def test_length_hint_empty(self):
looter = instaLooter.InstaLooter(profile="jkshksjdhfjkhdkfhk")
self.assertEqual(operator.length_hint(looter), 0)
looter = instaLooter.InstaLooter(hashtag="jkshksjdhfjkhdkfhk")
self.assertEqual(operator.length_hint(looter), 0)
def test_dump_json(self):
looter = instaLooter.InstaLooter(
self.tmpdir,
profile="instagram",
dump_json=True,
)
test_medias = list(
itertools.islice((m for m in looter.medias() if not m['is_video']),
3))
looter.download(media_count=3)
# Check all files were downloaded as expected
self.assertEqual(
sorted(os.listdir(self.tmpdir)),
sorted(f for media in test_medias for f in (
str("{}.jpg").format(media['id']),
str("{}.json").format(media['id']),
)))
# Check the metadata are OK
for media in test_medias:
with open(
os.path.join(self.tmpdir,
"{}.json").format(media['id'])) as f:
dump = json.load(f)
self.assertMediaEqual(media, dump)
def test_pr_122(self):
"""
Feature implemented by @susundberg.
Set the access time and modification time of a downloaded media
according to its IG date.
"""
looter = instaLooter.InstaLooter(
self.tmpdir, profile='franz_ferdinand', template='{code}')
# Test download_post
info = looter.get_post_info('BY77tSfBnRm')
looter.download_post('BY77tSfBnRm')
stat = os.stat(os.path.join(self.tmpdir, 'BY77tSfBnRm.jpg'))
self.assertEqual(stat.st_atime, info['date'])
self.assertEqual(stat.st_mtime, info['date'])
# Test download_pictures
pic = next(m for m in looter.medias() if not m['is_video'])
looter.download_pictures(media_count=1)
stat = os.stat(os.path.join(self.tmpdir, '{}.jpg'.format(pic['code'])))
self.assertEqual(stat.st_atime, pic['date'])
self.assertEqual(stat.st_mtime, pic['date'])
# Test download_videos
vid = next(m for m in looter.medias() if m['is_video'])
looter.download_videos(media_count=1)
stat = os.stat(os.path.join(self.tmpdir, '{}.mp4'.format(vid['code'])))
self.assertEqual(stat.st_atime, vid['date'])
self.assertEqual(stat.st_mtime, vid['date'])
def test_extended_dump(self):
looter = instaLooter.InstaLooter(
self.tmpdir,
profile="instagram",
dump_only=True,
extended_dump=True,
)
test_medias = list(
itertools.islice((m for m in looter.medias() if not m['is_video']),
3))
looter.download(media_count=3)
# Check all files were downloaded as expected
self.assertEqual(
sorted(os.listdir(self.tmpdir)),
sorted(
str("{}.json").format(media['id']) for media in test_medias))
# Check the metadata are OK
for media in test_medias:
with open(
os.path.join(self.tmpdir,
"{}.json").format(media['id'])) as f:
dump = json.load(f)
self.assertMediaEqual(media, dump)
# Check the dump was "extended"
self.assertIn('edge_media_to_comment', dump)
self.assertIn('edge_media_to_caption', dump)
def test_proxy_setup(self):
looter = instaLooter.InstaLooter(socks_port=9090)
self.assertTrue(hasattr(looter, 'tor_manager'))
self.assertTrue(
str(looter.session.proxies) == str(looter.tor_manager.proxies))
self.assertTrue(
isinstance(looter.session.hooks.get('response', None),
type(looter.tor_manager.call_for_new_ip)))
def test_issue_6(self):
"""
Checks that instaLooter does not iterate forever on a private
profile.
"""
looter = instaLooter.InstaLooter(self.tmpdir, profile="tldr")
with self.assertRaises(StopIteration):
next(looter.medias())
def _test(self):
looter = instaLooter.InstaLooter(self.tmpdir,
profile=profile,
get_videos=True)
looter.download(media_count=200)
self.assertEqual(len(os.listdir(self.tmpdir)),
min(200, int(looter.metadata['media']['count'])))
self.assertEqual(profile, looter.metadata['username'])
def test_issue_41(self):
"""
Feature request by @liorlior
Allow downloading only videos.
"""
looter = instaLooter.InstaLooter(self.tmpdir, profile="nintendo", videos_only=True)
looter.download(timeframe=(datetime.date(2017, 3, 10), )*2)
self.assertEqual(os.listdir(self.tmpdir), ["1467639884243493431.mp4"])
def test_template_1(self):
PROFILE = "therock"
looter = instaLooter.InstaLooter(self.tmpdir,
profile=PROFILE,
get_videos=True,
template='{username}')
looter.download(media_count=30, with_pbar=True)
for f in os.listdir(self.tmpdir):
self.assertTrue(f.startswith(PROFILE))
def test_template_1(self):
profile = "therock"
looter = instaLooter.InstaLooter(self.tmpdir,
profile=profile,
get_videos=True,
template='{username}-{id}')
looter.download(media_count=self.MEDIA_COUNT, with_pbar=False)
for f in os.listdir(self.tmpdir):
self.assertTrue(f.startswith(profile))
def test_resizer(self):
self.looter = instaLooter.InstaLooter(self.tmpdir,
url_generator=resizer(320),
profile="instagram")
self.looter.download(media_count=self.MEDIA_COUNT)
for img_file in glob.iglob(os.path.join(self.tmpdir, "*.jpg")):
with PIL.Image.open(img_file) as img:
width, height = img.size
self.assertEqual(max(width, height), 320)
def test_issue_12(self):
"""
Feature request by @paramjitrohit.
Allows downloading pictures and videos only within a timeframe.
"""
looter = instaLooter.InstaLooter(self.tmpdir, profile="slotfaceofficial")
timeframe = (datetime.date(2017, 2, 18),)*2
medias_in_timeframe = list(looter.medias(timeframe=timeframe))
self.assertEqual(len(medias_in_timeframe), 2)
def test_thumbnail(self):
self.looter = instaLooter.InstaLooter(self.tmpdir,
url_generator=thumbnail,
profile="therock")
self.looter.download(media_count=self.MEDIA_COUNT)
for img_file in glob.iglob(os.path.join(self.tmpdir, "*.jpg")):
with PIL.Image.open(img_file) as img:
width, height = img.size
self.assertEqual(width, height)
def test_issue_26(self):
"""
Feature request by @verafide.
Checks that pictures that are downloaded are not
resized.
"""
looter = instaLooter.InstaLooter(self.tmpdir)
looter.download_post("BO0XpEshejh")
filename = "1419863760138791137.jpg"
pic = PIL.Image.open(os.path.join(self.tmpdir, filename))
self.assertEqual(pic.size, (525, 612))
def test_length_hint(self):
looter = instaLooter.InstaLooter(self.tmpdir, profile="tide")
hint = operator.length_hint(looter)
# Check the post count is greater than 0
self.assertGreater(hint, 0)
# Download pictures and check if the count
# match (at most as many posts downloaded)
looter.download()
self.assertLessEqual(len(os.listdir(self.tmpdir)), hint)
def test_ip_changeability(self):
def get_ip(proxies):
r = requests.get('https://api.ipify.org?format=json',
proxies=proxies)
return json.loads(r.text)['ip']
looter = instaLooter.InstaLooter(socks_port=9090, change_ip_after=1)
old_ip = get_ip(looter.tor_manager.proxies)
looter.session.get('https://www.instagram.com/')
new_ip = get_ip(looter.tor_manager.proxies)
self.assertTrue(old_ip != new_ip)
def _test(self):
looter = instaLooter.InstaLooter(self.tmpdir,
profile=profile,
get_videos=True)
looter.download(media_count=200)
# We have to use GreaterEqual since multi media posts
# are counted as 1 but will download more than one
# picture / video
self.assertGreaterEqual(
len(os.listdir(self.tmpdir)),
min(200, int(looter.metadata['media']['count'])))
self.assertEqual(profile, looter.metadata['username'])
def test_issue_76(self):
"""
Thanks to @zeshuaro for reporting this bug.
Check that when downloading hashtags, the downloader
actually stops.
"""
looter = instaLooter.InstaLooter(self.tmpdir, hashtag="oulianov")
postcount = looter.__length_hint__() # operator.length_hint
for i, m in enumerate(looter.medias()):
if i > postcount:
self.fail("looter.medias() did not stop.")
def test_issue_15(self):
"""
Feature request by @MohamedIM.
Checks that videos are not downloaded several times if present
already in the destination directory.
"""
looter = instaLooter.InstaLooter(self.tmpdir, profile="instagram")
looter.download_videos(media_count=1)
video_file = next(glob.iglob(os.path.join(self.tmpdir, '*.mp4')))
mtime = os.stat(video_file).st_mtime
looter.download_videos(media_count=1) #, new_only=True)
self.assertEqual(mtime, os.stat(video_file).st_mtime)
def test_issue_94(self):
"""
Thanks to @jeanmarctst for raising this issue.
Make sure caption is properly extracted from images downloaded
from a post code and written to the metadata.
"""
looter = instaLooter.InstaLooter(
self.tmpdir, add_metadata=True, template='{code}')
looter.download_post('BY77tSfBnRm')
metadata = piexif.load(
os.path.join(self.tmpdir, 'BY77tSfBnRm.jpg'), True)
self.assertTrue(metadata['Exif']['UserComment'])
def test_issue_44(self):
"""
Thanks to @Bangaio64 for reporting this bug.
Checks that warn_windows does not trigger an exception.
"""
import instaLooter.utils
warnings.showwarning = instaLooter.utils.warn_windows
looter = instaLooter.InstaLooter(
directory=self.tmpdir,
profile="akjhdskjhfkjsdhfkjhdskjhfkjdshkfjhsdkjfdhkjdfshdfskhfd")
try:
looter.download()
except Exception:
self.fail()
def test_issue_66(self):
"""
Thanks to @douglasrizzo for reporting this bug.
Check that likescount and commentscount can be used
in filename templates without causing the program to
crash.
"""
looter = instaLooter.InstaLooter(
self.tmpdir, profile="zuck",
get_videos=True, add_metadata=True,
template='{id}-{likescount}-{commentscount}',
)
looter.download(media_count=10)
for image in os.listdir(self.tmpdir):
self.assertRegex(image, '[a-zA-Z0-9]*-[0-9]*-[0-9]*.(jpg|mp4)')
def test_issue_84(self):
"""
Thanks to @raphaelbernardino for reporting this bug.
Make sure private profiles with few pictures (less than a page worth)
raise the private warning as expected.
"""
with warnings.catch_warnings(record=True) as registry:
warnings.simplefilter('always')
looter = instaLooter.InstaLooter(profile="raphaelbernardino")
list(looter.medias())
self.assertEqual(
six.text_type(registry[0].message),
u"Profile raphaelbernardino is private, retry after logging in.")
def test_issue_42(self):
"""
Thanks to @MohamedIM for reporting this bug.
Checks that a multipost is successfully downloaded from
the CLI `post` option.
"""
looter = instaLooter.InstaLooter(self.tmpdir)
looter.download_post('BRW-j_dBI6F')
self.assertEqual(
set(os.listdir(self.tmpdir)), {
'1465633492745668095.mp4',
'1465633517836005761.mp4',
'1465633541559037966.mp4',
'1465633561523918792.mp4',
})
def test_issue_39(self):
"""
Feature request by @verafide
Checks that all pictures are downloaded from posts
with more than one picture.
"""
looter = instaLooter.InstaLooter(self.tmpdir)
looter.download_post("BRHecUuFhPl")
self.assertEqual(
set(os.listdir(self.tmpdir)), {
"1461270165803344956.jpg",
"1461270167497776767.jpg",
"1461270174435133336.jpg",
"1461270172581471925.jpg",
"1461270181565655668.jpg",
})
def test_issue_82(self):
"""
Thanks to @MohamedIM for reporting this bug.
Check that urls containing 'h-ak-igx' are not stripped from all
their parameters.
"""
looter = instaLooter.InstaLooter(self.tmpdir, template='{code}')
info = looter.get_post_info('BWOYSYQDCo5')
info['display_url'] = \
'https://ig-s-c-a.akamaihd.net/h-ak-igx/19764472_1586345694718446_4011887281420894208_n.jpg'
looter.get_post_info = lambda code: info
looter.download_post('BWOYSYQDCo5')
with open(os.path.join(self.tmpdir, 'BWOYSYQDCo5.jpg'), 'rb') as f:
self.assertNotIn(b'5xx Server Error', f.read())
def test_issue_57(self):
"""
Thanks to @VasiliPupkin256 for reporting this bug.
Checks that metadata can successfully extract caption
out of multiposts containing images.
"""
looter = instaLooter.InstaLooter(self.tmpdir,
profile="awwwwshoot_ob",
get_videos=True,
add_metadata=True)
looter._medias_queue = six.moves.queue.Queue()
looter._fill_media_queue()
while not looter._medias_queue.empty():
media = looter._medias_queue.get()
for key in ('caption', 'code', 'date'):
self.assertIn(key, media)
self.assertIsNotNone(key, media[key])
def test_hashtag_download(self):
looter = instaLooter.InstaLooter(self.tmpdir,
hashtag="python",
get_videos=True)
looter.download(media_count=200)
self.assertEqual(len(os.listdir(self.tmpdir)), 200)
def test_url_generator_nocallable(self):
with self.assertRaises(ValueError):
self.looter = instaLooter.InstaLooter(self.tmpdir,
profile="instagram",
url_generator=1)
def setUp(self):
super(TestUtils, self).setUp()
self.looter = instaLooter.InstaLooter()