Woohoo, I fixed the stupid bug that I thought was being caused by "Yesterday" and "Today". Turns out that people with special tags were messing up my grep statements.
I fixed the "Yesterday" and "Today" bugs. It's annoying, because I have to make the scripts compatible with two separate operating systems that have entirely different date functions. But it finally works!
I think I know where I'm going with this. I'm going to find the first day and the last day and make counter variables for each of the days (like a simple array), and just increment each when a date in their zone is found. Then the result will be a post per day chart. It'll be pretty easy (but not
as easy) to do this by hour. Any unit of time finer than that would require too much brain.
So what it's worth, I did a quick scan of Threads 3 and 4.
The slowest pages (50ppp, not what I'd prefer, but what the default for not logged in access, which the script is limited to) in this thread are p36 at 321min followed by p18 at 198min then p17 at 189min.
In comparison, Thread 3's slowest pages are p398 at 1027min, p135 at 654min and p321 at 483min. As you can see, as a thread gets older, it tends to slow down. There are fourteen pages in that thread that are slower than our thread's p36, and there are
fifty-four pages larger than our p18!
All this data is completely useless, but it's fun to write useless code now and again.
