Re: fileexist(f)

[Petr M. <mi...@ph...>]

>> What's the point of asking a question like "if fileexists", when
>> there's nothing useful that could be done with the answer, that we
> I simply don't believe that either of these functions does something that 
> gnuplot has to do, or even *can* do meaningfully.

The fileexist() is a long time on my done list, I needed that several times, 
so finally decided to commit it.

Some applications: number of fit files, availability of extracted 
experimental data files. This function is a good candidate for the group of
string-related functions for the "strings" patch which has become part of 
gnuplot recently.

Based on the discussion: it is fully sufficient to have fileexist() for 
testing whether the given file exist. Nothing more (no pipes, no loadpaths). 
Thus I still wish to commit it (unless there is really still big big 
opposition).


> My preferred method of dealing with this issue is to handle
> missing/empty files cleanly during the plotting process.

Gnuplot support files with palettes as well.


> Should I go ahead and apply the patch for 2D code,
> which is after all a major category of use?

Yes.


>> Trimmed from lines 1240-1280 of datafile.c:
>> =========================================================================
>>    /* filename cannot be static array! */
>>    gp_expand_tilde(&df_filename);
> This line makes sense only for regular files and thus it should be moved 
> to just above the "stat" function.

And this change should be committed too.

---
PM