lemmy/replaceable_schema.sql

-- This sets up the `r` schema, which contains things that can be safely dropped and replaced instead of being
-- changed using migrations.
--
-- Statements in this file may not create or modify things outside of the `r` schema (indicated by the `r.` prefix),
-- except for these things, which are associated with something other than a schema (usually a table):
--   * A trigger if the function name after `EXECUTE FUNCTION` is in `r` (dropping `r` drops the trigger)
--
-- The default schema is not temporarily set to `r` because it would not affect some things (such as triggers) which
-- makes it hard to tell if the rule above is being followed.
--
-- If you add something here that depends on something (such as a table) created in a new migration, then down.sql must use
-- `CASCADE` when dropping it. This doesn't need to be fixed in old migrations because the "replaceable-schema" migration
-- runs `DROP SCHEMA IF EXISTS r CASCADE` in down.sql.
BEGIN;

DROP SCHEMA IF EXISTS r CASCADE;

CREATE SCHEMA r;

-- Rank calculations
CREATE FUNCTION r.controversy_rank (upvotes numeric, downvotes numeric)
    RETURNS float
    LANGUAGE plpgsql
    IMMUTABLE PARALLEL SAFE
    AS $$
BEGIN
    IF downvotes <= 0 OR upvotes <= 0 THEN
        RETURN 0;
    ELSE
        RETURN (upvotes + downvotes) * CASE WHEN upvotes > downvotes THEN
            downvotes::float / upvotes::float
        ELSE
            upvotes::float / downvotes::float
        END;
    END IF;
END
$$;

-- This function creates statement-level triggers for all operation types. It's designed this way
-- because of these limitations:
--   * A trigger that uses transition tables can only handle 1 operation type.
--   * Transition tables must be relevant for the operation type (for example, `NEW TABLE` is
--     not allowed for a `DELETE` trigger)
--   * Transition tables are only provided to the trigger function, not to functions that it calls.
--
-- This function can only be called once per table. The command to run is given as the 2nd argument
-- and has access to these tables:
--   * `old_table` with old rows
--   * `new_table` with new rows
--   * `combined_transition_tables` with both old and new rows, with 2 columns:
--       1. `-1` for old rows and `1` for new rows, which can be used with `sum` to get the number
--          to add to a count
--       2. the old or new row as a composite value
CREATE PROCEDURE r.create_triggers (table_name text, command text)
LANGUAGE plpgsql
AS $a$
DECLARE
   defs text := $b$
        CREATE FUNCTION r.thing_delete_statement ()
            RETURNS TRIGGER
            LANGUAGE plpgsql
            AS $$
        BEGIN
            PERFORM (
                WITH
                    combined_transition_tables AS (
                        select_old_table
                    ),
                    trigger_result AS command
                SELECT
                    1
            );
            RETURN NULL;
        END
        $$;
        CREATE TRIGGER delete_statement
            AFTER DELETE ON thing REFERENCING OLD TABLE AS old_table
            FOR EACH STATEMENT
            EXECUTE FUNCTION r.thing_delete_statement ();
        CREATE FUNCTION r.thing_insert_statement ()
            RETURNS TRIGGER
            LANGUAGE plpgsql
            AS $$
        BEGIN
            PERFORM (
                WITH
                    combined_transition_tables AS (
                        select_new_table
                    ),
                    trigger_result AS command
                SELECT
                    1
            );
            RETURN NULL;
        END
        $$;
        CREATE TRIGGER insert_statement
            AFTER DELETE ON thing REFERENCING NEW TABLE AS new_table
            FOR EACH STATEMENT
            EXECUTE FUNCTION r.thing_insert_statement ();
        CREATE FUNCTION r.thing_update_statement ()
            RETURNS TRIGGER
            LANGUAGE plpgsql
            AS $$
        BEGIN
            PERFORM (
                WITH
                    combined_transition_tables AS (
                        select_old_table
                        UNION ALL
                        select_new_table
                    ),
                    trigger_result AS command
                SELECT
                    1
            );
            RETURN NULL;
        END
        $$;
        CREATE TRIGGER update_statement
            AFTER UPDATE ON thing REFERENCING OLD TABLE AS old_table NEW TABLE AS new_table
            FOR EACH STATEMENT
            EXECUTE FUNCTION r.thing_update_statement ();
    $b$;
BEGIN
    defs := replace(defs, 'select_old_table', $$
        SELECT -1 AS count_diff, old_table AS thing FROM old_table
    $$);
    defs := replace(defs, 'select_new_table', $$
        SELECT 1 AS count_diff, new_table AS thing FROM new_table
    $$);
    defs := replace(defs, 'thing', table_name);
    defs := replace(defs, 'command', format('(%s)', command));
    EXECUTE defs;
END
$a$;

-- Define functions
CREATE FUNCTION r.creator_id_from_post_aggregates (agg post_aggregates)
    RETURNS int RETURN agg.creator_id;

CREATE FUNCTION r.creator_id_from_comment_aggregates (agg comment_aggregates)
    RETURNS int RETURN (
        SELECT
            creator_id
        FROM
            comment
        WHERE
            comment.id = agg.comment_id LIMIT 1
);

-- Create triggers for both post and comments
CREATE PROCEDURE r.post_or_comment (thing_type text)
LANGUAGE plpgsql
AS $a$
BEGIN
    EXECUTE replace($b$
        -- When a thing is removed, resolve its reports
        CREATE FUNCTION r.resolve_reports_when_thing_removed ( )
            RETURNS TRIGGER
            LANGUAGE plpgsql
            AS $$
            BEGIN
                UPDATE
                    thing_report
                SET
                    resolved = TRUE, resolver_id = first_removal.mod_person_id, updated = first_removal.when_ FROM (
                        SELECT
                            thing_id, min(when_) AS when_ FROM new_removal
                        WHERE
                            new_removal.removed GROUP BY thing_id) AS first_removal
                        WHERE
                            report.thing_id = first_removal.thing_id
                            AND NOT report.resolved
                            AND COALESCE(report.updated < first_removal.when_, TRUE);
                RETURN NULL;
            END $$;
    CREATE TRIGGER resolve_reports
        AFTER INSERT ON mod_remove_thing REFERENCING NEW TABLE AS new_removal
        FOR EACH STATEMENT
        EXECUTE FUNCTION r.resolve_reports_when_thing_removed ( );
        -- When a thing gets a vote, update its aggregates and its creator's aggregates
        CALL r.create_triggers ('thing_like', $$
            WITH thing_diff AS (
                    UPDATE
                        thing_aggregates AS a
                    SET
                        score = a.score + diff.upvotes - diff.downvotes,
                        upvotes = a.upvotes + diff.upvotes,
                        downvotes = a.downvotes + diff.downvotes,
                        controversy_rank = controversy_rank ((a.upvotes + diff.upvotes)::numeric, (a.downvotes + diff.downvotes)::numeric)
                    FROM (
                        SELECT
                            (thing_like).thing_id,
                            sum(count_diff) FILTER (WHERE (thing_like).score = 1) AS upvotes,
                        sum(count_diff) FILTER (WHERE (thing_like).score != 1) AS downvotes
                FROM
                    combined_transition_tables
                    AS (count_diff bigint,
                    thing_like thing_like)
                GROUP BY
                    (thing_like).thing_id) AS diff
                WHERE
                    a.thing_id = diff.thing_id
                RETURNING
                    creator_id_from_thing_aggregates (a.*) AS creator_id,
                    diff.upvotes - diff.downvotes AS score)
        UPDATE
            person_aggregates AS a
        SET
            thing_score = a.thing_score + diff.score
        FROM (
            SELECT
                creator_id,
                sum(score) AS score
            FROM
                target_diff
            GROUP BY
                creator_id) AS diff
    WHERE
        a.person_id = diff.creator_id
            $$);
END
$a$;

CALL r.post_or_comment ('post');

CALL r.post_or_comment ('comment');

-- Create triggers that update counts in parent aggregates
CALL r.create_triggers('comment', $$
    WITH comment_group AS (
        SELECT
            (comment).post_id,
            (comment).creator_id,
            (comment).local,
            sum(count_diff) AS comments
        FROM
            combined_transition_tables
            AS (count_diff bigint,
            comment comment)
        WHERE
            NOT ((comment).deleted OR (comment).removed)
        GROUP BY
            GROUPING SETS ((comment).post_id,
            (comment).creator_id,
            (comment).local)
),
unused_person_aggregates_update_result AS (
    UPDATE
        person_aggregates AS a
    SET
        comment_count = a.comment_count + comment_group.comments
    FROM
        comment_group
    WHERE
        a.person_id = comment_group.creator_id
),
unused_site_aggregates_update_result AS (
    UPDATE
        site_aggregates AS a
    SET
        comments = a.comments + comment_group.comments
    FROM
        comment_group
    WHERE
        comment_group.local
),
post_diff AS (
    UPDATE
        post_aggregates AS a
    SET
        comments = a.comments + comment_group.comments,
        newest_comment_time = GREATEST (a.newest_comment_time, (
                SELECT
                    max(published)
                FROM new_table AS new_comment
                WHERE
                    a.post_id = new_comment.post_id LIMIT 1)),
        newest_comment_time_necro = GREATEST (a.newest_comment_time_necro, (
                SELECT
                    max(published)
                FROM new_table AS new_comment
                WHERE
                    a.post_id = new_comment.post_id
                    -- Ignore comments from the post's creator
                    AND a.creator_id != new_comment.creator_id
                    -- Ignore comments on old posts
                    AND a.published > (new_comment.published - '2 days'::interval)
                LIMIT 1))
    FROM
        comment_group,
        LATERAL (
            SELECT
                *
            FROM
                post
            WHERE
                a.post_id = post.id
            LIMIT 1) AS post
    WHERE
        a.post_id = comment_group.post_id
    RETURNING
        a.community_id,
        diff.comments,
        NOT (post.deleted
            OR post.removed) AS include_in_community_aggregates)
UPDATE
    community_aggregates AS a
SET
    comments = a.comments + diff.comments
FROM (
    SELECT
        community_id,
        sum(comments)
    FROM
        post_diff
    WHERE
        post_diff.include_in_community_aggregates
    GROUP BY
        community_id) AS diff
WHERE
    a.community_id = diff.community_id
    $$);

CALL r.create_triggers('post', $$
    WITH post_group AS (
        SELECT
            (post).community_id,
            (post).creator_id,
            (post).local,
            sum(count_diff) AS posts
        FROM
            combined_post_transition_tables
            AS (count_diff bigint,
            post post)
        WHERE
            NOT ((post).deleted OR (post).removed)
        GROUP BY
            GROUPING SETS ((post).community_id,
            (post).creator_id,
            (post).local)
),
unused_person_aggregates_update_result AS (
    UPDATE
        person_aggregates AS a
    SET
        post_count = a.post_count + post_group.posts
    FROM
        post_group
    WHERE
        a.person_id = post_group.creator_id
),
unused_site_aggregates_update_result AS (
    UPDATE
        site_aggregates AS a
    SET
        posts = a.posts + post_group.posts
    FROM
        post_group
    WHERE
        post_group.local)
UPDATE
    community_aggregates AS a
SET
    posts = a.posts + post_group.posts
FROM
    post_group
WHERE
    a.community_id = post_group.community_id
    $$);

CALL r.create_triggers('community', $$
    UPDATE
        site_aggregates AS a
    SET
        communities = a.communities + diff.communities
    FROM (
        SELECT
            sum(count_diff) AS communities
        FROM
            combined_transition_tables
            AS (count_diff bigint, community community)
        WHERE (community).local AND NOT ((community).deleted OR (community).removed)) AS diff
    $$);

CALL r.create_triggers('person', $$
    UPDATE
        site_aggregates AS a
    SET
        users = a.users + diff.users
    FROM (
        SELECT
            sum(count_diff) AS users
        FROM
            combined_transition_tables
            AS (count_diff bigint, person person)
        WHERE (person).local) AS diff
    $$);

-- For community_aggregates.comments, don't include comments of deleted or removed posts
CREATE FUNCTION r.update_comment_count_from_post ()
    RETURNS TRIGGER
    LANGUAGE plpgsql
    AS $$
BEGIN
    UPDATE
        community_aggregates AS a
    SET
        comments = a.comments + diff.comments
    FROM (
        SELECT
            old_post.community_id,
            sum((
                CASE WHEN new_post.deleted
                    AND new_post.removed THEN
                    -1
                ELSE
                    1
                END) * post_aggregates.comments) AS comments
        FROM
            new_post
            INNER JOIN old_post ON new_post.id = old_post.id
                AND (new_post.deleted
                    AND new_post.removed) != (old_post.deleted
                    AND old_post.removed),
                LATERAL (
                    SELECT
                        *
                    FROM
                        post_aggregates
                WHERE
                    post_id = new_post.id
                LIMIT 1) AS post_aggregates
        GROUP BY
            old_post.community_id) AS diff
WHERE
    a.community_id = diff.community_id;
    RETURN NULL;
END
$$;

CREATE TRIGGER comment_count
    AFTER UPDATE ON post REFERENCING OLD TABLE AS old_post NEW TABLE AS new_post
    FOR EACH STATEMENT
    EXECUTE FUNCTION r.update_comment_count_from_post ();

-- Count subscribers for local communities
CALL r.create_triggers ('community_follower', $$
    UPDATE
        community_aggregates AS a
    SET
        subscriber = a.subscribers + diff.subscribers
    FROM (
        SELECT
            (community_follower).community_id,
            sum(count_diff) AS subscribers
        FROM
            combine_transition_tables
            AS (count_diff bigint, community_follower community_follower)
        WHERE (
        SELECT
            local
        FROM
            community
        WHERE
            community.id = (community_follower).community_id
        LIMIT 1)
    GROUP BY
        (community_follower).community_id) AS diff
WHERE
    a.community_id = diff.community_id
    $$);

-- These triggers create and update rows in each aggregates table to match its associated table's rows.
-- Deleting rows and updating IDs are already handled by `CASCADE` in foreign key constraints.
CREATE FUNCTION r.comment_aggregates_from_comment ()
    RETURNS TRIGGER
    LANGUAGE plpgsql
    AS $$
BEGIN
    INSERT INTO comment_aggregates (comment_id, published)
    SELECT
        id,
        published
    FROM
        new_comment;
    RETURN NULL;
END
$$;

CREATE TRIGGER aggregates
    AFTER INSERT ON comment REFERENCING NEW TABLE AS new_comment
    FOR EACH STATEMENT
    EXECUTE FUNCTION r.comment_aggregates_from_comment ();

CREATE FUNCTION r.community_aggregates_from_community ()
    RETURNS TRIGGER
    LANGUAGE plpgsql
    AS $$
BEGIN
    INSERT INTO community_aggregates (community_id, published)
    SELECT
        community_id,
        published
    FROM
        new_community;
    RETURN NULL;
END
$$;

CREATE TRIGGER aggregates
    AFTER INSERT ON community REFERENCING NEW TABLE AS new_community
    FOR EACH STATEMENT
    EXECUTE FUNCTION r.community_aggregates_from_community ();

CREATE FUNCTION r.person_aggregates_from_person ()
    RETURNS TRIGGER
    LANGUAGE plpgsql
    AS $$
BEGIN
    INSERT INTO person_aggregates (person_id)
    SELECT
        id
    FROM
        new_person;
    RETURN NULL;
END
$$;

CREATE TRIGGER aggregates
    AFTER INSERT ON person REFERENCING NEW TABLE AS new_person
    FOR EACH STATEMENT
    EXECUTE FUNCTION r.person_aggregates_from_person ();

CREATE FUNCTION r.post_aggregates_from_post ()
    RETURNS TRIGGER
    LANGUAGE plpgsql
    AS $$
BEGIN
    INSERT INTO post_aggregates (post_id, published, newest_comment_time, newest_comment_time_necro, community_id, creator_id, instance_id, featured_community, featured_local)
    SELECT
        new_post.id,
        new_post.published,
        new_post.published,
        new_post.published,
        new_post.community_id,
        new_post.creator_id,
        community.instance_id,
        new_post.featured_community,
        new_post.featured_local
    FROM
        new_post,
        LATERAL (
            SELECT
                *
            FROM
                community
            WHERE
                community.id = new_post.community_id
            LIMIT 1) AS community;
    RETURN NULL;
END
$$;

CREATE TRIGGER aggregates
    AFTER INSERT ON post REFERENCING NEW TABLE AS new_post
    FOR EACH STATEMENT
    EXECUTE FUNCTION r.post_aggregates_from_post ();

CREATE FUNCTION r.post_aggregates_from_post_update ()
    RETURNS TRIGGER
    LANGUAGE plpgsql
    AS $$
BEGIN
    UPDATE
        post_aggregates
    SET
        featured_community = new_post.featured_community,
        featured_local = new_post.featured_local
    FROM
        new_post
    WHERE
        post_aggregates.post_id = new_post.id;
    RETURN NULL;
END
$$;

CREATE TRIGGER aggregates_update
    AFTER UPDATE ON post REFERENCING NEW TABLE AS new_post
    FOR EACH STATEMENT
    EXECUTE FUNCTION r.post_aggregates_from_post_update ();

CREATE FUNCTION r.site_aggregates_from_site ()
    RETURNS TRIGGER
    LANGUAGE plpgsql
    AS $$
BEGIN
    -- we only ever want to have a single value in site_aggregate because the site_aggregate triggers update all rows in that table.
    -- a cleaner check would be to insert it for the local_site but that would break assumptions at least in the tests
    IF (NOT EXISTS (
        SELECT
            1
        FROM
            site_aggregates)) THEN
        INSERT INTO site_aggregates (site_id)
            VALUES (NEW.id);
    END IF;
    RETURN NULL;
END
$$;

CREATE TRIGGER aggregates
    AFTER INSERT ON site
    FOR EACH ROW
    EXECUTE FUNCTION r.site_aggregates_from_site ();

COMMIT;