使用python在postgresql数据库中插入json数据问题

Question

main.py

data = []
with open('data.json') as f:
    for line in f:
        data.append(json.loads(line))
f.close()

fields = [
    'id', #integer
    'name', #varchar
    'log_date', #date
    'log_time', #timestamp
    'login', #timestamp
    'logout' #timestamp
]

for item in data:
    my_data = [item[field] for field in fields]
    insert_query = "INSERT INTO employee VALUES (%d, %s, %s, %s, %s, %s)"
    cur.execute(insert_query, tuple(my_data))

data.json

[
    {
        "id": 1,
        "name": "Prosenjit Das",
        "log_date": "2019-03-02",
        "log_time": "12:10:12.247257",
        "login": null,
        "logout": null
    },
    {
        "id": 2,
        "name": "Sudipto Rahman",
        "log_date": "2019-03-02",
        "log_time": "12:10:12.247257",
        "login": "11:26:45",
        "logout": "10:49:53"
    },
    {
        "id": 3,
        "name": "Trump Khatun",
        "log_date": "2019-03-02",
        "log_time": "12:10:12.247257",
        "login": null,
        "logout": null
    }
]

postgresql column fields

My database connect is okay. In that picture line 37 when i'm using dumps instead of loads then another problem is shown in line 50 that "Typeerror: string indices must be integers". Notice that here json format type is a list. This kinds of problem but not exactly i've seen many but properly doesn't work.

Thanks.

Answer 1

So couple of changes I would make here

with open('data.json') as f:
    data = json.load(f)

# no need to do f.close() since we are using a context manager

fields = [
    'id', #integer
    'name', #varchar
    'log_date', #date
    'log_time', #timestamp
    'login', #timestamp
    'logout' #timestamp
]

for item in data:
    my_data = [item[field] for field in fields]
    insert_query = "INSERT INTO employee (id, name, log_date, log_time, login, logout) VALUES (%s, %s, %s, %s, %s, %s)"

    # also ALL placeholders must be %s even if it is an integer
    cur.execute(insert_query, tuple(my_data))

Also if you are using the psycopg2 module for your DB actions you can do the following

from psycopg2.extras import execute_values

my_data = [tuple(item[field] for field in fields) for item in data]
insert_query = "INSERT INTO employee (id, name, log_date, log_time, login, logout) VALUES %s"
execute_values(cursor, insert_query, my_data) 

Answer 2

Load the json into a list of dicts once + removing an extra comma

import json

with open('data.json', 'r') as f:
    data = json.load(f)
# now  you can iterate and push to entries to DB

data.json

[
    {
        "id": 1,
        "name": "Prosenjit Das",
        "log_date": "2019-03-02",
        "log_time": "12:10:12.247257",
        "login": null,
        "logout": null
    },
    {
        "id": 2,
        "name": "Sudipto Rahman",
        "log_date": "2019-03-02",
        "log_time": "12:10:12.247257",
        "login": "11:26:45",
        "logout": "10:49:53"
    },
    {
        "id": 3,
        "name": "Trump Khatun",
        "log_date": "2019-03-02",
        "log_time": "12:10:12.247257",
        "login": null,
        "logout": null
    }, --> was removed
]