#!/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 struct import unpack
from sys import argv, exit, stderr, stdin, stdout


info = '''
id3pic [options...] [filepath/URI...]


Extract picture/thumbnail bytes from ID3/MP3 metadata, when available.

Any leading options can start with either single or double-dash:

    -h          show this help message
    -help       show this help message
'''

# handle standard help cmd-line options, quitting right away in that case
if len(argv) == 2 and argv[1] in ('-h', '--h', '-help', '--help'):
    print(info.strip(), file=stderr)
    exit(0)


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


def read_byte(src) -> int:
    b = src.read(1)
    return b[0] if b else -1


def skip_zstring(src) -> int:
    return match_byte(src, 0)


def skip_thumbnail_type_apic(src) -> int:
    if read_byte(src) < 0:
        raise Exception('failed to sync to thumbnail text-encoding')
    n = 1

    m = match_byte(src, ord('/'))
    if m < 0:
        raise Exception('failed to sync to thumbnail MIME-type')
    n += m
    m = skip_zstring(src)
    if m < 0:
        raise Exception('failed to sync to thumbnail MIME-type')
    n += m

    if read_byte(src) < 0:
        raise Exception('failed to sync to thumbnail picture type')
    n += 1

    m = skip_zstring(src)
    if m < 0:
        raise Exception('failed to sync to thumbnail comment')
    n += m

    return n


def match_byte(src, what: int) -> int:
    n = 0

    while True:
        b = src.read(1)
        if not b:
            return -1

        n += 1

        if b[0] == what:
            return n


def match_bytes(src, seq: bytes) -> int:
    i = 0
    n = 0
    end = len(seq)

    while True:
        b = src.read(1)
        if not b:
            return -1

        n += 1

        if b[0] == seq[i]:
            i += 1
        else:
            i = 0

        if i >= end:
            return n


def handle_apic(w, src) -> bool:
    # section-size seems stored as 4 little-endian bytes
    chunk = src.read(4)
    if not chunk:
        raise Exception('failed to read thumbnail-payload size')
    size = unpack('>I', chunk)[0]

    n = skip_thumbnail_type_apic(src)
    if n < 0:
        raise Exception('failed to sync to start of thumbnail data')
    size -= n

    # copy all thumbnail bytes
    w.write(src.read(size))
    return True


def handle_pic(w, src) -> bool:
    # http://www.unixgods.org/Ruby/ID3/docs/id3v2-00.html#PIC

    # thumbnail-payload-size seems stored as 3 big-endian bytes
    a = src.read(1)
    b = src.read(1)
    c = src.read(1)
    if not (a and b and c):
        raise Exception('failed to read thumbnail-payload size')
    size = (256 * 256) * a[0] + 256 * b[0] + c[0]

    # skip the text encoding
    src.read(5)

    # skip a null-delimited string
    while True:
        b = src.read(1)
        if not b:
            raise Exception('failed to read thumbnail-payload description')

        if b[0] == 0:
            break

    # copy all thumbnail bytes
    w.write(src.read(size))
    return True


def handle_id3_picture(w, src) -> bool:
    a = ord('A')
    p = ord('P')
    i = ord('I')
    c = ord('C')

    while True:
        chunk = src.read(1)
        if not chunk:
            break
        v = chunk[0]

        if v == a:
            if src.read(1)[0] != p:
                continue
            if src.read(1)[0] != i:
                continue
            if src.read(1)[0] != c:
                continue
            return handle_apic(w, src)

        if v == p:
            if src.read(1)[0] != i:
                continue
            if src.read(1)[0] != c:
                continue
            return handle_pic(w, src)

    return False


def handle_bytes(w, src) -> None:
    if not handle_id3_picture(w, src):
        raise Exception('no thumbnail found')


name = '-' if len(argv) == 1 else argv[1]
if len(argv) > 2:
    print('\x1b[31mmultiple inputs not allowed\x1b[0m', file=stderr)
    exit(1)

try:
    if seems_url(name):
        from urllib.request import urlopen

    # handle all named inputs given
    if name == '-':
        handle_bytes(stdout.buffer, stdin.buffer)
    elif seems_url(name):
        with urlopen(name) as inp:
            handle_bytes(stdout.buffer, inp)
    else:
        with open(name, mode='rb') as inp:
            handle_bytes(stdout.buffer, inp)
except BrokenPipeError:
    # quit quietly, instead of showing a confusing error message
    stderr.close()
except KeyboardInterrupt:
    exit(2)
except Exception as e:
    print(f'\x1b[31m{e}\x1b[0m', file=stderr)
    exit(1)