My 1billion row challenge solutions in various languages

More accurate file size estimate

Changed files
+9 -13
src
+9 -13
src/main/python/create_measurements.py
···
"""
Tries to estimate how large a file the test data will be
"""
-
max_string = float('-inf')
-
min_string = float('inf')
-
per_record_size = 0
-
record_size_unit = "bytes"
+
total_name_bytes = sum(len(s.encode("utf-8")) for s in weather_station_names)
+
avg_name_bytes = total_name_bytes / float(len(weather_station_names))
-
for station in weather_station_names:
-
if len(station) > max_string:
-
max_string = len(station)
-
if len(station) < min_string:
-
min_string = len(station)
-
per_record_size = ((max_string + min_string * 2) + len(",-123.4")) / 2
+
# avg_temp_bytes = sum(len(str(n / 10)) for n in range(-999, 1000)) / 1999
+
avg_temp_bytes = 4.400200100050025
+
+
# add 2 for separator and newline
+
avg_line_length = avg_name_bytes + avg_temp_bytes + 2
-
total_file_size = num_rows_to_create * per_record_size
-
human_file_size = convert_bytes(total_file_size)
+
human_file_size = convert_bytes(num_rows_to_create * avg_line_length)
-
return f"Estimated max file size is: {human_file_size}.\nTrue size is probably much smaller (around half)."
+
return f"Estimated max file size is: {human_file_size}."
def build_test_data(weather_station_names, num_rows_to_create):