NUL/CR munging in -p mode fixed -- time for 0.7.6? (was: problem with stripping NULs from input?)

[Matthias Andree]

On Thu, 24 Oct 2002, Allyn Fratkin wrote:

> >To achieve this, I changed the broken return semantics of the fgets
> >replacement to return the count of characters instead, the rest was
> >trivial, replacing strcpy by memcpy and fputs by fwrite.
> 
> 
> thanks for taking care of this.  sorry i didn't realize it sooner.

That's one of the shortcomings of the standard C library (the bigger one
is, of course, gets() -- it should have been eliminated ten years ago).
Not your fault -- you just used the tools that were there, and that's
not a bad thing usually; most programmers get trapped by this.

If you want something standalone (i. e. without lexer) in another
project, then also look at GNU libc's getline.

> >IMHO, it's time for 0.7.6 beta. What do you and the others think?
> 
> i'm all for it.  it seems like significant progress has been made just
> since the 0.7.5 release with things like the robinson algorithm.
> 
> i'll have to check again but i believe we can add hp-ux to the list of
> supported operating systems (as long as gcc/flex are used to compile,
> hp-ux's lex doesn't seem up to the task).

Why gcc? I have yet to see HP-UX's "ansic" break for me -- but I only
have remote (network)  access to an Agilent Logic Analyzer (with an
embedded version of HP-UX 10.20) which is switched off most of the time,
so I don't usually check HP-UX.

We might want to document the flex requirement if HP-UX's flex chokes
(and if it cannot easily be fixed -- I might want to try Solaris' lex as
well).

> thanks again.

You're welcome.