|
1
|
|
|
import os |
|
2
|
|
|
import re |
|
3
|
|
|
import sys |
|
4
|
|
|
from contextlib import contextmanager |
|
5
|
|
|
|
|
6
|
|
|
import requests |
|
7
|
|
|
from progressbar import (ProgressBar, Percentage, Bar, ETA, FileTransferSpeed, |
|
8
|
|
|
Timer, UnknownLength) |
|
9
|
|
|
from six.moves import zip, urllib |
|
|
|
|
|
|
10
|
|
|
from ..exceptions import NeedURLPrefix |
|
11
|
|
|
|
|
12
|
|
|
|
|
13
|
|
|
@contextmanager |
|
14
|
|
|
def progress_bar(name, maxval): |
|
15
|
|
|
"""Manages a progress bar for a download. |
|
16
|
|
|
|
|
17
|
|
|
Parameters |
|
18
|
|
|
---------- |
|
19
|
|
|
name : str |
|
20
|
|
|
Name of the downloaded file. |
|
21
|
|
|
maxval : int |
|
22
|
|
|
Total size of the download, in bytes. |
|
23
|
|
|
|
|
24
|
|
|
""" |
|
25
|
|
|
if maxval is not UnknownLength: |
|
26
|
|
|
widgets = ['{}: '.format(name), Percentage(), ' ', |
|
27
|
|
|
Bar(marker='=', left='[', right=']'), ' ', ETA(), ' ', |
|
28
|
|
|
FileTransferSpeed()] |
|
29
|
|
|
else: |
|
30
|
|
|
widgets = ['{}: '.format(name), ' ', Timer(), ' ', FileTransferSpeed()] |
|
31
|
|
|
bar = ProgressBar(widgets=widgets, max_value=maxval, fd=sys.stdout).start() |
|
32
|
|
|
try: |
|
33
|
|
|
yield bar |
|
34
|
|
|
finally: |
|
35
|
|
|
bar.update(maxval) |
|
36
|
|
|
bar.finish() |
|
37
|
|
|
|
|
38
|
|
|
|
|
39
|
|
|
def filename_from_url(url, path=None): |
|
|
|
|
|
|
40
|
|
|
"""Parses a URL to determine a file name. |
|
41
|
|
|
|
|
42
|
|
|
Parameters |
|
43
|
|
|
---------- |
|
44
|
|
|
url : str |
|
45
|
|
|
URL to parse. |
|
46
|
|
|
|
|
47
|
|
|
""" |
|
48
|
|
|
r = requests.get(url, stream=True) |
|
49
|
|
|
if 'Content-Disposition' in r.headers: |
|
50
|
|
|
filename = re.findall(r'filename=([^;]+)', |
|
51
|
|
|
r.headers['Content-Disposition'])[0].strip('"\"') |
|
52
|
|
|
else: |
|
53
|
|
|
filename = os.path.basename(urllib.parse.urlparse(url).path) |
|
54
|
|
|
return filename |
|
55
|
|
|
|
|
56
|
|
|
|
|
57
|
|
|
def download(url, file_handle, chunk_size=1024): |
|
58
|
|
|
"""Downloads a given URL to a specific file. |
|
59
|
|
|
|
|
60
|
|
|
Parameters |
|
61
|
|
|
---------- |
|
62
|
|
|
url : str |
|
63
|
|
|
URL to download. |
|
64
|
|
|
file_handle : file |
|
65
|
|
|
Where to save the downloaded URL. |
|
66
|
|
|
|
|
67
|
|
|
""" |
|
68
|
|
|
r = requests.get(url, stream=True) |
|
69
|
|
|
total_length = r.headers.get('content-length') |
|
70
|
|
|
if total_length is None: |
|
71
|
|
|
maxval = UnknownLength |
|
72
|
|
|
else: |
|
73
|
|
|
maxval = int(total_length) |
|
74
|
|
|
name = file_handle.name |
|
75
|
|
|
with progress_bar(name=name, maxval=maxval) as bar: |
|
76
|
|
|
for i, chunk in enumerate(r.iter_content(chunk_size)): |
|
77
|
|
|
if total_length: |
|
78
|
|
|
bar.update(i * chunk_size) |
|
79
|
|
|
file_handle.write(chunk) |
|
80
|
|
|
|
|
81
|
|
|
|
|
82
|
|
|
def ensure_directory_exists(directory): |
|
83
|
|
|
"""Create directory (with parents) if does not exist, raise on failure. |
|
84
|
|
|
|
|
85
|
|
|
Parameters |
|
86
|
|
|
---------- |
|
87
|
|
|
directory : str |
|
88
|
|
|
The directory to create |
|
89
|
|
|
|
|
90
|
|
|
""" |
|
91
|
|
|
if os.path.isdir(directory): |
|
92
|
|
|
return |
|
93
|
|
|
os.makedirs(directory) |
|
94
|
|
|
|
|
95
|
|
|
|
|
96
|
|
|
def default_downloader(directory, urls, filenames, url_prefix=None, |
|
97
|
|
|
clear=False): |
|
98
|
|
|
"""Downloads or clears files from URLs and filenames. |
|
99
|
|
|
|
|
100
|
|
|
Parameters |
|
101
|
|
|
---------- |
|
102
|
|
|
directory : str |
|
103
|
|
|
The directory in which downloaded files are saved. |
|
104
|
|
|
urls : list |
|
105
|
|
|
A list of URLs to download. |
|
106
|
|
|
filenames : list |
|
107
|
|
|
A list of file names for the corresponding URLs. |
|
108
|
|
|
url_prefix : str, optional |
|
109
|
|
|
If provided, this is prepended to filenames that |
|
110
|
|
|
lack a corresponding URL. |
|
111
|
|
|
clear : bool, optional |
|
112
|
|
|
If `True`, delete the given filenames from the given |
|
113
|
|
|
directory rather than download them. |
|
114
|
|
|
|
|
115
|
|
|
""" |
|
116
|
|
|
# Parse file names from URL if not provided |
|
117
|
|
|
for i, url in enumerate(urls): |
|
118
|
|
|
filename = filenames[i] |
|
119
|
|
|
if not filename: |
|
120
|
|
|
filename = filename_from_url(url) |
|
121
|
|
|
if not filename: |
|
122
|
|
|
raise ValueError("no filename available for URL '{}'".format(url)) |
|
123
|
|
|
filenames[i] = filename |
|
124
|
|
|
files = [os.path.join(directory, f) for f in filenames] |
|
125
|
|
|
|
|
126
|
|
|
if clear: |
|
127
|
|
|
for f in files: |
|
128
|
|
|
if os.path.isfile(f): |
|
129
|
|
|
os.remove(f) |
|
130
|
|
|
else: |
|
131
|
|
|
print('Downloading ' + ', '.join(filenames) + '\n') |
|
132
|
|
|
ensure_directory_exists(directory) |
|
133
|
|
|
|
|
134
|
|
|
for url, f, n in zip(urls, files, filenames): |
|
135
|
|
|
if not url: |
|
136
|
|
|
if url_prefix is None: |
|
137
|
|
|
raise NeedURLPrefix |
|
138
|
|
|
url = url_prefix + n |
|
139
|
|
|
with open(f, 'wb') as file_handle: |
|
140
|
|
|
download(url, file_handle) |
|
141
|
|
|
|
mila-udem /
fuel
Loading history...
It is generally discouraged to redefine built-ins as this makes code very hard to read.