Operations » Editor » Public Comments » Public Comments Archive

The spec says: "When utf16Width is 'variable', then on parsing an un-paired surrogate codepoint causes a decode error, which can be controlled via dfdl:encodingErrorPolicy described below."

Is this what the ICU library does?

We have generally tried to make it possible to use ICU when implementing DFDL, but does it in fact work this way? The alternative is that they are passed through and treated as 16-bit characters. Similarly, a string containing an unpaired surrogate codepoint could be unparsed.

This behavior needs to be clarified for parsing and unparsing, and for both C and Java ICU libraries if we are to have confidence we're not specifying something very hard to implement.