A sql lab filled with pokemon data
mpierse dd717478df Started final report преди 6 години
.idea part 3 complete преди 6 години
.DS_Store added tables преди 6 години
Queries Started final report преди 6 години
README.md Update README.md преди 6 години
pokemon_sql.zip added create statements to sql queries преди 6 години

README.md

MySQL Pokemon Report Queries

Goals

  • Become proficient selecting data out of a mysql database
  • Become comfortable performing a SQL join
  • Be able to format SQL output as a readable report

Instructions

Part 1: Importing data

Directions:

  • In intellij, connect to your local mysql instance
  • Create your pokemon schema
  • Unpack the pokemon_sql.zip files
  • One by one execute these files making sure to check your pokemon schema

From here you should have all the pokemon data in your mysql schema. Feel free to explore the data and perform a few select statements to see what the data looks like.

Part 2: Simple Selects and Counts

Directions: Write a sql query or sql queries that can answer the following questions

  • What are all the types of pokemon that a pokemon can have?
  • What is the name of the pokemon with id 45?
  • How many pokemon are there?
  • How many types are there?
  • How many pokemon have a secondary type?

Part 3: Joins and Groups

Directions: Write a sql query or sql queries that can answer the following questions

  • What is each pokemon's primary type?
  • What is Rufflet's secondary type?
  • What are the names of the pokemon that belong to the trainer with trainerID 303?
  • How many pokemon have a secondary type Poison
  • What are all the primary types and how many pokemon have that type?
  • How many pokemon at level 100 does each trainer with at least one level 100 pokemone have? (Hint: your query should not display a trainer
  • How many pokemon only belong to one trainer and no other?

Part 4: Final Report

Directions: Write a query that returns the following collumns:

Pokemon Name Trainer Name Level Primary Type Secondary Type
Pokemon's name Trainer's name Current Level Primary Type Name Secondary Type Name

Sort the data by finding out which trainer has the strongest pokemon so that this will act as a ranking of strongest to weakest trainer. You may interpret strongest in whatever way you want, but you will have to explain your decision.

Turning in this assignment

To turn in this assignment, create files for each part with at least one query for each question answered. Above each query include a comment with the question you were answering. Example:

 # How many characters are in the string 'Hello World!'
 SELECT CHAR_LENGTH('Hello World!') AS length_of_hello_world

For Part 4 specifically also leave a comment explaining how your query is deciding who the strongest trainer is

Once all of that is done, submit a pull request