1. Re: Error: file name for OPEN is too long

Matt Lewis wrote:
> Juergen Luethje wrote:
> > Matt Lewis wrote:
> > > It seems reasonable to pick these values up from the platform correct
> > > headers.
> > > 
> > > Matt
> > 
> > I agree.
> > Also, wouldn't it be better when Eu returns -1 (or maybe -2 ?), when
> > a path/file name is too long, rather than causing the program to crash?
> > For my program that crashed a few days ago, I'll write a wrapper for
> > open() with this behaviour.
> 
> That makes a lot of sense to me.  I didn't even really notice the RTFatal
> call there.  Perhaps Rob can chime in (if he remembers) about why it's
> a fatal error rather than a failed open?

I don't remember exactly.
I guess back in the old 8.3 file name era
the idea of a path greater than 250 or so characters seemed 
very weird, so I may have thought that it must be a 
bug in the user's program that he would want reported to him.

Feel free to change it.

> And since we have information about why the file failed, it does make sense
> to report this.  The only issue is that we will break code that tests:
> }}}
<eucode>
>    if fn = -1 then 
> </eucode>
{{{


I have a lot of code that tests for -1.

Regards,
   Rob Craig
   Rapid Deployment Software
   http://www.RapidEuphoria.com

new topic     » topic index » view message » categorize

Search



Quick Links

User menu

Not signed in.

Misc Menu