Closes#1714
This PR enables the use of an index as the iteration cursor for a point
or range deletion operation. Main changes:
- Use `Delete` opcode for the index that is iterating the rows - avoids
unnecessary seeking on that index, since it's already positioned
correctly
- Fix delete balancing; details below:
### current state
- a deletion may cause a btree rebalancing operation
- to get the cursor back to the right place after a rebalancing, we must
remember what the deleted key was and seek to it
- right now we are using `SeekOp::LT` to move to one slot BEFORE the
deleted key, so that if we delete rows in a loop, the following `Next()`
call will put us back into the right place
### problem
- When we delete multiple rows, we always iterate forwards. Using
`SeekOp::LT` implies backwards iteration, but it works OK for table
btrees since the cursor never remains on an internal node, because table
internal cells do not have payloads. However: this behavior is
problematic for indexes because we can effectively end up skipping
visiting a page entirely. Honestly: despite spending some debugging the
_old_ code, I still don't remember what exactly causes this to happen.
:) It's one of the `iter_dir` specific behaviors in `indexbtree_move_to`
or `get_prev_record()`, but I'm too tired to spend more time figuring it
out. I had the reason in my head before going on vacation, but it was
evicted from the cache it seems...
### solution
use `SeekOp::GE { eq_only: true }` instead and make the next call to
`Next()` a no-op instead. This has the same effect as SeekOp::LT +
next(), but without introducing bugs due to `LT` being implied backwards
iteration.
Reviewed-by: Nikita Sivukhin (@sivukhin)
Closes#2981
The `run_once()` name is just a historical accident. Furthermore, it now
started to appear elsewhere as well, so let's just call it IO::step() as we
should have from the beginning.
- Busy errors do not rollback the transaction
- Transaction takes snapshot of DB state after its first successful
access of the DB, not before its first attempt to access the DB
If upgrade from read to write transaction fails, don't roll back the
transaction. Instead restore the transaction into its original state,
which allows deferred transactions that have not read anything to
restart automatically.
Fixes#2984
Our current UPDATE/DELETE fuzz tests are coupled to the btree and do
not exercise the VDBE code paths at all, so a separate one makes sense.
This test repeats the following:
- Creates one table with n columns
- Creates (0..=n) indexes
- Executes UPDATE/DELETE statements
- Asserts that both sqlite and tursodb have the same DB state after each stmt
Using `usize` to compute file offsets caps us at ~16GB on 32-bit
systems. For example, with 4 KiB pages we can only address up to 1048576
pages; attempting the next page overflows a 32-bit usize and can wrap
the write offset, corrupting data. Switching our I/O APIs and offset
math to u64 avoids this overflow on 32-bit targets
Closes#2791
This patch brings a bunch of quality of life improvements to encryption:
1. Previously, we just let any string to be used as a key. I have
updated the `PRAGMA hexkey=''` to get the key in hex. I have also
renamed from `key`, because that will be used to get passphrase
2. Added `PRAGMA cipher` so that now users can select which cipher they
want to use (for now, either `aegis256` or `aes256gcm`)
3. We now set the encryption context when both cipher and key are set
I also updated tests to reflect this.
Reviewed-by: Preston Thorpe <preston@turso.tech>
Closes#2779
- transaction which was started with max_frame = 0 and max_frame_read_lock_index = 0
can write to the WAL and in this case it needs to read data back from WAL
- without cache spilling its hard to reproduce this issue for the turso-db now,
but I stumbled into this issue with sync-engine which do weird stuff with the WAL
which "simulates" cache spilling behaviour to some extent
Added `from_hex_string` which gets us `EncryptionKey` from a
hex string. Now we can use securely generated keys, like from openssl
$ openssl rand -hex 32
We had a bug where we were checking for duplicates in the DISTINCT
index based on both the result column count plus any ORDER BY columns
not present in the DISTINCT clause.
This is wrong, so fix it by only using the result columns for the
dedupe check.
We have to update the Transaction State before checking for the Schema
Cookie so that we can rollback the transaction later on correctly.
Closes#2535Closes#2549
This PR adds new `updates` column to the CDC table. This column holds
updated fields of the row in the following format:
```
[C boolean values where true set for changed columns]
[C values with updates where NULL is set for not-changed columns]
```
For example:
```
turso> UPDATE t SET y = 'turso', q = 'db' WHERE rowid = 1;
turso> SELECT bin_record_json_object('["x","y","z","q","x","y","z","q"]', updates) as updates FROM turso_cdc;
┌──────────────────────────────────────────────────────────────────┐
│ updates │
├──────────────────────────────────────────────────────────────────┤
│ {"x":0,"y":1,"z":0,"q":1,"x":null,"y":"turso","z":null,"q":"db"} │
└──────────────────────────────────────────────────────────────────┘
```
Also, this column works differently for `ALTER TABLE` statements where
update value for `sql` will be equal to the original `ALTER TABLE`:
```
turso> ALTER TABLE t ADD COLUMN t;
turso> SELECT bin_record_json_object('["type","name","tbl_name","rootpage","sql","type","name","tbl_name","rootpage","sql"]', updates) as updates FROM turso_cdc WHERE rowid = 2;
┌───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┐
│ updates │
├───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┤
│ {"type":0,"name":0,"tbl_name":0,"rootpage":0,"sql":1,"type":null,"name":null,"tbl_name":null,"rootpage":null,"sql":"ALTER TABLE t ADD COLUMN t;"} │
└───────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┘
```
This will help turso-db to implement logical replication which supports
both column-level updates and schema changes
Closes#2538
Use the same rusqlite version in every crate and use a bundled up-to-
date sqlite version
(the impetus for this PR is still me trying to figure out why sqlite in
the insert benchmark doesn't seem to be fsyncing, even when instructed)
Closes#2507
