def download_set(set_id, size_label=None, naming=None):
"""
Download the set with 'set_id' to the current directory.
@param set_id: str, id of the photo set
@param size_label: str|None, size to download (or None for largest available)
@param naming: str|append, photo naming strategy
"""
suffix = " ({})".format(size_label) if size_label else ""
pset = Flickr.Photoset(id=set_id)
photos = pset.getPhotos()
pagenum = 2
while True:
try:
page = pset.getPhotos(page=pagenum)
photos.extend(page)
pagenum += 1
except FlickrAPIError as ex:
if ex.code == 1:
break
raise
if not os.path.exists(pset.title):
os.mkdir(pset.title)
# Stores in a dictionary the photo titles already found and their count,
# if the --naming argument was set to 'increment'
duplicates = {}
for photo in photos:
if naming == 'append':
fname = '{0}/{1}-{2}{3}.jpg'.format(pset.title, photo.title,
photo.id, suffix)
elif naming == 'replace':
fname = '{0}/{1}{2}.jpg'.format(pset.title, photo.id, suffix)
elif naming == 'increment':
if duplicates.has_key(photo.title):
extra = '(' + str(duplicates[photo.title]) + ')'
duplicates[photo.title] += 1
else:
extra = ''
duplicates[photo.title] = 1
fname = '{0}/{1}{2}{3}.jpg'.format(pset.title, photo.title, suffix,
extra)
else:
fname = '{0}/{1}{2}.jpg'.format(pset.title, photo.title, suffix)
if os.path.exists(fname):
# TODO: Ideally we should check for file size / md5 here
# to handle failed downloads.
print 'Skipping {0}, as it exists already'.format(fname)
continue
print 'Saving: {0}'.format(fname)
photo.save(fname, size_label)
# Set file times to when the photo was taken
info = photo.getInfo()
taken = parser.parse(info['taken'])
taken_unix = time.mktime(taken.timetuple())
os.utime(fname, (taken_unix, taken_unix))
def order_photosets_by_taken_date():
logging.debug('Get photoset ids ordered by photo taken date desc.')
sql = """select ps.id
from f_photoset ps
inner join f_photosetphoto psp on ps.id = psp.photosetId
left join f_photosetnodate nd on ps.id = nd.id
inner join f_photo p on psp.photoId = p.id
group by ps.id, ps.title order by max(case when nd.id is not null then cast('1900-01-01' as datetime) else p.taken end) desc"""
setids = get_int_list_from_database(sql)
setids = [str(i) for i in setids]
logging.debug('Retrieved %s photoset ids.' % len(setids))
flickr_api.Photoset().orderSets(photoset_ids=','.join(setids))
logging.debug('Ordering photos inside sets. Getting list of sets.')
photosets = user.getPhotosets()
for photoset in photosets:
logging.debug('Getting photos of set %s and ordering them by taken date.' % photoset.title)
sql = """select p.id from f_photo p
inner join f_photosetphoto psp on p.id = psp.photoId
where psp.photosetId = %s
order by p.taken asc""" % photoset.id
photoids = get_int_list_from_database(sql)
photoids = [str(i) for i in photoids]
photoset.reorderPhotos(photo_ids=','.join(photoids))
def download_set(set_id, get_filename, size_label=None):
"""
Download the set with 'set_id' to the current directory.
@param set_id: str, id of the photo set
@param get_filename: Function, function that creates a filename for the photo
@param size_label: str|None, size to download (or None for largest available)
"""
pset = Flickr.Photoset(id=set_id)
download_list(pset, pset.title, get_filename, size_label)
def download_set(set_id, get_filename, size_label=None, skip_download=False, save_json=False):
"""
Download the set with 'set_id' to the current directory.
@param set_id: str, id of the photo set
@param get_filename: Function, function that creates a filename for the photo
@param size_label: str|None, size to download (or None for largest available)
@param skip_download: bool, do not actually download the photo
@param save_json: bool, save photo info as .json file
"""
pset = Flickr.Photoset(id=set_id)
download_list(pset, pset.title, get_filename, size_label, skip_download, save_json)
def check_flickr(self):
flickr.set_keys(api_key=settings.FLICKR_API_KEY, api_secret=settings.FLICKR_API_SECRET)
# photos from photoset
photoset = flickr.Photoset(id=settings.FLICKR_PHOTOSET_ID)
photos_from_photoset = photoset.getPhotos(extras=['url_m', 'url_o'], media='photos')
photos_from_photoset_to_save = [
Photo(
title=photo.title,
flickr_id=photo.id,
url_m=photo.url_m,
url_o=photo.url_o,
) for photo in photos_from_photoset
]
# photos by tag #int20h
photos_by_tag = flickr.Photo.search(
tags=settings.FLICKR_PARSE_TAG,
extras=['url_m', 'url_o'],
media='photos',
per_page=500,
)
photos_by_tag_to_save = [
Photo(
title=photo.title,
flickr_id=photo.id,
url_m=photo.url_m,
url_o=photo.url_o,
) for photo in photos_by_tag
]
# merge two unsaved lists into one
all_photos_to_save = photos_by_tag_to_save + photos_from_photoset_to_save
# save to db preventing duplicates
# we will need to wait for Django 2.2 to use bulk_create ignore_conflicts
all_photos_to_save_flickr_ids = {photo.flickr_id for photo in all_photos_to_save}
photos_already_exists_flickr_ids = set(
Photo.objects.filter(
flickr_id__in=list(all_photos_to_save_flickr_ids)
).values_list(
'flickr_id', flat=True,
)
)
# Note: we save flickr_id as integer in db
new_photos_to_save = [
photo for photo in all_photos_to_save if int(photo.flickr_id) not in photos_already_exists_flickr_ids
]
Photo.objects.bulk_create(new_photos_to_save)
return f'added: {len(new_photos_to_save)} Photos'
def get_photos_by_option(option, term):
if option == "s":
# try "minimalism"
print("Searching for photos with tag \"" + term + "\"")
return flickr_api.Walker(flickr_api.Photo.search, tags=term)
elif option == "u":
# try "alexcurrie"
print("Retrieving photos from user " + term)
user = flickr_api.Person.findByUserName(term)
return user.getPublicPhotos()
elif option == "a":
# try "72157669309212490"
print("Retrieving photos from album with ID " + term)
photoset = flickr_api.Photoset(id=term)
return photoset.getPhotos()
