convert.py

#!/usr/bin/env python3

import argparse
import csv
from math import log
from PIL import Image

ZOOM = 10
SCALE_MAX = 1_500_000

parser = argparse.ArgumentParser()
parser.add_argument("input", help = "Name of file to read from")
parser.add_argument("output", help = "Name of PNG to write to")
args = parser.parse_args()



# build a lookup table for colors
colortable = []

with open('extended-kindlmann-table-byte-1024.csv') as csvfile:
    reader = csv.reader(csvfile)
    next(reader) # skip header
    for row in reader:
        colortable.append((int(row[1]), int(row[2]), int(row[3])))

COLORTABLE_MAX = len(colortable) - 1

img = Image.new('RGB', (2**ZOOM, 2**ZOOM), "black")
pixels = img.load()
for line in open(args.input):
    splitline=line.split(' ', 1)
    tile=splitline[0].split('/', 1)
    x = int(tile[0])
    y = int(tile[1])
    hits = int(splitline[1])

    # the math gets tricky, because we need 0 hits to map to black (colortable[0]), but log(0) isn't defined. So we add 1 to all hit counts
    loghits = log(hits + 1)

    # We know 0 < loghits/log(SCALE_MAX) <= 1.

    lookup = min(int(round(loghits/log(SCALE_MAX)*COLORTABLE_MAX)), COLORTABLE_MAX)
    pixels[x,y] = colortable[lookup]

img.save(args.output)

extended-kindlmann-table-byte-1024.csv

generate_txt.sh

#!/bin/bash

#=======================================================================
# Specify the "start" and "end" time.
#=======================================================================
# Use format "HH:MM YYYY-MM-DD"
start="00:00 2023-01-01"
end="00:00 2023-05-31"
increment="+1 days"
#=======================================================================

# Convert start and end time to full Bash datetime format
# For example, it converts "00:00 2021-04-01" to "Thu Apr 1 00:00:00 GMT 2021"
start=$(date -ud "${start}")
end=$(date -ud "${end}")

# The below while statement will loop over each date between the start and
# end time. Each loop will increment the date by "+1 hours" (defined above).

# NOTE: The +%s in the first line converts the date to "seconds since 
# EPOC" which makes the comparison between start and end time possible.
# See bottom of page here: https://phoenixnap.com/kb/linux-date-command
while (( $(date -ud "${start}" +%s) <= $(date -ud "${end}" +%s) )); do

    current=$(date -ud "${start}" '+%F')
    tilelog --date "$current" --tile hourly/
   
    # Increment the value. This changes the value of `$start` every loop
    # with the next date.
    start=$(date -ud "${start} ${increment}")
done

maxreqs.py

#!/usr/bin/env python3

import argparse

parser = argparse.ArgumentParser()
parser.add_argument('files', type=argparse.FileType('r'), nargs='+')
args = parser.parse_args()

maxviews = 0
for f in args.files:
    
    for line in f:
        splitline=line.split(' ', 1)
        maxviews = max(maxviews, int(splitline[1]))

print(maxviews)

notes.txt

generate hourly files with modified tilelog
find maximum views with maxreqs.py and set constants in convert.py accordingly
create PNGs with for f in hourly/*.txt; do ./convert.py "$f" "hourly/$(basename $f).png"; done
turn PNGs to video with ffmpeg
cd hourly && ffmpeg -framerate 15 -pattern_type glob -i '*.png' -c:v libsvtav1 -crf 10 -b:v 5000k out.mp4

adjust ffmpeg as needed, still need to find good settings.