Re: Format of string output of a socket server

ASCII is the same no matter what byte encoding is used.
It's characters beyond ASCII you should worry about.
By definition any ASCII string is in UTF-8 encoding.
UNICODE code points 128-2047 are encoded in 2 bytes,
code points 2048-65535 (excluding the invalid code point
range of 55296-57343) are encoded in 3 bytes and the
rest (e.g. 65536-1114111) are encoded in 4 bytes.

For quick reference on UTF-8 here's the wikipedia page:

http://en.wikipedia.org/wiki/UTF-8

--
=====================================
Alexander Nickolov
Microsoft MVP [VC], MCSD
email: agnickolov@xxxxxxxx
MVP VC FAQ: http://www.mvps.org/vcfaq
=====================================

"Angus" <nospam@xxxxxxxxx> wrote in message
news:%23TKRV8y5GHA.4484@xxxxxxxxxxxxxxxxxxxxxxx
Well... think I am more confused now than when I asked the question ;)

I am talking about what the server will send. I am getting from these
comments that sending bytes (char) is OK. Basically a string response I
would send would be eg "my response\r\n" - ie byte chars followed by
carriage return line feed. I am supposing this is OK.

What for example does your standard POP3 server send? ASCII text just
like
I am saying here?

The client program can then convert to Unicode or whatever they see fit?

Angus


"Alexander Nickolov" <agnickolov@xxxxxxxx> wrote in message
news:OZSOk9x5GHA.3452@xxxxxxxxxxxxxxxxxxxxxxx
UNICODE is the only sane choice of course. However, don't
confuse with the Windows meaning of UNICODE which is
really only the UTF-16 representation of UNICODE. I'd
suggest you use UTF-8 representation of UNICODE to avoid
byte-ordering issues on the network. What you return to
your clients is up to you - you just need to do the appropriate
format conversion (e.g. MultiByteToWideChar to get UTF-16
for example).

--
=====================================
Alexander Nickolov
Microsoft MVP [VC], MCSD
email: agnickolov@xxxxxxxx
MVP VC FAQ: http://www.mvps.org/vcfaq
=====================================

"Angus" <nospam@xxxxxxxxx> wrote in message
news:eXxFh8m5GHA.400@xxxxxxxxxxxxxxxxxxxxxxx
Hello

I am writing a socket server to deliver telephony events to clients on
a
network. For example the telephony server might send out text to
connected
clients. Clients might be written in C/C++, Java, Visual Basic,
anything
in
fact which can talk to a socket.

My socket server is currently sending out char* . Do I have to worry
about
the format of string output? Should I be outputting Unicode? Some
other
format? Or would a C/C++ char* be OK? Will eg Java understand it? Do
they
use UTF-8 or something?

Angus








.

Relevant Pages

  • Re: Zeichenkodierung in der shell
    ... Erfinder zu benutzen - statt sie zu vergewaltigen - werden in der ... auf 8 Bit durch UTF-8? ... dass mit Unicode (egal welcher ... an bestimmten Stellen einfach ASCII _vorgeschrieben_ ist, ...
    (de.comp.os.unix.linux.misc)
  • Re: D2008 - VCL Makeover details?
    ... new TEncoding parameters so you can specify what format to use when loading/saving data (Ascii, UTF-7, UTF-8, Unicode, etc). ...
    (borland.public.delphi.non-technical)
  • Re: Representing futuristic English
    ... If I load up an ascii file in a unicode editor, ... UTF-8, it guesses UTF-8, which is incorrect, but close ...
    (rec.arts.sf.composition)
  • Re: Representing futuristic English
    ... > If I load up an ascii file in a unicode editor, ... > UTF-8, it guesses UTF-8, which is incorrect, but close ... > at once *guess* the ascii encoding, ...
    (rec.arts.sf.composition)
  • Re: Unicode Delphi Win32 - which approach
    ... I like the backwards compatibility aspects of UTF-8 vs UTF-16. ... The first 256 Unicode characters map to the ANSI character set. ... entire stream> but calling an API 100 times in a loop I can imagine. ... and explicitly contextualise every string. ...
    (borland.public.delphi.non-technical)