The side of the binary expression no longer needs to be stored in
`ConstraintInfo`, since the optimizer now guarantees that it is always
on the right. As a result, only the index of the corresponding constraint
needs to be preserved.
This change extends table-valued function support by allowing arguments
to be column references, not only literals.
Virtual tables can now reject a plan by returning an error from
best_index (e.g., when a TVF argument references a table that appears
later in the join order). The planner using this information excludes
invalid plans during join order search.
This change connects virtual tables with the query optimizer.
The optimizer now considers virtual tables during join order search
and invokes their best_index callbacks to determine feasible access
paths.
Currently, this is not a visible change, since none of the existing
extensions return information indicating that a plan is invalid.
The `best_index` implementation now returns a ResultCode along with the
IndexInfo. This allows it to signal specific outcomes, such as errors or
constraint violations. This change aligns better with SQLite’s xBestIndex
contract, where cases like missing constraints or invalid combinations of
constraints must not result in a valid plan.
Previously, there were two ways to indicate that a constraint should not
be passed to the filter function: setting `argv_index` to `None` or to
a value less than 1. This was redundant, so now only `None` is used.
Additional changes:
- Update IndexInfo documentation to clarify that constraint_usages must
have exact 1:1 correspondence with input ConstraintInfo array. The code
translating constraints into VFilter arguments heavily relies on this.
- Fix best_index implementation in test extension to comply with new
validation requirements by returning usage entry for each constraint
as nilskch points out in #1807, Rust 1.88.0 is stricter about
alignment.
because rust integers default to `i32`, we were casting a pointer
to an `i32` as a pointer to an `i64` causing a panic when dereferenced
due to misalignment as rust expects it to be 8 byte aligned.
The following code reproduces the leak, with memory usage increasing
over time:
```
#[tokio::main]
async fn main() {
let db = Builder::new_local(":memory:").build().await.unwrap();
let conn = db.connect().unwrap();
conn.execute("SELECT load_extension('./target/debug/liblimbo_series');", ())
.await
.unwrap();
loop {
conn.execute("SELECT * FROM generate_series(1,10,2);", ())
.await
.unwrap();
}
}
```
After switching to the system allocator, the leak becomes detectable
with Valgrind:
```
32,000 bytes in 1,000 blocks are definitely lost in loss record 24 of 24
at 0x538580F: malloc (vg_replace_malloc.c:446)
by 0x62E15FA: alloc::alloc::alloc (alloc.rs:99)
by 0x62E172C: alloc::alloc::Global::alloc_impl (alloc.rs:192)
by 0x62E1530: allocate (alloc.rs:254)
by 0x62E1530: alloc::alloc::exchange_malloc (alloc.rs:349)
by 0x62E0271: new<limbo_series::GenerateSeriesCursor> (boxed.rs:257)
by 0x62E0271: open_GenerateSeriesVTab (lib.rs:19)
by 0x425D8FA: limbo_core::VirtualTable::open (lib.rs:732)
by 0x4285DDA: limbo_core::vdbe::execute::op_vopen (execute.rs:890)
by 0x42351E8: limbo_core::vdbe::Program::step (mod.rs:396)
by 0x425C638: limbo_core::Statement::step (lib.rs:610)
by 0x40DB238: limbo::Statement::execute::{{closure}} (lib.rs:181)
by 0x40D9EAF: limbo::Connection::execute::{{closure}} (lib.rs:109)
by 0x40D54A1: example::main::{{closure}} (example.rs:26)
```
Interestingly, when using mimalloc, neither Valgrind nor mimalloc’s
internal statistics report the leak.
I keep having 3+ PR's in at the same time and always deal with crazy
conflicts because everything in the `ext` library is together in one
file.
This PR moves each category of extension into its own file, and
separates the `vfs` functionality in Core into the `ext/dynamic` module,
so that it can be more easily separated from wasm (or non feature =
"fs") targets to prevent build issues.
The only semantic changes made in this PR is the feature gating of vfs,
the rest is simply organizing and cleaning up imports.
Was unsure if `vfs` should be a feature on the `core` side too, or to
just enable it with the `fs` feature which seemed reasonable, as that
was already the current behavior. But let me know if we want it entirely
behind it's own feature.
Reviewed-by: Pere Diaz Bou <pere-altea@homail.com>
Closes#1124
