Re: [Openexr-devel] UNICODE support in openexr file I/O

[Drew Hess]

Bob Friesenhahn <address@hidden> writes:

> My understanding of UTF-8 is that some characters require multiple
> octets to represent so it is not so simple as simply assuming that
> strings are UTF-8.  The library needs to be prepared to handle UTF-8
> internally if it does any string parsing.  It needs to know for sure
> if the string is UTF-8.  Some existing strings may use extended
> characters (in the system's native character set) but not UTF-8.


As far as I know, IlmImf doesn't do any string processing that would
be susceptible to this problem (e.g., it doesn't search for
substrings), so I think the existing APIs are UTF-8-safe.


d