>> Actually what I ask is simpler, an HOURLY of QUARTER or FIVE which picks a
>> single (set of adjacent) hours, quarter, fives ... indipendent of the date
>> it finds in the log (if the log spans more than one day, just add things
>> up) and from launch time.
>>
>> (Am I making any sense?)
>
>Jeremy Wadsack wrote:
>Not to me. I'm confused between your use of terms like DAILY, HOURLY, QUARTER
>and FIVE which look like Analog report type commands. What does that have to do
>with the FROM and TO fields?
I *was* cryptic indeed.
It's really simple.
1) the QUARTER and FIVE report are inconsistently named, with respect to
the other time-based reports. Their proper names should be FULLQUARTER and
FULLFIVE, just as we have HOURLY and FULLHOURLY.
2) given 1), I can figure out that FROM and TO would allow the following
time specifications:
a) time FROM and TO a given absolute date (hopefully present in the log)
b) time FROM and TO a given relative date (relative to Analog launch time)
c) time FROM and TO a given timespan, of the kind already available: days
of the week, hours of day, quarters of hours, five minutes.
a) and b) are the accepted usage of FROM and TO
c) is just what Analog already does with days and hours, applied to
quarters and fives!
example: given a 28-day long log, I can do DAILY to know what happens on
a given, say, Monday. It's the DAILY (as opposed to FULLDAILY) report. Fine
Now I can also know what happens on the average 9-to-1Oam hour, it's the
HOURLY report (as opposed to FULLHOURLY)
*But*, I can't go any smaller than that. Why? Analog clearly knows how to
collapse a longer time period in congruence classes (sorry, my translation
of math italian) onto a given set (days, 24 hours), adding things up.
Actually this latter argument is more important than that regarding FROM
and TO. I would be comfortble to compute a proper QUARTER or FIVE report
(in the sense that it should work like HOURLY, not like FULLHOURLY as it
does now) to just check the quarter or fiveminutes that's important for me.
Just yhinking aloud, but I think it shouldn't be much stuff to put in. I
mean, the machinery is clearly already there. Just a matter of making it
also work at a finer grain than hours.
Ok. Now, it's back to my daily million lines, see you.
Walter Vannini
------------------------------------------------------------------------
This is the analog-help mailing list. To unsubscribe from this
mailing list, send mail to [EMAIL PROTECTED]
with "unsubscribe" in the main BODY OF THE MESSAGE.
List archived at http://www.mail-archive.com/[email protected]/
------------------------------------------------------------------------
