On 01/04/20 10:19, Stefan Raspl wrote:
> On 2020-03-31 23:45, Paolo Bonzini wrote:
>> On 31/03/20 22:00, Stefan Raspl wrote:
>>> @@ -1523,14 +1535,20 @@ def log(stats, opts, frmt, keys):
>>> """Prints statistics as reiterating key block, multiple value blocks."""
>>> line = 0
>>> banner_repeat = 20
>>> + banner_printed = False
>>> +
>>> while True:
>>> try:
>>> time.sleep(opts.set_delay)
>>> - if line % banner_repeat == 0:
>>> + if line % banner_repeat == 0 and not banner_printed:
>>> print(frmt.get_banner())
>>> - print(datetime.now().strftime("%Y-%m-%d %H:%M:%S") +
>>> - frmt.get_statline(keys, stats.get()))
>>> + banner_printed = True
>>
>> Can't skip_zero_records be handled here instead?
>>
>> values = stats.get()
>> if not opts.skip_zero_records or \
>> any((values[k].delta != 0 for k in keys):
>> statline = frmt.get_statline(keys, values)
>> print(datetime.now().strftime("%Y-%m-%d %H:%M:%S") + statline)
>
> I wanted to avoid such an extra check for performance reasons. Granted, I have
> to do something likewise (i.e. checking for non-zero values) in my patch for csv
> records, but at least for the standard format things are a bit less costly
> (avoiding an extra pass over the data).
I don't think it's a perceivable difference, and it simplifies the code
noticeably.
Paolo
