[Numpy-discussion] Inconsistent error messages.

[Christopher Barker]

Stéfan van der Walt wrote:
> 2009/5/23 Charles R Harris <charlesr.harris at gmail.com>:
>> In [9]: fromfile("empty.dat", count=10)

> Instead of throwing errors in these scenarios, we could just return
> the elements read and raise a warning?  This is consistent with most
> other file APIs I know and allows you to read blocks of data until the
> data runs out.

This was just discussed a week or two ago (look for messaged by me and 
Robert Kern)

fromfile needs some attention in general, but I think Robert an I (the 
only two that I know were paying attention to that discussion) agreed 
that there should be an API that says:

read up to n items from the file

Robert thought that should be the default, but I think that means 
everyone would be forced to check how many items they got every time 
they read, which is too much code and likely to be forgotten and lead to 
errors. So I think that an exception should be raised if you ask for n 
and get less than n, but that there should be a flag that says something 
like "max_items=n", that indicates that you'll take what you get.

I don't like a warning -- it's unconventional to catch those like 
exceptions -- if you want n items, and you haven't written code to 
handle less than n, you should get an exception. If you have written 
code to handle that, that you can use the flag.

-Chris



-- 
Christopher Barker, Ph.D.
Oceanographer

NOAA/OR&R/HAZMAT         (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception