This commit updates to the latest btree and rtree.
The rtree algorithm has been modified in `tidwall/rtree@v1.7`
which now keeps internal and leaf rect sorted by the min-x
coordinate. This make for much faster (up to 50%) faster
searches and replacements, but slightly slower inserts.
Because of the R-tree update, the tests needed to be updated to
account for the change in order for undeterministic WITHIN and
INTERSECTS commands.
Each MATCH is inclusive OR, thus
WITHIN fleet MATCH train* truck* BOUNDS 33 -112 34 -113
will find all trains and trucks that within the provides bounds.
The returned distance value for the kNN test was failing on a
Apple M1 machine. The test expected a hardcoded value.
amd64: 13053.885940801563
apple: 13053.885940801567
Not sure why the difference between the two cpus but I changed
the test to not compare for exact equality.
This commit allows for buffering any GeoJSON object.
For example:
INTERSECTS fleet BUFFER 1000 OBJECT {...LineString...}
This will buffer add a 1 kilometer buffer to a linesting and
search the 'fleet' collection for all objects that
intersect the buffered linestring.
This commit also allows for performing INTERSECTS with a POINT
type. Thus allowing for a polygon-over-point operation, which is
an inverted point-in-polygon.
This commit changes the logic for managing the expiration of
objects in the database.
Before: There was a server-wide hashmap that stored the
collection key, id, and expiration timestamp for all objects
that had a TTL. The hashmap was occasionally probed at 20
random positions, looking for objects that have expired. Those
expired objects were immediately deleted, and if there was 5
or more objects deleted, then the probe happened again, with
no delay. If the number of objects was less than 5 then the
there was a 1/10th of a second delay before the next probe.
Now: Rather than a server-wide hashmap, each collection has
its own ordered priority queue that stores objects with TTLs.
Rather than probing, there is a background routine that
executes every 1/10th of a second, which pops the expired
objects from the collection queues, and deletes them.
The collection/queue method is a more stable approach than
the hashmap/probing method. With probing, we can run into
major cache misses for some cases where there is wide
TTL duration, such as in the hours or days. This may cause
the system to occasionally fall behind, leaving should-be
expired objects in memory. Using a queue, there is no
cache misses, all objects that should be expired will be
right away, regardless of the TTL durations.
Fixes#616
Disabled some fence tests due to them hanging and
timing out in the Github Actions workflow, which is
something that did not happen with Travis CI.
I suspect that there's something up with Go routines
and or num of CPUs in the GH virtual environment.
Tests pass locally though
This commit fixes a case where a roaming geofence will not fire
a "faraway" event when it's supposed to.
The fix required rewriting the nearby/faraway detection logic. It
is now much more accurate and takes overall less memory, but it's
also a little slower per operation because each object proximity
is checked twice per update. Once to compare the old object's
surrounding, and once to evaulated the new object. The two lists
are then used to generate accurate "nearby" and "faraway" results.
