Sql Row & Cell

[the-moisrex]

• [ ] Lazy api calling (don't call the driver apis when you don't need to)
• [ ] As a tuple
• [ ] operator T: loop through the types and get the first column that its type matches the specified type if not, then the first type that is convertible to it
• [ ] Allocator access
• [x] column size
• [ ] column name based on index
• [ ] column iterator (a column for this query not a column for table)
• [x] operator[] get a column
• [x] remove cell iteratable
• [x] Do we need "rows" or should they be just "statements"?

cell

• [ ] .type and an enum for the valid types
• [x] .category
• [ ] is column primary key
• [ ] is not null ( or maybe inside a .flags func? )
• [x] is null
• [x] is_[type] functions
• [x] as string
• [x] as number
• [ ] as<>
• [ ] as blob ( and blob type )
• [x] operator ==
• [x] operator <=>
• [x] operator T
• [x] rename column to cell
• [ ] Maybe rename cell to field?

[the-moisrex]

Row

• rows, columns, and cells
• a data view or owner or retriever but nothing more

Statement

• represents the whole query
• we can use the "bind" function and a-likes
• can be in a state that's not executed yet

Both

• the user has to use .rows() on stmt to get the rows which executes the query first