Re: MinGW isprint test

[Philip Nienhuis]

Philip Nienhuis wrote:
> Rik wrote:
> > - mappers.cc: assert (isprint (charset), result) differs in column 10.
> > \r\n versus \n line endings?
> >
> > 8/28/12
> >
> > Philip,
> >
> > This one should be easy to check. Can you try the following at the
> > command line?
> >
> > isprint ("\n")
> > isprint ("\r")
> >
> > On Unix systems these are both false. It sounds like character 10
> > ("\n") may be considered printable no MinGW.
>
> Both give 0.
>
> For completeness here are the relevant results from fntest.log:
>
> assert (isprint (charset),result) expected
> Columns 1 through 22:
>
> 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
> :
> but got
> Columns 1 through 22:
>
> 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0
>
> But:
> octave.exe:2> isprint ( "\t")
> ans = 1
>
> so it is TAB that is considered printable. Sorry for my miscounting.
> BTW isprint ("\t") does the same on octave-3.6.2_MinGW and 3.4.3_MinGW.
> Apparently the test is wrong.

That should read: on MinGW the result of isprint ("\t") seems wrong.

I tried to find the cause but quickly got lost in the sheer number of 
related macros and modules so I gave up.

Philip