[Bug]: Incorrect Quali Times - Duplicated from other Quali Session

[theOehrly]

Description

In Azerbaijan 2023 Qualifying session, some drivers are listed with equal times for Q1/Q2 or Q2/Q3 where one of these times is incorrect.

Data from the admin interface suggest that all the correct times are available in the database, and this is caused by a bug when responses are generated.

Affected drivers (in this session):

• position 3, Perez, Q1/Q2
• position 4, Sainz, Q2/Q3
• position 5, Hamilton, Q2/Q3

Other sessions may be affected as well.

Affected URL

https://api.jolpi.ca/ergast/f1/2023/4/qualifying/

To Reproduce

Expected Behavior

The correct lap time should be returned for each Quali session.

Supporting Data / Screenshots

Example: Sainz is listed with identical Q2 and Q3 time, where the Q3 time is correct and the Q2 time is incorrect.

{
      "number": "55",
      "position": "4",
      "Driver": {
          "driverId": "sainz",
          "permanentNumber": "55",
          "code": "SAI",
          "url": "http://en.wikipedia.org/wiki/Carlos_Sainz_Jr.",
          "givenName": "Carlos",
          "familyName": "Sainz",
          "dateOfBirth": "1994-09-01",
          "nationality": "Spanish"
      },
      "Constructor": {
          "constructorId": "ferrari",
          "url": "http://en.wikipedia.org/wiki/Scuderia_Ferrari",
          "name": "Ferrari",
          "nationality": "Italian"
      },
      "Q1": "1:42.197",
      "Q2": "1:41.016",
      "Q3": "1:41.016"
  },

Our Admin interface shows

[jolpica]

This is a data issue:
The is_entry_fastest_lap field determines which laps we show, in this case we had a duplicate lap one in Q3 one in Q2. This could be related to the data bug we get where we sometimes can't figure out if a qualifying lap is in Q1,Q2 or Q3.

Fixed this for Sainz in issue above, we'll need to do this on a case by case basis for other issues.

[theOehrly]

I'll reopen this as a reminder, since I actually forgot about this problem for a bit.

I've done some work on fixing the underlying parser issues. This still needs to be discussed, and we need to figure out if my fix is a good solution in the end. If yes, we can reload the data for the affected sessions. Else, I'll look into manually fixing the data.