Variant #2.
When Histogram::point_selectivity() sees that the point value of interest
falls into one bucket, it tries to guess whether the bucket has many
different (unpopular) values or a few popular values. (The number of
rows is fixed, as it's a Height-balanced histogram).
The basis for this guess is the "width" of the value range the bucket
covers. Buckets covering wider value ranges are assumed to contain
values with proportionally lower frequencies.
This is just a [brave] guesswork. For a very narrow bucket, it may
produce an estimate that's larger than total #rows in the bucket
or even in the whole table.
Remove the guesswork and replace it with basic logic: return
either the per-table average selectivity of col=const, or selectivity
of one bucket, whichever is lower.
This patch is the result of running
run-clang-tidy -fix -header-filter=.* -checks='-*,modernize-use-equals-default' .
Code style changes have been done on top. The result of this change
leads to the following improvements:
1. Binary size reduction.
* For a -DBUILD_CONFIG=mysql_release build, the binary size is reduced by
~400kb.
* A raw -DCMAKE_BUILD_TYPE=Release reduces the binary size by ~1.4kb.
2. Compiler can better understand the intent of the code, thus it leads
to more optimization possibilities. Additionally it enabled detecting
unused variables that had an empty default constructor but not marked
so explicitly.
Particular change required following this patch in sql/opt_range.cc
result_keys, an unused template class Bitmap now correctly issues
unused variable warnings.
Setting Bitmap template class constructor to default allows the compiler
to identify that there are no side-effects when instantiating the class.
Previously the compiler could not issue the warning as it assumed Bitmap
class (being a template) would not be performing a NO-OP for its default
constructor. This prevented the "unused variable warning".
Other changes:
- In test_quick_select(), assume that if table->used_stats_records is 0
then the table has 0 rows.
- Fixed prepare_simple_select() to populate table->used_stat_records
- Enusre that set_statistics_for_tables() doesn't cause used_stats_records
to be 0 when using stat_tables.
- To get blackhole to work with replication, set stats.records to 2 so
that test_quick_select() doesn't assume the table is empty.
Followup: remove this line from get_column_range_cardinality()
set_if_bigger(res, col_stats->get_avg_frequency());
and make sure it is only used with the binary histograms.
For JSON histograms, it makes the estimates unnecessarily imprecise.
Previous JSON parser was using an API which made the parsing
inefficient: the same JSON contents was parsed again and again.
Switch to using a lower-level parsing API which allows to do
parsing in an efficient way.
When loading the histogram, use table->field[N], not table->s->field[N].
When we used the latter we would corrupt the fields's default value. One
of the consequences of that would be that AUTO_INCREMENT fields would
stop working correctly.
The problem was introduced in fix for MDEV-26724. That patch has made it
possible for histogram collection to fail. In particular, it fails for
non-assigned characters.
When histogram construction fails, we also abort the computation of
COUNT(DISTINCT). When we try to use the value, we get valgrind failures.
Switched the code to abort the statistics collection in this case.
Part#3:
- make json_escape() return different errors on conversion error
and on out-of-space condition.
- Make histogram code handle conversion errors.
Factor the code that updates count, count_distinct,
count_distinct_single_occurrence into class Basic_stats_collector
Change from Histogram_builder and its descendant Histogram_builder_json
to Histogram_builder (the interface), and Histogram_binary_builder,
Histogram_json_builder.
In Histogram_json_builder, do not forget to collect the right bound
of the right-most bucket.
