Re: [Bug-readline] [PATCH] readline/histfile.c: Check and retry write() operation in history_truncate_file()

[Chen Gang]

And I guess, we can simply bypass this discussion, because the new fix
can process write() well and let history_truncate_file() never return -1.

Thanks.

On 06/23/2014 10:59 PM, Chen Gang wrote:
> On 06/23/2014 09:57 PM, Chet Ramey wrote:
>> On 6/21/14, 10:01 PM, Chen Gang wrote:
>>
>>>> history_truncate_file will never return -1.
>>>>
>>>
>>> Hmm... do you mean:
>>>
>>>   "for regular file, write() never return 0, if parameter 'count' > 0?"
>>>
> 
> I am not quite sure whether it is true, in my experience, it should be
> true, but I have no any proofs for it (if you have, welcome to supply).
> 
> thank.
> 
>>>   or
>>>
>>>   "if write() return 0, can also return 0 to history_truncate_file()?".
>>
> 
> For me, if write() could return 0, when it happened, we had to process the
> case within history_truncate_file(), could not only return 0 to indicate
> all things go on well.
> 
> 
>> Both of those things are true, but neither is what I said above.
>>
> 
> If both of those things are not true, what your originally said above
> are not true, either.
> 
> 
> Thanks.
> 


-- 
Chen Gang

Open share and attitude like air water and life which God blessed