[GHC] #710: library reorganisation

Bulat Ziganshin bulat.ziganshin at gmail.com
Fri Apr 28 04:50:06 EDT 2006


Hello Ketil,

Friday, April 28, 2006, 10:52:17 AM, you wrote:

> Bulat Ziganshin <bulat.ziganshin at gmail.com> writes:

>> sorry for repetition, but ByteString library in its current state
>> still don't replaces PackedString in functionality, because it don't
>> support full Unicode range of chars

> What would be required for it to replace PackedString? (If that is a
> goal?)  If I understand correctly, PS is an array of Word32, and
> ByteString is (obviously) an array of Word8.  Would it be sufficient
> if there was a 'Char' interface supporting all of Unicode (the obvious
> candidate encoding being UTF-8), or must it support UCS-2 and UCS-4
> directly? 

IMHO, because PackedString is anyway abstract and DON'T support any way
to see it's internal representation, any implementation that supports
full unicode range, would be enough. it may be ucs4, utf8, or even
ByteString+String (that selects representation depending on presence
of non-Latin1 characters in string)

support of specific encodings, such as utf16 or ucs4, will be great
for special purposes (as i said, ucs4 allows fastest processing with
support for full Unicode range, while utf16 is great for working
directly with windows filenames), but that is another question. i just
want to point that omitting PackedString may create problems for the
people that use it


one more suggestion about standardizing modules:

ByteString gives access to Word8 strings and it provides ability to
work with memory regions directly

PackedString should just allow to work with some compact String
representation without any information about it's internal
representation, so it can be implemented using any above-given
approach

PackedString.UTF8 should work with memory areas that contains
UTF8-packed strings and give direct access to such memory areas

PackedString.UTF16 should work with memory areas that contains
UTF16-packed strings and, again, give direct access to such memory areas

The same for PackedString.UCS4...

So, PackedString implementation should just import one of these
modules and then rexxport only functions that are
representation-independent. If someone just wants fast and compact
strings, he should import this module, if he want some specific
representation, he should import representation-specific module



-- 
Best regards,
 Bulat                            mailto:Bulat.Ziganshin at gmail.com



More information about the Glasgow-haskell-users mailing list