[yt-dlp.git] / yt_dlp / downloader / mhtml.py

# coding: utf-8
from __future__ import unicode_literals

import io
import quopri
import re
import uuid

from .fragment import FragmentFD
from ..utils import (
    escapeHTML,
    formatSeconds,
    srt_subtitles_timecode,
    urljoin,
)
from ..version import __version__ as YT_DLP_VERSION


class MhtmlFD(FragmentFD):
    FD_NAME = 'mhtml'

    _STYLESHEET = """\
html, body {
    margin: 0;
    padding: 0;
    height: 100vh;
}

html {
    overflow-y: scroll;
    scroll-snap-type: y mandatory;
}

body {
    scroll-snap-type: y mandatory;
    display: flex;
    flex-flow: column;
}

body > figure {
    max-width: 100vw;
    max-height: 100vh;
    scroll-snap-align: center;
}

body > figure > figcaption {
    text-align: center;
    height: 2.5em;
}

body > figure > img {
    display: block;
    margin: auto;
    max-width: 100%;
    max-height: calc(100vh - 5em);
}
"""
    _STYLESHEET = re.sub(r'\s+', ' ', _STYLESHEET)
    _STYLESHEET = re.sub(r'\B \B|(?<=[\w\-]) (?=[^\w\-])|(?<=[^\w\-]) (?=[\w\-])', '', _STYLESHEET)

    @staticmethod
    def _escape_mime(s):
        return '=?utf-8?Q?' + (b''.join(
            bytes((b,)) if b >= 0x20 else b'=%02X' % b
            for b in quopri.encodestring(s.encode('utf-8'), header=True)
        )).decode('us-ascii') + '?='

    def _gen_cid(self, i, fragment, frag_boundary):
        return '%u.%s@yt-dlp.github.io.invalid' % (i, frag_boundary)

    def _gen_stub(self, *, fragments, frag_boundary, title):
        output = io.StringIO()

        output.write((
            '<!DOCTYPE html>'
            '<html>'
            '<head>'
            ''  '<meta name="generator" content="yt-dlp {version}">'
            ''  '<title>{title}</title>'
            ''  '<style>{styles}</style>'
            '<body>'
        ).format(
            version=escapeHTML(YT_DLP_VERSION),
            styles=self._STYLESHEET,
            title=escapeHTML(title)
        ))

        t0 = 0
        for i, frag in enumerate(fragments):
            output.write('<figure>')
            try:
                t1 = t0 + frag['duration']
                output.write((
                    '<figcaption>Slide #{num}: {t0} – {t1} (duration: {duration})</figcaption>'
                ).format(
                    num=i + 1,
                    t0=srt_subtitles_timecode(t0),
                    t1=srt_subtitles_timecode(t1),
                    duration=formatSeconds(frag['duration'], msec=True)
                ))
            except (KeyError, ValueError, TypeError):
                t1 = None
                output.write((
                    '<figcaption>Slide #{num}</figcaption>'
                ).format(num=i + 1))
            output.write('<img src="cid:{cid}">'.format(
                cid=self._gen_cid(i, frag, frag_boundary)))
            output.write('</figure>')
            t0 = t1

        return output.getvalue()

    def real_download(self, filename, info_dict):
        fragment_base_url = info_dict.get('fragment_base_url')
        fragments = info_dict['fragments'][:1] if self.params.get(
            'test', False) else info_dict['fragments']
        title = info_dict.get('title', info_dict['format_id'])
        origin = info_dict.get('webpage_url', info_dict['url'])

        ctx = {
            'filename': filename,
            'total_frags': len(fragments),
        }

        self._prepare_and_start_frag_download(ctx, info_dict)

        extra_state = ctx.setdefault('extra_state', {
            'header_written': False,
            'mime_boundary': str(uuid.uuid4()).replace('-', ''),
        })

        frag_boundary = extra_state['mime_boundary']

        if not extra_state['header_written']:
            stub = self._gen_stub(
                fragments=fragments,
                frag_boundary=frag_boundary,
                title=title
            )

            ctx['dest_stream'].write((
                'MIME-Version: 1.0\r\n'
                'From: <nowhere@yt-dlp.github.io.invalid>\r\n'
                'To: <nowhere@yt-dlp.github.io.invalid>\r\n'
                'Subject: {title}\r\n'
                'Content-type: multipart/related; '
                ''  'boundary="{boundary}"; '
                ''  'type="text/html"\r\n'
                'X.yt-dlp.Origin: {origin}\r\n'
                '\r\n'
                '--{boundary}\r\n'
                'Content-Type: text/html; charset=utf-8\r\n'
                'Content-Length: {length}\r\n'
                '\r\n'
                '{stub}\r\n'
            ).format(
                origin=origin,
                boundary=frag_boundary,
                length=len(stub),
                title=self._escape_mime(title),
                stub=stub
            ).encode('utf-8'))
            extra_state['header_written'] = True

        for i, fragment in enumerate(fragments):
            if (i + 1) <= ctx['fragment_index']:
                continue

            fragment_url = urljoin(fragment_base_url, fragment['path'])
            success, frag_content = self._download_fragment(ctx, fragment_url, info_dict)
            if not success:
                continue

            mime_type = b'image/jpeg'
            if frag_content.startswith(b'\x89PNG\r\n\x1a\n'):
                mime_type = b'image/png'
            if frag_content.startswith((b'GIF87a', b'GIF89a')):
                mime_type = b'image/gif'
            if frag_content.startswith(b'RIFF') and frag_content[8:12] == 'WEBP':
                mime_type = b'image/webp'

            frag_header = io.BytesIO()
            frag_header.write(
                b'--%b\r\n' % frag_boundary.encode('us-ascii'))
            frag_header.write(
                b'Content-ID: <%b>\r\n' % self._gen_cid(i, fragment, frag_boundary).encode('us-ascii'))
            frag_header.write(
                b'Content-type: %b\r\n' % mime_type)
            frag_header.write(
                b'Content-length: %u\r\n' % len(frag_content))
            frag_header.write(
                b'Content-location: %b\r\n' % fragment_url.encode('us-ascii'))
            frag_header.write(
                b'X.yt-dlp.Duration: %f\r\n' % fragment['duration'])
            frag_header.write(b'\r\n')
            self._append_fragment(
                ctx, frag_header.getvalue() + frag_content + b'\r\n')

        ctx['dest_stream'].write(
            b'--%b--\r\n\r\n' % frag_boundary.encode('us-ascii'))
        self._finish_frag_download(ctx, info_dict)
        return True
Commit	Line	Data
cdb19aa4	1	# coding: utf-8
	2	from __future__ import unicode_literals
	3
	4	import io
	5	import quopri
	6	import re
	7	import uuid
	8
	9	from .fragment import FragmentFD
	10	from ..utils import (
	11	escapeHTML,
	12	formatSeconds,
	13	srt_subtitles_timecode,
	14	urljoin,
	15	)
	16	from ..version import __version__ as YT_DLP_VERSION
	17
	18
	19	class MhtmlFD(FragmentFD):
	20	FD_NAME = 'mhtml'
	21
	22	_STYLESHEET = """\
	23	html, body {
	24	margin: 0;
	25	padding: 0;
	26	height: 100vh;
	27	}
	28
	29	html {
	30	overflow-y: scroll;
	31	scroll-snap-type: y mandatory;
	32	}
	33
	34	body {
	35	scroll-snap-type: y mandatory;
	36	display: flex;
	37	flex-flow: column;
	38	}
	39
	40	body > figure {
	41	max-width: 100vw;
	42	max-height: 100vh;
	43	scroll-snap-align: center;
	44	}
	45
	46	body > figure > figcaption {
	47	text-align: center;
	48	height: 2.5em;
	49	}
	50
	51	body > figure > img {
	52	display: block;
	53	margin: auto;
	54	max-width: 100%;
	55	max-height: calc(100vh - 5em);
	56	}
	57	"""
	58	_STYLESHEET = re.sub(r'\s+', ' ', _STYLESHEET)
	59	_STYLESHEET = re.sub(r'\B \B\|(?<=[\w\-]) (?=[^\w\-])\|(?<=[^\w\-]) (?=[\w\-])', '', _STYLESHEET)
	60
	61	@staticmethod
	62	def _escape_mime(s):
	63	return '=?utf-8?Q?' + (b''.join(
	64	bytes((b,)) if b >= 0x20 else b'=%02X' % b
65	for b in quopri.encodestring(s.encode('utf-8'), header=True)
66	)).decode('us-ascii') + '?='
67
68	def _gen_cid(self, i, fragment, frag_boundary):
69	return '%u.%s@yt-dlp.github.io.invalid' % (i, frag_boundary)
70
71	def _gen_stub(self, *, fragments, frag_boundary, title):
72	output = io.StringIO()
73
74	output.write((
75	'<!DOCTYPE html>'
76	'<html>'
77	'<head>'
78	'' '<meta name="generator" content="yt-dlp {version}">'
79	'' '<title>{title}</title>'
80	'' '<style>{styles}</style>'
81	'<body>'
82	).format(
83	version=escapeHTML(YT_DLP_VERSION),
84	styles=self._STYLESHEET,
85	title=escapeHTML(title)
86	))
87
88	t0 = 0
89	for i, frag in enumerate(fragments):
90	output.write('<figure>')
91	try:
92	t1 = t0 + frag['duration']
93	output.write((
94	'<figcaption>Slide #{num}: {t0} – {t1} (duration: {duration})</figcaption>'
95	).format(
96	num=i + 1,
97	t0=srt_subtitles_timecode(t0),
98	t1=srt_subtitles_timecode(t1),
99	duration=formatSeconds(frag['duration'], msec=True)
100	))
101	except (KeyError, ValueError, TypeError):
102	t1 = None
103	output.write((
104	'<figcaption>Slide #{num}</figcaption>'
105	).format(num=i + 1))
106	output.write('<img src="cid:{cid}">'.format(
107	cid=self._gen_cid(i, frag, frag_boundary)))
108	output.write('</figure>')
109	t0 = t1
110
111	return output.getvalue()
112
113	def real_download(self, filename, info_dict):
114	fragment_base_url = info_dict.get('fragment_base_url')
115	fragments = info_dict['fragments'][:1] if self.params.get(
116	'test', False) else info_dict['fragments']
d76991ab	117	title = info_dict.get('title', info_dict['format_id'])
d76991ab	118	origin = info_dict.get('webpage_url', info_dict['url'])
cdb19aa4	119
	120	ctx = {
	121	'filename': filename,
	122	'total_frags': len(fragments),
	123	}
	124
3ba7740d	125	self._prepare_and_start_frag_download(ctx, info_dict)
cdb19aa4	126
	127	extra_state = ctx.setdefault('extra_state', {
	128	'header_written': False,
	129	'mime_boundary': str(uuid.uuid4()).replace('-', ''),
	130	})
	131
	132	frag_boundary = extra_state['mime_boundary']
	133
	134	if not extra_state['header_written']:
	135	stub = self._gen_stub(
	136	fragments=fragments,
	137	frag_boundary=frag_boundary,
	138	title=title
	139	)
	140
	141	ctx['dest_stream'].write((
	142	'MIME-Version: 1.0\r\n'
	143	'From: <nowhere@yt-dlp.github.io.invalid>\r\n'
	144	'To: <nowhere@yt-dlp.github.io.invalid>\r\n'
	145	'Subject: {title}\r\n'
	146	'Content-type: multipart/related; '
	147	'' 'boundary="{boundary}"; '
	148	'' 'type="text/html"\r\n'
	149	'X.yt-dlp.Origin: {origin}\r\n'
	150	'\r\n'
	151	'--{boundary}\r\n'
	152	'Content-Type: text/html; charset=utf-8\r\n'
	153	'Content-Length: {length}\r\n'
	154	'\r\n'
	155	'{stub}\r\n'
	156	).format(
	157	origin=origin,
	158	boundary=frag_boundary,
	159	length=len(stub),
	160	title=self._escape_mime(title),
	161	stub=stub
	162	).encode('utf-8'))
	163	extra_state['header_written'] = True
	164
	165	for i, fragment in enumerate(fragments):
	166	if (i + 1) <= ctx['fragment_index']:
	167	continue
	168
	169	fragment_url = urljoin(fragment_base_url, fragment['path'])
	170	success, frag_content = self._download_fragment(ctx, fragment_url, info_dict)
	171	if not success:
	172	continue
	173
	174	mime_type = b'image/jpeg'
	175	if frag_content.startswith(b'\x89PNG\r\n\x1a\n'):
	176	mime_type = b'image/png'
	177	if frag_content.startswith((b'GIF87a', b'GIF89a')):
	178	mime_type = b'image/gif'
	179	if frag_content.startswith(b'RIFF') and frag_content[8:12] == 'WEBP':
	180	mime_type = b'image/webp'
	181
	182	frag_header = io.BytesIO()
	183	frag_header.write(
	184	b'--%b\r\n' % frag_boundary.encode('us-ascii'))
	185	frag_header.write(
	186	b'Content-ID: <%b>\r\n' % self._gen_cid(i, fragment, frag_boundary).encode('us-ascii'))
	187	frag_header.write(
	188	b'Content-type: %b\r\n' % mime_type)
	189	frag_header.write(
190	b'Content-length: %u\r\n' % len(frag_content))
191	frag_header.write(
192	b'Content-location: %b\r\n' % fragment_url.encode('us-ascii'))
193	frag_header.write(
194	b'X.yt-dlp.Duration: %f\r\n' % fragment['duration'])
195	frag_header.write(b'\r\n')
196	self._append_fragment(
197	ctx, frag_header.getvalue() + frag_content + b'\r\n')
198
199	ctx['dest_stream'].write(
200	b'--%b--\r\n\r\n' % frag_boundary.encode('us-ascii'))
3ba7740d	201	self._finish_frag_download(ctx, info_dict)
cdb19aa4	202	return True