CODEWITHSUNDEEP
pythonmysqlfastapipydantic
comlink
comlink

Reputation: 27

Fast API - MySQL - SQLAlchemy: Get ValidationError 'Invalid type' in Pydantic Model

I'm fairly new to this framework, but i'm working on a timesheet application (API) for a small company.

The issue i have is that when i select a time value from mysql i got the following error in the pydantic module:

pydantic.error_wrappers.ValidationError: 2 validation errors for TimeSheetRange 
response -> 0 -> total   
invalid type; expected time, string, bytes, int or float (type=type_error) 
response -> 1 -> total   
invalid type; expected time, string, bytes, int or float (type=type_error)

I have tried a lot of workarounds to get the time to string from mysql with to_char() which also failed, but i rather want to know what the issue is here as i would like to use the data type time...

schemas.py

from typing import List, Optional
from datetime import date, time, datetime

from pydantic import BaseModel
from . import models

...


class TimeSheetRange(BaseModel):
    user_id: int
    start: date
    total: time

    class Config:
        orm_mode = True

crud.py

def get_timesheet_by_date_range_and_user(db: Session, user_id=int, date_start=datetime, date_end=datetime):
    return db.query(models.TimeSheet.user_id, \
                    func.date(models.TimeSheet.start).label('start'), \
                    func.sec_to_time(func.sum(func.timediff(models.TimeSheet.end,models.TimeSheet.start))).label('total') ) \
            .filter(models.TimeSheet.user_id == user_id) \
            .filter(func.date(models.TimeSheet.start) >= date_start) \
            .filter(func.date(models.TimeSheet.end) <= date_end) \
            .group_by(func.date(models.TimeSheet.start)) \
            .all()

main.py

@app.get("/timesheets/range/", response_model=List[schemas.TimeSheetRange])
    def read_timesheets(user_id: int, date_start:date = datetime.now().date(), date_end:date = datetime.now().date(), db: Session = Depends(get_db)):
        timesheets = crud.get_timesheet_by_date_range_and_user(db, user_id=user_id, date_start=date_start, date_end=date_end)
        return timesheets

And the SQL Query that is build with a result:

SELECT 
    timesheet.user_id AS timesheet_user_id, 
    date(timesheet.start) AS start,
    sec_to_time(sum(timediff(timesheet.end, timesheet.start))) AS total  
FROM 
    timesheet  
WHERE 
    timesheet.user_id = 7 
AND  
    date(timesheet.start) >= '2020-09-20' 
AND 
    date(timesheet.end) <= '2020-09-22' 
GROUP BY 
    date(timesheet.start);
+-------------------+------------+----------+
| timesheet_user_id | start      | total    |
+-------------------+------------+----------+
|                 7 | 2020-09-20 | 17:50:47 |
|                 7 | 2020-09-21 | 18:21:11 |
+-------------------+------------+----------+

Is there something wrong with the datatype time or am i missing something really stupid?

Upvotes: 2

Views: 4773

Answers (1)

TheDataFox
TheDataFox

Reputation: 334

I think the issue here is that the Pydantic model you are using is expecting a time object type, when it should be expecting a timedelta type as your sql statement seems to indicate that you are calculating a time difference for the total column.

So you Pydantic schemas.py file should read:

from typing import List, Optional
from datetime import date, datetime, timedelta

from pydantic import BaseModel
from . import models

...


class TimeSheetRange(BaseModel):
    user_id: int
    start: date
    total: timedelta

    class Config:
        orm_mode = True

Upvotes: 1

Related Questions