SELECT statements

A SELECT statement specifies an input table using FROM [table]
- SELECT [columns] FROM [table] WHERE [condition] ORDER BY [order];
- SELECT [expression] AS [name], [expression] AS name;
- We can optionally use [column] AS [name] to rename the input column in our new table.
- Column descriptions determine how each input row is projected to a result row.
- A subset of the rows can be selected (i.e. filtered) using WHERE [condition]
- An ordering can be declared using ORDER BY [column]
To save the subset of the table we generate using select, we must concatenate the select statement to the CREATE TABLE [name] statement using AS.
- CREATE TABLE [name] AS [SELECT statement]
To Generate a table by stupid brute force:

  
CREATE TABLE [name] AS
SELECT [value] AS [column name], [value] AS [column name] UNION
SELECT [VALUE], [value] UNION
SELECT [VALUE], [value];

Joining two tables

Tables A & B are joined by a comma (or JOIN) to form a new table of all the combinations of the rows of A with the rows of B
To format these new tables to reduce ambiguity (duplicate table names) we may assign an alias to each column name.

  
SELECT 
A.column1 as a_col1,
A.column2 as a_col2,
B.column1 as b_col1,
B.column2 as b_col2,
FROM A, B;

  
INSERT INTO [table name] VALUES ([value for col1], [value for col2])

To join a table with itself, (to create combinations of our original data) we use aliases to distinguish between the same columns of the same table.

  
SELECT [alias1].[column] as [name1], [alias2].[column] as [name2],
FROM [table name] as [alias1], [table name] as [alias2],
WHERE [CONDITION]

  
SELECT parent from parents, dogs WHERE child = name AND fur = "curly"

Expressions may contain function calls and arithmetic operators
- Combine valiues: +, -, *, /, %, and, or
- Transformvalues: abs, round
- Compare values: <, <=, >, >=, =
String avlues can be combined to form longer strings