from glob import glob import os from os.path import join from pathlib import Path import click from app.settings import types from app.utils import click_utils from app.settings import app_cfg as cfg from app.utils import logger_utils import dlib import pandas as pd from PIL import Image, ImageOps, ImageFilter from app.utils import file_utils, im_utils log = logger_utils.Logger.getLogger() @click.command() @click.option('-i', '--input', 'opt_fp_in', required=True, help='Input directory') @click.option('-o', '--output', 'opt_fp_out', help='Output directory') @click.option('--slice', 'opt_slice', type=(int, int), default=(None, None), help='Slice list of files') @click.option('-d', '--delay', 'opt_delay', default=None, type=int, help='Delay between API calls to prevent rate-limiting') @click.option('--checkpoints', 'opt_checkpoints', is_flag=True, help='Save checkpoints') @click.option('--api_key', 'opt_api_key', envvar='FLICKR_API_KEY') @click.option('--api_secret', 'opt_api_secret', envvar='FLICKR_API_SECRET') @click.option('--checkpoint_interval', 'opt_ckpt_interval', default=10000, help='Save checkpoint interval') @click.pass_context def cli(ctx, opt_fp_in, opt_fp_out, opt_slice, opt_api_key, opt_api_secret, opt_delay, opt_checkpoints, opt_ckpt_interval): """Appends Flickr API info to CSV""" from tqdm import tqdm from glob import glob import time import flickr_api # pip install flickr_api from flickr_api.flickrerrors import FlickrAPIError # ------------------------------------------------- # process if not opt_api_key or not opt_api_secret: log.error('source .env vars for Flickr API and try again') return # init Flickr API flickr_api.set_keys(api_key=opt_api_key, api_secret=opt_api_secret) # reqd in CSV df_ids = pd.read_csv(opt_fp_in) if opt_slice: df_ids = df_ids[opt_slice[0]:opt_slice[1]] log.info('Processing: {:,} items'.format(len(df_ids))) # iterate MegaFace IDs identities = [] tqdm.pandas() for idx, df_id in tqdm(df_ids.iterrows(), total=len(df_ids)): # a = flickr_api.Person(id='123456789@N01') df_id_dict = dict(df_id) # append relevant data try: person = flickr_api.Person(id=df_id['nsid']) info = person.getInfo() df_id_dict.update( { 'user_name': info.get('username', ''), 'location': info.get('location', ''), 'real_name': info.get('realname', ''), 'time_zone': info.get('timezone', {}).get('timezone_id', ''), 'time_first_photo': info.get('photos_info', {}).get('firstdatetaken'), 'photos_count': info.get('photos_info', {}).get('count'), 'description': info.get('description', ''), 'id': info.get('id'), 'path_alias': info.get('path_alias', ''), 'is_pro': info.get('ispro', ''), 'url_photos': info.get('photosurl', ''), 'url_profile': info.get('photosurl', ''), 'url_mobile': info.get('mobileurl', ''), }) identities.append(df_id_dict) except FlickrAPIError as e: log.error(e) if opt_checkpoints: if (idx + 1) % opt_ckpt_interval == 0: df = pd.DataFrame.from_dict(identities) fpp_out = Path(opt_fp_out) opt_fp_out_ckpt = join(fpp_out.parent, '{}_ckpt_{}.csv'.format(fpp_out.stem, file_utils.zpad(idx + 1))) log.info('Saving checkpoint {:,} to {}'.format(idx + 1, opt_fp_out_ckpt)) df.to_csv(opt_fp_out_ckpt, index=False) if opt_delay: time.sleep(opt_delay) df = pd.DataFrame.from_dict(identities) df.to_csv(opt_fp_out, index=False) log.info('Wrote: {:,} lines to {}'.format(len(df), opt_fp_out)) """ Example API data: {'id': '7124086@N07', 'nsid': '7124086@N07', 'ispro': 1, 'can_buy_pro': 0, 'iconserver': '2325', 'iconfarm': 3, 'path_alias': 'shirleylin', 'has_stats': '1', 'pro_badge': 'standard', 'expire': '0', 'username': 'ShirleyLin', 'realname': 'Shirley Lin', 'location': 'Fremont, California, US', 'timezone': {'label': 'Pacific Time (US & Canada); Tijuana', 'offset': '-08:00', 'timezone_id': 'PST8PDT'}, 'description': '', 'photosurl': 'https://www.flickr.com/photos/shirleylin/', 'profileurl': 'https://www.flickr.com/people/shirleylin/', 'mobileurl': 'https://m.flickr.com/photostream.gne?id=7102756', 'photos_info': {'firstdatetaken': '2004-05-24 12:12:15', 'firstdate': '1172556588', 'count': 9665}} """