Database and SQLAlchemy

In this blog we will explore using programs with data, focused on Databases. We will use SQLite Database to learn more about using Programs with Data. Use Debugging through these examples to examine Objects created in Code.

  • College Board talks about ideas like

    • Program Usage. "iterative and interactive way when processing information"
    • Managing Data. "classifying data are part of the process in using programs", "data files in a Table"
    • Insight "insight and knowledge can be obtained from ... digitally represented information"
    • Filter systems. 'tools for finding information and recognizing patterns"
    • Application. "the preserve has two databases", "an employee wants to count the number of book"

  • PBL, Databases, Iterative/OOP

    • Iterative. Refers to a sequence of instructions or code being repeated until a specific end result is achieved
    • OOP. A computer programming model that organizes software design around data, or objects, rather than functions and logic
    • SQL. Structured Query Language, abbreviated as SQL, is a language used in programming, managing, and structuring data

Imports and Flask Objects

Defines and key object creations

  • Comment on where you have observed these working? Provide a defintion of purpose.
    1. Flask app object
    2. SQLAlchemy db object


Response:

  • The code below were present in the Tri 2 final project's Flask portfolio.
  • The Flask app object "implements a WSGI application". A WSGI application is used by web servers to send requests to applications that are written with Python.
  • In the code below, the SQLAlchemy db object is created. The class is used to integrate SQLAlchemy to Flask. The object is created so that Flask will support it when needed (as opposed to doing:


app = Flask(__name__)
db = SQLAlchemy(app)

where the the instance is bound to Flask so that methods can be used at anytime.)

"""
These imports define the key objects
"""

from flask import Flask
from flask_sqlalchemy import SQLAlchemy

"""
These object and definitions are used throughout the Jupyter Notebook.
"""

# Setup of key Flask object (app)
app = Flask(__name__)
# Setup SQLAlchemy object and properties for the database (db)
database = 'sqlite:///sqlite.db'  # path and filename of database
app.config['SQLALCHEMY_TRACK_MODIFICATIONS'] = False
app.config['SQLALCHEMY_DATABASE_URI'] = database
app.config['SECRET_KEY'] = 'SECRET_KEY'
db = SQLAlchemy()


# This belongs in place where it runs once per project
db.init_app(app)

Model Definition

Define columns, initialization, and CRUD methods for users table in sqlite.db

  • Comment on these items in the class, purpose and definition.
    • class User The user class is the overarching branch that contains things that can define the user. There is a constructor called __init__ in the class.
    • db.Model inheritance db.Model is seen as a parameter in the User class. This is the database model, and the functions in the model are inherited into the User class in order to build it.
    • init method This is the constructor of the User class.
    • @property, @<column>.setter @property is another name for a getter. @column.setter is the setter.
    • create, read, update, delete methods @peroperty relates to create, while @column.setter relates to create or update.


Response

  • The columns in the sqlite.db table can be associated with the variables defined in the code below. In this case, the columns/variables are:_name, _uid, _password, and _dob.
  • Initialization can be found in the __init__ constructor. Within the constructor, the name, uid, password, and dob is initialized to the values specified by the user. If the password is not specified, it is set to the default value of 123qwerty.
  • CRUD methods can be seen in the four methods at the end of the code (create, read, update, and delete)
""" database dependencies to support sqlite examples """
import datetime
from datetime import datetime
import json

from sqlalchemy.exc import IntegrityError
from werkzeug.security import generate_password_hash, check_password_hash


''' Tutorial: https://www.sqlalchemy.org/library.html#tutorials, try to get into a Python shell and follow along '''

# Define the User class to manage actions in the 'users' table
# -- Object Relational Mapping (ORM) is the key concept of SQLAlchemy
# -- a.) db.Model is like an inner layer of the onion in ORM
# -- b.) User represents data we want to store, something that is built on db.Model
# -- c.) SQLAlchemy ORM is layer on top of SQLAlchemy Core, then SQLAlchemy engine, SQL
class User(db.Model):
    __tablename__ = 'users'  # table name is plural, class name is singular

    # Define the User schema with "vars" from object
    id = db.Column(db.Integer, primary_key=True)
    _name = db.Column(db.String(255), unique=False, nullable=False)
    _uid = db.Column(db.String(255), unique=True, nullable=False)
    _password = db.Column(db.String(255), unique=False, nullable=False)
    _dob = db.Column(db.Date)

    # constructor of a User object, initializes the instance variables within object (self)
    def __init__(self, name, uid, password="123qwerty", dob=datetime.today()):
        self._name = name    # variables with self prefix become part of the object, 
        self._uid = uid
        self.set_password(password)
        if isinstance(dob, str):  # not a date type     
            dob = date=datetime.today()
        self._dob = dob

    # a name getter method, extracts name from object
    @property
    def name(self):
        return self._name
    
    # a setter function, allows name to be updated after initial object creation
    @name.setter
    def name(self, name):
        self._name = name
    
    # a getter method, extracts uid from object
    @property
    def uid(self):
        return self._uid
    
    # a setter function, allows uid to be updated after initial object creation
    @uid.setter
    def uid(self, uid):
        self._uid = uid
        
    # check if uid parameter matches user id in object, return boolean
    def is_uid(self, uid):
        return self._uid == uid
    
    @property
    def password(self):
        return self._password[0:10] + "..." # because of security only show 1st characters

    # update password, this is conventional method used for setter
    def set_password(self, password):
        """Create a hashed password."""
        self._password = generate_password_hash(password, method='sha256')

    # check password parameter against stored/encrypted password
    def is_password(self, password):
        """Check against hashed password."""
        result = check_password_hash(self._password, password)
        return result
    
    # dob property is returned as string, a string represents date outside object
    @property
    def dob(self):
        dob_string = self._dob.strftime('%m-%d-%Y')
        return dob_string
    
    # dob setter, verifies date type before it is set or default to today
    @dob.setter
    def dob(self, dob):
        if isinstance(dob, str):  # not a date type     
            dob = date=datetime.today()
        self._dob = dob
    
    # age is calculated field, age is returned according to date of birth
    @property
    def age(self):
        today = datetime.today()
        return today.year - self._dob.year - ((today.month, today.day) < (self._dob.month, self._dob.day))
    
    # output content using str(object) is in human readable form
    # output content using json dumps, this is ready for API response
    def __str__(self):
        return json.dumps(self.read())

    # CRUD create/add a new record to the table
    # returns self or None on error
    def create(self):
        try:
            # creates a person object from User(db.Model) class, passes initializers
            db.session.add(self)  # add prepares to persist person object to Users table
            db.session.commit()  # SqlAlchemy "unit of work pattern" requires a manual commit
            return self
        except IntegrityError:
            db.session.remove()
            return None

    # CRUD read converts self to dictionary
    # returns dictionary
    def read(self):
        return {
            "id": self.id,
            "name": self.name,
            "uid": self.uid,
            "dob": self.dob,
            "age": self.age,
        }

    # CRUD update: updates user name, password, phone
    # returns self
    def update(self, name="", uid="", password=""):
        print("SELF: " + str(self))
        """only updates values with length"""
        if len(name) > 0:
            self.name = name
        if len(uid) > 0:
            self.uid = uid
        if len(password) > 0:
            self.set_password(password)
        print(self)
        db.session.commit()
        return self

    # CRUD delete: remove self
    # None
    def delete(self):
        db.session.delete(self)
        db.session.commit()
        return None

Initial Data

Uses SQLALchemy db.create_all() to initialize rows into sqlite.db

  • Comment on how these work?
    1. Create All Tables from db Object
    2. User Object Constructors
    3. Try / Except

Response

  • create_all() creates the database.
  • The user objects are initialized with a call to the constructor in the User class. This initializes the name, uid, etc. variables.
  • try/except works by executing the code in the try statement. If there are errors with the try clause, the program will execute the code in the except. In this example, the users in the users list are added into the Users table. However, if the users are already present in the table or the database has an error (such as not existing), the program instead outputs a message saying that the user already exists in the table or there is an error with the database.
"""Database Creation and Testing """


# Builds working data for testing
def initUsers():
    with app.app_context():
        """Create database and tables"""
        db.create_all()
        """Tester data for table"""
        u1 = User(name='Thomas Edison', uid='toby', password='123toby', dob=datetime(1847, 2, 11))
        u2 = User(name='Nikola Tesla', uid='niko', password='123niko')
        u3 = User(name='Alexander Graham Bell', uid='lex', password='123lex')
        u4 = User(name='Eli Whitney', uid='whit', password='123whit')
        u5 = User(name='Indiana Jones', uid='indi', dob=datetime(1920, 10, 21))
        u6 = User(name='Marion Ravenwood', uid='raven', dob=datetime(1921, 10, 21))


        users = [u1, u2, u3, u4, u5, u6]

        """Builds sample user/note(s) data"""
        for user in users:
            try:
                '''add user to table'''
                object = user.create()
                print(f"Created new uid {object.uid}")
            except:  # error raised if object nit created
                '''fails with bad or duplicate data'''
                print(f"Records exist uid {user.uid}, or error.")
                
initUsers()

Created new uid toby
Created new uid niko
Created new uid lex
Created new uid whit
Created new uid indi
Created new uid raven

Check for given Credentials in users table in sqlite.db

Use of ORM Query object and custom methods to identify user to credentials uid and password

  • Comment on purpose of following
    1. User.query.filter_by
    2. user.password

Response

  • User.query.filter_by searches through the database based on the attributes specified. The .first() method retrieves the first result found in the database. You can also use .all(), which will retrieve all instances found in the database (see code example below)
  • .is_password() checks to see if the user's password matches the password specified. This is achieved by calling the imported check_password_hash() method and specifying the hash of the user's password (found in the database) and the inputted password (the password you want to check against the hash).
def find_by_uid(uid):
    with app.app_context():
        user = User.query.filter_by(_uid=uid).first()
    return user # returns user object

# Check credentials by finding user and verify password
def check_credentials(uid, password):
    # query email and return user record
    user = find_by_uid(uid)
    if user == None:
        return False
    if (user.is_password(password)):
        return True
    return False
        
check_credentials("indi", "123qwerty")

True

My notes

.query.filter_by can end with .all() instead of first(). However, this returns the objects in a list format, so to get the actual contents of the object, you need to output the element by specifying the index of the list (ex: print(user[0]))

def find_by_uid2(uid):
    with app.app_context():
        user = User.query.filter_by(_uid=uid).all()
    print(user)
    print(user[0])
    print(user[0].uid)

        
find_by_uid2("indi")

[<User 5>]
{"id": 5, "name": "Indiana Jones", "uid": "indi", "dob": "10-21-1920", "age": 102}
indi

Create a new User in table in Sqlite.db

Uses SQLALchemy and custom user.create() method to add row.

  • Comment on purpose of following
    1. user.find_by_uid() and try/except
    2. user = User(...)
    3. user.dob and try/except
    4. user.create() and try/except

Response

  • The program first takes in the uid that is specified by the user, and stores it in the variable uid. This variable is passed into the method find_by_uid() to see if the user id already exists in the database. The try/except clause takes in the user object and attempts to execute the read() method on it. If successful, the program stops, because that means that the user was found in the database and therefore could not be added into the database. If the user was not present in the database, the try statement would result in an error, therefore executing the statement in the except clause, which says pass, or to keep executing the code below.
  • user = User() initializes a User object. Although the dob was not specified, the constructor contains a default value for the dob parameter. Alternatively, you can specify the dob variable before you initialize the User object, and then add it as an attribute to the User object (see in code example below)
  • user.dob is a setter that sets the birthday of the user. The setter is defined with the @dob.setter decorator. The try/except checks to see if the date that the user inputted is a valid date.
  • user.create() creates the user in the try statement. The try statement attempts to read the user object after the user is created. If this fails, that means that the user object was not successfully added to the database, so an error message will be printed.
def create():
    # optimize user time to see if uid exists
    uid = input("Enter your user id:")
    user = find_by_uid(uid)
    try:
        print("Found\n", user.read())
        return
    except:
        pass # keep going
    
    # request value that ensure creating valid object
    name = input("Enter your name:")
    password = input("Enter your password")
    
    # Initialize User object before date
    user = User(name=name, 
                uid=uid, 
                password=password
                )
    
    # create user.dob, fail with today as dob
    dob = input("Enter your date of birth 'YYYY-MM-DD'")
    try:
        user.dob = datetime.strptime(dob, '%Y-%m-%d').date()
    except ValueError:
        user.dob = datetime.today()
        print(f"Invalid date {dob} require YYYY-mm-dd, date defaulted to {user.dob}")
           
    # write object to database
    with app.app_context():
        try:
            object = user.create()
            print("Created\n", object.read())
        except:  # error raised if object not created
            print("Unknown error uid {uid}")
        
create()

Created
 {'id': 7, 'name': 'a', 'uid': 'a', 'dob': '03-25-2023', 'age': 0}

Alternate create() method 

# Inputs, Try/Except, and SQLAlchemy work together to build a valid database object
def create():
    # optimize user time to see if uid exists
    uid = input("Enter your user id:")
    user = find_by_uid(uid)
    try:
        print("Found\n", user.read())
        return
    except:
        pass # keep going
    
    # request value that ensure creating valid object
    name = input("Enter your name:")
    password = input("Enter your password")

    dob = input("Enter your date of birth 'YYYY-MM-DD'")
    try:
        dob = datetime.strptime(dob, '%Y-%m-%d').date()
    except ValueError:
        dob = datetime.today()
        print(f"Invalid date {dob} require YYYY-mm-dd, date defaulted to {user.dob}")
        
    # Initialize User object before date
    user = User(name=name, 
                uid=uid, 
                password=password,
                dob=dob
                )
    
    # create user.dob, fail with today as dob
    
           
    # write object to database
    with app.app_context():
        try:
            object = user.create()
            print("Created\n", object.read())
        except:  # error raised if object not created
            print("Unknown error uid {uid}")
        
create()

Created
 {'id': 8, 'name': 'bob2', 'uid': 'bob2', 'dob': '03-25-2023', 'age': 0}

Reading users table in sqlite.db

Uses SQLALchemy query.all method to read data

  • Comment on purpose of following
    1. User.query.all
    2. json_ready assignment, google List Comprehension

Response

  • As noted above, the .all() method returns all of the entries in the table that match the query filter (in this case, there is no filter; therefore, all entries are obtained)

  • json_ready calls the .read() method on all of the objects in the table list. List comprehension allows an easier way to iterate through statements in the list. The format of a list comprehension is:[expression for-each-loop if-loop].
    Expression is related to the item in the iteration and also the outcome of the element in the list.

    In this case, the list comprehension adds the JSON output of the users data from the database into the json_ready list.

# SQLAlchemy extracts all users from database, turns each user into JSON
def read():
    with app.app_context():
        table = User.query.all()
    json_ready = [user.read() for user in table] # "List Comprehensions", for each user add user.read() to list
    return json_ready

read()

[{'id': 1,
  'name': 'Thomas Edison',
  'uid': 'toby',
  'dob': '02-11-1847',
  'age': 176},
 {'id': 2,
  'name': 'Nikola Tesla',
  'uid': 'niko',
  'dob': '03-25-2023',
  'age': 0},
 {'id': 3,
  'name': 'Alexander Graham Bell',
  'uid': 'lex',
  'dob': '03-25-2023',
  'age': 0},
 {'id': 4,
  'name': 'Eli Whitney',
  'uid': 'whit',
  'dob': '03-25-2023',
  'age': 0},
 {'id': 5,
  'name': 'Indiana Jones',
  'uid': 'indi',
  'dob': '10-21-1920',
  'age': 102},
 {'id': 6,
  'name': 'Marion Ravenwood',
  'uid': 'raven',
  'dob': '10-21-1921',
  'age': 101},
 {'id': 7, 'name': 'Bob', 'uid': 'bob', 'dob': '03-25-2023', 'age': 0},
 {'id': 8, 'name': 'bob2', 'uid': 'bob2', 'dob': '03-25-2023', 'age': 0}]

Hacks

  • Add this Blog to you own Blogging site. In the Blog add notes and observations on each code cell.
  • Change blog to your own database.
  • Add additional CRUD
    • Add Update functionality to this blog.
    • Add Delete functionality to this blog.

Update

Note: The code below does not actually update the sqlite.db. My suspicion is that it has something to do with how the sqlite.db is added into fastpages.

However, based on the output below, you can see that the user, in fact, does change. By passing the user object with self, the code above in the update() method is able to change its attributes (in this case, changing the name from a to b). However, for some reason, db.session.commit() does not save the updated user object, even though this documentation supports it.

I tried the same thing on Tri 2's project, and the update is successful. This indicates that the table not being able to update may be due to something with fastpages.

def update():
    # optimize user time to see if uid exists
    uid = input("Enter your user id:")
    
    user = find_by_uid(uid)
    try:
        print("Found\n", user.read())
        return
    except:
        pass # keep going
    
    
    print("USER: " + str(user))
    
    # request value that ensure creating valid object
    name = input("Enter your name:")
    password = input("Enter your password")
    
    
         
          
    # write object to database
    with app.app_context():
        try:
            object = user.update(name=name, password=password, uid=uid)
            print("Created\n", object.read())
        except:  # error raised if object not created
            print("Unknown error uid {uid}")
        
update()

USER: {"id": 7, "name": "a", "uid": "a", "dob": "03-25-2023", "age": 0}
SELF: {"id": 7, "name": "a", "uid": "a", "dob": "03-25-2023", "age": 0}
{"id": 7, "name": "b", "uid": "a", "dob": "03-25-2023", "age": 0}
Created
 {'id': 7, 'name': 'b', 'uid': 'a', 'dob': '03-25-2023', 'age': 0}

Delete 

def delete():
    # optimize user time to see if uid exists
    uid = input("Enter your user id:")
    
    user = find_by_uid(uid)
   
    
    # write object to database
    with app.app_context():
        user.delete()
        
delete()