Re: [encoding] "13.2.2 iso-2022-jp encoder" (#15)

I made a research on how some encoders do treat bare ESC, SI and SO.

Browsers:
 Submit the form on the web page below and observe the resulting querystring.

    header -> Content-Type: text/html;charset=ISO-2022-jp
    <form>
    <input type=text name=p1 value="xxx&#x1B;yyy">
    <input type=text name=p2 value="xxx&#x0E;yyy">
    <input type=text name=p3 value="xxx&#x0F;yyy">
    <input type=submit value=send>
    </form>

 IE, Firefox:
  Output them to the encoded bytes.
  Resulting querystring: ?p1=xxx%1Byyy&p2=xxx%0Eyyy&p3=xxx%0Fyyy

 Chrome:
  Convert them to HTML character references.
  Resulting querystring: ?p1=xxx%26%2327%3Byyy&p2=xxx%26%2314%3Byyy&p3=xxx%26%2315%3Byyy

 Safari:
  Remove them and the rest.
  Resulting querystring: ?p1=xxx&p2=xxx&p3=xxx

Java, ASP.NET:
 Output them to the encoded bytes.

Test Environments:
 Java 1.8.0_31
 ASP.NET 4.0.30319
 Firefox 42.0 on Windows 7
 Chrome 47.0.2526.80 m on Windows 7
 IE 10 (10.0.9200.17566) on Windows 7
 Safari 9.0.2 (10601.3.9) on Mac OS X 10.10.5

Only WebKit based browsers seem not to output ESC, SI and SO to the encoded bytes.


---
Reply to this email directly or view it on GitHub:
https://github.com/whatwg/encoding/issues/15#issuecomment-164139068