Re: Encoding single-byte tests

On 02/09/2014 11:35, "Martin J. Dürst" wrote:
> Hello Anne,
>
> Many thanks for your analysis. I have cc'ed Jungshik who hopefully can
> give us some info from a Chrome point of view. If you know others with
> Chrome or Safari to help with the issues below, please cc or contact them.

Fwiw, I updated the information in the summaries of the test page 
results at 
www.w3.org/International/tests/repository/encoding/indexes/results-aliases 
to make it clearer what is going on (ie. that the Firefox problems are 
separate from the Chrome/Safari/Opera ones, and what the issues are for 
all.)

See, in particular,
http://www.w3.org/International/tests/repository/encoding/indexes/results-aliases#windows-1253
http://www.w3.org/International/tests/repository/encoding/indexes/results-aliases#windows-874

It seems to me that the problem in windows-1253 is simply a bug, given 
that FFFD rather than a passthrough appears in rows D and F, and in all 
other encodings except windows-874.

As for windows-874, given that this is behaviour is unique to this 
encoding only (producing PUA characters rather than FFFD), I suspect 
that it would be better to fix it or leave it as an exception, rather 
than change the index. Note that Firefox does what the index expects, 
and that this is a standard deviation for IE.

RI

Received on Tuesday, 2 September 2014 11:01:49 UTC