Fetch top 100 popular photos from Flickr

fetch-by-tag.py

import re
import json
import os
from urllib.request import urlopen, urlretrieve
from glob import glob

config = {
	'api_key': 'YOUR_API_KEY',
	'sort': 'interestingness-desc',
	'per_page': 100
}
api_call = 'http://api.flickr.com/services/rest/?method=flickr.photos.search&format=json&nojsoncallback=1&api_key=%(api_key)s&extras=original_format&sort=%(sort)s&per_page=%(per_page)s&tags=%(tags)s'
url_format = 'http://farm%(farm)s.static.flickr.com/%(server)s/%(id)s_%(secret)s.jpg'
url_format_o = 'http://farm%(farm)s.static.flickr.com/%(server)s/%(id)s_%(originalsecret)s_o.%(originalformat)s'

for tag in ['boy', 'children', 'crowd', 'face', 'girl', 'kid', 'man', 'men', 'people', 'person', 'portrait', 'woman', 'women']:
	cfg = config.copy()
	cfg['tags'] = tag
	data = urlopen(api_call % cfg).read().decode('utf-8')
	obj = json.loads(data)
	if not os.path.exists(tag):
		os.mkdir(tag)
	for p in obj['photos']['photo']:
		if p.get('originalsecret'):
			url = url_format_o % p
		else:
			url = url_format % p
		name = os.path.join(tag, os.path.basename(url))
		urlretrieve(url, name)

fetch-geo.py

import re
import json
import os
from urllib.request import urlopen, urlretrieve
 
config = {
	'api_key': 'YOUR_API_KEY',
	'sort': 'interestingness-desc',
	'per_page': 100
}
api_call = 'https://api.flickr.com/services/rest/?method=flickr.photos.search&format=json&nojsoncallback=1&api_key=%(api_key)s&extras=original_format&sort=%(sort)s&per_page=%(per_page)s&has_geo=1'
url_format = 'http://farm%(farm)s.static.flickr.com/%(server)s/%(id)s_%(secret)s.jpg'
url_format_o = 'http://farm%(farm)s.static.flickr.com/%(server)s/%(id)s_%(originalsecret)s_o.%(originalformat)s'

def process(p):
	if p.get('originalsecret') is None:
		return
	url = url_format_o % p
	name = os.path.basename(url)
	if os.path.exists(name):
		return
	print 'downloading %s' % name
	urlretrieve(url, name)

if __name__ == '__main__':
	from multiprocessing import Pool
	pool = Pool(8)

	page = 1
	try:
		while True:
			print 'page %d' % page
			cfg = config.copy()
			cfg['page'] = page
			data = urlopen(api_call % cfg).read().decode('utf-8')
			obj = json.loads(data)
			if not obj['photos']['photo']:
				break
			pool.map(process, obj['photos']['photo'])
			page += 1
	except:
		pool.close()