Viva Prep

SimpleCascade parent ke saath child pe automatic action (generally delete).

Company delete → uski drives bhi delete, warna orphan rows / FK error.
SQLAlchemy: db.relationship('Drive', cascade='all, delete-orphan', backref='company')
SQL: ForeignKey(..., ondelete='CASCADE')

Cascade sirf delete nahi: save-update, merge, refresh-expire, expunge, delete-orphan.
Cascade Update: parent PK change to child FK update (rare, PK mat badlo).
Parent delete jab children active hon: constraint error, ya cascade se children gayab, ya pehle children handle karo — design choice examiner poochta hai.
CSS cascading alag topic hai (styles inherit/override).

SimpleNetwork tab mein password dikhna HTTPS pe encrypted transit hai, DevTools local dikhata hai. Kabhi query string mein password mat bhejo — POST body. Production HTTPS mandatory.

SimpleBase template pe navbar/footer, child pages {% extends %} + {% block %} se content bharte hain. DRY.

Example

{# base.html #}
{% block content %}{% endblock %}

{# home.html #}
{% extends 'base.html' %}
{% block content %}
  <h1>Home</h1>
{% endblock %}

SimpleClient (Flask .delay), Broker (Redis), Worker. Optional Beat + result backend.

Viva line'Flask produce, Redis transport, worker consume.'

include task modules, app context ke andar DB session. Windows pe eventlet/gevent issues — docs se command.

Celery Python ka distributed task queue — HTTP request ko 10s email/CSV ke liye mat rokna.

Flow: Flask .delay() → Redis broker → Worker execute. Beat alag process hai (clock): crontab padhke due time pe queue mein daalta hai. Beat khud mail nahi bhejta; Worker haath hai.

MAD2 typical: daily reminder (Beat), monthly report (Beat), user Export CSV (button, Beat nahi). JS async/await browser event loop hai, Celery server-side process — mix mat karo.

Demo: tin terminals — redis-server, celery worker, celery beat — plus Flask. MailHog :8025 inbox. Result backend se task SUCCESS poll optional.

Project example: create_app ke baad celery.conf.update. tasks import taaki register hon.

Viva tip

Viva line'Flask task produce karta hai, Redis transport, worker consume.'

Simple:class object/array/string dynamic CSS.

Purpose dynamic class. Error red.

Directives Vue ke special attributes hain — HTML ko superpower: v-if/v-show, v-for, v-bind (:), v-on (@), v-model.

v-if DOM se hataata/jodta. v-show sirf CSS display. Rare/permission: v-if. Bar-bar toggle: v-show. v-model two-way form. v-bind one-way JS→attribute.

:key v-for pe zaroori warna Vue galat reuse. v-html XSS. Secret UI ke liye sirf v-if kaafi nahi — API 403 bhi.

MAD1 Jinja server HTML. MAD2 yeh directives browser mein reactive.

Example

<div :class="{ 'text-danger': error, 'fw-bold': important }">Msg</div>

Project example: :class="{ 'text-danger': error }"

SimpleDono RAM cache. Redis zyada smart: lists, hashes, pub/sub, persistence, Celery broker.

Memcached: simple key-value, yahi. Redis: data structures + expire + optional disk dump.

MAD2 Redis choose — cache aur broker ek hi process. Memcached Celery nahi chalaata.

Redis in-memory key-value store hai — fridge pe sticky note, poori kitchen (SQL database) har baar mat kholo.

MAD2 mein Redis do kaam: (1) cache — same venues/shows/dashboard JSON baar-baar DB se nahi (2) Celery broker — Flask task queue mein daalta hai, worker uthata hai.

Flow: cache.get(key) → hit pe return, miss pe SQLAlchemy + SET with timeout. Admin write ke baad cache.delete, warna stale list.

Source of truth database hi hai. Redis crash/band = cache miss (app chalni chahiye) + queue ruk sakti hai. Important bookings sirf Redis mein mat rakho.

Alag: Redis ≠ Memcached (Celery broker nahi). Redis ≠ localStorage (woh browser). redis-cli ping → PONG viva proof.

Project example: Tumhare config mein CACHE_TYPE RedisCache, CELERY broker redis://. Memcached nahi.

Viva tipMAD2 mein Redis isliye — cache + broker ek hi tool. Memcached Celery nahi chalaata.

SimpleSlots = dabbe ke andar parent apna HTML daal de. Props data, slots markup.

Default slot, named slots. App.vue router-view layout jaisa outlet.

Agar nahi use kiya: 'layout router-view se kiya' valid.

Component = reusable UI block (Navbar, Card). Props parent→child data, emit child→parent event. Slots parent HTML inner content.

computed cache + reactive derive (filtered list). watch side-effect (query change pe API). methods events. ref() primitive, reactive() object. Vuex/Pinia global store; localStorage persist, store nahi.

SFC: template + script + style ek .vue file. Parent-child exam favourite: list parent, row child :item + @delete.

Example

<div class="card"><slot>Default</slot></div>
<Card><p>Custom body</p></Card>

Project example: Card.vue <slot>Default</slot>. Parent custom body.

SimpleSQLAlchemy MAD1 ka common ORM. Flask-SQLAlchemy wrapper se db.Model, db.session.

Example

user = User(email='a@b.com')
db.session.add(user)
db.session.commit()
users = User.query.filter_by(role='admin').all()

SimpleLifecycle = component ki zindagi ke stages. API call tab jab DOM ready ho — mounted.

created: instance bani, this data, DOM nahi. mounted/onMounted: page pe aa gaya — fetch, Chart.js, setInterval.
updated: data change ke baad. unmounted: cleanup (clearInterval) warna memory leak.

Dashboards onMounted isliye: canvas/API tab valid. created mein $el nahi.

Composition: onMounted(() => {}) setup ke andar.

Example

export default {
  async mounted() {
    const r = await fetch('/api/venues')
    this.venues = await r.json()
  }
}

Project example: Venues.vue mounted fetch list. Clock component unmounted pe interval clear.

Viva tipcreated: data setup, DOM nahi. mounted/onMounted: $el mil gaya — fetch yahi. Dashboards isliye onMounted use karti hain.

SimpleCode explain formula: input kya aaya → auth/validation → DB change → response/UI.

Viva tipFunction signature + 3-4 important lines + edge case.

SimpleSQLite file-based DB — setup easy, MAD1 ke liye perfect. Production heavy traffic pe Postgres better.

Example

SQLALCHEMY_DATABASE_URI = 'sqlite:///app.db'

SimpleVirtualenv isolated Python env — project deps system Python ko nahi bigaadte.

Example

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

SimpleVue = UI ka smart assistant. Data badlo, screen khud update. MAD1 Jinja har click pe naya HTML.

Progressive frontend framework. Components, reactivity, directives.

MAD2 SPA: ek index.html, Vue Router pages, Flask sirf JSON.

Fayde: easy, SFC, docs. Vs React: templates HTML-like, course sikhata hai. Vs plain JS: baar-baar DOM mat chhedo.

Example

<!-- HelloWorld.vue -->
<template>
  <h1>{{ msg }}</h1>
</template>
<script>
export default { data() { return { msg: 'Hello' } } }
</script>

Project example: Navbar.vue + views (Login, Dashboard). data() / setup() reactive. fetch APIs.

Viva tipFayde: easy, SFC, reactivity. Vs React: chhota learning curve. Vs plain JS: UI state auto-update.

SimpleReactive: data change → UI auto. Vue/React. jQuery manual.

Why Vue. Lose reactivity Vue2 new field. Vue3 proxy better.

Pehle 1-line definition, phir related MAD2 layer (Vue SFC / Flask route / models.py / tasks.py / Redis).

4-line formula: input kahan se aaya, auth/validation kaunsa, DB ya cache ya Celery, kya return.

Related mix mat karo — cache ≠ Celery, authentication ≠ authorization, v-if ≠ v-show, Beat ≠ Worker. Follow-up ke liye ek difference ready rakho.

Project example: Change data, list updates.

SimpleMemoize = function(args) ka answer copy. Caching broader (HTTP, Redis, page).

Fibonacci textbook. Flask @cache.memoize. cached whole view.

Redis in-memory key-value store hai — fridge pe sticky note, poori kitchen (SQL database) har baar mat kholo.

MAD2 mein Redis do kaam: (1) cache — same venues/shows/dashboard JSON baar-baar DB se nahi (2) Celery broker — Flask task queue mein daalta hai, worker uthata hai.

Flow: cache.get(key) → hit pe return, miss pe SQLAlchemy + SET with timeout. Admin write ke baad cache.delete, warna stale list.

Source of truth database hi hai. Redis crash/band = cache miss (app chalni chahiye) + queue ruk sakti hai. Important bookings sirf Redis mein mat rakho.

Alag: Redis ≠ Memcached (Celery broker nahi). Redis ≠ localStorage (woh browser). redis-cli ping → PONG viva proof.

Project example: get_user(id) memoize. venues list cached.

SimplePrecomputed hash→password tables. Unique salt per user se useless. Slow hash.

Why hashing + salt viva combo.

Pehle 1-line definition, phir related MAD2 layer (Vue SFC / Flask route / models.py / tasks.py / Redis).

4-line formula: input kahan se aaya, auth/validation kaunsa, DB ya cache ya Celery, kya return.

Related mix mat karo — cache ≠ Celery, authentication ≠ authorization, v-if ≠ v-show, Beat ≠ Worker. Follow-up ke liye ek difference ready rakho.

Project example: Werkzeug salts included in hash string.

Simplev-cloak Vue compile hone tak raw {{ }} hide karta hai. CSS: [v-cloak] { display: none; }

SimpleCode explain formula: input kya aaya → auth/validation → DB change → response/UI.

Viva tipFunction signature + 3-4 important lines + edge case.

SimpleMemoize args-aware. cached view/URL. Prefixes differ.

Same Redis. Different key recipe.

Redis in-memory key-value store hai — fridge pe sticky note, poori kitchen (SQL database) har baar mat kholo.

MAD2 mein Redis do kaam: (1) cache — same venues/shows/dashboard JSON baar-baar DB se nahi (2) Celery broker — Flask task queue mein daalta hai, worker uthata hai.

Flow: cache.get(key) → hit pe return, miss pe SQLAlchemy + SET with timeout. Admin write ke baad cache.delete, warna stale list.

Source of truth database hi hai. Redis crash/band = cache miss (app chalni chahiye) + queue ruk sakti hai. Important bookings sirf Redis mein mat rakho.

Alag: Redis ≠ Memcached (Celery broker nahi). Redis ≠ localStorage (woh browser). redis-cli ping → PONG viva proof.

Project example: Two decorators two files explain.

SimpleCelery background tasks (email, CSV, reports) async chalaata hai. Redis broker/queue + cache.

User 'Export' click → Celery task queue → Redis message → worker CSV banaata hai → user baad mein download.
Saath isliye: Celery ko broker chahiye, Redis tez in-memory. MAD2 topic zyada, MAD1 mein optional.
Priority: task queues/routing. Caching: Redis mein key-value, DB hits kam.
Pub/Sub: publisher channel pe message, subscribers sunte hain — Redis yeh bhi karta hai.

SimpleMAD1 mein common: Werkzeug generate_password_hash — default usually pbkdf2:sha256. Apna code dikhao.

Example

from werkzeug.security import generate_password_hash
print(generate_password_hash('secret'))  # pbkdf2:sha256:...

SimpleMAD1 mein common: Werkzeug generate_password_hash — default usually pbkdf2:sha256. Apna code dikhao.

Example

from werkzeug.security import generate_password_hash
print(generate_password_hash('secret'))  # pbkdf2:sha256:...

SimpleAJAX page reload ke bina server se data laata hai. Aaj kal fetch() + Promises.

fetch('/api/users').then(r => r.json()).then(data => { ... })
Promise: future value (pending/fulfilled/rejected). fetch Promise return karta hai.
SPA (Vue): ek HTML, baaki routes JS se, partial update. MAD1 generally full page render (Jinja) — submit pe poora refresh.
async/await Promise ka clean syntax.

SimplePehle 1-line definition, phir apne MAD1 project mein file dikhao, phir chhota example.

Suggested flow1) Concept kya hai?
2) Mere code mein kahan?
3) Example / demo
4) Edge case

Example

# related file: models.py / routes / template
# formula: request → check/auth → DB → render/redirect

Viva tipExact line yaad nahi to honestly related part dikhao. Bluff mat karo.

SimpleCache = pehli baar hisaab karke copy rakh lo, doosri baar copy dikha do.

Baar-baar same expensive kaam (DB query, counting bookings) mat karo. Result Redis jaise store mein timeout ke saath rakh do.

Cache hit: key mil gayi, DB skip. Cache miss: DB se lao, SET karo, return karo.

Tradeoff: speed vs stale data. Admin naya venue add kare aur cache 50s ka ho to user purani list dekh sakta hai — isliye write pe invalidation zaroori hai.

Browser cache, Flask-Caching, Redis alag layers hain. MAD2 examiner Redis + @cache.cached code dekhta hai.

Example

@cache.cached(timeout=50, query_string=True)
@app.route('/api/shows')
def shows():
    return jsonify([s.serialize() for s in Show.query.all()])

Project example: Admin dashboard counts / venues list cache. Create/update/delete ke baad cache.delete('venues').

Viva tipprint('DB queried') function ke andar laga ke miss/hit demo karo — hit pe print nahi chalega.

SimpleImprove ideas: caching, better validation, tests, pagination, email reminders, accessibility, mobile UI, logging.

Viva tip1-2 concrete features bolo jo time milne pe add karte — vague 'AI' mat.

SimpleDemo script pehle se practice karo, 5-7 minute, saari roles.

Suggested flow

1) Register/login user 2) Admin login — CRUD 3) Approval/blacklist 4) Booking/apply 5) Search 6) Validation edge (slots 0, duplicate) 7) Logout
Bolte-bolte dikhao: 'Yahan overbooking rokta hoon'. Code tab kholna jab poochhein.
Crash ho to debug calmly: terminal error, typo, DB path.
Mandatory features skip mat karna — examiner list tick karta hai.

SimpleDemo script pehle se practice karo, 5-7 minute, saari roles.

Suggested flow

1) Register/login user 2) Admin login — CRUD 3) Approval/blacklist 4) Booking/apply 5) Search 6) Validation edge (slots 0, duplicate) 7) Logout
Bolte-bolte dikhao: 'Yahan overbooking rokta hoon'. Code tab kholna jab poochhein.
Crash ho to debug calmly: terminal error, typo, DB path.
Mandatory features skip mat karna — examiner list tick karta hai.

SimpleCelery background tasks (email, CSV, reports) async chalaata hai. Redis broker/queue + cache.

User 'Export' click → Celery task queue → Redis message → worker CSV banaata hai → user baad mein download.
Saath isliye: Celery ko broker chahiye, Redis tez in-memory. MAD2 topic zyada, MAD1 mein optional.
Priority: task queues/routing. Caching: Redis mein key-value, DB hits kam.
Pub/Sub: publisher channel pe message, subscribers sunte hain — Redis yeh bhi karta hai.

SimpleDemo script pehle se practice karo, 5-7 minute, saari roles.

Suggested flow

1) Register/login user 2) Admin login — CRUD 3) Approval/blacklist 4) Booking/apply 5) Search 6) Validation edge (slots 0, duplicate) 7) Logout
Bolte-bolte dikhao: 'Yahan overbooking rokta hoon'. Code tab kholna jab poochhein.
Crash ho to debug calmly: terminal error, typo, DB path.
Mandatory features skip mat karna — examiner list tick karta hai.

SimpleAPI = Application Programming Interface — do systems predefined rules se baat karti hain. REST API HTTP + JSON use karti hai.

Controller/route HTML bhi return kar sakta hai. REST API generally JSON return karti hai, frontend alag ho (Vue/Postman).

Example

@app.route('/api/users/<int:id>')
def get_user(id):
    u = User.query.get_or_404(id)
    return jsonify(id=u.id, name=u.name)

MAD1 HTML apps mein 'API' kabhi-kabhi routes ko hi keh dete hain. Agar Vue nahi to bolo: server-rendered Flask, JSON APIs optional.
Boolean return: technically jsonify(True) ho sakta hai, lekin meaningful JSON object better. Flask view None return kare to error.
Test: Postman se GET/POST.
Parsing: request.get_json() body se, request.args query se.

SimpleCDN se Vue script tag, CLI se full project (webpack/vite, SFC).

CDN: <script src='https://unpkg.com/vue@3'></script> — setup 1 minute, MAD2 proto ke liye theek, offline toot sakta hai.
CLI (vue create / Vite): .vue files, components, router, build step. Production apps ke liye better.
Standalone/SFC components CLI/Vite mein milte hain.

Examiner code pe ungli rakh ke poochta hai — panic mat karo, 4-line formula bolo.

1) Yeh file/line kya role hai (model / route / template / config)?
2) Input kahan se aata hai (form, URL <id>, session, JSON)?
3) DB pe kya query/write?
4) Output kya (render_template / redirect / jsonify)?

Example

@app.route('/book/<int:id>', methods=['POST']) — logged-in user session se, trek id URL se, Booking add, slots--, redirect history.

Unknown line: imports (request, url_for, flash), decorators, if checks, commit — har keyword ka reason ready rakho.

SimpleORM = Object Relational Mapping. Database tables ko Python classes/objects se map karta hai, raw SQL kam likhni padti hai.

Bina ORM: cursor.execute('SELECT * FROM user WHERE id=?', [5])
ORM se: User.query.get(5) ya db.session.get(User, 5)

Example model

class User(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    email = db.Column(db.String(120), unique=True)

FaydePython mein soch sakte ho, SQL injection se protection (parameterized), relationships easy, DB switch thoda aasan.

Nuksaancomplex queries slow/abstract, seekhne ki layer extra, kabhi raw SQL phir bhi chahiye.

MAD1: SQLAlchemy (Flask-SQLAlchemy). Tables db.create_all() ya migrations se banti hain.

SimpleCamera pe clearly IITM/college ID dikhao — naam photo match.

Pehle se ID haath mein rakho. Glare mat aane do. Examiner screenshot/verify karta hai. Iske baad GitHub + project run.

Yeh live coding / notepad task hai. Chhota, correct Flask snippet likho.

from flask import Flask, render_template, request, redirect, url_for
@app.route('/do', methods=['GET','POST'])
def do():
    if request.method == 'POST':
        # request.form.get('field') ya request.args.get('q')
        # db.session.add(...); commit
        return redirect(url_for('dashboard'))
    items = Model.query.all()
    return render_template('page.html', items=items)

Jinja: {% for x in items %}<li>{{ x.name }}</li>{% endfor %}
ID URL: @app.route('/item/<int:id>')

SimpleAuthentication poochta hai "tu kaun hai?", Authorization poochta hai "tujhe yeh kaam karne ki permission hai?"

Authentication: login. Username + password check karke prove karte ho ki tum genuine user ho. Success pe session/token milta hai.
Authorization: uske baad role check. Admin hi parking lot delete kar sakta hai, normal user nahi.

Example (Flask):
# Authentication

if user and check_password_hash(user.password, form_password):
    session['user_id'] = user.id

# Authorization
if session.get('role') != 'admin':
    abort(403)  # logged in ho, lekin allowed nahi

Project example: student login = authentication. Student admin dashboard nahi khol sakta = authorization.
Validation alag cheez hai (form sahi bhara hai ya nahi, jaise age 18+).

SimpleCSS 3 tarikon se lagti hai, priority Inline > Internal > External (same specificity pe).

Inline: style='color:red' element pe. Highest priority, lekin reuse nahi. Debugging mushkil.
Internal: <style> tag HTML head mein. Us page tak limited.
External: style.css file. Large project ke liye best — ek jagah change, saari pages update.

Example same button pe teeno:
/* external */ .btn { color: blue; }
/* internal */ .btn { color: green; }
<!-- inline --> <button class='btn' style='color:red'> → red jeetega

ID (#box) class (.btn) se strong, class element (button) se strong.
!important sabko override karta hai — avoid karo.
Large project: external CSS preferred.

SimpleFlask default port 5000. Change: app.run(port=8000) ya env PORT, ya flask run --port 8000.

if __name__ == '__main__':
    app.run(debug=True, host='0.0.0.0', port=8000)

Do apps ek saath: alag ports. Occupied port: error, doosra number try.

SimpleJinja2 Flask ka template engine hai. HTML ke andar Python-jaisa data print/loop/if kar sakte ho.

Bina Jinja har user ke liye alag HTML file. Jinja se ek template + data.

{{ user.name }} print, {% for x in items %} loop, {% if %} condition, {% extends %} layout.

Example routereturn render_template('dash.html', treks=treks)

dash.html: {% for t in treks %}<li>{{ t.name }}</li>{% endfor %}

HTML files templates/ folder mein isliye kyunki Flask default wahan dhundhta hai. render_template Jinja ko run karta hai.
Boolean Jinja handle karta hai: {% if user.is_admin %}.

SimpleGitHub repo public/private policy ke hisaab, MAD-1 collaborator add, commits dikhao.

Repo → Settings → Collaborators → instructor added.
Commits: git log / GitHub History — regular commits better, ek dum 1 commit suspicious.
README + report mein AI usage likha hona chahiye agar use kiya.

SimpleChecksum file ka hash (jaise SHA256) — file change hui ya nahi verify.

Portal pe submitted ZIP ka checksum diya hota hai. Local file ka hash nikaalo, match hona chahiye.
Linux: sha256sum project.zip
Mismatch = galat file / corrupt download. Examiner integrity check karta hai ki tumhari submitted copy hi run ho rahi hai.

SimpleDemo script pehle se practice karo, 5-7 minute, saari roles.

Suggested flow

1) Register/login user 2) Admin login — CRUD 3) Approval/blacklist 4) Booking/apply 5) Search 6) Validation edge (slots 0, duplicate) 7) Logout
Bolte-bolte dikhao: 'Yahan overbooking rokta hoon'. Code tab kholna jab poochhein.
Crash ho to debug calmly: terminal error, typo, DB path.
Mandatory features skip mat karna — examiner list tick karta hai.

Yeh project-specific feature add/modify hai. 4 layers touch karo: model, route, template, validation.

1) Data chahiye to models.py mein column/relationship + DB update (ALTER / migrate / create_all).
2) Route: GET pe form/page, POST pe request.form se save, flash + redirect.
3) Template: input name=... Jinja se {{ obj.field }} display.
4) Rule: duplicate, range, role check backend pe.

Example naya field 'city':

user.city = db.Column(db.String(50))
user.city = request.form.get('city')
<input name='city'>  ...  {{ user.city }}

Simplemodels.py har table ki class hai — columns, PK/FK, relationships. Yeh Model layer hai.

Viva mein file khol ke bolo:
1) Kaunsi classes/tables hain (User, Role-specific, Booking...)
2) Har table ka kaam 1 line
3) Relationships: 1-1 / 1-M / M-M, backref, cascade, lazy
4) Constraints: unique email, nullable, default status

Example skeleton

class User(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    email = db.Column(db.String(120), unique=True, nullable=False)
    role = db.Column(db.String(20), default='user')
    bookings = db.relationship('Booking', backref='user', cascade='all, delete-orphan')

ER diagram: rectangles entities, lines relationships, PK/FK mark. Schema = yahi structure.
Tables create: db.create_all() ya migrations. SQLite file SQL viewer se dikha sakte ho.

Examiner code pe ungli rakh ke poochta hai — panic mat karo, 4-line formula bolo.

1) Yeh file/line kya role hai (model / route / template / config)?
2) Input kahan se aata hai (form, URL <id>, session, JSON)?
3) DB pe kya query/write?
4) Output kya (render_template / redirect / jsonify)?

Example

@app.route('/book/<int:id>', methods=['POST']) — logged-in user session se, trek id URL se, Booking add, slots--, redirect history.

Unknown line: imports (request, url_for, flash), decorators, if checks, commit — har keyword ka reason ready rakho.

SimpleBlueprint Flask app ko modules mein baant ta hai — auth, admin, user alag files.

admin_bp = Blueprint('admin', __name__, url_prefix='/admin')
@admin_bp.route('/dashboard')

def dash(): ...
app.register_blueprint(admin_bp)

Bina blueprint: sab routes ek app.py — chhote MAD1 ke liye theek. Alternative: multiple modules import karke app pe route attach, ya create_app factory.
Bada project: Blueprints = clean controllers.

SimpleBlueprint Flask app ko modules mein baant ta hai — auth, admin, user alag files.

admin_bp = Blueprint('admin', __name__, url_prefix='/admin')
@admin_bp.route('/dashboard')

def dash(): ...
app.register_blueprint(admin_bp)

Bina blueprint: sab routes ek app.py — chhote MAD1 ke liye theek. Alternative: multiple modules import karke app pe route attach, ya create_app factory.
Bada project: Blueprints = clean controllers.

SimpleORM = Object Relational Mapping. Database tables ko Python classes/objects se map karta hai, raw SQL kam likhni padti hai.

Bina ORM: cursor.execute('SELECT * FROM user WHERE id=?', [5])
ORM se: User.query.get(5) ya db.session.get(User, 5)

Example model

class User(db.Model):
    id = db.Column(db.Integer, primary_key=True)
    email = db.Column(db.String(120), unique=True)

FaydePython mein soch sakte ho, SQL injection se protection (parameterized), relationships easy, DB switch thoda aasan.

Nuksaancomplex queries slow/abstract, seekhne ki layer extra, kabhi raw SQL phir bhi chahiye.

MAD1: SQLAlchemy (Flask-SQLAlchemy). Tables db.create_all() ya migrations se banti hain.

SimpleCascade parent ke saath child pe automatic action (generally delete).

Company delete → uski drives bhi delete, warna orphan rows / FK error.
SQLAlchemy: db.relationship('Drive', cascade='all, delete-orphan', backref='company')
SQL: ForeignKey(..., ondelete='CASCADE')

Cascade sirf delete nahi: save-update, merge, refresh-expire, expunge, delete-orphan.
Cascade Update: parent PK change to child FK update (rare, PK mat badlo).
Parent delete jab children active hon: constraint error, ya cascade se children gayab, ya pehle children handle karo — design choice examiner poochta hai.
CSS cascading alag topic hai (styles inherit/override).

SimpleRBAC = Role-Based Access Control — permission role se (admin/staff/user), har user pe alag nahi.

user.role = 'admin'
if current_user.role != 'admin': abort(403)

Same login page, role ke hisaab dashboard. Multiple roles: roles table Many-to-Many, ya comma field (behtar normalized table).
Unapproved staff: is_approved flag, login pe check.
Flask-Security/UserMixin is_authenticated, is_active, get_id dete hain.

SimpleRBAC = Role-Based Access Control — permission role se (admin/staff/user), har user pe alag nahi.

user.role = 'admin'
if current_user.role != 'admin': abort(403)

Same login page, role ke hisaab dashboard. Multiple roles: roles table Many-to-Many, ya comma field (behtar normalized table).
Unapproved staff: is_approved flag, login pe check.
Flask-Security/UserMixin is_authenticated, is_active, get_id dete hain.

SimplePOST nayi cheez create karta hai, PUT existing cheez ko replace/update karta hai.

POST /books → naya book add. Har baar naya ID banega. Idempotent nahi — 2 baar POST = 2 records.
PUT /books/5 → id=5 wale book ko poora update. Same PUT dubara bhejo to same result (idempotent).

PUT se create bhi ho sakta hai (agar ID client decide kare), lekin REST convention: create = POST, replace = PUT, partial = PATCH.

Example

@app.route('/user', methods=['POST'])          # register naya user
@app.route('/user/<int:id>', methods=['PUT'])  # poora profile replace

HTML <form> PUT nahi bhejta, isliye Flask apps mein edit pe bhi POST use hota hai. Examiner ko yeh reason clearly bolo.

Yeh MAD-1 viva question tumhare project ke context se linked hai. Short answer is structure se do:

Pehle 1 line definition/purpose, phir tumhare project ka example, phir chhota code/flow.

Typical Flask flowBrowser request → @app.route controller → SQLAlchemy model/DB → Jinja template response.

Auth: session['user_id'] + role check. CRUD: add/commit, query.all, field update, delete.
Live change maange to: related model → route → template, phir browser refresh karke dikhao.

Example (generic create):
obj = Model(name=request.form.get('name'))

db.session.add(obj); db.session.commit()
return redirect(url_for('list_view'))

SimplePOST nayi cheez create karta hai, PUT existing cheez ko replace/update karta hai.

POST /books → naya book add. Har baar naya ID banega. Idempotent nahi — 2 baar POST = 2 records.
PUT /books/5 → id=5 wale book ko poora update. Same PUT dubara bhejo to same result (idempotent).

PUT se create bhi ho sakta hai (agar ID client decide kare), lekin REST convention: create = POST, replace = PUT, partial = PATCH.

Example

@app.route('/user', methods=['POST'])          # register naya user
@app.route('/user/<int:id>', methods=['PUT'])  # poora profile replace

HTML <form> PUT nahi bhejta, isliye Flask apps mein edit pe bhi POST use hota hai. Examiner ko yeh reason clearly bolo.

Simple 4-step: Model → DB → Form → Route → Template. Yeh sabse common live-coding hai.

1) models.py: age = db.Column(db.Integer, nullable=True)
2) DB: SQLite viewer se column, ya DB delete karke create_all, ya ALTER TABLE user ADD COLUMN age INTEGER; production mein Flask-Migrate.
3) register.html: <input type='number' name='age' min='18' required>
4) route: user.age = request.form.get('age', type=int); commit
5) dashboard: {{ user.age }} admin table mein <td>{{ u.age }}</td>

Print terminal: print('AGE', request.form.get('age'))
Phone/city same pattern, String(20). Openings on Drive: integer column + form + create_drive route.