cvs commit: hugs98/src MkDepend.in builtin.c char.h compiler.c
errors.c input.c iomonad.c machdep.c machine.c opts.c output.c
printer.c hugs98/src/dotnet Makefile hugs98/src/msc Makefile
hugs98/src/winhugs Makefile
ross at glass.cse.ogi.edu
ross at glass.cse.ogi.edu
Tue Dec 2 04:15:55 EST 2003
ross 2003/12/02 04:15:55 PST
Modified files:
src MkDepend.in builtin.c char.h compiler.c
errors.c input.c iomonad.c machdep.c
machine.c opts.c output.c printer.c
src/dotnet Makefile
src/msc Makefile
src/winhugs Makefile
Log:
Another refactoring, in preparation for introducing byte encoding of
Unicode in I/O and strings (including text storage). Byte encoding of
strings is a compromise: ideally we'd use wide chars internally, but this
would cost more space, and would require a lot more changes to the code.
This commit routes character operations on streams and strings through
the following macros (currently just byte versions):
int MAX_CHAR_ENCODING
Char FPutChar(Char c, FILE *f)
Char FGetChar(FILE *f)
void AddChar(Char c, String &sp)
Char ExtractChar(String &sp)
Bool charIsRepresentable(Char c)
We assume that any encoding will be an extension of ASCII, so it's OK
to bypass these if you know you're dealing with ASCII.
Revision Changes Path
1.10 +8 -7 hugs98/src/MkDepend.in
1.72 +3 -3 hugs98/src/builtin.c
1.3 +52 -0 hugs98/src/char.h
1.24 +8 -5 hugs98/src/compiler.c
1.5 +3 -2 hugs98/src/errors.c
1.78 +35 -34 hugs98/src/input.c
1.66 +31 -15 hugs98/src/iomonad.c
1.107 +4 -3 hugs98/src/machdep.c
1.21 +5 -3 hugs98/src/machine.c
1.14 +3 -2 hugs98/src/opts.c
1.34 +9 -7 hugs98/src/output.c
1.14 +3 -3 hugs98/src/printer.c
1.20 +8 -7 hugs98/src/dotnet/Makefile
1.49 +8 -7 hugs98/src/msc/Makefile
1.35 +8 -7 hugs98/src/winhugs/Makefile
More information about the Cvs-hugs
mailing list