Esempio n. 1
0
def shares_from_url(url, request_start_date):

    params = {
        'count': 1000,
        # 'includeHistory': 'true',
        'link': url,
        'platforms': 'facebook',
        'sortBy': 'total_interactions',
        'startDate': request_start_date,
        'token': CROWDTANGLE_API_KEY
    }

    api_endpoint = 'links'

    response = requests.get('{}/{}'.format(CROWDTANGLE_API_URL, api_endpoint),
                            params).json()

    shares = []
    if response['status'] == 200:
        if not response.get('result'):
            logger.warning('Crowdtangle data returned is empty.')
            return shares

        for post in response['result']['posts']:
            account = post['account']
            shares.append({
                'account': {
                    'crowdtangleIdentifier': str(account['id']),
                    'facebookIdentifier': str(account['platformId']),
                    'logoUrl': account['profileImage'],
                    'name': account['name'],
                    'url': account['url']
                },
                'post': {
                    'crowdtangleIdentifier':
                    str(post['id']),
                    'facebookIdentifier':
                    str(post['platformId']),
                    'publishedDate':
                    datetime.strptime(post['date'], '%Y-%m-%d %H:%M:%S'),
                    'url':
                    post['postUrl'],
                }
            })
    else:
        logger.error(
            f'Error in fetching from Crowdtangle: {response.get("message", "Unknown exception.")}'
        )
        logger.warning('Returning empty interaction data')

    sleep(30)
    return shares
Esempio n. 2
0
from flask import Flask

from sqlalchemy_api_handler import logger
from utils.jobs import get_all_jobs, write_jobs_to_file, remove_oldest_jobs_file
from utils.setup import setup

CLOCK_APP = Flask(__name__)

setup(CLOCK_APP, with_jobs=True)

if __name__ == '__main__':
    # CLOCK_APP.async_scheduler.start()
    CLOCK_APP.background_scheduler.start()
    # atexit.register(lambda: CLOCK_APP.async_scheduler.shutdown())
    atexit.register(CLOCK_APP.background_scheduler.shutdown)
    print_jobs = True

    try:
        while True:
            if print_jobs:
                jobs = get_all_jobs(CLOCK_APP)
                write_jobs_to_file(jobs)
                remove_oldest_jobs_file()
            time.sleep(60)
    except (KeyboardInterrupt, SystemExit):
        logger.warning('Scheduler interupted')
        print_jobs = False
        # CLOCK_APP.async_scheduler.shutdown()
        CLOCK_APP.background_scheduler.shutdown()
Esempio n. 3
0
import os
from datetime import datetime
import requests
from sqlalchemy_api_handler import logger
from urllib.parse import urlencode

from utils.date import strftime
from utils.config import IS_DEVELOPMENT

BUZZSUMO_API_KEY = os.environ.get('BUZZSUMO_API_KEY')
BUZZSUMO_API_URL = "http://api.buzzsumo.com/search"

if BUZZSUMO_API_KEY is None:
    logger.warning('BUZZSUMO_API_KEY is not defined in the env')

DEVELOPMENT_TRENDINGS = [{
    'id':
    123,
    'published_date':
    1585956625,
    'subdomain':
    'www.cnn.com',
    'tags':
    'climate',
    'thumbnail':
    'https://cdn.cnn.com/cnnnext/dam/assets/170708175538-05-trump-abroad-0708-super-tease.jpg',
    'title':
    "Donald Trump buried a climate change report because 'I don't believe it'",
    'total_facebook_shares':
    1000,
    'total_shares':
Esempio n. 4
0
def sync_for(
    name,
    formula=None,
    max_records=None,
    session=None,
    sync_to_airtable=False
):
    if session is None:
        session = requests.Session()

    rows = request_airtable_rows(
        SCIENCE_FEEDBACK_AIRTABLE_BASE_ID,
        NAME_TO_AIRTABLE[name],
        filter_by_formula=formula,
        max_records=max_records,
        session=session
    )

    entities = []
    if rows:
        logger.info(f'syncing table {NAME_TO_AIRTABLE[name]}')
    else:
        logger.info(f'nothing to sync for table {NAME_TO_AIRTABLE[name]}')

    for (index, row) in enumerate(rows):
        try:
            entity = entity_from_row_for(name, row, index)
            if entity:
                entities.append(entity)
                row['Synced time input'] = datetime.now().isoformat()
            else:
                row['Synced time input'] = 'ERROR'
        except KeyError as exception:
            logger.warning(f'Error while trying to create entity from row at table {NAME_TO_AIRTABLE[name]}')
            logger.error(f'KeyError {exception}: {row}')
            row['Synced time input'] = 'ERROR'
        except Exception as exception:
            logger.warning(f'Error while trying to create entity from row at table {NAME_TO_AIRTABLE[name]}')
            logger.error(f'Unexpected error: {exception} - {sys.exc_info()[0]} at {row}')
            row['Synced time input'] = 'ERROR'

    def _update_10_rows_from_index(i):
        records = [{'id': row['airtableId'], 'fields': {'Synced time input': row['Synced time input']}} for row in rows[i: i + 10]]
        res = update_airtable_rows(
            SCIENCE_FEEDBACK_AIRTABLE_BASE_ID,
            NAME_TO_AIRTABLE[name],
            {'records': records},
            session=session
        )

        if res.status_code != 200:
            logger.error(f'code: {res.status_code}, error: {res.content}')

    try:
        # Sync verdict status from wordpress
        if name == 'verdict' and formula is not None:
            entities = claim_verdicts_from_airtable(verdicts_to_sync=entities)

        # Sync related contents for appearances
        if name == 'appearance' and formula is not None:
            for entity in entities:
                sync_content(entity.quotingContent)

        # Set the time synced so that the status in airtable is "Synced"
        if sync_to_airtable:
            for i in range(0, len(rows), 10):
                try:
                    ApiHandler.save(*entities[i:i + 10])
                    _update_10_rows_from_index(i)

                except Exception as exception:
                    logger.warning(f'Error while trying to save 10 entities at table {NAME_TO_AIRTABLE[name]}')
                    logger.error(f'Unexpected error: {exception} - {sys.exc_info()[0]}')
                    for index in range(i, i + 10):
                        rows[index]['Synced time input'] = 'BATCH ERROR'
                    _update_10_rows_from_index(i)

    except Exception as exception:
        logger.warning(f'Error while trying to save entities at table {NAME_TO_AIRTABLE[name]}')
        logger.error(f'Unexpected error: {exception} - {sys.exc_info()[0]}')
Esempio n. 5
0
import os
from time import sleep

from datetime import datetime
import requests
from sqlalchemy_api_handler import logger

CROWDTANGLE_API_URL = 'https://api.crowdtangle.com'
CROWDTANGLE_API_KEY = os.environ.get('CROWDTANGLE_API_KEY')

if CROWDTANGLE_API_KEY is None:
    logger.warning('CROWDTANGLE_API_KEY is not defined in the env!')


def shares_from_url(url, request_start_date):

    params = {
        'count': 1000,
        # 'includeHistory': 'true',
        'link': url,
        'platforms': 'facebook',
        'sortBy': 'total_interactions',
        'startDate': request_start_date,
        'token': CROWDTANGLE_API_KEY
    }

    api_endpoint = 'links'

    response = requests.get('{}/{}'.format(CROWDTANGLE_API_URL, api_endpoint),
                            params).json()