In case an ORDER BY column exactly matches a result column in the SELECT,
the insertion of the result column into the ORDER BY sorter can be skipped
because it's already necessarily inserted as a sorting column.
For this reason we have a mapping to know what index a given result column
has in the order by sorter.
This commit makes that mapping much simpler.
Previously, with the `index_experimental` feature enabled, the query in
the added test would enter an infinite loop. This happened because
`label_grouping_agg_step` pointed to a constant argument that was moved
to the end of the program. As a result, the aggregation loop would jump
to the constant, then return to the start of the main loop, rewind the
index, and re-enter the aggregation loop—causing it to repeat
indefinitely.
Previously, queries like:
```
SELECT
CASE WHEN c0 != 'x' THEN group_concat(c1, ',') ELSE 'x' END
FROM t0
GROUP BY c0;
```
would return incorrect results because c0 was not copied during the
aggregation loop into a register accessible to the logic processing the
grouped results (e.g., the CASE WHEN expression in this example).
The same issue applied to expressions in the HAVING and ORDER BY clauses.
Previously, the logic for collecting non-aggregate columns was duplicated
across multiple locations and implemented inconsistently. This caused a
bug that was revealed by the refactoring in this commit (see the added
test).
Again found when fuzzing nested where clause subqueries:
Aggregate registers need to be NULLed at the start because the same
registers might be reused on another invocation of a subquery, and if
they are not NULLed, the 2nd invocation of the same subquery will have
values left over from the first invocation.
Reviewed-by: Preston Thorpe (@PThorpe92)
Closes#1614
Currently we have this:
program.alloc_cursor_id(Option<String>, CursorType)`
where the String is the table's name or alias ('users' or 'u' in
the query).
This is problematic because this can happen:
`SELECT * FROM t WHERE EXISTS (SELECT * FROM t)`
There are two cursors, both with identifier 't'. This causes a bug
where the program will use the same cursor for both the main query
and the subquery, since they are keyed by 't'.
Instead introduce `CursorKey`, which is a combination of:
1. `TableInternalId`, and
2. index name (Option<String> -- in case of index cursors.
This should provide key uniqueness for cursors:
`SELECT * FROM t WHERE EXISTS (SELECT * FROM t)`
here the first 't' will have a different `TableInternalId` than the
second `t`, so there is no clash.
Currently we have some usages of LIMIT where the actual limit counter
is initialized next to the DecrJumpZero instruction, and then
`program.mark_last_insn_constant()` is used to hoist the counter
initialization to the beginning of the program.
This is very fragile, and already FROM clause subquery handling works
around this with a hack (removed in this PR), and (upcoming) WHERE clause
subqueries would also run into problems because of this, because the LIMIT
might need to be initialized once for every iteration of the subquery.
This PR removes those usages for LIMIT, and LIMIT processing is now more
intuitive:
- limit counter is now initialized at the start of the query processing
- a function init_limit() is extracted to do this for select/update/delete
Currently our "table id"/"table no"/"table idx" references always
use the direct index of the `TableReference` in the plan, e.g. in
`SelectPlan::table_references`. For example:
```rust
Expr::Column { table: 0, column: 3, .. }
```
refers to the 0'th table in the `table_references` list.
This is a fragile approach because it assumes the table_references
list is stable for the lifetime of the query processing. This has so
far been the case, but there exist certain query transformations,
e.g. subquery unnesting, that may fold new table references from
a subquery (which has its own table ref list) into the table reference
list of the parent.
If such a transformation is made, then potentially all of the Expr::Column
references to tables will become invalid. Consider this example:
```sql
-- Assume tables: users(id, age), orders(user_id, amount)
-- Get total amount spent per user on orders over $100
SELECT u.id, sub.total
FROM users u JOIN
(SELECT user_id, SUM(amount) as total
FROM orders o
WHERE o.amount > 100
GROUP BY o.user_id) sub
WHERE u.id = sub.user_id
-- Before subquery unnesting:
-- Main query table_references: [users, sub]
-- u.id refers to table 0, column 0
-- sub.total refers to table 1, column 1
--
-- Subquery table_references: [orders]
-- o.user_id refers to table 0, column 0
-- o.amount refers to table 0, column 1
--
-- After unnesting and folding subquery tables into main query,
-- the query might look like this:
SELECT u.id, SUM(o.amount) as total
FROM users u JOIN orders o ON u.id = o.user_id
WHERE o.amount > 100
GROUP BY u.id;
-- Main query table_references: [users, orders]
-- u.id refers to table index 0 (correct)
-- o.amount refers to table index 0 (incorrect, should be 1)
-- o.user_id refers to table index 0 (incorrect, should be 1)
```
We could ofc traverse every expression in the subquery and rewrite
the table indexes to be correct, but if we instead use stable identifiers
for each table reference, then all the column references will continue
to be correct.
Hence, this PR introduces a `TableInternalId` used in `TableReference`
as well as `Expr::Column` and `Expr::Rowid` so that this kind of query
transformations can happen with less pain.
Reviewable commit by commit. CI failures are not related.
Adds support for e.g. `select first_name, sum(distinct age),
count(distinct age), avg(distinct age) from users group by 1`
Implementation details:
- Creates an ephemeral index per distinct aggregate, and jumps over the
accumulation step if a duplicate is found
Closes#1507
we had an incorrect optimization in `eliminate_orderby_like_groupby()`
where it could remove e.g. the first term of the ORDER BY if it matched
the first GROUP BY term and the result set was naturally ordered by that
term. this is invalid. see e.g.:
```sql
main branch - BAD: removes the `ORDER BY id` term because the results are naturally ordered by id.
However, this results in sorting the entire thing by last name only!
limbo> select id, last_name, count(1) from users GROUP BY 1,2 order by id, last_name desc limit 3;
┌──────┬───────────┬───────────┐
│ id │ last_name │ count (1) │
├──────┼───────────┼───────────┤
│ 6235 │ Zuniga │ 1 │
├──────┼───────────┼───────────┤
│ 8043 │ Zuniga │ 1 │
├──────┼───────────┼───────────┤
│ 944 │ Zimmerman │ 1 │
└──────┴───────────┴───────────┘
after fix - GOOD:
limbo> select id, last_name, count(1) from users GROUP BY 1,2 order by id, last_name desc limit 3;
┌────┬───────────┬───────────┐
│ id │ last_name │ count (1) │
├────┼───────────┼───────────┤
│ 1 │ Foster │ 1 │
├────┼───────────┼───────────┤
│ 2 │ Salazar │ 1 │
├────┼───────────┼───────────┤
│ 3 │ Perry │ 1 │
└────┴───────────┴───────────┘
I also refactored sorters to always use the ast `SortOrder` instead of boolean vectors, and use the `compare_immutable()` utility we use inside btrees too.
Closes#1365
