[javascript] bugs? Problems dealing with UTF-16 strings

[Mathew]

I'm having trouble encoding UTF-16 strings using both doc.createDataObject() and util.streamFromString(). In the example below, I'm encoding an arbitrary UTF-16 string with JSON.stringify() first, then comparing this string with what comes from the above two functions. I think that util.stringFromStream() is including the 'FEFF' byte order mark as the first character (bug).

I get the following:
doc.getDataObjectContents() I get an additional first character, but only if the text included is actually utf-16.
util.streamFromString() seems to fail entirely on characters with two unicode code units.

Am I doing something wrong?

Code: Select all

// test string from stream

function testStringStreamUTF16(str) {
  const cName = 'tempTest.txt';
  const tempDoc = app.newDoc();
  tempDoc.createDataObject({
    cName,
    cValue: JSON.stringify(str,null,1),
    cMIMEType: 'utf-16'
  });
  function outDiffs(q,p) {
    try {
      console.println([q.length, p.length]);
      for (let i in q) {
        if (q.charCodeAt(i) !== p.charCodeAt(i)) console.println(i + ": " + [q.charAt(i),p.charAt(i)] + " : " + [q.charCodeAt(i).toString(16), p.charCodeAt(i).toString(16)]);
      }
    } catch (e) {
      console.println(e);
    }
  }
  
  const q = JSON.stringify(str,null,1);
  let p = util.stringFromStream(tempDoc.getDataObjectContents(cName), 'utf-16');
  console.println("data object");
  outDiffs(q,p);
  
  p = util.stringFromStream(util.streamFromString(q, 'utf-16'), 'utf-16');
  console.println("stream from string");
  outDiffs(q,p);
}
  
testStringStreamUTF16([`ヸ摭㝑ů᪌ڬม疯ⅲॼᲥㆬ₌兼ৡ⏵c拀㒧刾⁊炨ᚃ屚Ñዘ㚧㨯晢຋Ⳃᠽઔڹ㒥޲戔က瀣‸∂勁⿨稫䂥ٌ̦ቝ偸懠㑧瀪悇ᅀۙ⍣愇䗆㘲扡ࢩ瀥⥦疼况ငⷲ瑆懑剒ᮥ⌣䛆े≦䠭備怠࣎㢌Ạᐣ倨儞懲ᅧ簱䤁ੀኯ䲠჌䗁翆䑆ؚհ癴庨凌ᜄ䕶摻慤䠱╗⁐捆Ɉ⯧≅拐ϐ䊢ᐨ凚ສ憤ʼ姴䌂喓㑚ĩ䎠䜨ṟᅞᲀ䢠崴䌮ᐮ澵⯙甦።ଂ坒ŵ愎ᬫ‬䐈䉈疢琭戆 㑨ˊ䋄桀\"宁߃挈䣏ਯᄰ䬀`]);