List view

• Goal 4: Auto-Index

  Try to auto-create indexes on columns based on the JSON data that populates the table, and based on the searches on a table. If a table is consistently searched on its `start_date` column, then auto-add an index on that column. Same for multi-column indexes.

  No due date
  0% complete0 open0 closed

• Goal 3: Search table with JSON object

  easily search for data matching an input JSON object. figure out how to easily support wildcard text search, and < > for numeric and date columns

  No due date
  0% complete0 open0 closed

• Goal 2: validate table structure

  Given a JSON object, validate the construction of a mysql table, including sub-tables, foreign keys, and data types. Return a JSON object describing what edits should be made (new columns, change types, add NULLABLE, etc) to the existing table if any

  No due date
  0% complete0 open0 closed

• Goal 1 Notes: Create Table and Insert Data

  The idea is to take a json object, and insert auto-generate the table and/or columns necessary to insert the data into mysql. given a json object and target table name: 1. if the table doesn't exist then: - loop through the json object, collecting type information for each key value - generate a table with the table name and columns - start back at (1) 2. else, if the table exists, then: - verify that the table contains columns for all of the keys in the json object - if not, then add the appropriate columns (never delete/rename), start back at 2 - determine the unique key of the json object "id" unless otherwise specified - determine if a row with that id already exists, - if so, then update the row with the json object - otherwise insert a new row if the JSON object contains other objects as its property values, then flatten them into columns, so: `{ "foo" : { "bar1" : "baz1", "bar2" : "baz2"} }` is equivalant to: `{"foo_bar1" : "baz1", "foo_bar2" : "baz2" }` if a property contains an array for a value, then each of those objects should be added to their own mysql table, with a foreign key to the original table's row. so: `{ "foo" : [{ "bar" : "baz1"}, {"bar" : "baz2"}] }` would auto generate a table `tablename_foo`, with foreign key column: `foo_id` that points to the original foo object. `tablename_foo` would also contain the data column `bar`, and 2 rows for each `baz1` and `baz2` would be inserted.

  No due date

  •0/2 issues closed
  0% complete2 open0 closed