* * Check that password `password` matches the hash for user of id `ID` * Done in one query: * read hash: SELECT hash FROM AUTH_TABLE WHERE id=ID LIMIT 1 */.

* * Check that a secret `secret` for some user of id `ID` matches * Done in one query: * read secret: SELECT secret FROM SECRET_TABLE WHERE id=ID LIMIT 1 */.

* * Connect to a database, given a connection string * If the connection fails a ping, this function wil panic with the err * The conenction string should look something like * user:pass@tcp(addr)/table */.

* * Create a secret for some user of id `ID` * Any existing secret for that user is destroyed * Done in one query: * update secret REPLACE INTO SECRET_TABLE (id, secret) VALUES ID, new_secret */.

* * Create a token for some user of id `ID` that expires in 24 hours * Any existing token for that user is destroyed * Done in one query: * update secret REPLACE INTO TOKEN_TABLE (id, token, created) VALUES ID, new_token, now */.

* * Delete some content of id `ID` * Uses 1 querie * delete content: DELETE FROM CONTENT_TABLE WHERE id=ID LIMIT 1 */.

* * Delete some user from USER_TABLE * Uses 1 query: * delete user: DELETE FROM USER_TABLE WHERE id=ID LIMIT 1 */.

* * Ping the database, and return any error * useful for health checks */.

* * Increment the post count of user of id `ID` by one * Done in one query * increment: UPDATE USER_TABLE SET post_count=post_count+1 WHERE id=ID */.

* * Get whether or not a user is banned, either by a permanent ban, or an expirable ban * Done in one query */.

* * Same as ReadManyContent but for some author of id `ID` * Uses 2 queries * get content: SELECT * FROM CONTENT_TABLE ORDER BY created DESC LIMIT offset, count * queries from: getManyTags */.

* * Read a slice of bans of a user * Done in one query */.

* * Read `count` number of contents, before content of id `before` * If the first set of content should be read, `before` may be empty * Newest posts are returned first * Uses 2 queries * get content: SELECT * FROM CONTENT_TABLE ORDER BY created DESC LIMIT offset, count * queries from: getManyTags */.

* * Read a slice of unresolved reports (ie, the mod queue) by order of most recent * Done in one query */.

* * Read a single ban of id `ID` * Done in one query */.

* * Read some content of id `ID` * Uses 2 queries * get content: SELECT * FROM CONTENT_TABLE WHERE id=ID LIMIT 1 * get tags: SELECT tag FROM TAG_TABLE WHERE id=ID */.

* * Lookup single report by it's ID * Done in one query */.

* * Read some user of id `ID` from USER_TABLE * Uses 1 query * read user: SELECT * FROM USER_TABLE WHERE id=ID LIMIT 1 */.

* * Read some user of email `email` from USER_TABLE * Works in the same way as ReadSingleUser, but with email * Uses 1 query * read user: SELECT * FROM USER_TABLE WHERE email=email LIMIT 1 */.

* * Read some user of email `email` from USER_TABLE * Works in the same way as ReadSingleUser, but with nick * Uses 1 query * read user: SELECT * FROM USER_TABLE WHERE nick=nick LIMIT 1 */.

* * Read information about some token `token` * Returns who it belongs to, and whether or not it's valid * done in one query: * read token: SELECT id, created FROM TOKEN_TABLE WHERE token=? LIMIT 1 */.

* * Revoke the secret of some user of id `ID` * Done in one query: * delete row: DELETE FROM SECRET_TABLE WHERE id=ID LIMIT 1 */.

* * Revoke some token `token` * Done in one query: * delete row: DELETE FROM TOKEN_TABLE WHERE token=token LIMIT 1 */.

* * Revoke the token of some user of id `ID` * Done in one query: * delete row: DELETE FROM TOKEN_TABLE WHERE id=ID LIMIT 1 */.

* * Set a password `password` for some user of id `ID` * Done in one query: * write row: REPLACE INTO AUTH_TABLE (id, hash) VALUES (ID, hash(password)) */.

* * Write some content `content` to the table CONTENT_TABLE * Uses 3 query * write content: REPLACE INTO CONTENT_TABLE (keys...) VALUES (values...) * queries from: setTags * Returns error, if any */.

* * Create or update a report for some user * Done in one query */.

* * Write some user `user` into USER_TABLE * Uses 1 query * write user: REPLACE INTO USER_TABLE (keys...) VALUES (values...) * Returns error, if any */.