BuildABase

School project

It should be able to do CRUD operations on all tables, and maintain constraints fx. for ID.
Only IDs of a table should be represented in other tables, if at all.
We may alter the database structure.
We could programatically create indexes for all tables, but it may not make sense for all of them, it depends on how often we expect to read versus write to it, as writing means recreating the index.

File structure

tConst: Primary key for all title files, except title_principals.
nConst: Primary key for name_basics tConst + nConst: Primary key for title_principals

For values missing in a row, like deathYear for those alive, an \N symbol is used in its place.
Some IDs are missing, and those lines are not empty, so sometimes the step size between IDs is one, other times two and yet other times maybe 4.
This inconsistency adds up, so even though the IDs start at 1 on line 1, ID 9193140 is on line 5364380 in title_basics!

Index

ByteOffset + PrimaryKey (cannot use line numbers as each row doesn't have the same byte size)

Project structure

Suggested structure for the database files, so we don't have to have them on GitHub and so our path and name of them is consistent for programming purposes?
They're all called data.tsv by default so they need to be renamed.

/database - 7 files:

1. database/name_basics.tsv
2. database/title_akas.tsv
3. database/title_basics.tsv
4. database/title_crew.tsv
5. database/title_episode.tsv
6. database/title_principals.tsv
7. database/title_ratings.tsv