def scrape_images(profile, num, startdate, enddate, out_dir):
"""
Function to scrape images from Instagram using Instalooter
Parameters:
----------
profile: name of the Instagram profile.
num: number of images to download.
startdate: Most recent date from which you wanna start downloading.
enddate: Date to specify the end of timeframe.
Returns:
--------
output: The path of the output directory
"""
#output = 'images_'+str(startdate)+ '_' +str(enddate)
looter = ProfileLooter(profile)
#if not os.path.exists(output):
# os.makedirs(output)
#output = "./"+output
looter.download(output, media_count=int(num), timeframe=(startdate, enddate))
return output
class TestLogin(unittest.TestCase):
def setUp(self):
self.looter = ProfileLooter(USERNAME, template="test")
self.destfs = fs.memoryfs.MemoryFS()
def tearDown(self):
self.destfs.close()
def test_login(self):
self.assertFalse(self.looter.logged_in())
self.assertRaises(RuntimeError, self.looter.medias)
self.assertFalse(self.looter._cachefs.exists(self.looter._COOKIE_FILE))
try:
self.looter.login(USERNAME, PASSWORD)
self.assertTrue(self.looter.logged_in())
self.assertTrue(
self.looter._cachefs.exists(self.looter._COOKIE_FILE))
self.assertTrue(next(self.looter.medias()))
finally:
self.looter.logout()
self.assertFalse(
self.looter._cachefs.exists(self.looter._COOKIE_FILE))
def test_download(self):
try:
self.looter.login(USERNAME, PASSWORD)
self.looter.download(self.destfs)
self.assertTrue(self.destfs.exists('test.jpg'))
self.assertEqual(self.destfs.getbytes('test.jpg')[6:10], b'JFIF')
finally:
self.looter.logout()
def test_issue_041(self):
"""
Feature request by @liorlior
Allow downloading only videos.
"""
looter = ProfileLooter("nintendo", videos_only=True, session=self.session)
day = datetime.date(2017, 3, 10)
looter.download(self.destfs, timeframe=[day, day])
self.assertTrue(self.destfs.isfile("1467639884243493431.mp4"))
async def instagram_dl(igdl):
""" To downloading photos from instagram account """
uname = igdl.pattern_match.group(1)
input_str = TEMP_DOWNLOAD_DIRECTORY
if not os.path.exists(input_str):
os.makedirs(input_str)
try:
await igdl.edit(f"`Getting info.....`")
looter = ProfileLooter(uname)
looter.download('TEMP_DOWNLOAD_DIRECTORY', media_count=5)
except ValueError:
await igdl.edit(
f"**Account {uname} Not Found.**\nPlease enter correct username.")
return
except RuntimeError:
await igdl.edit(f"**Can't Catch Media.**\nAccount {uname} is Private.")
return
await igdl.edit("Processing ...")
lst_of_files = []
for r, d, f in os.walk(input_str):
for file in f:
lst_of_files.append(os.path.join(r, file))
for file in d:
lst_of_files.append(os.path.join(r, file))
LOGS.info(lst_of_files)
uploaded = 0
countf = "{}".format(len(lst_of_files))
count = int(countf)
if count == 0:
await igdl.edit(
"**No Media Found**\nSorry this account doesn't have any content")
else:
await igdl.edit(
"Found {} files. Uploading will start soon. Please wait!".format(
len(lst_of_files)))
for single_file in lst_of_files:
if os.path.exists(single_file):
# https://stackoverflow.com/a/678242/4723940
caption_rts = os.path.basename(single_file)
c_time = time.time()
if not caption_rts.lower().endswith(".mp4"):
await igdl.client.send_file(
igdl.chat_id,
single_file,
caption=f"[{uname}](https://instagram.com/{uname})",
force_document=True,
allow_cache=False,
progress_callback=lambda d, t: asyncio.get_event_loop(
).create_task(
progress(d, t, igdl, c_time, "Uploading...",
single_file)))
os.remove(single_file)
def test_issue_041(self):
"""Feature request by @liorlior
Allow downloading only videos.
"""
looter = ProfileLooter("nintendo", videos_only=True, session=self.session)
day = datetime.date(2017, 3, 10)
with contexter.Contexter() as ctx:
ctx << mock.patch.object(looter, 'pages', MockPages('nintendo'))
looter.download(self.destfs, timeframe=[day, day])
self.assertEqual(self.destfs.listdir("/"), ["1467639884243493431.mp4"])
def test_issue_014(self):
"""Feature request by @JFLarsen.
Allows customizing filenames using a template following Python
`.format()` minilanguage.
"""
looter = ProfileLooter("nintendo", template="{username}.{id}", session=self.session)
with contexter.Contexter() as ctx:
ctx << mock.patch.object(looter, 'pages', MockPages('nintendo'))
looter.download(self.destfs, media_count=5)
for f in self.destfs.scandir("/"):
self.assertTrue(f.name.startswith('nintendo.'))
def test_issue_022(self):
"""
Thanks to @kuchenmitsahne for reporting this bug.
Checks that using ``{datetime}`` in the template does not put
a Windows forbidden character in the filename.
"""
FORBIDDEN = set('<>:"/\|?*')
looter = ProfileLooter(
"mysteryjets", template="{datetime}", session=self.session)
looter.download(self.destfs, media_count=5)
for f in self.destfs.scandir("/"):
self.assertFalse(FORBIDDEN.intersection(f.name))
def test_issue_022(self):
"""
Thanks to @kuchenmitsahne for reporting this bug.
Checks that using ``{datetime}`` in the template does not put
a Windows forbidden character in the filename.
"""
FORBIDDEN = set('<>:"/\|?*')
looter = ProfileLooter("nintendo", template="{datetime}", session=self.session)
with contexter.Contexter() as ctx:
ctx << mock.patch.object(looter, 'pages', MockPages('nintendo'))
looter.download(self.destfs, media_count=5)
for f in self.destfs.scandir("/"):
self.assertFalse(FORBIDDEN.intersection(f.name))
def test_issue_014(self):
"""
Feature request by @JFLarsen.
Allows customizing filenames using a template following Python
`.format()` minilanguage.
"""
looter = ProfileLooter(
"mysteryjets", template="{username}.{id}", session=self.session)
looter.download(self.destfs, media_count=5)
for f in self.destfs.scandir("/"):
self.assertTrue(f.name.startswith('mysteryjets'))
def test_issue_009(self):
"""
Thanks to @kurtmaia for reporting this bug.
Checks that adding metadata to pictures downloaded from a hashtag
works as well.
"""
looter = ProfileLooter("fluoxetine", add_metadata=True, session=self.session)
looter.download(self.destfs, media_count=10)
for f in self.destfs.listdir("/"):
exif = piexif.load(self.destfs.getbytes(f))
self.assertTrue(exif['Exif']) # Date & Caption
self.assertTrue(exif['0th']) # Image creator
def test_issue_066(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 = ProfileLooter(
"zuck", get_videos=True, add_metadata=True,
template='{id}-{likescount}-{commentscount}',
session=self.session)
looter.download(self.destfs, media_count=10)
for image in self.destfs.listdir("/"):
self.assertRegex(image, '[a-zA-Z0-9]*-[0-9]*-[0-9]*.(jpg|mp4)')
def scrapeVideos(username = "",
password = "",
output_folder = "",
days = 1):
print("Starting Scraping")
L = instaloader.Instaloader()
# Login or load session for loader
L.login(username, password)
profile = instaloader.Profile.from_username(L.context, username)
following = profile.get_followees()
print(following)
for profile in following:
acc = profile.username
looter = ProfileLooter(acc, videos_only=True, template="{id}-{username}-{width}-{height}")
if not looter.logged_in():
looter.login(username, password)
print("Scraping From Account: " + acc)
today = datetime.date.today()
timeframe = (today, today - dateutil.relativedelta.relativedelta(days=days))
numDowloaded = looter.download(output_folder, media_count=30, timeframe=timeframe)
print("Downloaded " + str(numDowloaded) + " videos successfully")
print("")
def _profile_images(username, destination='.instagram'):
if not InstaLooter._logged_in():
InstaLooter._login(username, password)
looter = ProfileLooter(username)
n = looter.download(destination=destination, new_only=True)
path = pathlib.Path(destination)
return list(path.glob('*'))
class TestLogin(unittest.TestCase):
@classmethod
def setUpClass(cls):
cls.session = requests.Session()
InstaLooter._user_agent = cls.session.headers["User-Agent"]
@classmethod
def tearDownClass(cls):
cls.session.close()
del InstaLooter._user_agent
def setUp(self):
self.looter = ProfileLooter(USERNAME, template="test")
self.destfs = fs.memoryfs.MemoryFS()
def tearDown(self):
self.destfs.close()
def test_login(self):
self.assertFalse(self.looter.logged_in())
self.assertRaises(RuntimeError, self.looter.medias)
self.assertFalse(self.looter._cachefs.exists(self.looter._COOKIE_FILE))
try:
self.looter.login(USERNAME, PASSWORD)
self.assertTrue(self.looter.logged_in())
self.assertTrue(
self.looter._cachefs.exists(self.looter._COOKIE_FILE))
self.assertTrue(next(self.looter.medias()))
finally:
self.looter.logout()
self.assertFalse(
self.looter._cachefs.exists(self.looter._COOKIE_FILE))
def test_download(self):
try:
self.looter.login(USERNAME, PASSWORD)
self.looter.download(self.destfs)
self.assertTrue(self.destfs.exists('test.jpg'))
self.assertEqual(self.destfs.getbytes('test.jpg')[6:10], b'JFIF')
finally:
self.looter.logout()
def post(self, target, number_media):
looter = ProfileLooter(target)
looter.download('Pictures/' + target, media_count=int(number_media))
return "ok", 201
def download(username: str, image_dir: str) -> None:
"""Downloading images from instagram."""
logging.info("Downloading...")
looter = ProfileLooter(username, template="insta-{datetime}-{id}")
looter.download(image_dir)
logging.info("Done downloading.")
from instalooter.looters import ProfileLooter
import datetime
import dateutil.relativedelta
# instalooter_test downloads videos posted by daquan in the last month
# Instanciate
looter = ProfileLooter("daquan",
videos_only=True,
template="{id}-{username}-{width}-{height}")
looter.login("", "")
today = datetime.date.today()
thismonth = (today, today - dateutil.relativedelta.relativedelta(days=28))
looter.download('./Memes_December_4', media_count=50, timeframe=thismonth)
def test_profile(self, profile, **kwargs):
looter = ProfileLooter(profile, session=self.session, **kwargs)
looter.download(self.destfs, media_count=self.MEDIA_COUNT)
self.assertGreaterEqual(len(self.destfs.listdir("/")),
self.MEDIA_COUNT)
from instalooter.looters import ProfileLooter
instagram_account_name = "eleanor3069"
# Grab newest instagram posts
looter = ProfileLooter(instagram_account_name)
looter.download('./subjectimg/', media_count=5)
print("Finished downloading photos")
from instalooter.looters import ProfileLooter
from os import listdir
import pandas as pd
instagram_account_name = "iampaintingrobot"
# Grab newest instagram post data
photorecord_dir = './instabotimagerecords/'
photorecord_csv_dir = photorecord_dir + 'procdimagelist.csv'
total_photo_list = listdir(photorecord_dir)
looter = ProfileLooter(instagram_account_name, dump_only=True)
looter.download(photorecord_dir, media_count=1)
print("Finished downloading photo data")
csvdf = pd.read_csv(photorecord_csv_dir)
non_matching_file_list = [
file_name for file_name in total_photo_list
if not any(file_name in proc_name for proc_name in csvdf['name'])
]
print("Non matching files: ")
print(non_matching_file_list)
if len(non_matching_file_list) == 1:
# Move the file if necessary
def checkInsta(domain, foldername):
lt = ProfileLooter(domain)
lt.download(foldername)
