = parseFloat(strBuf.join(''));
if (isNaN(value)) {
error('Invalid floating point number: ' + value);
}
}
return value;
},
getString: function Lexer_getString() {
var numParen = 1;
var done = false;
var str = '';
var strBuf = this.strBuf;
strBuf.length = 0;
var ch = this.nextChar();
while (true) {
@ -427,14 +451,14 @@ var Lexer = (function LexerClosure() {
break;
case 0x28: // '('
++numParen;
str += '(';
strBuf.push('(');
break;
case 0x29: // ')'
if (--numParen === 0) {
this.nextChar(); // consume strings ')'
done = true;
} else {
str += ')';
strBuf.push(')');
}
break;
case 0x5C: // '\\'
@ -445,24 +469,24 @@ var Lexer = (function LexerClosure() {
done = true;
break;
case 0x6E: // 'n'
str += '\n';
strBuf.push('\n');
break;
case 0x72: // 'r'
str += '\r';
strBuf.push('\r');
break;
case 0x74: // 't'
str += '\t';
strBuf.push('\t');
break;
case 0x62: // 'b'
str += '\b';
strBuf.push('\b');
break;
case 0x66: // 'f'
str += '\f';
strBuf.push('\f');
break;
case 0x5C: // '\'
case 0x28: // '('
case 0x29: // ')'
str += String.fromCharCode(ch);
strBuf.push(String.fromCharCode(ch));
break;
case 0x30: case 0x31: case 0x32: case 0x33: // '0'-'3'
case 0x34: case 0x35: case 0x36: case 0x37: // '4'-'7'
@ -478,17 +502,17 @@ var Lexer = (function LexerClosure() {
}
}
str += String.fromCharCode(x);
strBuf.push(String.fromCharCode(x));
break;
case 0x0A: case 0x0D: // LF, CR
break;
default:
str += String.fromCharCode(ch);
strBuf.push(String.fromCharCode(ch));
break;
}
break;
default:
str += String.fromCharCode(ch);
strBuf.push(String.fromCharCode(ch));
break;
}
if (done) {
@ -498,10 +522,12 @@ var Lexer = (function LexerClosure() {
ch = this.nextChar();
}
}
return str;
return strBuf.join('');
},
getName: function Lexer_getName() {
var str = '', ch;
var ch;
var strBuf = this.strBuf;
strBuf.length = 0;
while ((ch = this.nextChar()) >= 0 && !specialChars[ch]) {
if (ch === 0x23) { // '#'
ch = this.nextChar();
@ -510,23 +536,23 @@ var Lexer = (function LexerClosure() {
var x2 = toHexDigit(this.nextChar());
if (x2 == -1)
error('Illegal digit in hex char in name: ' + x2);
str += String.fromCharCode((x << 4) | x2);
strBuf.push(String.fromCharCode((x << 4) | x2));
} else {
str += '#';
str += String.fromCharCode(ch);
strBuf.push('#', String.fromCharCode(ch));
}
} else {
str += String.fromCharCode(ch);
strBuf.push(String.fromCharCode(ch));
}
}
if (str.length > 128) {
if (strBuf.length > 128) {
error('Warning: name token is longer than allowed by the spec: ' +
str.length);
strBuf.length);
}
return new Name(str);
return new Name(strBuf.join(''));
},
getHexString: function Lexer_getHexString() {
var str = '';
var strBuf = this.strBuf;
strBuf.length = 0;
var ch = this.currentChar;
var isFirstHex = true;
var firstDigit;
@ -556,13 +582,13 @@ var Lexer = (function LexerClosure() {
ch = this.nextChar();
continue;
}
str += String.fromCharCode((firstDigit << 4) | secondDigit);
strBuf.push(String.fromCharCode((firstDigit << 4) | secondDigit));
}
isFirstHex = !isFirstHex;
ch = this.nextChar();
}
}
return str;
return strBuf.join('');
},
getObj: function Lexer_getObj() {
// skip whitespace and comments