Exam Review

Exam details

24 questions (100 points) with material from each of the 6 lectures
Same format as the assignments
No large problems (e.g. Scratch program, web page)
Most problems can be answered in a few sentences – NOT paragraphs!
Not allowed to get help from humans (or relatively intelligent chatbots). All other sources are yours to use!
No communication with the staff about the exam

Variables: x = 50
Boolean expressions: x < 100
Conditionals: if ... else if ... else
Loops
- for loops: repeat a set number of times
- while loops: repeat while some Boolean expression is true
Functions
- Reusable pieces of code that do a specific task
- Abstraction – don’t need to know how it was implemented to use it

Compiled languages
- text –> compiler –> machine code
- Generally better performing
Interpreted languages
- text –> interpreter –> byte code –> virtual machine
- Generally easier to work with, especially for prototyping
In reality, this is an oversimplification – nothing is black and white

Again, if you have to create one, be EXTREMELY precise
- Note: “precise” does not equal verbose
- Simple is generally better/easier

IP - Internet Protocol
- Your unique address on the internet
- IPv4 vs IPv6
  - 32-bit vs 128-bit
DHCP - Dynamic Host Configuration Protocol
- Assigns your device an IP address (ex: 10.75.234.190)
DNS - Domain Name Service
- Lookup service that can translate a domain name into an IP address
- “google.com” –> 172.217.12.132
TCP - Transmission Control Protocol
- Used to move data between routers with minimal loss
- Data is divided into discrete packets
- Uses ports to identify the type of traffic
  - http: 80
  - https: 443
  - smtp: 587
UDP - User Datagram Protocol
- Prioritizes speed at the expense of some data loss
- Good for real-time applications
HTTP - Hypertext Transfer Protocol
- Used to ask for and receive web pages
- Envelope analogy for requests
  - Need both a recipient address and a return address (headers)
  - Data “inside” the envelope
- Response codes can tell you what happened
  - 200: OK
  - 404: Not found

HTML - Hypertext Markup Language
- Markup language vs. programming language
- Tells us what to display and some basics about how to do it
- Uses tags to do this
  - Tags use angle brackets: <body>
  - Tags (usually) need to be closed: </body>
- Tags can have attributes
  - class
  - id
CSS - Cascading Style Sheets
- Provides robust control over style, layout, formatting, etc.
- Uses selectors to choose HTML elements
  - Text on its own gets HTML tags
  - . used for classes
  - # used for ids
- Many different attribute: parameter; combinations
  - text-align: center;
  - color: darkred;
JavaScript
- Makes websites more interactive and responsive
- Code runs in the client’s web browser
- Uses the Document Object Model (DOM) to interact with the HTML page
  - Tree-like hierarchy of HTML elements
Separation of concerns
- Likely better to keep different languages in different files

Combination of all languages, tools, frameworks, databases, etc. used to build an app or service

Front end
- User-facing portion of the web page or application
- For web apps: HTML, CSS, JavaScript, JS frameworks
Back end
- Server-side portion of the application
- Handles logic, interaction with front end and databases, etc.
- Python, Ruby, Java, C++, etc. and related frameworks

Programs that provide big-picture functionality allowing you to focus on the parts unique to your app/site
When using a framework, you must fit your code into theirs
Contrasts somewhat with a library
- Still someone else’s code that solves a problem
- But you get to decide how to use it
A framework uses your code; your code uses libraries
Delineation can often be a bit fuzzy

SQL
- Organizes data into tables
  - Think spreadsheet-like organization
- Relational database
  - Uses primary and foreign keys to join data across multiple tables – normalization
  - Avoids duplication of data and means changes only need to be made in one place
  - Needs additional tables to bridge multiple smaller tables
- Columns must be assigned a data type
  - Specific names depent on database software, but include some form of text, number, and date fields, among others
  - Common types
    - CHAR – fixed length, always uses full amount of space
    - VARCHAR – variable length
    - INT
    - REAL or FLOAT
    - DATETIME
    - etc.
NoSQL
- Non-tabular data storage
  - More like a document than a spreadsheet
  - Data stored in objects (example: key/value pairs in MongoDB)
Database operations (CRUD)
- Create
- Read
- Update
- Delete
Tradeoff: representation of floating point numbers
- More accuracy (numbers after the decimal), more space needed
- Less accuracy, less space needed

Platform-specific languages
- Android: Kotlin, Java
- iOS: Swift, Objective-C
- Takes advantage of all device features and is generally faster
- Need 2 separate code bases if developing for both platforms
Platform-independent applications
- Possible to make a web app that looks and feels similar to a native app
- Tools available for writing code in JavaScript, C#
- May not perform as well or look the same as native applications

Vertical scaling – get bigger, faster hardware
Horizontal scaling – get MORE hardware
- Can become complex, as servers and load balancers are needed
  - Hardware must check in with each other to make sure everything is OK – “heartbeats”
- Offers elasticity, as you can turn off a server if it isn’t needed anymore
Database scaling
- Sharding – dividing data across DBs
- Replication – multiple copies of the data
If possible, design to avoid having a single point of failure

Infrastructure (IaaS)
- Purchase time/space on someone else’s hardware
- Vitrual Machines allow multiple “computers” to run on the same hardware
Platform (PaaS)
- Abstract away much of the hardware – someone else will deal with it for you
- Allows you to focus on your application, but you have less control
Software (SaaS)
- Don’t even worry about making the software, just use it!
- Gmail, Google Docs, Office 365, etc.