aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/downloader/http.py
blob: f79e6a99587cdc7d13ba210e528512a1e572e4ea (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
import os
import time

from .common import FileDownloader
from ..utils import (
    compat_urllib_request,
    compat_urllib_error,
    ContentTooShortError,

    encodeFilename,
    sanitize_open,
    format_bytes,
)


class HttpFD(FileDownloader):
    _TEST_FILE_SIZE = 10241

    def real_download(self, filename, info_dict):
        url = info_dict['url']
        tmpfilename = self.temp_name(filename)
        stream = None

        # Do not include the Accept-Encoding header
        headers = {'Youtubedl-no-compression': 'True'}
        if 'user_agent' in info_dict:
            headers['Youtubedl-user-agent'] = info_dict['user_agent']
        if 'http_referer' in info_dict:
            headers['Referer'] = info_dict['http_referer']
        basic_request = compat_urllib_request.Request(url, None, headers)
        request = compat_urllib_request.Request(url, None, headers)

        is_test = self.params.get('test', False)

        if is_test:
            request.add_header('Range', 'bytes=0-%s' % str(self._TEST_FILE_SIZE - 1))

        # Establish possible resume length
        if os.path.isfile(encodeFilename(tmpfilename)):
            resume_len = os.path.getsize(encodeFilename(tmpfilename))
        else:
            resume_len = 0

        open_mode = 'wb'
        if resume_len != 0:
            if self.params.get('continuedl', False):
                self.report_resuming_byte(resume_len)
                request.add_header('Range', 'bytes=%d-' % resume_len)
                open_mode = 'ab'
            else:
                resume_len = 0

        count = 0
        retries = self.params.get('retries', 0)
        while count <= retries:
            # Establish connection
            try:
                data = self.ydl.urlopen(request)
                break
            except (compat_urllib_error.HTTPError, ) as err:
                if (err.code < 500 or err.code >= 600) and err.code != 416:
                    # Unexpected HTTP error
                    raise
                elif err.code == 416:
                    # Unable to resume (requested range not satisfiable)
                    try:
                        # Open the connection again without the range header
                        data = self.ydl.urlopen(basic_request)
                        content_length = data.info()['Content-Length']
                    except (compat_urllib_error.HTTPError, ) as err:
                        if err.code < 500 or err.code >= 600:
                            raise
                    else:
                        # Examine the reported length
                        if (content_length is not None and
                                (resume_len - 100 < int(content_length) < resume_len + 100)):
                            # The file had already been fully downloaded.
                            # Explanation to the above condition: in issue #175 it was revealed that
                            # YouTube sometimes adds or removes a few bytes from the end of the file,
                            # changing the file size slightly and causing problems for some users. So
                            # I decided to implement a suggested change and consider the file
                            # completely downloaded if the file size differs less than 100 bytes from
                            # the one in the hard drive.
                            self.report_file_already_downloaded(filename)
                            self.try_rename(tmpfilename, filename)
                            self._hook_progress({
                                'filename': filename,
                                'status': 'finished',
                            })
                            return True
                        else:
                            # The length does not match, we start the download over
                            self.report_unable_to_resume()
                            resume_len = 0
                            open_mode = 'wb'
                            break
            # Retry
            count += 1
            if count <= retries:
                self.report_retry(count, retries)

        if count > retries:
            self.report_error(u'giving up after %s retries' % retries)
            return False

        data_len = data.info().get('Content-length', None)

        # Range HTTP header may be ignored/unsupported by a webserver
        # (e.g. extractor/scivee.py, extractor/bambuser.py).
        # However, for a test we still would like to download just a piece of a file.
        # To achieve this we limit data_len to _TEST_FILE_SIZE and manually control
        # block size when downloading a file.
        if is_test and (data_len is None or int(data_len) > self._TEST_FILE_SIZE):
            data_len = self._TEST_FILE_SIZE

        if data_len is not None:
            data_len = int(data_len) + resume_len
            min_data_len = self.params.get("min_filesize", None)
            max_data_len = self.params.get("max_filesize", None)
            if min_data_len is not None and data_len < min_data_len:
                self.to_screen(u'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
                return False
            if max_data_len is not None and data_len > max_data_len:
                self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
                return False

        data_len_str = format_bytes(data_len)
        byte_counter = 0 + resume_len
        block_size = self.params.get('buffersize', 1024)
        start = time.time()
        while True:
            # Download and write
            before = time.time()
            data_block = data.read(block_size if not is_test else min(block_size, data_len - byte_counter))
            after = time.time()
            if len(data_block) == 0:
                break
            byte_counter += len(data_block)

            # Open file just in time
            if stream is None:
                try:
                    (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
                    assert stream is not None
                    filename = self.undo_temp_name(tmpfilename)
                    self.report_destination(filename)
                except (OSError, IOError) as err:
                    self.report_error(u'unable to open for writing: %s' % str(err))
                    return False
            try:
                stream.write(data_block)
            except (IOError, OSError) as err:
                self.to_stderr(u"\n")
                self.report_error(u'unable to write data: %s' % str(err))
                return False
            if not self.params.get('noresizebuffer', False):
                block_size = self.best_block_size(after - before, len(data_block))

            # Progress message
            speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
            if data_len is None:
                eta = percent = None
            else:
                percent = self.calc_percent(byte_counter, data_len)
                eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
            self.report_progress(percent, data_len_str, speed, eta)

            self._hook_progress({
                'downloaded_bytes': byte_counter,
                'total_bytes': data_len,
                'tmpfilename': tmpfilename,
                'filename': filename,
                'status': 'downloading',
                'eta': eta,
                'speed': speed,
            })

            if is_test and byte_counter == data_len:
                break

            # Apply rate limit
            self.slow_down(start, byte_counter - resume_len)

        if stream is None:
            self.to_stderr(u"\n")
            self.report_error(u'Did not get any data blocks')
            return False
        stream.close()
        self.report_finish(data_len_str, (time.time() - start))
        if data_len is not None and byte_counter != data_len:
            raise ContentTooShortError(byte_counter, int(data_len))
        self.try_rename(tmpfilename, filename)

        # Update file modification time
        if self.params.get('updatetime', True):
            info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None))

        self._hook_progress({
            'downloaded_bytes': byte_counter,
            'total_bytes': byte_counter,
            'filename': filename,
            'status': 'finished',
        })

        return True