Unclear statistics from count endpoint

[borya_g]

Hi,
Can you please help me to understand the results of the next query:

premium_search_args = load_credentials("~/.twitter_keys.yaml",
                                       yaml_key="search_tweets_api",
                                       env_overwrite=False)
rule = {"query": "#BeDeadpools2nd OR #Deadpool OR #Deadpool2 OR #WalkingDeadpool OR @deadpoolmovie", "bucket": "day"} 
counts = collect_results(rule, result_stream_args=premium_search_args)

I received the next results:

[{'timePeriod': '201806060000', 'count': 1894}, {'timePeriod': '201806050000', 'count': 9400}, {'timePeriod': '201806040000', 'count': 13181},
{'timePeriod': '201806030000', 'count': 15936}, {'timePeriod': '201806020000', 'count': 24676}, {'timePeriod': '201806010000', 'count': 28975},
{'timePeriod': '201805310000', 'count': 45992}, {'timePeriod': '201805300000', 'count': 13830}, {'timePeriod': '201805290000', 'count': 25131},
{'timePeriod': '201805280000', 'count': 38660}, {'timePeriod': '201805270000', 'count': 55530}, {'timePeriod': '201805260000', 'count': 30699},
{'timePeriod': '201805250000', 'count': 43110}, {'timePeriod': '201805240000', 'count': 30912}, {'timePeriod': '201805230000', 'count': 40654},
{'timePeriod': '201805220000', 'count': 44777}, {'timePeriod': '201805210000', 'count': 80401}, {'timePeriod': '201805200000', 'count': 90817},
{'timePeriod': '201805190000', 'count': 0}, {'timePeriod': '201805180000', 'count': 112479}, {'timePeriod': '201805170000', 'count': 102260},
{'timePeriod': '201805160000', 'count': 111577}, {'timePeriod': '201805150000', 'count': 106105}, {'timePeriod': '201805140000', 'count': 54840},
{'timePeriod': '201805130000', 'count': 41614}, {'timePeriod': '201805120000', 'count': 34084}, {'timePeriod': '201805110000', 'count': 69753}, 
{'timePeriod': '201805100000', 'count': 103061}, {'timePeriod': '201805090000', 'count': 63232}, {'timePeriod': '201805080000', 'count': 45802},
{'timePeriod': '201805070000', 'count': 63416}] 

For date 2018-05-19 the count is 0, but I can see the tweets with those hashtags from this date.
What is wrong with the counts ?
Best,
Boris

[KvdohertyQU]

Hi @borya_g,

We believe we’ve fixed what was causing this behavior, so you should see the count data populating correctly now. Please feel free to run this search again and let us know if you’re seeing the same behaviour.

Best,
Kevin

[borya_g]

I have another issue with Premium API and it looks similar for me like the previous one from this topic :
I run the next request :
query_text = “”“TAG (“Jeremy Renner” OR “Ed Helms” OR “Jake Johnson” OR “Jeff Tomsic” OR “Broken Road Productions” OR “New Line Cinema” OR “Comedy” OR “MOVIE” OR “FILM”)”""
since_date = “201805200000”
until_date = “201806201200”
premium_search_args = load_credentials("~/.twitter_keys.yaml",
yaml_key=“search_tweets_api”,
env_overwrite=False)
pt_rule = query_text + " -is:retweet lang:en "
rule = gen_rule_payload(pt_rule, results_per_call=500,
from_date=since_date, to_date=until_date, count_bucket=None,
tag=None,stringify=True)
print (rule)
tweets_list = collect_results(rule,
max_results=40000,
result_stream_args=premium_search_args)
And results looks inconsistent with my request
TweetID TweetText
9.98166285275091E+017 / PERFECT SUMMER READING! In this epic thriller @KipAuthor blends #political #intrigue #homelandsecurity #blackops… https://t.co/qUH61pR3kg
9.98152073614123E+017 @Siddh_Bhadoria @nanditakhaire @Jayajaitly yes cleavage showing gives roles…fr god sake learn diff bw actress and… https://t.co/kpc2AoIZpl
9.98096611149398E+017 “Blu-ray
New Releases in Anime
#10: Black Butler: Book of the Atlantic - Movie (Blu-ray/DVD Combo + UV) ~ Brina Pale… https://t.co/oeQxDL7FnV”
9.98087365527724E+017 The best part about #Sanju getting a blockbuster tag MIGHT see a lot of people watching the Extraordinary, Underrat… https://t.co/uN7UDRSnZI
9.98051524734603E+017 @BrackAttack78 Well apparently in the new movie Sheriff Brackett arrests him after he falls off the balcony. Loomis… https://t.co/YzqplJsFQf
9.98051321604444E+017 “Step 1: Watch a Different Movie” Read #WorryNoMoreBook => https://t.co/5RkTgfIZ4g https://t.co/3IYfU6JaWR
In some of the tweets the word “tag” is missing but this word is a part of my query.
Can you please explain me the reason of it ?
Best,
Boris

[KvdohertyQU]

In those example above where “tag” isn’t immediately visible in the text field, it appears that the keyword search for TAG was matching in the full_text field and the url’s of the Tweet metadata. As an example, for Tweet 9.98166285275091E+01, the word TAG is in the url of the Tweet.