> String Type:
>
> * Has null terminator, for interopability with older Win32 code. We
> couldn't give a hoot about interop with Win32 code :).
might be usefull for debugging though...
> * Each char, is indeed, a fixed length UTF-16 (Unicode) bit sequence
> (2 bytes in length).
> o We should decide: UTF-16 or UTF-32 as a standard? Point for
> a meeting? UTF-16 has a smaller footprint, UTF-32 hits a
> much larger market.
UTF-16 would make it easier to serialize data which has been saved by
other non sharp OS applications...
> * Has a m_firstChar field. In the article this is a char, but I
> assume that it is a char*.
well maybe it's actually closer to
struct
{
bool interned; // set when string is interned / pooled
int capacity;
int length;
fixed char[] characters;
}
this would only require one memory block for your string instead of two..
ofcourse with a char* it's easier to implement null strings...
> * Two lengths are stored, one with the null terminator (no longer
> needed) and one that is the actual length.
> o Obviously drop the first.
I think "m_arrayLength" is the capacity of the string.. you basically
allocate a little more memory than you need, this saves you from
allocating new memory every time you add a single character..
> o The second is an Signed Int32. Why did they use signed ints?
> Can anyone come up with a good argument for this? Maybe we
> should use a UInt32 instead.
I can't imagine why you would need the sign bit...
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
SharpOS-Developers mailing list
SharpOS-Developers@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sharpos-developers
