#!/usr/bin/python3

# The MIT License (MIT)
#
# Copyright © 2024 pacman64
#
# Permission is hereby granted, free of charge, to any person obtaining a copy
# of this software and associated documentation files (the “Software”), to deal
# in the Software without restriction, including without limitation the rights
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
# copies of the Software, and to permit persons to whom the Software is
# furnished to do so, subject to the following conditions:
#
# The above copyright notice and this permission notice shall be included in
# all copies or substantial portions of the Software.
#
# THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
# SOFTWARE.


from multiprocessing import Pool
from sys import argv, exit, stderr, stdin, stdout
from typing import Dict


info = '''
coby [filepaths/URIs...]

COunt BYtes finds various byte-related stats for the files/URIs given. When
given no named inputs, it uses standard input by default.
'''

# a leading help-option arg means show the help message and quit
if len(argv) == 2 and argv[1] in ('-h', '--h', '-help', '--help'):
    print(info.strip(), file=stderr)
    exit(0)


def fail(msg, code: int = 1) -> None:
    'Show the error message given, and quit the app right away.'
    print(f'\x1b[31m{msg}\x1b[0m', file=stderr)
    exit(code)


def count_bytes(src) -> Dict[str, int]:
    lines = 0
    crlf = 0
    trails = 0
    tally = [0] * 256

    # counting lines with trailing spaces needs remembering the previous 2
    # bytes, as the last trailing space in a line can come either before a
    # single line-feed byte, or a CRLF byte-pair
    prev2 = 0
    prev1 = 0

    for chunk in src:
        for b in chunk:
            tally[b] += 1
            is10 = b == 10 # 10 is ord('\n')
            crlf += is10 and prev1 == 13 # 13 is ord('\r')
            trails += (is10 and prev1 == 32) or (prev1 == 13 and prev2 == 32)
            # notice how the last 2 bytes are remembered even across chunks
            prev2 = prev1
            prev1 = b

    n = sum(tally)
    lines = tally[ord('\n')]
    if lines == 0 and n > 0:
        lines += 1

    return {
        'n': sum(tally),
        'lines': lines,
        'lf': tally[ord('\n')],
        'crlf': crlf,
        'tabs': tally[ord('\t')],
        'spaces': tally[ord(' ')],
        'trails': trails,
        'nulls': tally[0],
        'fulls': tally[255],
        'highs': sum(tally[128:]),
    }


def seems_url(s: str) -> bool:
    protocols = ('https://', 'http://', 'file://', 'ftp://', 'data:')
    return any(s.startswith(p) for p in protocols)


def handle_named_input(path: str) -> Dict[str, int]:
    if path == '-':
        return count_bytes(stdin.buffer)

    if seems_url(path):
        with urlopen(path) as inp:
            return count_bytes(inp)

    with open(path, mode='rb') as inp:
        return count_bytes(inp)


header = '''
name\tbytes\tlines\tlf\tcrlf\ttabs\tspaces\ttrails\tnulls\tfulls\thighs
'''.strip()


def show_results(name, counts) -> None:
    stdout.write(name)
    stdout.write('\t')
    stdout.write(str(counts['n']))
    for k in header.split('\t')[2:]:
        stdout.write('\t')
        stdout.write(str(counts[k]))
    stdout.write('\n')


try:
    args = argv[1:]
    if args.count('-') > 1:
        msg = 'reading from `-` (standard input) more than once not allowed'
        raise ValueError(msg)

    if any(seems_url(e) for e in args):
        from urllib.request import urlopen

    # given no named inputs, just use stdin
    if len(args) == 0:
        args = ['-']

    # show header line right away, to reassure users something's happening
    stdout.write(header)
    stdout.write('\n')

    if len(args) == 1:
        # don't bother starting multiple interpreters for a single input
        results = [handle_named_input(args[0])]
    else:
        # vastly speed-up script by handling multiple inputs concurrently
        with Pool(processes=min(4, len(args))) as pool:
            results = pool.map(handle_named_input, args)

    for name, counts in zip(args, results):
        show_results(name, counts)
except BrokenPipeError:
    # quit quietly, instead of showing a confusing error message
    stderr.close()
except KeyboardInterrupt:
    exit(2)
except Exception as e:
    fail(e, 1)