[m-rev.] for review: Add web browser-based term browsing in the debugger.

[Julien Fischer]

Hi Peter,

On Tue, 15 Aug 2017, Peter Wang wrote:

> On Tue, 15 Aug 2017 15:04:55 +1000 (AEST), Julien Fischer <jfischer at opturion.com> wrote:
>>
>> Hi Peter,
>>
>> On Tue, 15 Aug 2017, Peter Wang wrote:
>>
>>> I am considering removing the browse --xml command.
>>> Does it still work,
>>
>> Presumably.  (I don't have any objections to removing it however.)
>
> The xsltproc command fails for me (and I don't know how to fix it).
> Also, I thought XUL was being deprecated, but perhaps that doesn't mean
> what I think it means.

Ask Paul, he works for Mozilla now.  (FWIW, so did I.)

...

>>> +:- pred put_unicode_escape(io.output_stream::in, char::in,
>>> +    io::di, io::uo) is det.
>>> +
>>> +put_unicode_escape(Stream, Char, !State) :-
>>> +    CodePoint = char.to_int(Char),
>>> +    ( if CodePoint > 0xFFFF then
>>> +        code_point_to_utf16_surrogates(CodePoint, LS, TS),
>>> +        put_hex_digits(Stream, LS, !State),
>>> +        put_hex_digits(Stream, TS, !State)
>>> +    else
>>> +        put_hex_digits(Stream, CodePoint, !State)
>>> +    ).
>>> +
>>> +:- pred code_point_to_utf16_surrogates(int::in, int::out, int::out) is det.
>>> +
>>> +code_point_to_utf16_surrogates(CodePoint, LS, TS) :-
>>> +    AdjustedCodePoint = CodePoint - 0x10000,
>>> +    LS = 0xD800 + (AdjustedCodePoint >> 10),
>>> +    TS = 0xDC00 + (AdjustedCodePoint /\ 0x3FF).
>>
>> Do we not provide this predicate (or the equivalent) in the standard
>> library?
>
> Yes, char.to_utf16.

I guess the disadvantage of the library version is that it returns a
list; which is not so helpful if you know the the code point is above
0xffff (as here).

Julien.