Window functions such as row_number() always return a value higher than
the previously returned value for tuples in any given window partition.
Traditionally queries such as;
SELECT * FROM (
SELECT *, row_number() over (order by c) rn
FROM t
) t WHERE rn <= 10;
were executed fairly inefficiently. Neither the query planner nor the
executor knew that once rn made it to 11 that nothing further would match
the outer query's WHERE clause. It would blindly continue until all
tuples were exhausted from the subquery.
Here we implement means to make the above execute more efficiently.
This is done by way of adding a pg_proc.prosupport function to various of
the built-in window functions and adding supporting code to allow the
support function to inform the planner if the window function is
monotonically increasing, monotonically decreasing, both or neither. The
planner is then able to make use of that information and possibly allow
the executor to short-circuit execution by way of adding a "run condition"
to the WindowAgg to allow it to determine if some of its execution work
can be skipped.
This "run condition" is not like a normal filter. These run conditions
are only built using quals comparing values to monotonic window functions.
For monotonic increasing functions, quals making use of the btree
operators for <, <= and = can be used (assuming the window function column
is on the left). You can see here that once such a condition becomes false
that a monotonic increasing function could never make it subsequently true
again. For monotonically decreasing functions the >, >= and = btree
operators for the given type can be used for run conditions.
The best-case situation for this is when there is a single WindowAgg node
without a PARTITION BY clause. Here when the run condition becomes false
the WindowAgg node can simply return NULL. No more tuples will ever match
the run condition. It's a little more complex when there is a PARTITION
BY clause. In this case, we cannot return NULL as we must still process
other partitions. To speed this case up we pull tuples from the outer
plan to check if they're from the same partition and simply discard them
if they are. When we find a tuple belonging to another partition we start
processing as normal again until the run condition becomes false or we run
out of tuples to process.
When there are multiple WindowAgg nodes to evaluate then this complicates
the situation. For intermediate WindowAggs we must ensure we always
return all tuples to the calling node. Any filtering done could lead to
incorrect results in WindowAgg nodes above. For all intermediate nodes,
we can still save some work when the run condition becomes false. We've
no need to evaluate the WindowFuncs anymore. Other WindowAgg nodes cannot
reference the value of these and these tuples will not appear in the final
result anyway. The savings here are small in comparison to what can be
saved in the top-level WingowAgg, but still worthwhile.
Intermediate WindowAgg nodes never filter out tuples, but here we change
WindowAgg so that the top-level WindowAgg filters out tuples that don't
match the intermediate WindowAgg node's run condition. Such filters
appear in the "Filter" clause in EXPLAIN for the top-level WindowAgg node.
Here we add prosupport functions to allow the above to work for;
row_number(), rank(), dense_rank(), count(*) and count(expr). It appears
technically possible to do the same for min() and max(), however, it seems
unlikely to be useful enough, so that's not done here.
Bump catversion
Author: David Rowley
Reviewed-by: Andy Fan, Zhihong Yu
Discussion: https://postgr.es/m/CAApHDvqvp3At8++yF8ij06sdcoo1S_b2YoaT9D4Nf+MObzsrLQ@mail.gmail.com
These clauses allow the user to specify how data from nested paths are
joined, allowing considerable freedom in shaping the tabular output of
JSON_TABLE.
PLAN DEFAULT allows the user to specify the global strategies when
dealing with sibling or child nested paths. The is often sufficient to
achieve the necessary goal, and is considerably simpler than the full
PLAN clause, which allows the user to specify the strategy to be used
for each named nested path.
Nikita Glukhov
Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup, Erik Rijkers, Zhihong Yu,
Himanshu Upadhyaya, Daniel Gustafsson, Justin Pryzby.
Discussion: https://postgr.es/m/7e2cb85d-24cf-4abb-30a5-1a33715959bd@postgrespro.ru
This feature allows jsonb data to be treated as a table and thus used in
a FROM clause like other tabular data. Data can be selected from the
jsonb using jsonpath expressions, and hoisted out of nested structures
in the jsonb to form multiple rows, more or less like an outer join.
Nikita Glukhov
Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup, Erik Rijkers, Zhihong Yu (whose
name I previously misspelled), Himanshu Upadhyaya, Daniel Gustafsson,
Justin Pryzby.
Discussion: https://postgr.es/m/7e2cb85d-24cf-4abb-30a5-1a33715959bd@postgrespro.ru
When evaluating a query with a multi-column GROUP BY clause using sort,
the cost may be heavily dependent on the order in which the keys are
compared when building the groups. Grouping does not imply any ordering,
so we're allowed to compare the keys in arbitrary order, and a Hash Agg
leverages this. But for Group Agg, we simply compared keys in the order
as specified in the query. This commit explores alternative ordering of
the keys, trying to find a cheaper one.
In principle, we might generate grouping paths for all permutations of
the keys, and leave the rest to the optimizer. But that might get very
expensive, so we try to pick only a couple interesting orderings based
on both local and global information.
When planning the grouping path, we explore statistics (number of
distinct values, cost of the comparison function) for the keys and
reorder them to minimize comparison costs. Intuitively, it may be better
to perform more expensive comparisons (for complex data types etc.)
last, because maybe the cheaper comparisons will be enough. Similarly,
the higher the cardinality of a key, the lower the probability we’ll
need to compare more keys. The patch generates and costs various
orderings, picking the cheapest ones.
The ordering of group keys may interact with other parts of the query,
some of which may not be known while planning the grouping. E.g. there
may be an explicit ORDER BY clause, or some other ordering-dependent
operation, higher up in the query, and using the same ordering may allow
using either incremental sort or even eliminate the sort entirely.
The patch generates orderings and picks those minimizing the comparison
cost (for various pathkeys), and then adds orderings that might be
useful for operations higher up in the plan (ORDER BY, etc.). Finally,
it always keeps the ordering specified in the query, on the assumption
the user might have additional insights.
This introduces a new GUC enable_group_by_reordering, so that the
optimization may be disabled if needed.
The original patch was proposed by Teodor Sigaev, and later improved and
reworked by Dmitry Dolgov. Reviews by a number of people, including me,
Andrey Lepikhov, Claudio Freire, Ibrar Ahmed and Zhihong Yu.
Author: Dmitry Dolgov, Teodor Sigaev, Tomas Vondra
Reviewed-by: Tomas Vondra, Andrey Lepikhov, Claudio Freire, Ibrar Ahmed, Zhihong Yu
Discussion: https://postgr.es/m/7c79e6a5-8597-74e8-0671-1c39d124c9d6%40sigaev.ru
Discussion: https://postgr.es/m/CA%2Bq6zcW_4o2NC0zutLkOJPsFt80megSpX_dVRo6GK9PC-Jx_Ag%40mail.gmail.com
This Patch introduces three SQL standard JSON functions:
JSON() (incorrectly mentioned in my commit message for f4fb45d15c)
JSON_SCALAR()
JSON_SERIALIZE()
JSON() produces json values from text, bytea, json or jsonb values, and
has facilitites for handling duplicate keys.
JSON_SCALAR() produces a json value from any scalar sql value, including
json and jsonb.
JSON_SERIALIZE() produces text or bytea from input which containis or
represents json or jsonb;
For the most part these functions don't add any significant new
capabilities, but they will be of use to users wanting standard
compliant JSON handling.
Nikita Glukhov
Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup, Erik Rijkers, Zihong Yu,
Himanshu Upadhyaya, Daniel Gustafsson, Justin Pryzby.
Discussion: https://postgr.es/m/cd0bb935-0158-78a7-08b5-904886deac4b@postgrespro.ru
This introduces the SQL/JSON functions for querying JSON data using
jsonpath expressions. The functions are:
JSON_EXISTS()
JSON_QUERY()
JSON_VALUE()
All of these functions only operate on jsonb. The workaround for now is
to cast the argument to jsonb.
JSON_EXISTS() tests if the jsonpath expression applied to the jsonb
value yields any values. JSON_VALUE() must return a single value, and an
error occurs if it tries to return multiple values. JSON_QUERY() must
return a json object or array, and there are various WRAPPER options for
handling scalar or multi-value results. Both these functions have
options for handling EMPTY and ERROR conditions.
Nikita Glukhov
Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup, Erik Rijkers, Zihong Yu,
Himanshu Upadhyaya, Daniel Gustafsson, Justin Pryzby.
Discussion: https://postgr.es/m/cd0bb935-0158-78a7-08b5-904886deac4b@postgrespro.ru
This patch intrdocuces the SQL standard IS JSON predicate. It operates
on text and bytea values representing JSON as well as on the json and
jsonb types. Each test has an IS and IS NOT variant. The tests are:
IS JSON [VALUE]
IS JSON ARRAY
IS JSON OBJECT
IS JSON SCALAR
IS JSON WITH | WITHOUT UNIQUE KEYS
These are mostly self-explanatory, but note that IS JSON WITHOUT UNIQUE
KEYS is true whenever IS JSON is true, and IS JSON WITH UNIQUE KEYS is
true whenever IS JSON is true except it IS JSON OBJECT is true and there
are duplicate keys (which is never the case when applied to jsonb values).
Nikita Glukhov
Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup, Erik Rijkers, Zihong Yu,
Himanshu Upadhyaya, Daniel Gustafsson, Justin Pryzby.
Discussion: https://postgr.es/m/cd0bb935-0158-78a7-08b5-904886deac4b@postgrespro.ru
MERGE performs actions that modify rows in the target table using a
source table or query. MERGE provides a single SQL statement that can
conditionally INSERT/UPDATE/DELETE rows -- a task that would otherwise
require multiple PL statements. For example,
MERGE INTO target AS t
USING source AS s
ON t.tid = s.sid
WHEN MATCHED AND t.balance > s.delta THEN
UPDATE SET balance = t.balance - s.delta
WHEN MATCHED THEN
DELETE
WHEN NOT MATCHED AND s.delta > 0 THEN
INSERT VALUES (s.sid, s.delta)
WHEN NOT MATCHED THEN
DO NOTHING;
MERGE works with regular tables, partitioned tables and inheritance
hierarchies, including column and row security enforcement, as well as
support for row and statement triggers and transition tables therein.
MERGE is optimized for OLTP and is parameterizable, though also useful
for large scale ETL/ELT. MERGE is not intended to be used in preference
to existing single SQL commands for INSERT, UPDATE or DELETE since there
is some overhead. MERGE can be used from PL/pgSQL.
MERGE does not support targetting updatable views or foreign tables, and
RETURNING clauses are not allowed either. These limitations are likely
fixable with sufficient effort. Rewrite rules are also not supported,
but it's not clear that we'd want to support them.
Author: Pavan Deolasee <pavan.deolasee@gmail.com>
Author: Álvaro Herrera <alvherre@alvh.no-ip.org>
Author: Amit Langote <amitlangote09@gmail.com>
Author: Simon Riggs <simon.riggs@enterprisedb.com>
Reviewed-by: Peter Eisentraut <peter.eisentraut@enterprisedb.com>
Reviewed-by: Andres Freund <andres@anarazel.de> (earlier versions)
Reviewed-by: Peter Geoghegan <pg@bowt.ie> (earlier versions)
Reviewed-by: Robert Haas <robertmhaas@gmail.com> (earlier versions)
Reviewed-by: Japin Li <japinli@hotmail.com>
Reviewed-by: Justin Pryzby <pryzby@telsasoft.com>
Reviewed-by: Tomas Vondra <tomas.vondra@enterprisedb.com>
Reviewed-by: Zhihong Yu <zyu@yugabyte.com>
Discussion: https://postgr.es/m/CANP8+jKitBSrB7oTgT9CY2i1ObfOt36z0XMraQc+Xrz8QB0nXA@mail.gmail.com
Discussion: https://postgr.es/m/CAH2-WzkJdBuxj9PO=2QaO9-3h3xGbQPZ34kJH=HukRekwM-GZg@mail.gmail.com
Discussion: https://postgr.es/m/20201231134736.GA25392@alvherre.pgsql
This patch introduces the SQL/JSON standard constructors for JSON:
JSON()
JSON_ARRAY()
JSON_ARRAYAGG()
JSON_OBJECT()
JSON_OBJECTAGG()
For the most part these functions provide facilities that mimic
existing json/jsonb functions. However, they also offer some useful
additional functionality. In addition to text input, the JSON() function
accepts bytea input, which it will decode and constuct a json value from.
The other functions provide useful options for handling duplicate keys
and null values.
This series of patches will be followed by a consolidated documentation
patch.
Nikita Glukhov
Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup, Erik Rijkers, Zihong Yu,
Himanshu Upadhyaya, Daniel Gustafsson, Justin Pryzby.
Discussion: https://postgr.es/m/cd0bb935-0158-78a7-08b5-904886deac4b@postgrespro.ru
This introduces some of the building blocks used by the SQL/JSON
constructor and query functions. Specifically, it provides node
executor and grammar support for the FORMAT JSON [ENCODING foo]
clause, and values decorated with it, and for the RETURNING clause.
The following SQL/JSON patches will leverage these.
Nikita Glukhov (who probably deserves an award for perseverance).
Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup, Erik Rijkers, Zihong Yu,
Himanshu Upadhyaya, Daniel Gustafsson, Justin Pryzby.
Discussion: https://postgr.es/m/cd0bb935-0158-78a7-08b5-904886deac4b@postgrespro.ru
This introduces some of the building blocks used by the SQL/JSON
constructor and query functions. Specifically, it provides node
executor and grammar support for the FORMAT JSON [ENCODING foo]
clause, and values decorated with it, and for the RETURNING clause.
The following SQL/JSON patches will leverage these.
Nikita Glukhov (who probably deserves an award for perseverance).
Reviewers have included (in no particular order) Andres Freund, Alexander
Korotkov, Pavel Stehule, Andrew Alsup. Erik Rijkers, Zihong Yu and
Himanshu Upadhyaya.
Discussion: https://postgr.es/m/cd0bb935-0158-78a7-08b5-904886deac4b@postgrespro.ru
This adds to database objects the same version tracking that collation
objects have. There is a new pg_database column datcollversion that
stores the version, a new function
pg_database_collation_actual_version() to get the version from the
operating system, and a new subcommand ALTER DATABASE ... REFRESH
COLLATION VERSION.
This was not originally added together with pg_collation.collversion,
since originally version tracking was only supported for ICU, and ICU
on a database-level is not currently supported. But we now have
version tracking for glibc (since PG13), FreeBSD (since PG14), and
Windows (since PG13), so this is useful to have now.
Reviewed-by: Julien Rouhaud <rjuju123@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/f0ff3190-29a3-5b39-a179-fa32eee57db6%40enterprisedb.com
In logical replication mode, a WalSender is supposed to be able
to execute any regular SQL command, as well as the special
replication commands. Poor design of the replication-command
parser caused it to fail in various cases, notably:
* semicolons embedded in a command, or multiple SQL commands
sent in a single message;
* dollar-quoted literals containing odd numbers of single
or double quote marks;
* commands starting with a comment.
The basic problem here is that we're trying to run repl_scanner.l
across the entire input string even when it's not a replication
command. Since repl_scanner.l does not understand all of the
token types known to the core lexer, this is doomed to have
failure modes.
We certainly don't want to make repl_scanner.l as big as scan.l,
so instead rejigger stuff so that we only lex the first token of
a non-replication command. That will usually look like an IDENT
to repl_scanner.l, though a comment would end up getting reported
as a '-' or '/' single-character token. If the token is a replication
command keyword, we push it back and proceed normally with repl_gram.y
parsing. Otherwise, we can drop out of exec_replication_command()
without examining the rest of the string.
(It's still theoretically possible for repl_scanner.l to fail on
the first token; but that could only happen if it's an unterminated
single- or double-quoted string, in which case you'd have gotten
largely the same error from the core lexer too.)
In this way, repl_gram.y isn't involved at all in handling general
SQL commands, so we can get rid of the SQLCmd node type. (In
the back branches, we can't remove it because renumbering enum
NodeTag would be an ABI break; so just leave it sit there unused.)
I failed to resist the temptation to clean up some other sloppy
coding in repl_scanner.l while at it. The only externally-visible
behavior change from that is it now accepts \r and \f as whitespace,
same as the core lexer.
Per bug #17379 from Greg Rychlewski. Back-patch to all supported
branches.
Discussion: https://postgr.es/m/17379-6a5c6cfb3f1f5e77@postgresql.org
A new option "FOR ALL TABLES IN SCHEMA" in Create/Alter Publication allows
one or more schemas to be specified, whose tables are selected by the
publisher for sending the data to the subscriber.
The new syntax allows specifying both the tables and schemas. For example:
CREATE PUBLICATION pub1 FOR TABLE t1,t2,t3, ALL TABLES IN SCHEMA s1,s2;
OR
ALTER PUBLICATION pub1 ADD TABLE t1,t2,t3, ALL TABLES IN SCHEMA s1,s2;
A new system table "pg_publication_namespace" has been added, to maintain
the schemas that the user wants to publish through the publication.
Modified the output plugin (pgoutput) to publish the changes if the
relation is part of schema publication.
Updates pg_dump to identify and dump schema publications. Updates the \d
family of commands to display schema publications and \dRp+ variant will
now display associated schemas if any.
Author: Vignesh C, Hou Zhijie, Amit Kapila
Syntax-Suggested-by: Tom Lane, Alvaro Herrera
Reviewed-by: Greg Nancarrow, Masahiko Sawada, Hou Zhijie, Amit Kapila, Haiying Tang, Ajin Cherian, Rahila Syed, Bharath Rupireddy, Mark Dilger
Tested-by: Haiying Tang
Discussion: https://www.postgresql.org/message-id/CALDaNm0OANxuJ6RXqwZsM1MSY4s19nuH3734j4a72etDwvBETQ@mail.gmail.com
The command is supported for physical slots for now, and returns the
type of slot, its restart_lsn and its restart_tli.
This will be useful for an upcoming patch related to pg_receivewal, to
allow the tool to be able to stream from the position of a slot, rather
than the last WAL position flushed by the backend (as reported by
IDENTIFY_SYSTEM) if the archive directory is found as empty, which would
be an advantage in the case of switching to a different archive
locations with the same slot used to avoid holes in WAL segment
archives.
Author: Ronan Dunklau
Reviewed-by: Kyotaro Horiguchi, Michael Paquier, Bharath Rupireddy
Discussion: https://postgr.es/m/18708360.4lzOvYHigE@aivenronan
The Value node struct is a weird construct. It is its own node type,
but most of the time, it actually has a node type of Integer, Float,
String, or BitString. As a consequence, the struct name and the node
type don't match most of the time, and so it has to be treated
specially a lot. There doesn't seem to be any value in the special
construct. There is very little code that wants to accept all Value
variants but nothing else (and even if it did, this doesn't provide
any convenient way to check it), and most code wants either just one
particular node type (usually String), or it accepts a broader set of
node types besides just Value.
This change removes the Value struct and node type and replaces them
by separate Integer, Float, String, and BitString node types that are
proper node types and structs of their own and behave mostly like
normal node types.
Also, this removes the T_Null node tag, which was previously also a
possible variant of Value but wasn't actually used outside of the
Value contained in A_Const. Replace that by an isnull field in
A_Const.
Reviewed-by: Dagfinn Ilmari Mannsåker <ilmari@ilmari.org>
Reviewed-by: Kyotaro Horiguchi <horikyota.ntt@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/5ba6bc5b-3f95-04f2-2419-f8ddb4c046fb@enterprisedb.com
"Result Cache" was never a great name for this node, but nobody managed
to come up with another name that anyone liked enough. That was until
David Johnston mentioned "Node Memoization", which Tom Lane revised to
just "Memoize". People seem to like "Memoize", so let's do the rename.
Reviewed-by: Justin Pryzby
Discussion: https://postgr.es/m/20210708165145.GG1176@momjian.us
Backpatch-through: 14, where Result Cache was introduced
This adds support for writing CREATE FUNCTION and CREATE PROCEDURE
statements for language SQL with a function body that conforms to the
SQL standard and is portable to other implementations.
Instead of the PostgreSQL-specific AS $$ string literal $$ syntax,
this allows writing out the SQL statements making up the body
unquoted, either as a single statement:
CREATE FUNCTION add(a integer, b integer) RETURNS integer
LANGUAGE SQL
RETURN a + b;
or as a block
CREATE PROCEDURE insert_data(a integer, b integer)
LANGUAGE SQL
BEGIN ATOMIC
INSERT INTO tbl VALUES (a);
INSERT INTO tbl VALUES (b);
END;
The function body is parsed at function definition time and stored as
expression nodes in a new pg_proc column prosqlbody. So at run time,
no further parsing is required.
However, this form does not support polymorphic arguments, because
there is no more parse analysis done at call time.
Dependencies between the function and the objects it uses are fully
tracked.
A new RETURN statement is introduced. This can only be used inside
function bodies. Internally, it is treated much like a SELECT
statement.
psql needs some new intelligence to keep track of function body
boundaries so that it doesn't send off statements when it sees
semicolons that are inside a function body.
Tested-by: Jaime Casanova <jcasanov@systemguards.com.ec>
Reviewed-by: Julien Rouhaud <rjuju123@gmail.com>
Discussion: https://www.postgresql.org/message-id/flat/1c11f1eb-f00c-43b7-799d-2d44132c02d7@2ndquadrant.com
Here we add a new executor node type named "Result Cache". The planner
can include this node type in the plan to have the executor cache the
results from the inner side of parameterized nested loop joins. This
allows caching of tuples for sets of parameters so that in the event that
the node sees the same parameter values again, it can just return the
cached tuples instead of rescanning the inner side of the join all over
again. Internally, result cache uses a hash table in order to quickly
find tuples that have been previously cached.
For certain data sets, this can significantly improve the performance of
joins. The best cases for using this new node type are for join problems
where a large portion of the tuples from the inner side of the join have
no join partner on the outer side of the join. In such cases, hash join
would have to hash values that are never looked up, thus bloating the hash
table and possibly causing it to multi-batch. Merge joins would have to
skip over all of the unmatched rows. If we use a nested loop join with a
result cache, then we only cache tuples that have at least one join
partner on the outer side of the join. The benefits of using a
parameterized nested loop with a result cache increase when there are
fewer distinct values being looked up and the number of lookups of each
value is large. Also, hash probes to lookup the cache can be much faster
than the hash probe in a hash join as it's common that the result cache's
hash table is much smaller than the hash join's due to result cache only
caching useful tuples rather than all tuples from the inner side of the
join. This variation in hash probe performance is more significant when
the hash join's hash table no longer fits into the CPU's L3 cache, but the
result cache's hash table does. The apparent "random" access of hash
buckets with each hash probe can cause a poor L3 cache hit ratio for large
hash tables. Smaller hash tables generally perform better.
The hash table used for the cache limits itself to not exceeding work_mem
* hash_mem_multiplier in size. We maintain a dlist of keys for this cache
and when we're adding new tuples and realize we've exceeded the memory
budget, we evict cache entries starting with the least recently used ones
until we have enough memory to add the new tuples to the cache.
For parameterized nested loop joins, we now consider using one of these
result cache nodes in between the nested loop node and its inner node. We
determine when this might be useful based on cost, which is primarily
driven off of what the expected cache hit ratio will be. Estimating the
cache hit ratio relies on having good distinct estimates on the nested
loop's parameters.
For now, the planner will only consider using a result cache for
parameterized nested loop joins. This works for both normal joins and
also for LATERAL type joins to subqueries. It is possible to use this new
node for other uses in the future. For example, to cache results from
correlated subqueries. However, that's not done here due to some
difficulties obtaining a distinct estimation on the outer plan to
calculate the estimated cache hit ratio. Currently we plan the inner plan
before planning the outer plan so there is no good way to know if a result
cache would be useful or not since we can't estimate the number of times
the subplan will be called until the outer plan is generated.
The functionality being added here is newly introducing a dependency on
the return value of estimate_num_groups() during the join search.
Previously, during the join search, we only ever needed to perform
selectivity estimations. With this commit, we need to use
estimate_num_groups() in order to estimate what the hit ratio on the
result cache will be. In simple terms, if we expect 10 distinct values
and we expect 1000 outer rows, then we'll estimate the hit ratio to be
99%. Since cache hits are very cheap compared to scanning the underlying
nodes on the inner side of the nested loop join, then this will
significantly reduce the planner's cost for the join. However, it's
fairly easy to see here that things will go bad when estimate_num_groups()
incorrectly returns a value that's significantly lower than the actual
number of distinct values. If this happens then that may cause us to make
use of a nested loop join with a result cache instead of some other join
type, such as a merge or hash join. Our distinct estimations have been
known to be a source of trouble in the past, so the extra reliance on them
here could cause the planner to choose slower plans than it did previous
to having this feature. Distinct estimations are also fairly hard to
estimate accurately when several tables have been joined already or when a
WHERE clause filters out a set of values that are correlated to the
expressions we're estimating the number of distinct value for.
For now, the costing we perform during query planning for result caches
does put quite a bit of faith in the distinct estimations being accurate.
When these are accurate then we should generally see faster execution
times for plans containing a result cache. However, in the real world, we
may find that we need to either change the costings to put less trust in
the distinct estimations being accurate or perhaps even disable this
feature by default. There's always an element of risk when we teach the
query planner to do new tricks that it decides to use that new trick at
the wrong time and causes a regression. Users may opt to get the old
behavior by turning the feature off using the enable_resultcache GUC.
Currently, this is enabled by default. It remains to be seen if we'll
maintain that setting for the release.
Additionally, the name "Result Cache" is the best name I could think of
for this new node at the time I started writing the patch. Nobody seems
to strongly dislike the name. A few people did suggest other names but no
other name seemed to dominate in the brief discussion that there was about
names. Let's allow the beta period to see if the current name pleases
enough people. If there's some consensus on a better name, then we can
change it before the release. Please see the 2nd discussion link below
for the discussion on the "Result Cache" name.
Author: David Rowley
Reviewed-by: Andy Fan, Justin Pryzby, Zhihong Yu, Hou Zhijie
Tested-By: Konstantin Knizhnik
Discussion: https://postgr.es/m/CAApHDvrPcQyQdWERGYWx8J%2B2DLUNgXu%2BfOSbQ1UscxrunyXyrQ%40mail.gmail.com
Discussion: https://postgr.es/m/CAApHDvq=yQXr5kqhRviT2RhNKwToaWr9JAN5t+5_PzhuRJ3wvg@mail.gmail.com
This removes "Add Result Cache executor node". It seems that something
weird is going on with the tracking of cache hits and misses as
highlighted by many buildfarm animals. It's not yet clear what the
problem is as other parts of the plan indicate that the cache did work
correctly, it's just the hits and misses that were being reported as 0.
This is especially a bad time to have the buildfarm so broken, so
reverting before too many more animals go red.
Discussion: https://postgr.es/m/CAApHDvq_hydhfovm4=izgWs+C5HqEeRScjMbOgbpC-jRAeK3Yw@mail.gmail.com
Here we add a new executor node type named "Result Cache". The planner
can include this node type in the plan to have the executor cache the
results from the inner side of parameterized nested loop joins. This
allows caching of tuples for sets of parameters so that in the event that
the node sees the same parameter values again, it can just return the
cached tuples instead of rescanning the inner side of the join all over
again. Internally, result cache uses a hash table in order to quickly
find tuples that have been previously cached.
For certain data sets, this can significantly improve the performance of
joins. The best cases for using this new node type are for join problems
where a large portion of the tuples from the inner side of the join have
no join partner on the outer side of the join. In such cases, hash join
would have to hash values that are never looked up, thus bloating the hash
table and possibly causing it to multi-batch. Merge joins would have to
skip over all of the unmatched rows. If we use a nested loop join with a
result cache, then we only cache tuples that have at least one join
partner on the outer side of the join. The benefits of using a
parameterized nested loop with a result cache increase when there are
fewer distinct values being looked up and the number of lookups of each
value is large. Also, hash probes to lookup the cache can be much faster
than the hash probe in a hash join as it's common that the result cache's
hash table is much smaller than the hash join's due to result cache only
caching useful tuples rather than all tuples from the inner side of the
join. This variation in hash probe performance is more significant when
the hash join's hash table no longer fits into the CPU's L3 cache, but the
result cache's hash table does. The apparent "random" access of hash
buckets with each hash probe can cause a poor L3 cache hit ratio for large
hash tables. Smaller hash tables generally perform better.
The hash table used for the cache limits itself to not exceeding work_mem
* hash_mem_multiplier in size. We maintain a dlist of keys for this cache
and when we're adding new tuples and realize we've exceeded the memory
budget, we evict cache entries starting with the least recently used ones
until we have enough memory to add the new tuples to the cache.
For parameterized nested loop joins, we now consider using one of these
result cache nodes in between the nested loop node and its inner node. We
determine when this might be useful based on cost, which is primarily
driven off of what the expected cache hit ratio will be. Estimating the
cache hit ratio relies on having good distinct estimates on the nested
loop's parameters.
For now, the planner will only consider using a result cache for
parameterized nested loop joins. This works for both normal joins and
also for LATERAL type joins to subqueries. It is possible to use this new
node for other uses in the future. For example, to cache results from
correlated subqueries. However, that's not done here due to some
difficulties obtaining a distinct estimation on the outer plan to
calculate the estimated cache hit ratio. Currently we plan the inner plan
before planning the outer plan so there is no good way to know if a result
cache would be useful or not since we can't estimate the number of times
the subplan will be called until the outer plan is generated.
The functionality being added here is newly introducing a dependency on
the return value of estimate_num_groups() during the join search.
Previously, during the join search, we only ever needed to perform
selectivity estimations. With this commit, we need to use
estimate_num_groups() in order to estimate what the hit ratio on the
result cache will be. In simple terms, if we expect 10 distinct values
and we expect 1000 outer rows, then we'll estimate the hit ratio to be
99%. Since cache hits are very cheap compared to scanning the underlying
nodes on the inner side of the nested loop join, then this will
significantly reduce the planner's cost for the join. However, it's
fairly easy to see here that things will go bad when estimate_num_groups()
incorrectly returns a value that's significantly lower than the actual
number of distinct values. If this happens then that may cause us to make
use of a nested loop join with a result cache instead of some other join
type, such as a merge or hash join. Our distinct estimations have been
known to be a source of trouble in the past, so the extra reliance on them
here could cause the planner to choose slower plans than it did previous
to having this feature. Distinct estimations are also fairly hard to
estimate accurately when several tables have been joined already or when a
WHERE clause filters out a set of values that are correlated to the
expressions we're estimating the number of distinct value for.
For now, the costing we perform during query planning for result caches
does put quite a bit of faith in the distinct estimations being accurate.
When these are accurate then we should generally see faster execution
times for plans containing a result cache. However, in the real world, we
may find that we need to either change the costings to put less trust in
the distinct estimations being accurate or perhaps even disable this
feature by default. There's always an element of risk when we teach the
query planner to do new tricks that it decides to use that new trick at
the wrong time and causes a regression. Users may opt to get the old
behavior by turning the feature off using the enable_resultcache GUC.
Currently, this is enabled by default. It remains to be seen if we'll
maintain that setting for the release.
Additionally, the name "Result Cache" is the best name I could think of
for this new node at the time I started writing the patch. Nobody seems
to strongly dislike the name. A few people did suggest other names but no
other name seemed to dominate in the brief discussion that there was about
names. Let's allow the beta period to see if the current name pleases
enough people. If there's some consensus on a better name, then we can
change it before the release. Please see the 2nd discussion link below
for the discussion on the "Result Cache" name.
Author: David Rowley
Reviewed-by: Andy Fan, Justin Pryzby, Zhihong Yu
Tested-By: Konstantin Knizhnik
Discussion: https://postgr.es/m/CAApHDvrPcQyQdWERGYWx8J%2B2DLUNgXu%2BfOSbQ1UscxrunyXyrQ%40mail.gmail.com
Discussion: https://postgr.es/m/CAApHDvq=yQXr5kqhRviT2RhNKwToaWr9JAN5t+5_PzhuRJ3wvg@mail.gmail.com
This patch makes two closely related sets of changes:
1. For UPDATE, the subplan of the ModifyTable node now only delivers
the new values of the changed columns (i.e., the expressions computed
in the query's SET clause) plus row identity information such as CTID.
ModifyTable must re-fetch the original tuple to merge in the old
values of any unchanged columns. The core advantage of this is that
the changed columns are uniform across all tables of an inherited or
partitioned target relation, whereas the other columns might not be.
A secondary advantage, when the UPDATE involves joins, is that less
data needs to pass through the plan tree. The disadvantage of course
is an extra fetch of each tuple to be updated. However, that seems to
be very nearly free in context; even worst-case tests don't show it to
add more than a couple percent to the total query cost. At some point
it might be interesting to combine the re-fetch with the tuple access
that ModifyTable must do anyway to mark the old tuple dead; but that
would require a good deal of refactoring and it seems it wouldn't buy
all that much, so this patch doesn't attempt it.
2. For inherited UPDATE/DELETE, instead of generating a separate
subplan for each target relation, we now generate a single subplan
that is just exactly like a SELECT's plan, then stick ModifyTable
on top of that. To let ModifyTable know which target relation a
given incoming row refers to, a tableoid junk column is added to
the row identity information. This gets rid of the horrid hack
that was inheritance_planner(), eliminating O(N^2) planning cost
and memory consumption in cases where there were many unprunable
target relations.
Point 2 of course requires point 1, so that there is a uniform
definition of the non-junk columns to be returned by the subplan.
We can't insist on uniform definition of the row identity junk
columns however, if we want to keep the ability to have both
plain and foreign tables in a partitioning hierarchy. Since
it wouldn't scale very far to have every child table have its
own row identity column, this patch includes provisions to merge
similar row identity columns into one column of the subplan result.
In particular, we can merge the whole-row Vars typically used as
row identity by FDWs into one column by pretending they are type
RECORD. (It's still okay for the actual composite Datums to be
labeled with the table's rowtype OID, though.)
There is more that can be done to file down residual inefficiencies
in this patch, but it seems to be committable now.
FDW authors should note several API changes:
* The argument list for AddForeignUpdateTargets() has changed, and so
has the method it must use for adding junk columns to the query. Call
add_row_identity_var() instead of manipulating the parse tree directly.
You might want to reconsider exactly what you're adding, too.
* PlanDirectModify() must now work a little harder to find the
ForeignScan plan node; if the foreign table is part of a partitioning
hierarchy then the ForeignScan might not be the direct child of
ModifyTable. See postgres_fdw for sample code.
* To check whether a relation is a target relation, it's no
longer sufficient to compare its relid to root->parse->resultRelation.
Instead, check it against all_result_relids or leaf_result_relids,
as appropriate.
Amit Langote and Tom Lane
Discussion: https://postgr.es/m/CA+HiwqHpHdqdDn48yCEhynnniahH78rwcrv1rEX65-fsZGBOLQ@mail.gmail.com
Allow defining extended statistics on expressions, not just just on
simple column references. With this commit, expressions are supported
by all existing extended statistics kinds, improving the same types of
estimates. A simple example may look like this:
CREATE TABLE t (a int);
CREATE STATISTICS s ON mod(a,10), mod(a,20) FROM t;
ANALYZE t;
The collected statistics are useful e.g. to estimate queries with those
expressions in WHERE or GROUP BY clauses:
SELECT * FROM t WHERE mod(a,10) = 0 AND mod(a,20) = 0;
SELECT 1 FROM t GROUP BY mod(a,10), mod(a,20);
This introduces new internal statistics kind 'e' (expressions) which is
built automatically when the statistics object definition includes any
expressions. This represents single-expression statistics, as if there
was an expression index (but without the index maintenance overhead).
The statistics is stored in pg_statistics_ext_data as an array of
composite types, which is possible thanks to 79f6a942bd.
CREATE STATISTICS allows building statistics on a single expression, in
which case in which case it's not possible to specify statistics kinds.
A new system view pg_stats_ext_exprs can be used to display expression
statistics, similarly to pg_stats and pg_stats_ext views.
ALTER TABLE ... ALTER COLUMN ... TYPE now treats indexes the same way it
treats indexes, i.e. it drops and recreates the statistics. This means
all statistics are reset, and we no longer try to preserve at least the
functional dependencies. This should not be a major issue in practice,
as the functional dependencies actually rely on per-column statistics,
which were always reset anyway.
Author: Tomas Vondra
Reviewed-by: Justin Pryzby, Dean Rasheed, Zhihong Yu
Discussion: https://postgr.es/m/ad7891d2-e90c-b446-9fe2-7419143847d7%40enterprisedb.com
This adds a new executor node named TID Range Scan. The query planner
will generate paths for TID Range scans when quals are discovered on base
relations which search for ranges on the table's ctid column. These
ranges may be open at either end. For example, WHERE ctid >= '(10,0)';
will return all tuples on page 10 and over.
To support this, two new optional callback functions have been added to
table AM. scan_set_tidrange is used to set the scan range to just the
given range of TIDs. scan_getnextslot_tidrange fetches the next tuple
in the given range.
For AMs were scanning ranges of TIDs would not make sense, these functions
can be set to NULL in the TableAmRoutine. The query planner won't
generate TID Range Scan Paths in that case.
Author: Edmund Horner, David Rowley
Reviewed-by: David Rowley, Tomas Vondra, Tom Lane, Andres Freund, Zhihong Yu
Discussion: https://postgr.es/m/CAMyN-kB-nFTkF=VA_JPwFNo08S0d-Yk0F741S2B7LDmYAi8eyA@mail.gmail.com
Invent new RawParseModes that allow the core grammar to handle
pl/pgsql expressions and assignments directly, and thereby get rid
of a lot of hackery in pl/pgsql's parser. This moves a good deal
of knowledge about pl/pgsql into the core code: notably, we have to
invent a CoercionContext that matches pl/pgsql's (rather dubious)
historical behavior for assignment coercions. That's getting away
from the original idea of pl/pgsql as an arm's-length extension of
the core, but really we crossed that bridge a long time ago.
The main advantage of doing this is that we can now use the core
parser to generate FieldStore and/or SubscriptingRef nodes to handle
assignments to pl/pgsql variables that are records or arrays. That
fixes a number of cases that had never been implemented in pl/pgsql
assignment, such as nested records and array slicing, and it allows
pl/pgsql assignment to support the datatype-specific subscripting
behaviors introduced in commit c7aba7c14.
There are cosmetic benefits too: when a syntax error occurs in a
pl/pgsql expression, the error report no longer includes the confusing
"SELECT" keyword that used to get prefixed to the expression text.
Also, there seem to be some small speed gains.
Discussion: https://postgr.es/m/4165684.1607707277@sss.pgh.pa.us
When commit bd3daddaf introduced AlternativeSubPlans, I had some
ambitions towards allowing the choice of subplan to change during
execution. That has not happened, or even been thought about, in the
ensuing twelve years; so it seems like a failed experiment. So let's
rip that out and resolve the choice of subplan at the end of planning
(in setrefs.c) rather than during executor startup. This has a number
of positive benefits:
* Removal of a few hundred lines of executor code, since
AlternativeSubPlans need no longer be supported there.
* Removal of executor-startup overhead (particularly, initialization
of subplans that won't be used).
* Removal of incidental costs of having a larger plan tree, such as
tree-scanning and copying costs in the plancache; not to mention
setrefs.c's own costs of processing the discarded subplans.
* EXPLAIN no longer has to print a weird (and undocumented)
representation of an AlternativeSubPlan choice; it sees only the
subplan actually used. This should mean less confusion for users.
* Since setrefs.c knows which subexpression of a plan node it's
working on at any instant, it's possible to adjust the estimated
number of executions of the subplan based on that. For example,
we should usually estimate more executions of a qual expression
than a targetlist expression. The implementation used here is
pretty simplistic, because we don't want to expend a lot of cycles
on the issue; but it's better than ignoring the point entirely,
as the executor had to.
That last point might possibly result in shifting the choice
between hashed and non-hashed EXISTS subplans in a few cases,
but in general this patch isn't meant to change planner choices.
Since we're doing the resolution so late, it's really impossible
to change any plan choices outside the AlternativeSubPlan itself.
Patch by me; thanks to David Rowley for review.
Discussion: https://postgr.es/m/1992952.1592785225@sss.pgh.pa.us
0f5ca02f53 introduces 3 new keywords. It appears to be too much for relatively
small feature. Given now we past feature freeze, it's already late for
discussion of the new syntax. So, revert.
Discussion: https://postgr.es/m/28209.1586294824%40sss.pgh.pa.us
This commit adds following optional clause to BEGIN and START TRANSACTION
commands.
WAIT FOR LSN lsn [ TIMEOUT timeout ]
New clause pospones transaction start till given lsn is applied on standby.
This clause allows user be sure, that changes previously made on primary would
be visible on standby.
New shared memory struct is used to track awaited lsn per backend. Recovery
process wakes up backend once required lsn is applied.
Author: Ivan Kartyshov, Anna Akenteva
Reviewed-by: Craig Ringer, Thomas Munro, Robert Haas, Kyotaro Horiguchi
Reviewed-by: Masahiko Sawada, Ants Aasma, Dmitry Ivanov, Simon Riggs
Reviewed-by: Amit Kapila, Alexander Korotkov
Discussion: https://postgr.es/m/0240c26c-9f84-30ea-fca9-93ab2df5f305%40postgrespro.ru
Incremental Sort is an optimized variant of multikey sort for cases when
the input is already sorted by a prefix of the requested sort keys. For
example when the relation is already sorted by (key1, key2) and we need
to sort it by (key1, key2, key3) we can simply split the input rows into
groups having equal values in (key1, key2), and only sort/compare the
remaining column key3.
This has a number of benefits:
- Reduced memory consumption, because only a single group (determined by
values in the sorted prefix) needs to be kept in memory. This may also
eliminate the need to spill to disk.
- Lower startup cost, because Incremental Sort produce results after each
prefix group, which is beneficial for plans where startup cost matters
(like for example queries with LIMIT clause).
We consider both Sort and Incremental Sort, and decide based on costing.
The implemented algorithm operates in two different modes:
- Fetching a minimum number of tuples without check of equality on the
prefix keys, and sorting on all columns when safe.
- Fetching all tuples for a single prefix group and then sorting by
comparing only the remaining (non-prefix) keys.
We always start in the first mode, and employ a heuristic to switch into
the second mode if we believe it's beneficial - the goal is to minimize
the number of unnecessary comparions while keeping memory consumption
below work_mem.
This is a very old patch series. The idea was originally proposed by
Alexander Korotkov back in 2013, and then revived in 2017. In 2018 the
patch was taken over by James Coleman, who wrote and rewrote most of the
current code.
There were many reviewers/contributors since 2013 - I've done my best to
pick the most active ones, and listed them in this commit message.
Author: James Coleman, Alexander Korotkov
Reviewed-by: Tomas Vondra, Andreas Karlsson, Marti Raudsepp, Peter Geoghegan, Robert Haas, Thomas Munro, Antonin Houska, Andres Freund, Alexander Kuzmenkov
Discussion: https://postgr.es/m/CAPpHfdscOX5an71nHd8WSUH6GNOCf=V7wgDaTXdDd9=goN-gfA@mail.gmail.com
Discussion: https://postgr.es/m/CAPpHfds1waRZ=NOmueYq0sx1ZSCnt+5QJvizT8ndT2=etZEeAQ@mail.gmail.com
Specifically, this patch allows ALTER TYPE to:
* Change the default TOAST strategy for a toastable base type;
* Promote a non-toastable type to toastable;
* Add/remove binary I/O functions for a type;
* Add/remove typmod I/O functions for a type;
* Add/remove a custom ANALYZE statistics functions for a type.
The first of these can be done by the type's owner; all the others
require superuser privilege since misuse could cause problems.
The main motivation for this patch is to allow extensions to
upgrade the feature sets of their data types, so the set of
alterable properties is biased towards that use-case. However
it's also true that changing some other properties would be
a lot harder, as they get baked into physical storage and/or
stored expressions that depend on the type.
Along the way, refactor GenerateTypeDependencies() to make it easier
to call, refactor DefineType's volatility checks so they can be shared
by AlterType, and teach typcache.c that it might have to reload data
from the type's pg_type row, a scenario it never handled before.
Also rearrange alter_type.sgml a bit for clarity (put the
composite-type operations together).
Tomas Vondra and Tom Lane
Discussion: https://postgr.es/m/20200228004440.b23ein4qvmxnlpht@development
This follows multiple complains from Peter Geoghegan, Andres Freund and
Alvaro Herrera that this issue ought to be dug more before actually
happening, if it happens.
Discussion: https://postgr.es/m/20191226144606.GA5659@alvherre.pgsql
The following renaming is done so as source files related to index
access methods are more consistent with table access methods (the
original names used for index AMs ware too generic, and could be
confused as including features related to table AMs):
- amapi.h -> indexam.h.
- amapi.c -> indexamapi.c. Here we have an equivalent with
backend/access/table/tableamapi.c.
- amvalidate.c -> indexamvalidate.c.
- amvalidate.h -> indexamvalidate.h.
- genam.c -> indexgenam.c.
- genam.h -> indexgenam.h.
This has been discussed during the development of v12 when table AM was
worked on, but the renaming never happened.
Author: Michael Paquier
Reviewed-by: Fabien Coelho, Julien Rouhaud
Discussion: https://postgr.es/m/20191223053434.GF34339@paquier.xyz
When building statistics, we need to decide how many rows to sample and
how accurate the resulting statistics should be. Until now, it was not
possible to explicitly define statistics target for extended statistics
objects, the value was always computed from the per-attribute targets
with a fallback to the system-wide default statistics target.
That's a bit inconvenient, as it ties together the statistics target set
for per-column and extended statistics. In some cases it may be useful
to require larger sample / higher accuracy for extended statics (or the
other way around), but with this approach that's not possible.
So this commit introduces a new command, allowing to specify statistics
target for individual extended statistics objects, overriding the value
derived from per-attribute targets (and the system default).
ALTER STATISTICS stat_name SET STATISTICS target_value;
When determining statistics target for an extended statistics object we
first look at this explicitly set value. When this value is -1, we fall
back to the old formula, looking at the per-attribute targets first and
then the system default. This means the behavior is backwards compatible
with older PostgreSQL releases.
Author: Tomas Vondra
Discussion: https://postgr.es/m/20190618213357.vli3i23vpkset2xd@development
Reviewed-by: Kirk Jamison, Dean Rasheed
This introduces the concept of table access methods, i.e. CREATE
ACCESS METHOD ... TYPE TABLE and
CREATE TABLE ... USING (storage-engine).
No table access functionality is delegated to table AMs as of this
commit, that'll be done in following commits.
Subsequent commits will incrementally abstract table access
functionality to be routed through table access methods. That change
is too large to be reviewed & committed at once, so it'll be done
incrementally.
Docs will be updated at the end, as adding them incrementally would
likely make them less coherent, and definitely is a lot more work,
without a lot of benefit.
Table access methods are specified similar to index access methods,
i.e. pg_am.amhandler returns, as INTERNAL, a pointer to a struct with
callbacks. In contrast to index AMs that struct needs to live as long
as a backend, typically that's achieved by just returning a pointer to
a constant struct.
Psql's \d+ now displays a table's access method. That can be disabled
with HIDE_TABLEAM=true, which is mainly useful so regression tests can
be run against different AMs. It's quite possible that this behaviour
still needs to be fine tuned.
For now it's not allowed to set a table AM for a partitioned table, as
we've not resolved how partitions would inherit that. Disallowing
allows us to introduce, if we decide that's the way forward, such a
behaviour without a compatibility break.
Catversion bumped, to add the heap table AM and references to it.
Author: Haribabu Kommi, Andres Freund, Alvaro Herrera, Dimitri Golgov and others
Discussion:
https://postgr.es/m/20180703070645.wchpu5muyto5n647@alap3.anarazel.dehttps://postgr.es/m/20160812231527.GA690404@alvherre.pgsqlhttps://postgr.es/m/20190107235616.6lur25ph22u5u5av@alap3.anarazel.dehttps://postgr.es/m/20190304234700.w5tmhducs5wxgzls@alap3.anarazel.de
For a long time, indxpath.c has had the ability to extract derived (lossy)
index conditions from certain operators such as LIKE. For just as long,
it's been obvious that we really ought to make that capability available
to extensions. This commit finally accomplishes that, by adding another
API for planner support functions that lets them create derived index
conditions for their functions. As proof of concept, the hardwired
"special index operator" code formerly present in indxpath.c is pushed
out to planner support functions attached to LIKE and other relevant
operators.
A weak spot in this design is that an extension needs to know OIDs for
the operators, datatypes, and opfamilies involved in the transformation
it wants to make. The core-code prototypes use hard-wired OID references
but extensions don't have that option for their own operators etc. It's
usually possible to look up the required info, but that may be slow and
inconvenient. However, improving that situation is a separate task.
I want to do some additional refactorization around selfuncs.c, but
that also seems like a separate task.
Discussion: https://postgr.es/m/15193.1548028093@sss.pgh.pa.us
