Open dbetchkal opened 7 years ago
Obviously the most expedient way to complete this analysis would be to run through the desired time range on each day and just calculate the intervals that happen in that block, listing them!
This handles both issues:
If NVSPL data exist for the entire day, the day can be calculated, else, drop the day from the list of those to be calculated.
Since noise-free intervals begin and terminate with noise events, check to see if any noise events begin or end in the desired time range, and use start_bound, noise events, and end_bound to determine intervals.
Allow the user to define a time range (for example an arbitrary daytime range like 07:00 - 19:00) such that intervals within the time range can be listed/summarized.
To do this, we must determine for interval i:
for the day i begins,_ does i begin earlier than the starting bound? if so, truncate i to start with the starting bound.
for the day i begins, does i begin after the ending bound? if so, check if i ends after the next beginning bound. if so, truncate i to the next starting bound. else, drop the interval from the analysis.
THEN
Then we'd also want to handle for time periods when data were not being collected. (Incidentally this is an issue for ALL NFI functions as they currently stand and might be worth addressing first.)