The inspiration for this feature is I was playing around with o3-mini and had it write me a full Twitter clone: https://chatgpt.com/share/679ec506-cf10-8006-b8eb-92d74422ed90

Full code here: https://gist.github.com/simonw/da5cc6107690726794cfd59354d8c705

When I ran that it created a database with this schema:

CREATE TABLE users (
          id INTEGER PRIMARY KEY AUTOINCREMENT,
          username TEXT UNIQUE NOT NULL,
          password TEXT NOT NULL,
          created_at DATETIME DEFAULT CURRENT_TIMESTAMP
        );
CREATE TABLE sqlite_sequence(name,seq);
CREATE TABLE tweets (
          id INTEGER PRIMARY KEY AUTOINCREMENT,
          user_id INTEGER NOT NULL,
          content TEXT NOT NULL,
          created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
          FOREIGN KEY(user_id) REFERENCES users(id)
        );
CREATE TABLE follows (
          id INTEGER PRIMARY KEY AUTOINCREMENT,
          follower_id INTEGER NOT NULL,
          following_id INTEGER NOT NULL,
          UNIQUE(follower_id, following_id),
          FOREIGN KEY(follower_id) REFERENCES users(id),
          FOREIGN KEY(following_id) REFERENCES users(id)
        );

Browsing this demo database in Datasette gave me the idea: https://lite.datasette.io/?url=https%3A%2F%2Fgist.githubusercontent.com%2Fsimonw%2F476bbac3be77fd05282b2f9e205055f4%2Fraw%2F42942339c8135b6e96d1069e207a58b130b7764f%2Ftwitter.db#/twitter/tweets

Before this change:

After this change:

Uh oh!

label_column_for_table() should spot tables with a single unique text column #2458

Description

Metadata

Metadata

Assignees

Labels

Projects

Milestone

Relationships

Development

Issue actions