Broken JS minifing pdf.js

[andreacfm]

Attempting to minify mozilla/pdf.js brakes js. ( see mozilla/pdf.js#2479) Using the file I get this error:

SyntaxError: octal literals and octal escape sequences are deprecated [Break On This Error]

...-b);var w=n.yMax||y,E=-n.yMin||-b;return"\0$ô\0\0\0Š»\0\0\0ŒŠ»\0\0ß\01...

Compression is made on rails asset pipeline.

[lautis]

Works for me with Uglifier 2.0.0.

[phillbaker]

As a follow up to this, @lautis can you clarify if you used any specific options that allowed you to compile and then use pdf.js?

The specific issue I get (same as above) is that uglifier 2.1.0, with default options compiles

// This is one example of problematic encoding currently around L20107 of https://github.com/mozilla/pdf.js/blob/gh-pages/build/pdf.js
    return '\x00\x03' + // version
           '\x02\x24' + // xAvgCharWidth
           '\x01\xF4' + // usWeightClass
           '\x00\x05' + // usWidthClass
           '\x00\x00' + // fstype (0 to let the font loads via font-face on IE)
           '\x02\x8A' + // ySubscriptXSize
           '\x02\xBB' + // ySubscriptYSize
           '\x00\x00' + // ySubscriptXOffset
           '\x00\x8C' + // ySubscriptYOffset
           '\x02\x8A' + // ySuperScriptXSize
           '\x02\xBB' + // ySuperScriptYSize
           '\x00\x00' + // ySuperScriptXOffset
           '\x01\xDF' + // ySuperScriptYOffset
           '\x00\x31' + // yStrikeOutSize
           '\x01\x02' + // yStrikeOutPosition
           '\x00\x00' + // sFamilyClass
           '\x00\x00\x06' +
           String.fromCharCode(properties.fixedPitch ? 0x09 : 0x00) +
           '\x00\x00\x00\x00\x00\x00' + // Panose
           string32(ulUnicodeRange1) + // ulUnicodeRange1 (Bits 0-31)
           string32(ulUnicodeRange2) + // ulUnicodeRange2 (Bits 32-63)
           string32(ulUnicodeRange3) + // ulUnicodeRange3 (Bits 64-95)
           string32(ulUnicodeRange4) + // ulUnicodeRange4 (Bits 96-127)
           '\x2A\x32\x31\x2A' + // achVendID
           string16(properties.italicAngle ? 1 : 0) + // fsSelection
           string16(firstCharIndex ||
                    properties.firstChar) + // usFirstCharIndex
           string16(lastCharIndex || properties.lastChar) +  // usLastCharIndex
           string16(typoAscent) + // sTypoAscender
           string16(typoDescent) + // sTypoDescender
           '\x00\x64' + // sTypoLineGap (7%-10% of the unitsPerEM value)
           string16(winAscent) + // usWinAscent
           string16(winDescent) + // usWinDescent
           '\x00\x00\x00\x00' + // ulCodePageRange1 (Bits 0-31)
           '\x00\x00\x00\x00' + // ulCodePageRange2 (Bits 32-63)
           string16(properties.xHeight) + // sxHeight
           string16(properties.capHeight) + // sCapHeight
           string16(0) + // usDefaultChar
           string16(firstCharIndex || properties.firstChar) + // usBreakChar
           '\x00\x03';  // usMaxContext

to

return"\0$
ô\0\0\0Š»\0\0\0ŒŠ»\0\0
ß\01
\0\0\0\0"+String.fromCharCode(e.fixedPitch?9:0)+"\0\0\0\0\0\0"+d(i)+d(n)+d(c)+d(s)+"*21*"+h(e.italicAngle?1:0)+h(o||e.firstChar)+h(f||e.lastChar)+h(b)+h(y)+"\0d"+h(k)+h(w)+"\0\0\0\0"+"\0\0\0\0"+h(e.xHeight)+h(e.capHeight)+h(0)+h(o||e.firstChar)+"\0"

This seems to be an encoding/escaping issue. So I tried with { :output => { :ascii_only => true } }, which generates for the same line

return"\0$
\xf4\0\0\0\x8a\xbb\0\0\0\x8c\x8a\xbb\0\0
\xdf\01
\0\0\0\0"+String.fromCharCode(e.fixedPitch?9:0)+"\0\0\0\0\0\0"+d(i)+d(n)+d(c)+d(s)+"*21*"+h(e.italicAngle?1:0)+h(o||e.firstChar)+h(f||e.lastChar)+h(b)+h(y)+"\0d"+h(k)+h(w)+"\0\0\0\0"+"\0\0\0\0"+h(e.xHeight)+h(e.capHeight)+h(0)+h(o||e.firstChar)+"\0"

Both of these generated js files throw "Error: SyntaxError: octal literals and octal escape sequences are deprecated" when used.

Any help would be appreciated.

CC https://github.com/mozilla/pdf.js/issues/2479

[lautis]

You're right, even the latest uglifier seems to result in this behaviour. I misunderstood the description or managed to try this out with a JS interpreter that ignores strict mode. Anyway, I've opened a PR for this bug in UglifyJS2.

[phillbaker]

Thanks @lautis.

For reference, I'd also point out: https://github.com/ndbroadbent/turbo-sprockets-rails3/pull/64#issuecomment-18654913, which is a solution to Uglify compiling CodeMirror, however, in that case { :output => { :ascii_only => true } } does work. It looks like the difference might be hexadecimal escape characters vs. octal?

[lautis]

I had similar issues with characters outside Basic Multilingual Plane using Node 0.4 as JavaScript runtime. Newer Nodes can handle 4-byte UTF-8, but I haven't checked out if those still get corrupted and where the corruption happens.

[lautis]

Fixed in 2.2.0.

[phillbaker]

Awesome! Much appreciated.

[andreacfm]

Is not solved from my side. Even with 2.2.1 I get

SyntaxError: invalid property id
..."ۇٴ",ٸ:"يٴ",ำ:"ํา",ຳ:"ໍາ",ໜ:"ຫນ",ໝ:"ຫມ",ཷ:"ྲཱྀ",ཹ:"ླཱྀ",ẚ:"aʾ","᾽":" ̓","᾿":" ̓","...

[brendandc]

I hit the same issue with 2.2.1, running jruby 1.7.4 on windows, rails 4, asset pipeline and uglifier as the compressor. I'm curious what your environment is @andreacfm , maybe there's a similarity there with us vs. what others are doing..

My content is also in a js object, with keys/values like: "プリウス" : "トヨタ", "アクア" : "トヨタ", "ヴィッツ" : "トヨタ",

[lautis]

Sorry, I dont' have a Windows box to test so I can't verify this. There could be issues with non-ASCII characters and Windows Script Host. Installing therubyracer or therubyrhino could help.

[fwoeck]

I see the same effects as @andreacfm - with an ubuntu node.js v0.10.18 and the uglifier gem 2.2.1. For reference here is the uncompressed pdf.worker.js and the uglified version:

http://dokmatic-cdn.s3.amazonaws.com/pdf.worker.js http://dokmatic-cdn.s3.amazonaws.com/pdf.worker-ae4b62e34ddaa13ac163e8ff46f487fb.js

the errors stem from the lines around #16500 of the original.

[andreacfm]

My issue happens on ubuntu and osx. I did not test on win. Il giorno 21/set/2013 17:19, "Frank Wöckener" notifications@github.com ha scritto:

I see the same effects as @andreacfm https://github.com/andreacfm - with an ubuntu node.js v0.10.18 and the uglifier gem 2.2.1. For reference here is the uncompressed pdf.worker.js and the uglified version:

http://dokmatic-cdn.s3.amazonaws.com/pdf.worker.js

http://dokmatic-cdn.s3.amazonaws.com/pdf.worker-ae4b62e34ddaa13ac163e8ff46f487fb.js

the errors stem from the lines around #16500 of the original.

— Reply to this email directly or view it on GitHubhttps://github.com/lautis/uglifier/issues/47#issuecomment-24864252 .

[lautis]

Thanks for the sample files. There seems to be an issue with object keys in UglifyJS. I'll update Uglifier when UglifyJS has a fix for this.

As a workaround, you could try to use the quote_keys output option, Uglifier.new(output: {quote_keys: true}). This at least produces valid output.

[jasonwebster]

Is there an issue filed with them for this? I looked and couldn't find anything.

[lautis]

I believe this has been addressed in mishoo/UglifyJS2#308. I've pulled the latest UglifyJS master to Uglifier, this should fix the issues with Unicode keys.

[andreacfm]

When do you think the changes will be pushed to rubygems?

Thanks

On Oct 19, 2013, at 6:26 PM, Ville Lautanala notifications@github.com wrote:

I believe this has been addressed in mishoo/UglifyJS2#308. I've pulled the latest UglifyJS master to Uglifier, this should fix the issues with Unicode keys.

— Reply to this email directly or view it on GitHub.

Andrea Campolonghi acampolonghi@gmail.com

[lautis]

I just pushed 2.3.0 to rubygems.

[andreacfm]

Thanks

Ville Lautanala wrote:

I just pushed 2.3.0 to rubygems.

— Reply to this email directly or view it on GitHub https://github.com/lautis/uglifier/issues/47#issuecomment-27143476.