Re: cat fionread usage bug?

[Jim Meyering]

Pádraig Brady wrote:
> Jim Meyering wrote:
>> Pádraig Brady wrote:
>>> I was just looking at this line in cat.c: http://url.ie/1aq1
>>>
>>>     if (input_pending)
>>>         write_pending (outbuf, &bpout);
>>>
>>> Shouldn't that be?
>>>
>>>     if (!input_pending)
>>>         write_pending (outbuf, &bpout);
>>
>> Oh!  You're right.
>> That's a bug (mine): disabled optimization.
>>
>>> The checkin I think is 790892db:
>>
>> Yes, indeed.  Here's the URL:
>>
>>     http://git.sv.gnu.org/gitweb/?p=coreutils.git;a=commitdiff;h=790892db
>>
>> Have you tried to demonstrate the unwanted behavior?
>
> The following should output 1
>
> (printf 1; sleep 2; printf 3) | timeout 1s cat -v | wc -c

Nice.

> If I change it as suggested above it does.
> Nobody has complained in the last 2 years,
> but I guess it's worth keeping the functionality.

I agree.
Would you like to handle it?