Skip to content

fswair/MentoDB

Repository files navigation

MentoDB

Sqlite3 based powerful database project.

Install with pip just one step

pip3 install mentodb

also; PyPI Page for MentoDB

Requirements:

  • Python 3.9.6 or greater version

  • pydantic -> pip install pydantic

  • pandas -> pip install pandas

  • numpy -> pip install numpy

  • Import these two module before start:

from pydantic import BaseModel
from pydantic.dataclasses import dataclass

Working with Base Models

The following code demonstrates how to work with base models in Python using pydantic and dataclasses.

Creating a Model Extended from Base Model

The following code creates a model named MyModel that extends from BaseModel:

@dataclass
class MyModel(BaseModel):
    id: int
    name: str
    job: str
    price: int

Creating a SQL Table with a Model

Here's an example of how to create a SQL table with the MyModel model:

# Initialize a connection with MentoConnection (similar to "sqlite3.Connection")
con = MentoConnection("./database/new.db", check_same_thread=False)

# Create a database cursor with the connection object.
cursor = Mento(con)

# Create a table with the following structure: (id int, name text, job text, price int)
cursor.create("sample_table", model=MyModel)

Using Primary Key and Unique Column Matches When Creating Tables

Primary Key:

@dataclass
class PrimaryKeySample(BaseModel):
    id: PrimaryKey(int)
    name: str
    age: int
    price: int

# Create a table with the following structure: (id int primary key, name text, age int, price int)
cursor.create("primary_sample", model=PrimaryKeySample)

Unique Matches:

@dataclass
class Sample(BaseModel):
    id: PrimaryKey(int)
    name: str
    age: int
    price: int
    check_match: UniqueMatch("id", "name")

# Create a table with unique match control.
cursor.create("unique_matches_sample", model=Sample)

# Set the check_model parameter to check if there are matches.
# If the table has matched data, the insert process will be stopped.
cursor.check_model = Sample

Data Statements

Create

  • Create a table if it does not already exist:
cursor.create("sample", model=Sample)
  • Create a table without checking if it already exists:
cursor.create("sample", model=Sample, exists_check=False)
  • Create multiple tables:
cursor.create_many(dict(first=MyModel, second=PrimaryKeySample, third=Sample))

Insert

cursor.insert(
    "sample",
    data=dict(id=1, name="fswair", age=18, price=4250),
    # If your model has a UniqueMatch control and you want to check matches, set the model with the check_model keyword argument.
    check_model=Sample
    )

Select

  • Return all rows as a list of dictionaries:
cursor.select("sample")
# Output: [{id: 1, name: fswair, age: 18, price: 4250}]

SELECT

The following are the different methods for returning data from the table using the select statement in the cursor object:

  • cursor.select("sample"): Returns all rows as list[dict] -> [{id: 1, name: fswair, age: 18, price: 4250}]

  • cursor.select("sample", where={"id": 1, "name": "fswair"}): Returns all rows matched with the where condition. The condition looks like (in SQL): SELECT * FROM TABLE WHERE id = 1 AND name = 'fswair'.

  • cursor.select("sample", where={"id": 1, "name": "fswair"}, order_by="id"): Returns all rows matched with the where condition sorted as ORDER BY. The condition looks like (in SQL): SELECT * FROM TABLE WHERE id = 1 AND name = 'fswair' ORDER BY id.

  • cursor.select("sample", select_column="id"): Returns all row's id columns as list[dict] -> [{id: 1}, {id: 2}]

  • cursor.select("sample", filter=lambda id: id % 3 == 0): Returns all rows matched with the lambda filter (lambda arg must be column name). Example output: list[dict] -> [{id: 3, name: fswair, age: 18, price: 4250}].

  • cursor.select("sample", regexp={"id": ["\d{1,3}"]}): Returns all rows matched with regexp patterns (regexp dict must be one key as column name, value could be pattern or list of patterns). Example output: list[dict] -> [{id: 999, name: fswair, age: 18, price: 4250}].

Response Formatters for Select Statement

The following are the response formatters for the select statement:

  • cursor.select("table", as_json=True): Returns data as JSON.

  • cursor.select("table", as_dataframe=True): Returns data as a DataFrame (using Pandas).

  • cursor.select("table", as_dataframe=True).to_csv(): Returns data as a CSV file.

  • cursor.select("table", model=Sample, as_model=True): Returns object list (accessible with attributes).

UPDATE

The following updates the data matched with the where condition:

cursor.update(
    "sample",
    set_data=dict(name="fswair-up", age=20),
    where={"id": 1, "name": "fswair"}
    )

About

MentoDB is powerful database engine for sqlite3. You have many options to use, specially basic things, also lambda filters, regular expressions included.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages