POST stats/jobs/accounts/:account_id endpoint not returning all data

[HarrisonVanOort]

Using the twitter-python-ads-sdk (https://github.com/twitterdev/twitter-python-ads-sdk), I’m attempting to asynchronously retrieve PROMOTED_TWEET data by first querying the POST stats/jobs/accounts/:account_id endpoint to retrieve job_ids, and then the GET stats/jobs/accounts/:account_id endpoint to download the gzip data. When making the POST request, I am passing in >1 entity_ids and >1 metric_groups, but am only receiving back a job_id for 1 entity_id and 1 metric_group. For example, the code looks like this…

from twitter_ads.http import Request
from twitter_ads.client import Client

client = Client(...)
account_id = '18ce54o6agw'

params = {
    'start_time': '2018-10-01T04:00:00Z',
    'end_time': '2018-10-07T04:00:00Z',
    'entity': 'PROMOTED_TWEET',
    'entity_ids': ['1urqsy', '26jsvt'],
    'granularity': 'DAY',
    'metric_groups': ['BILLING', 'ENGAGEMENT'],
    'placement': 'ALL_ON_TWITTER'
}

resource = '/4/stats/jobs/accounts/{account_id}'.format(account_id=account_id)
response = Request(client, 'post', resource, params=params).perform()

Then, response.body only returns the following (notice it is only for the first entity_id and the first metric_group):

{'request': {'params': {'start_time': '2018-10-01T04:00:00Z',
   'entity_ids': ['1urqsy'],
   'end_time': '2018-10-07T04:00:00Z',
   'placement': 'ALL_ON_TWITTER',
   'granularity': 'DAY',
   'entity': 'PROMOTED_TWEET',
   'metric_groups': ['BILLING']}},
 'data': {'start_time': '2018-10-01T04:00:00Z',
  'segmentation_type': None,
  'url': None,
  'id_str': '1072523786702876672',
  'entity_ids': ['1urqsy'],
  'end_time': '2018-10-07T04:00:00Z',
  'country': None,
  'placement': 'ALL_ON_TWITTER',
  'id': 1072523786702876672,
  'expires_at': None,
  'account_id': '18ce54o6agw',
  'status': 'PROCESSING',
  'granularity': 'DAY',
  'entity': 'PROMOTED_TWEET',
  'created_at': '2018-12-11T16:09:31Z',
  'platform': None,
  'updated_at': '2018-12-11T16:09:31Z',
  'metric_groups': ['BILLING']}}

So the question is, how can I make the request with multiple entity_ids and multiple metric_groups, and receive back a list of job_ids?

[jrsyo]

Hi @HarrisonVanOort ,

If you look at the Class of Analytics (code):
It normalizes some of the parameters before making a call which is for entity_ids and metric_groups:

twitterdev/twitter-python-ads-sdk/blob/master/twitter_ads/resource.py#L254

As you can see, it expects comma-separated strings rather than a list. I haven’t actually tested it yet, but I assume you can try if it does the trick.

Best,
Shohei

[HarrisonVanOort]

Hi @jrsyo,

Thanks for the reply. Unfortunately I already tried that with no success. For example, changing the parameters to this…

params = {
    'start_time': '2018-10-01T04:00:00Z',
    'end_time': '2018-10-07T04:00:00Z',
    'entity': 'PROMOTED_TWEET',
    'entity_ids': '1urqsy, 26jsvt',
    'granularity': 'DAY',
    'metric_groups': 'BILLING',
    'placement': 'ALL_ON_TWITTER'
    }

Gives me the following error:

[{'code': 'INVALID_PARAMETER', 'message': 'Expected an id, got " 26jsvt" for entity_ids', 'parameter': 'entity_ids'}]

[juanshishido]

@HarrisonVanOort: You might want to try to making entity_ids a list.

---

So the question is, how can I make the request with multiple entity_ids and multiple metric_groups, and receive back a list of job_ids?

If you have 20 or less entity IDs specified, there will be a single job ID.

---

Also, if you’re using the SDK is it possible to use the built-in methods to do what you want? If so, please take a look at this example for fetching analytics—both synchronous and asynchronous.

[HarrisonVanOort]

Thanks @juanshishido.

In my first post I mentioned that I provided entity_ids as a list, and that it did not work. When providing entity_ids as a list, I receive a response with data for only the first entity_id. That was my reason for posting the question. @jrsyo then mentioned to try a comma separated string, which also unfortunately did not work.

Also, what example are you referring to?

[juanshishido]

Here is the example: https://github.com/twitterdev/twitter-python-ads-sdk/blob/master/examples/analytics.py.

[juanshishido]

@jrsyo then mentioned to try a comma separated string, which also unfortunately did not work.

For what it’s worth, this worked for me.

>>> params = {
        'start_time': '2018-10-01',
        'end_time': '2018-10-07',
        'entity': 'PROMOTED_TWEET',
        'entity_ids': '5tddn,619jl',
        'granularity': 'DAY',
        'metric_groups': 'BILLING,ENGAGEMENT',
        'placement': 'ALL_ON_TWITTER'
    }
>>> resource = '/4/stats/jobs/accounts/{account_id}'.format(account_id=ACCOUNT_ID)
>>> response = Request(client, 'post', resource, params=params).perform()
>>> response.body
{'data': {'account_id': '18ce54d4x5t',
  'country': None,
  'created_at': '2018-12-17T22:18:45Z',
  'end_time': '2018-10-07T07:00:00Z',
  'entity': 'PROMOTED_TWEET',
  'entity_ids': ['5tddn', '619jl'],
  'expires_at': None,
  'granularity': 'DAY',
  'id': 1074791033962082304,
  'id_str': '1074791033962082304',
  'metric_groups': ['BILLING', 'ENGAGEMENT'],
  'placement': 'ALL_ON_TWITTER',
  'platform': None,
  'segmentation_type': None,
  'start_time': '2018-10-01T07:00:00Z',
  'status': 'PROCESSING',
  'updated_at': '2018-12-17T22:18:45Z',
  'url': None},
 'request': {'params': {'end_time': '2018-10-07T07:00:00Z',
   'entity': 'PROMOTED_TWEET',
   'entity_ids': ['5tddn', '619jl'],
   'granularity': 'DAY',
   'metric_groups': ['BILLING', 'ENGAGEMENT'],
   'placement': 'ALL_ON_TWITTER',
   'start_time': '2018-10-01T07:00:00Z'}}}

It’s likely that you have a space between 1urqsy and 26jsvt.

[HarrisonVanOort]

That did it. Thanks!