TODO
TODONE
- Run queries for ReadingDepth and Popups schemas.
- Document the results on this task and on the "Extraneous EL events investigation summary [draft]" document.
For the ReadingDepth schema:
select substring( userAgent, instr(userAgent,'"browser_family": "') + 19, instr(substring(userAgent, instr(userAgent,'"browser_family": "') + 19 ), '"' ) - 1 ) as BrowserFamily, count(*) AS Dupes from ( select userAgent, count(*) as nDupes from log.ReadingDepth_16325045 where timestamp like '2017051%' and event_pageToken is not null and event_action = 'pageLoaded' group by userAgent, event_pageToken having nDupes > 1 ) as dupeList group by BrowserFamily order by Dupes desc; +------------------+-------+ | BrowserFamily | Dupes | +------------------+-------+ | Chrome | 463 | | Chrome Mobile | 33 | | Safari | 15 | | Mobile Safari | 11 | | Firefox | 11 | | IE | 4 | | Edge | 3 | | Samsung Internet | 2 | | AppEngine-Google | 2 | | Yandex Browser | 1 | | Python Requests | 1 | | IE Mobile | 1 | +------------------+-------+
Here's a dump of the data that I've collected while investigation whether the frequency of logged events triggers duplication.
select timestamp as t, event_testID as 'Test ID', event_testLength as 'Test Length (Expected)', count(*) as 'Test Length (Actual)' from log.TestEventDuplication_16757884 group by event_testID; +----------------+------------------+------------------------+----------------------+ | t | Test ID | Test Length (Expected) | Test Length (Actual) | +----------------+------------------+------------------------+----------------------+ | 20170526083002 | 1d79cb8ed87c8bda | 50 | 50 | | 20170514053424 | 265fbd258eb06c3c | 20 | 20 | | 20170514072830 | 3c016715ca71fe09 | 50 | 50 | | 20170514070226 | 63bd7c82e4739056 | 50 | 50 | | 20170526082851 | a41956a47e5dca74 | 50 | 50 | | 20170514053637 | d65f157cc11ae213 | 20 | 20 | | 20170514053654 | ea4fae0f8bf7c06d | 20 | 20 | +----------------+------------------+------------------------+----------------------+