tldr_find_too_long.py

#!/usr/bin/env python3

import os

def calc_lengths():
    max_lengths = {}

    for root in os.listdir('.'):
        if not root.startswith('pages'):
            continue

        for platform in os.listdir(root):
            for page in os.listdir(os.path.join(root, platform)):
                fname = os.path.join(root, platform, page)

                with open(fname, encoding='utf8') as f:
                    max_length = 0

                    for line in f:
                        line = line.rstrip('\n')
                        if line.startswith('>') or line.startswith('-'):
                            max_length = max(max_length, len(line))

                    max_lengths[fname] = max_length

    return max_lengths

def find_too_long(max_lengths, limit, old_found={}):
    found = {}

    for fname, max_length in max_lengths.items():
        diff = max_length - limit
        if diff > 0:
            if fname not in old_found:
                found[fname] = (max_length, diff)

    n = len(found)

    if (n <= 50):
        print('Limit %d:' % limit, n, '{}page{} over the limit.'.format('more ' if old_found else '', 's' if n == 0 or n > 1 else ''))
        for fname, (max_length, diff) in sorted(found.items(), key=lambda x: (-x[1][0], x[0])):
            print('{:<50s}  {} ({} over)'.format(fname.replace('\\', '/'), max_length, diff))
    else:
        print('Limit %d:' % limit, n, '{}pages over the limit (avoiding listing).'.format('more ' if old_found else ''))

    print('-' * 65)

    old_found.update(found)

ml = calc_lengths()

for limit in (250, 240, 220, 200, 180, 150, 140, 100, 80):
    find_too_long(ml, limit)