[clean-list] Re: Prospects for Unicode Support?

Fri, 12 Apr 2002 10:18:56 EDT

--part1_173.68c192c.29e846d0_boundary
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable

19940804

Marco Kesseler wrote:

<What kind of things are you thinking about?
- Unicode literals?
- Basic operations (concat, substring, search, ...)?
- Character classification (isletter, isdigit, ...)?
- Conversions to/from other codepages?
- Encodings (UTF8, UTF16, UCS-2, ...)?
- Rendering (GUI, Printing)?
- Having Unicode variants of (IO) lib functions?
- Having Unicode identifiers in Clean itself?
=20
What would be real show-stoppers? What would be nice to have, but not=20
essential?>

Well, basically, functional programming (or any other programming paradigm=20
for that matter) should not be limited to the ASCII character set. Were it=20
not for the absence of universally accepted "global" encoding standards, I=20
think all programming system would be made to cater for as much writing=20
systems and languages as possible, or at least with facilities for the users=
=20
to adapt the development environments to their particular needs. Now that a=20
comprehensive standard is being introduced with Unicode, it is only natural=20
that some of us who use non Western writing systems should wish to enjoy the=
=20
benefits of such a marvelous development system as Clean.

Of the above list of features (Unicode literals, basic operations, character=
=20
classifications, etc) all are wished for. But, realisticaly, this can only b=
e=20
a long term objective.

Some features have higher priority than others; especially, the possibility=20
of having Unicode literals, or that of using the basic operations over=20
relevant Unicode characters (in addition to the 'Char' BasicType, we should=20
have the equivalent of the 'WChar' in some other programming languages, ...)=
.

The Clean development team are best placed to answer the feasibility issues=20
of the extension of Clean. To begin with, the possibility of addressing all=20
Unicode characters together with basic operations should be provided; on thi=
s=20
basis, users could develop specific IO or other libraries as needed, leading=
=20
to their incremental integration and the transition of Clean towards a truly=
=20
global development environment, in say ten or twenty years time ...

Best regards.

Wel=E9 Negga  =20

--part1_173.68c192c.29e846d0_boundary
Content-Type: text/html; charset="ISO-8859-1"
Content-Transfer-Encoding: quoted-printable

<HTML>19940804
 
 Marco Kesseler wrote:
 
 &lt;What kind of things are you thinking about?
 - Unicode literals?
 - Basic operations (concat, substring, search, ...)?<=
FONT COLOR=3D"#000000" SIZE=3D3 FAMILY=3D"SANSSERIF" FACE=3D"Arial" LANG=
=3D"0">
 - Character classification (isletter, isdigit, ...)?<=
FONT COLOR=3D"#000000" SIZE=3D3 FAMILY=3D"SANSSERIF" FACE=3D"Arial" LANG=
=3D"0">
 - Conversions to/from other codepages?
 - Encodings (UTF8, UTF16, UCS-2, ...)?
 - Rendering (GUI, Printing)?
 - Having Unicode variants of (IO) lib functions?<FONT=
 COLOR=3D"#000000" SIZE=3D3 FAMILY=3D"SANSSERIF" FACE=3D"Arial" LANG=3D"0">
 - Having Unicode identifiers in Clean itself?
 =20
 What would be real show-stoppers? What would be nice to have=
, but not essential?&gt;
 
 Well, basically, functional programming (or any other programming paradi=
gm for that matter) should not be limited to the ASCII character set. Were i=
t not for the absence of universally accepted "global" encoding standards, I=
 think all programming system would be made to cater for as much writing sys=
tems and languages as possible, or at least with facilities for the users to=
 adapt the development environments to their particular needs. Now that a co=
mprehensive standard is being introduced with Unicode, it is only natural th=
at some of us who use non Western writing systems should wish to enjoy the b=
enefits of such a marvelous development system as Clean.
 
 Of the above list of features (Unicode literals, basic operations, chara=
cter classifications, etc) all are wished for. But, realisticaly, this can o=
nly be a long term objective.
 
 Some features have higher priority than others; especially, the possibil=
ity of having Unicode literals, or that of using the basic operations over r=
elevant Unicode characters (in addition to the 'Char' BasicType, we should h=
ave the equivalent of the 'WChar' in some other programming languages, ...).
 
 The Clean development team are best placed to answer the feasibility iss=
ues of the extension of Clean. To begin with, the possibility of addressing=20=
all Unicode characters together with basic operations should be provided; on=
 this basis, users could develop specific IO or other libraries as needed, l=
eading to their incremental integration and the transition of Clean towards=20=
a truly global development environment, in say ten or twenty years time ...
 
 Best regards.
 
 Wel=E9 Negga &nbsp;&nbsp;
 </HTML>

--part1_173.68c192c.29e846d0_boundary--