mirror of
https://github.com/ClickHouse/ClickHouse.git
synced 2024-11-26 01:22:04 +00:00
Merge branch 'master' into pipes-and-creating-sets
This commit is contained in:
Nikolai Kochetov
commit
c761970bc5
193
CHANGELOG.md
193
CHANGELOG.md
@ -1,3 +1,196 @@
|
||||
## ClickHouse release 20.7
|
||||
|
||||
### ClickHouse release v20.7.2.30-stable, 2020-08-31
|
||||
|
||||
#### Backward Incompatible Change
|
||||
|
||||
* Function `modulo` (operator `%`) with at least one floating point number as argument will calculate remainder of division directly on floating point numbers without converting both arguments to integers. It makes behaviour compatible with most of DBMS. This also applicable for Date and DateTime data types. Added alias `mod`. This closes [#7323](https://github.com/ClickHouse/ClickHouse/issues/7323). [#12585](https://github.com/ClickHouse/ClickHouse/pull/12585) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Deprecate special printing of zero Date/DateTime values as `0000-00-00` and `0000-00-00 00:00:00`. [#12442](https://github.com/ClickHouse/ClickHouse/pull/12442) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* The function `groupArrayMoving*` was not working for distributed queries. It's result was calculated within incorrect data type (without promotion to the largest type). The function `groupArrayMovingAvg` was returning integer number that was inconsistent with the `avg` function. This fixes [#12568](https://github.com/ClickHouse/ClickHouse/issues/12568). [#12622](https://github.com/ClickHouse/ClickHouse/pull/12622) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Add sanity check for MergeTree settings. If the settings are incorrect, the server will refuse to start or to create a table, printing detailed explanation to the user. [#13153](https://github.com/ClickHouse/ClickHouse/pull/13153) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Protect from the cases when user may set `background_pool_size` to value lower than `number_of_free_entries_in_pool_to_execute_mutation` or `number_of_free_entries_in_pool_to_lower_max_size_of_merge`. In these cases ALTERs won't work or the maximum size of merge will be too limited. It will throw exception explaining what to do. This closes [#10897](https://github.com/ClickHouse/ClickHouse/issues/10897). [#12728](https://github.com/ClickHouse/ClickHouse/pull/12728) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
|
||||
#### New Feature
|
||||
|
||||
* Polygon dictionary type that provides efficient "reverse geocoding" lookups - to find the region by coordinates in a dictionary of many polygons (world map). It is using carefully optimized algorithm with recursive grids to maintain low CPU and memory usage. [#9278](https://github.com/ClickHouse/ClickHouse/pull/9278) ([achulkov2](https://github.com/achulkov2)).
|
||||
* Added support of LDAP authentication for preconfigured users ("Simple Bind" method). [#11234](https://github.com/ClickHouse/ClickHouse/pull/11234) ([Denis Glazachev](https://github.com/traceon)).
|
||||
* Introduce setting `alter_partition_verbose_result` which outputs information about touched parts for some types of `ALTER TABLE ... PARTITION ...` queries (currently `ATTACH` and `FREEZE`). Closes [#8076](https://github.com/ClickHouse/ClickHouse/issues/8076). [#13017](https://github.com/ClickHouse/ClickHouse/pull/13017) ([alesapin](https://github.com/alesapin)).
|
||||
* Add `bayesAB` function for bayesian-ab-testing. [#12327](https://github.com/ClickHouse/ClickHouse/pull/12327) ([achimbab](https://github.com/achimbab)).
|
||||
* Added `system.crash_log` table into which stack traces for fatal errors are collected. This table should be empty. [#12316](https://github.com/ClickHouse/ClickHouse/pull/12316) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Added http headers `X-ClickHouse-Database` and `X-ClickHouse-Format` which may be used to set default database and output format. [#12981](https://github.com/ClickHouse/ClickHouse/pull/12981) ([hcz](https://github.com/hczhcz)).
|
||||
* Add `minMap` and `maxMap` functions support to `SimpleAggregateFunction`. [#12662](https://github.com/ClickHouse/ClickHouse/pull/12662) ([Ildus Kurbangaliev](https://github.com/ildus)).
|
||||
* Add setting `allow_non_metadata_alters` which restricts to execute `ALTER` queries which modify data on disk. Disabled be default. Closes [#11547](https://github.com/ClickHouse/ClickHouse/issues/11547). [#12635](https://github.com/ClickHouse/ClickHouse/pull/12635) ([alesapin](https://github.com/alesapin)).
|
||||
* A function `formatRow` is added to support turning arbitrary expressions into a string via given format. It's useful for manipulating SQL outputs and is quite versatile combined with the `columns` function. [#12574](https://github.com/ClickHouse/ClickHouse/pull/12574) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Add `FROM_UNIXTIME` function for compatibility with MySQL, related to [12149](https://github.com/ClickHouse/ClickHouse/issues/12149). [#12484](https://github.com/ClickHouse/ClickHouse/pull/12484) ([flynn](https://github.com/ucasFL)).
|
||||
* Allow Nullable types as keys in MergeTree tables if `allow_nullable_key` table setting is enabled. https://github.com/ClickHouse/ClickHouse/issues/5319. [#12433](https://github.com/ClickHouse/ClickHouse/pull/12433) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Integration with [COS](https://intl.cloud.tencent.com/product/cos). [#12386](https://github.com/ClickHouse/ClickHouse/pull/12386) ([fastio](https://github.com/fastio)).
|
||||
* Add mapAdd and mapSubtract functions for adding/subtracting key-mapped values. [#11735](https://github.com/ClickHouse/ClickHouse/pull/11735) ([Ildus Kurbangaliev](https://github.com/ildus)).
|
||||
|
||||
#### Bug Fix
|
||||
|
||||
* Fix premature `ON CLUSTER` timeouts for queries that must be executed on a single replica. Fixes [#6704](https://github.com/ClickHouse/ClickHouse/issues/6704), [#7228](https://github.com/ClickHouse/ClickHouse/issues/7228), [#13361](https://github.com/ClickHouse/ClickHouse/issues/13361), [#11884](https://github.com/ClickHouse/ClickHouse/issues/11884). [#13450](https://github.com/ClickHouse/ClickHouse/pull/13450) ([alesapin](https://github.com/alesapin)).
|
||||
* Fix crash in mark inclusion search introduced in https://github.com/ClickHouse/ClickHouse/pull/12277. [#14225](https://github.com/ClickHouse/ClickHouse/pull/14225) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Fix race condition in external dictionaries with cache layout which can lead server crash. [#12566](https://github.com/ClickHouse/ClickHouse/pull/12566) ([alesapin](https://github.com/alesapin)).
|
||||
* Fix visible data clobbering by progress bar in client in interactive mode. This fixes [#12562](https://github.com/ClickHouse/ClickHouse/issues/12562) and [#13369](https://github.com/ClickHouse/ClickHouse/issues/13369) and [#13584](https://github.com/ClickHouse/ClickHouse/issues/13584) and fixes [#12964](https://github.com/ClickHouse/ClickHouse/issues/12964). [#13691](https://github.com/ClickHouse/ClickHouse/pull/13691) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fixed incorrect sorting order for `LowCardinality` columns when ORDER BY multiple columns is used. This fixes [#13958](https://github.com/ClickHouse/ClickHouse/issues/13958). [#14223](https://github.com/ClickHouse/ClickHouse/pull/14223) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)).
|
||||
* Removed hardcoded timeout, which wrongly overruled `query_wait_timeout_milliseconds` setting for cache-dictionary. [#14105](https://github.com/ClickHouse/ClickHouse/pull/14105) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)).
|
||||
* Fixed wrong mount point in extra info for `Poco::Exception: no space left on device`. [#14050](https://github.com/ClickHouse/ClickHouse/pull/14050) ([tavplubix](https://github.com/tavplubix)).
|
||||
* Fix wrong query optimization of select queries with `DISTINCT` keyword when subqueries also have `DISTINCT` in case `optimize_duplicate_order_by_and_distinct` setting is enabled. [#13925](https://github.com/ClickHouse/ClickHouse/pull/13925) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Fixed potential deadlock when renaming `Distributed` table. [#13922](https://github.com/ClickHouse/ClickHouse/pull/13922) ([tavplubix](https://github.com/tavplubix)).
|
||||
* Fix incorrect sorting for `FixedString` columns when ORDER BY multiple columns is used. Fixes [#13182](https://github.com/ClickHouse/ClickHouse/issues/13182). [#13887](https://github.com/ClickHouse/ClickHouse/pull/13887) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix potentially lower precision of `topK`/`topKWeighted` aggregations (with non-default parameters). [#13817](https://github.com/ClickHouse/ClickHouse/pull/13817) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fix reading from MergeTree table with INDEX of type SET fails when compared against NULL. This fixes [#13686](https://github.com/ClickHouse/ClickHouse/issues/13686). [#13793](https://github.com/ClickHouse/ClickHouse/pull/13793) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Fix step overflow in function `range()`. [#13790](https://github.com/ClickHouse/ClickHouse/pull/13790) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fixed `Directory not empty` error when concurrently executing `DROP DATABASE` and `CREATE TABLE`. [#13756](https://github.com/ClickHouse/ClickHouse/pull/13756) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Add range check for `h3KRing` function. This fixes [#13633](https://github.com/ClickHouse/ClickHouse/issues/13633). [#13752](https://github.com/ClickHouse/ClickHouse/pull/13752) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix race condition between DETACH and background merges. Parts may revive after detach. This is continuation of [#8602](https://github.com/ClickHouse/ClickHouse/issues/8602) that did not fix the issue but introduced a test that started to fail in very rare cases, demonstrating the issue. [#13746](https://github.com/ClickHouse/ClickHouse/pull/13746) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix logging Settings.Names/Values when `log_queries_min_type` greater than `QUERY_START`. [#13737](https://github.com/ClickHouse/ClickHouse/pull/13737) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fix incorrect message in `clickhouse-server.init` while checking user and group. [#13711](https://github.com/ClickHouse/ClickHouse/pull/13711) ([ylchou](https://github.com/ylchou)).
|
||||
* Do not optimize `any(arrayJoin())` to `arrayJoin()` under `optimize_move_functions_out_of_any`. [#13681](https://github.com/ClickHouse/ClickHouse/pull/13681) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fixed possible deadlock in concurrent `ALTER ... REPLACE/MOVE PARTITION ...` queries. [#13626](https://github.com/ClickHouse/ClickHouse/pull/13626) ([tavplubix](https://github.com/tavplubix)).
|
||||
* Fixed the behaviour when sometimes cache-dictionary returned default value instead of present value from source. [#13624](https://github.com/ClickHouse/ClickHouse/pull/13624) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)).
|
||||
* Fix secondary indices corruption in compact parts (compact parts is an experimental feature). [#13538](https://github.com/ClickHouse/ClickHouse/pull/13538) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Fix wrong code in function `netloc`. This fixes [#13335](https://github.com/ClickHouse/ClickHouse/issues/13335). [#13446](https://github.com/ClickHouse/ClickHouse/pull/13446) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix error in `parseDateTimeBestEffort` function when unix timestamp was passed as an argument. This fixes [#13362](https://github.com/ClickHouse/ClickHouse/issues/13362). [#13441](https://github.com/ClickHouse/ClickHouse/pull/13441) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix invalid return type for comparison of tuples with `NULL` elements. Fixes [#12461](https://github.com/ClickHouse/ClickHouse/issues/12461). [#13420](https://github.com/ClickHouse/ClickHouse/pull/13420) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix wrong optimization caused `aggregate function any(x) is found inside another aggregate function in query` error with `SET optimize_move_functions_out_of_any = 1` and aliases inside `any()`. [#13419](https://github.com/ClickHouse/ClickHouse/pull/13419) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Fix possible race in `StorageMemory`. [#13416](https://github.com/ClickHouse/ClickHouse/pull/13416) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix empty output for `Arrow` and `Parquet` formats in case if query return zero rows. It was done because empty output is not valid for this formats. [#13399](https://github.com/ClickHouse/ClickHouse/pull/13399) ([hcz](https://github.com/hczhcz)).
|
||||
* Fix select queries with constant columns and prefix of primary key in `ORDER BY` clause. [#13396](https://github.com/ClickHouse/ClickHouse/pull/13396) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Fix `PrettyCompactMonoBlock` for clickhouse-local. Fix extremes/totals with `PrettyCompactMonoBlock`. Fixes [#7746](https://github.com/ClickHouse/ClickHouse/issues/7746). [#13394](https://github.com/ClickHouse/ClickHouse/pull/13394) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fixed deadlock in system.text_log. [#12452](https://github.com/ClickHouse/ClickHouse/pull/12452) ([alexey-milovidov](https://github.com/alexey-milovidov)). It is a part of [#12339](https://github.com/ClickHouse/ClickHouse/issues/12339). This fixes [#12325](https://github.com/ClickHouse/ClickHouse/issues/12325). [#13386](https://github.com/ClickHouse/ClickHouse/pull/13386) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)).
|
||||
* Fixed `File(TSVWithNames*)` (header was written multiple times), fixed `clickhouse-local --format CSVWithNames*` (lacks header, broken after [#12197](https://github.com/ClickHouse/ClickHouse/issues/12197)), fixed `clickhouse-local --format CSVWithNames*` with zero rows (lacks header). [#13343](https://github.com/ClickHouse/ClickHouse/pull/13343) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fix segfault when function `groupArrayMovingSum` deserializes empty state. Fixes [#13339](https://github.com/ClickHouse/ClickHouse/issues/13339). [#13341](https://github.com/ClickHouse/ClickHouse/pull/13341) ([alesapin](https://github.com/alesapin)).
|
||||
* Throw error on `arrayJoin()` function in `JOIN ON` section. [#13330](https://github.com/ClickHouse/ClickHouse/pull/13330) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Fix crash in `LEFT ASOF JOIN` with `join_use_nulls=1`. [#13291](https://github.com/ClickHouse/ClickHouse/pull/13291) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Fix possible error `Totals having transform was already added to pipeline` in case of a query from delayed replica. [#13290](https://github.com/ClickHouse/ClickHouse/pull/13290) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* The server may crash if user passed specifically crafted arguments to the function `h3ToChildren`. This fixes [#13275](https://github.com/ClickHouse/ClickHouse/issues/13275). [#13277](https://github.com/ClickHouse/ClickHouse/pull/13277) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix potentially low performance and slightly incorrect result for `uniqExact`, `topK`, `sumDistinct` and similar aggregate functions called on Float types with `NaN` values. It also triggered assert in debug build. This fixes [#12491](https://github.com/ClickHouse/ClickHouse/issues/12491). [#13254](https://github.com/ClickHouse/ClickHouse/pull/13254) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix assertion in KeyCondition when primary key contains expression with monotonic function and query contains comparison with constant whose type is different. This fixes [#12465](https://github.com/ClickHouse/ClickHouse/issues/12465). [#13251](https://github.com/ClickHouse/ClickHouse/pull/13251) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Return passed number for numbers with MSB set in function roundUpToPowerOfTwoOrZero(). It prevents potential errors in case of overflow of array sizes. [#13234](https://github.com/ClickHouse/ClickHouse/pull/13234) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fix function if with nullable constexpr as cond that is not literal NULL. Fixes [#12463](https://github.com/ClickHouse/ClickHouse/issues/12463). [#13226](https://github.com/ClickHouse/ClickHouse/pull/13226) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix assert in `arrayElement` function in case of array elements are Nullable and array subscript is also Nullable. This fixes [#12172](https://github.com/ClickHouse/ClickHouse/issues/12172). [#13224](https://github.com/ClickHouse/ClickHouse/pull/13224) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix DateTime64 conversion functions with constant argument. [#13205](https://github.com/ClickHouse/ClickHouse/pull/13205) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fix parsing row policies from users.xml when names of databases or tables contain dots. This fixes https://github.com/ClickHouse/ClickHouse/issues/5779, https://github.com/ClickHouse/ClickHouse/issues/12527. [#13199](https://github.com/ClickHouse/ClickHouse/pull/13199) ([Vitaly Baranov](https://github.com/vitlibar)).
|
||||
* Fix access to `redis` dictionary after connection was dropped once. It may happen with `cache` and `direct` dictionary layouts. [#13082](https://github.com/ClickHouse/ClickHouse/pull/13082) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Fix wrong index analysis with functions. It could lead to some data parts being skipped when reading from `MergeTree` tables. Fixes [#13060](https://github.com/ClickHouse/ClickHouse/issues/13060). Fixes [#12406](https://github.com/ClickHouse/ClickHouse/issues/12406). [#13081](https://github.com/ClickHouse/ClickHouse/pull/13081) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Fix error `Cannot convert column because it is constant but values of constants are different in source and result` for remote queries which use deterministic functions in scope of query, but not deterministic between queries, like `now()`, `now64()`, `randConstant()`. Fixes [#11327](https://github.com/ClickHouse/ClickHouse/issues/11327). [#13075](https://github.com/ClickHouse/ClickHouse/pull/13075) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix crash which was possible for queries with `ORDER BY` tuple and small `LIMIT`. Fixes [#12623](https://github.com/ClickHouse/ClickHouse/issues/12623). [#13009](https://github.com/ClickHouse/ClickHouse/pull/13009) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix `Block structure mismatch` error for queries with `UNION` and `JOIN`. Fixes [#12602](https://github.com/ClickHouse/ClickHouse/issues/12602). [#12989](https://github.com/ClickHouse/ClickHouse/pull/12989) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Corrected `merge_with_ttl_timeout` logic which did not work well when expiration affected more than one partition over one time interval. (Authored by @excitoon). [#12982](https://github.com/ClickHouse/ClickHouse/pull/12982) ([Alexander Kazakov](https://github.com/Akazz)).
|
||||
* Fix columns duplication for range hashed dictionary created from DDL query. This fixes [#10605](https://github.com/ClickHouse/ClickHouse/issues/10605). [#12857](https://github.com/ClickHouse/ClickHouse/pull/12857) ([alesapin](https://github.com/alesapin)).
|
||||
* Fix unnecessary limiting for the number of threads for selects from local replica. [#12840](https://github.com/ClickHouse/ClickHouse/pull/12840) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix rare bug when `ALTER DELETE` and `ALTER MODIFY COLUMN` queries executed simultaneously as a single mutation. Bug leads to an incorrect amount of rows in `count.txt` and as a consequence incorrect data in part. Also, fix a small bug with simultaneous `ALTER RENAME COLUMN` and `ALTER ADD COLUMN`. [#12760](https://github.com/ClickHouse/ClickHouse/pull/12760) ([alesapin](https://github.com/alesapin)).
|
||||
* Wrong credentials being used when using `clickhouse` dictionary source to query remote tables. [#12756](https://github.com/ClickHouse/ClickHouse/pull/12756) ([sundyli](https://github.com/sundy-li)).
|
||||
* Fix `CAST(Nullable(String), Enum())`. [#12745](https://github.com/ClickHouse/ClickHouse/pull/12745) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fix performance with large tuples, which are interpreted as functions in `IN` section. The case when user writes `WHERE x IN tuple(1, 2, ...)` instead of `WHERE x IN (1, 2, ...)` for some obscure reason. [#12700](https://github.com/ClickHouse/ClickHouse/pull/12700) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Fix memory tracking for input_format_parallel_parsing (by attaching thread to group). [#12672](https://github.com/ClickHouse/ClickHouse/pull/12672) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Fix wrong optimization `optimize_move_functions_out_of_any=1` in case of `any(func(<lambda>))`. [#12664](https://github.com/ClickHouse/ClickHouse/pull/12664) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Fixed [#10572](https://github.com/ClickHouse/ClickHouse/issues/10572) fix bloom filter index with const expression. [#12659](https://github.com/ClickHouse/ClickHouse/pull/12659) ([Winter Zhang](https://github.com/zhang2014)).
|
||||
* Fix SIGSEGV in StorageKafka when broker is unavailable (and not only). [#12658](https://github.com/ClickHouse/ClickHouse/pull/12658) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Add support for function `if` with `Array(UUID)` arguments. This fixes [#11066](https://github.com/ClickHouse/ClickHouse/issues/11066). [#12648](https://github.com/ClickHouse/ClickHouse/pull/12648) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* CREATE USER IF NOT EXISTS now doesn't throw exception if the user exists. This fixes https://github.com/ClickHouse/ClickHouse/issues/12507. [#12646](https://github.com/ClickHouse/ClickHouse/pull/12646) ([Vitaly Baranov](https://github.com/vitlibar)).
|
||||
* Exception `There is no supertype...` can be thrown during `ALTER ... UPDATE` in unexpected cases (e.g. when subtracting from UInt64 column). This fixes [#7306](https://github.com/ClickHouse/ClickHouse/issues/7306). This fixes [#4165](https://github.com/ClickHouse/ClickHouse/issues/4165). [#12633](https://github.com/ClickHouse/ClickHouse/pull/12633) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix possible `Pipeline stuck` error for queries with external sorting. Fixes [#12617](https://github.com/ClickHouse/ClickHouse/issues/12617). [#12618](https://github.com/ClickHouse/ClickHouse/pull/12618) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix error `Output of TreeExecutor is not sorted` for `OPTIMIZE DEDUPLICATE`. Fixes [#11572](https://github.com/ClickHouse/ClickHouse/issues/11572). [#12613](https://github.com/ClickHouse/ClickHouse/pull/12613) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix the issue when alias on result of function `any` can be lost during query optimization. [#12593](https://github.com/ClickHouse/ClickHouse/pull/12593) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Remove data for Distributed tables (blocks from async INSERTs) on DROP TABLE. [#12556](https://github.com/ClickHouse/ClickHouse/pull/12556) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* Now ClickHouse will recalculate checksums for parts when file `checksums.txt` is absent. Broken since [#9827](https://github.com/ClickHouse/ClickHouse/issues/9827). [#12545](https://github.com/ClickHouse/ClickHouse/pull/12545) ([alesapin](https://github.com/alesapin)).
|
||||
* Fix bug which lead to broken old parts after `ALTER DELETE` query when `enable_mixed_granularity_parts=1`. Fixes [#12536](https://github.com/ClickHouse/ClickHouse/issues/12536). [#12543](https://github.com/ClickHouse/ClickHouse/pull/12543) ([alesapin](https://github.com/alesapin)).
|
||||
* Fixing race condition in live view tables which could cause data duplication. LIVE VIEW is an experimental feature. [#12519](https://github.com/ClickHouse/ClickHouse/pull/12519) ([vzakaznikov](https://github.com/vzakaznikov)).
|
||||
* Fix backwards compatibility in binary format of `AggregateFunction(avg, ...)` values. This fixes [#12342](https://github.com/ClickHouse/ClickHouse/issues/12342). [#12486](https://github.com/ClickHouse/ClickHouse/pull/12486) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix crash in JOIN with dictionary when we are joining over expression of dictionary key: `t JOIN dict ON expr(dict.id) = t.id`. Disable dictionary join optimisation for this case. [#12458](https://github.com/ClickHouse/ClickHouse/pull/12458) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Fix overflow when very large LIMIT or OFFSET is specified. This fixes [#10470](https://github.com/ClickHouse/ClickHouse/issues/10470). This fixes [#11372](https://github.com/ClickHouse/ClickHouse/issues/11372). [#12427](https://github.com/ClickHouse/ClickHouse/pull/12427) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* kafka: fix SIGSEGV if there is a message with error in the middle of the batch. [#12302](https://github.com/ClickHouse/ClickHouse/pull/12302) ([Azat Khuzhin](https://github.com/azat)).
|
||||
|
||||
#### Improvement
|
||||
|
||||
* Keep smaller amount of logs in ZooKeeper. Avoid excessive growing of ZooKeeper nodes in case of offline replicas when having many servers/tables/inserts. [#13100](https://github.com/ClickHouse/ClickHouse/pull/13100) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Now exceptions forwarded to the client if an error happened during ALTER or mutation. Closes [#11329](https://github.com/ClickHouse/ClickHouse/issues/11329). [#12666](https://github.com/ClickHouse/ClickHouse/pull/12666) ([alesapin](https://github.com/alesapin)).
|
||||
* Add `QueryTimeMicroseconds`, `SelectQueryTimeMicroseconds` and `InsertQueryTimeMicroseconds` to `system.events`, along with system.metrics, processes, query_log, etc. [#13028](https://github.com/ClickHouse/ClickHouse/pull/13028) ([ianton-ru](https://github.com/ianton-ru)).
|
||||
* Added `SelectedRows` and `SelectedBytes` to `system.events`, along with system.metrics, processes, query_log, etc. [#12638](https://github.com/ClickHouse/ClickHouse/pull/12638) ([ianton-ru](https://github.com/ianton-ru)).
|
||||
* Added `current_database` information to `system.query_log`. [#12652](https://github.com/ClickHouse/ClickHouse/pull/12652) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Allow `TabSeparatedRaw` as input format. [#12009](https://github.com/ClickHouse/ClickHouse/pull/12009) ([hcz](https://github.com/hczhcz)).
|
||||
* Now `joinGet` supports multi-key lookup. [#12418](https://github.com/ClickHouse/ClickHouse/pull/12418) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Allow `*Map` aggregate functions to work on Arrays with NULLs. Fixes [#13157](https://github.com/ClickHouse/ClickHouse/issues/13157). [#13225](https://github.com/ClickHouse/ClickHouse/pull/13225) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Avoid overflow in parsing of DateTime values that will lead to negative unix timestamp in their timezone (for example, `1970-01-01 00:00:00` in Moscow). Saturate to zero instead. This fixes [#3470](https://github.com/ClickHouse/ClickHouse/issues/3470). This fixes [#4172](https://github.com/ClickHouse/ClickHouse/issues/4172). [#12443](https://github.com/ClickHouse/ClickHouse/pull/12443) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* AvroConfluent: Skip Kafka tombstone records - Support skipping broken records [#13203](https://github.com/ClickHouse/ClickHouse/pull/13203) ([Andrew Onyshchuk](https://github.com/oandrew)).
|
||||
* Fix wrong error for long queries. It was possible to get syntax error other than `Max query size exceeded` for correct query. [#13928](https://github.com/ClickHouse/ClickHouse/pull/13928) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix data race in `lgamma` function. This race was caught only in `tsan`, no side effects really happened. [#13842](https://github.com/ClickHouse/ClickHouse/pull/13842) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Fix a 'Week'-interval formatting for ATTACH/ALTER/CREATE QUOTA-statements. [#13417](https://github.com/ClickHouse/ClickHouse/pull/13417) ([vladimir-golovchenko](https://github.com/vladimir-golovchenko)).
|
||||
* Now broken parts are also reported when encountered in compact part processing. Compact parts is an experimental feature. [#13282](https://github.com/ClickHouse/ClickHouse/pull/13282) ([Amos Bird](https://github.com/amosbird)).
|
||||
* Fix assert in `geohashesInBox`. This fixes [#12554](https://github.com/ClickHouse/ClickHouse/issues/12554). [#13229](https://github.com/ClickHouse/ClickHouse/pull/13229) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix assert in `parseDateTimeBestEffort`. This fixes [#12649](https://github.com/ClickHouse/ClickHouse/issues/12649). [#13227](https://github.com/ClickHouse/ClickHouse/pull/13227) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Minor optimization in Processors/PipelineExecutor: breaking out of a loop because it makes sense to do so. [#13058](https://github.com/ClickHouse/ClickHouse/pull/13058) ([Mark Papadakis](https://github.com/markpapadakis)).
|
||||
* Support TRUNCATE table without TABLE keyword. [#12653](https://github.com/ClickHouse/ClickHouse/pull/12653) ([Winter Zhang](https://github.com/zhang2014)).
|
||||
* Fix explain query format overwrite by default, issue https://github.com/ClickHouse/ClickHouse/issues/12432. [#12541](https://github.com/ClickHouse/ClickHouse/pull/12541) ([BohuTANG](https://github.com/BohuTANG)).
|
||||
* Allow to set JOIN kind and type in more standad way: `LEFT SEMI JOIN` instead of `SEMI LEFT JOIN`. For now both are correct. [#12520](https://github.com/ClickHouse/ClickHouse/pull/12520) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Changes default value for `multiple_joins_rewriter_version` to 2. It enables new multiple joins rewriter that knows about column names. [#12469](https://github.com/ClickHouse/ClickHouse/pull/12469) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Add several metrics for requests to S3 storages. [#12464](https://github.com/ClickHouse/ClickHouse/pull/12464) ([ianton-ru](https://github.com/ianton-ru)).
|
||||
* Use correct default secure port for clickhouse-benchmark with `--secure` argument. This fixes [#11044](https://github.com/ClickHouse/ClickHouse/issues/11044). [#12440](https://github.com/ClickHouse/ClickHouse/pull/12440) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Rollback insertion errors in `Log`, `TinyLog`, `StripeLog` engines. In previous versions insertion error lead to inconsisent table state (this works as documented and it is normal for these table engines). This fixes [#12402](https://github.com/ClickHouse/ClickHouse/issues/12402). [#12426](https://github.com/ClickHouse/ClickHouse/pull/12426) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Implement `RENAME DATABASE` and `RENAME DICTIONARY` for `Atomic` database engine - Add implicit `{uuid}` macro, which can be used in ZooKeeper path for `ReplicatedMergeTree`. It works with `CREATE ... ON CLUSTER ...` queries. Set `show_table_uuid_in_table_create_query_if_not_nil` to `true` to use it. - Make `ReplicatedMergeTree` engine arguments optional, `/clickhouse/tables/{uuid}/{shard}/` and `{replica}` are used by default. Closes [#12135](https://github.com/ClickHouse/ClickHouse/issues/12135). - Minor fixes. - These changes break backward compatibility of `Atomic` database engine. Previously created `Atomic` databases must be manually converted to new format. Atomic database is an experimental feature. [#12343](https://github.com/ClickHouse/ClickHouse/pull/12343) ([tavplubix](https://github.com/tavplubix)).
|
||||
* Separated `AWSAuthV4Signer` into different logger, removed excessive `AWSClient: AWSClient` from log messages. [#12320](https://github.com/ClickHouse/ClickHouse/pull/12320) ([Vladimir Chebotarev](https://github.com/excitoon)).
|
||||
* Better exception message in disk access storage. [#12625](https://github.com/ClickHouse/ClickHouse/pull/12625) ([alesapin](https://github.com/alesapin)).
|
||||
* Better exception for function `in` with invalid number of arguments. [#12529](https://github.com/ClickHouse/ClickHouse/pull/12529) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Fix error message about adaptive granularity. [#12624](https://github.com/ClickHouse/ClickHouse/pull/12624) ([alesapin](https://github.com/alesapin)).
|
||||
* Fix SETTINGS parse after FORMAT. [#12480](https://github.com/ClickHouse/ClickHouse/pull/12480) ([Azat Khuzhin](https://github.com/azat)).
|
||||
* If MergeTree table does not contain ORDER BY or PARTITION BY, it was possible to request ALTER to CLEAR all the columns and ALTER will stuck. Fixed [#7941](https://github.com/ClickHouse/ClickHouse/issues/7941). [#12382](https://github.com/ClickHouse/ClickHouse/pull/12382) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Avoid re-loading completion from the history file after each query (to avoid history overlaps with other client sessions). [#13086](https://github.com/ClickHouse/ClickHouse/pull/13086) ([Azat Khuzhin](https://github.com/azat)).
|
||||
|
||||
#### Performance Improvement
|
||||
|
||||
* Lower memory usage for some operations up to 2 times. [#12424](https://github.com/ClickHouse/ClickHouse/pull/12424) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Optimize PK lookup for queries that match exact PK range. [#12277](https://github.com/ClickHouse/ClickHouse/pull/12277) ([Ivan Babrou](https://github.com/bobrik)).
|
||||
* Slightly optimize very short queries with `LowCardinality`. [#14129](https://github.com/ClickHouse/ClickHouse/pull/14129) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Slightly improve performance of aggregation by UInt8/UInt16 keys. [#13091](https://github.com/ClickHouse/ClickHouse/pull/13091) and [#13055](https://github.com/ClickHouse/ClickHouse/pull/13055) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Push down `LIMIT` step for query plan (inside subqueries). [#13016](https://github.com/ClickHouse/ClickHouse/pull/13016) ([Nikolai Kochetov](https://github.com/KochetovNicolai)).
|
||||
* Parallel primary key lookup and skipping index stages on parts, as described in [#11564](https://github.com/ClickHouse/ClickHouse/issues/11564). [#12589](https://github.com/ClickHouse/ClickHouse/pull/12589) ([Ivan Babrou](https://github.com/bobrik)).
|
||||
* Converting String-type arguments of function "if" and "transform" into enum if `set optimize_if_transform_strings_to_enum = 1`. [#12515](https://github.com/ClickHouse/ClickHouse/pull/12515) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Replaces monotonic functions with its argument in `ORDER BY` if `set optimize_monotonous_functions_in_order_by=1`. [#12467](https://github.com/ClickHouse/ClickHouse/pull/12467) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Add order by optimization that rewrites `ORDER BY x, f(x)` with `ORDER by x` if `set optimize_redundant_functions_in_order_by = 1`. [#12404](https://github.com/ClickHouse/ClickHouse/pull/12404) ([Artem Zuikov](https://github.com/4ertus2)).
|
||||
* Allow pushdown predicate when subquery contains `WITH` clause. This fixes [#12293](https://github.com/ClickHouse/ClickHouse/issues/12293) [#12663](https://github.com/ClickHouse/ClickHouse/pull/12663) ([Winter Zhang](https://github.com/zhang2014)).
|
||||
* Improve performance of reading from compact parts. Compact parts is an experimental feature. [#12492](https://github.com/ClickHouse/ClickHouse/pull/12492) ([Anton Popov](https://github.com/CurtizJ)).
|
||||
* Attempt to implement streaming optimization in `DiskS3`. DiskS3 is an experimental feature. [#12434](https://github.com/ClickHouse/ClickHouse/pull/12434) ([Vladimir Chebotarev](https://github.com/excitoon)).
|
||||
|
||||
#### Build/Testing/Packaging Improvement
|
||||
|
||||
* Use `shellcheck` for sh tests linting. [#13200](https://github.com/ClickHouse/ClickHouse/pull/13200) [#13207](https://github.com/ClickHouse/ClickHouse/pull/13207) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Add script which set labels for pull requests in GitHub hook. [#13183](https://github.com/ClickHouse/ClickHouse/pull/13183) ([alesapin](https://github.com/alesapin)).
|
||||
* Remove some of recursive submodules. See [#13378](https://github.com/ClickHouse/ClickHouse/issues/13378). [#13379](https://github.com/ClickHouse/ClickHouse/pull/13379) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Ensure that all the submodules are from proper URLs. Continuation of [#13379](https://github.com/ClickHouse/ClickHouse/issues/13379). This fixes [#13378](https://github.com/ClickHouse/ClickHouse/issues/13378). [#13397](https://github.com/ClickHouse/ClickHouse/pull/13397) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Added support for user-declared settings, which can be accessed from inside queries. This is needed when ClickHouse engine is used as a component of another system. [#13013](https://github.com/ClickHouse/ClickHouse/pull/13013) ([Vitaly Baranov](https://github.com/vitlibar)).
|
||||
* Added testing for RBAC functionality of INSERT privilege in TestFlows. Expanded tables on which SELECT is being tested. Added Requirements to match new table engine tests. [#13340](https://github.com/ClickHouse/ClickHouse/pull/13340) ([MyroTk](https://github.com/MyroTk)).
|
||||
* Fix timeout error during server restart in the stress test. [#13321](https://github.com/ClickHouse/ClickHouse/pull/13321) ([alesapin](https://github.com/alesapin)).
|
||||
* Now fast test will wait server with retries. [#13284](https://github.com/ClickHouse/ClickHouse/pull/13284) ([alesapin](https://github.com/alesapin)).
|
||||
* Function `materialize()` (the function for ClickHouse testing) will work for NULL as expected - by transforming it to non-constant column. [#13212](https://github.com/ClickHouse/ClickHouse/pull/13212) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix libunwind build in AArch64. This fixes [#13204](https://github.com/ClickHouse/ClickHouse/issues/13204). [#13208](https://github.com/ClickHouse/ClickHouse/pull/13208) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Even more retries in zkutil gtest to prevent test flakiness. [#13165](https://github.com/ClickHouse/ClickHouse/pull/13165) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Small fixes to the RBAC TestFlows. [#13152](https://github.com/ClickHouse/ClickHouse/pull/13152) ([vzakaznikov](https://github.com/vzakaznikov)).
|
||||
* Fixing `00960_live_view_watch_events_live.py` test. [#13108](https://github.com/ClickHouse/ClickHouse/pull/13108) ([vzakaznikov](https://github.com/vzakaznikov)).
|
||||
* Improve cache purge in documentation deploy script. [#13107](https://github.com/ClickHouse/ClickHouse/pull/13107) ([alesapin](https://github.com/alesapin)).
|
||||
* Rewrote some orphan tests to gtest. Removed useless includes from tests. [#13073](https://github.com/ClickHouse/ClickHouse/pull/13073) ([Nikita Mikhaylov](https://github.com/nikitamikhaylov)).
|
||||
* Added tests for RBAC functionality of `SELECT` privilege in TestFlows. [#13061](https://github.com/ClickHouse/ClickHouse/pull/13061) ([Ritaank Tiwari](https://github.com/ritaank)).
|
||||
* Rerun some tests in fast test check. [#12992](https://github.com/ClickHouse/ClickHouse/pull/12992) ([alesapin](https://github.com/alesapin)).
|
||||
* Fix MSan error in "rdkafka" library. This closes [#12990](https://github.com/ClickHouse/ClickHouse/issues/12990). Updated `rdkafka` to version 1.5 (master). [#12991](https://github.com/ClickHouse/ClickHouse/pull/12991) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix UBSan report in base64 if tests were run on server with AVX-512. This fixes [#12318](https://github.com/ClickHouse/ClickHouse/issues/12318). Author: @qoega. [#12441](https://github.com/ClickHouse/ClickHouse/pull/12441) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Fix UBSan report in HDFS library. This closes [#12330](https://github.com/ClickHouse/ClickHouse/issues/12330). [#12453](https://github.com/ClickHouse/ClickHouse/pull/12453) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Check an ability that we able to restore the backup from an old version to the new version. This closes [#8979](https://github.com/ClickHouse/ClickHouse/issues/8979). [#12959](https://github.com/ClickHouse/ClickHouse/pull/12959) ([alesapin](https://github.com/alesapin)).
|
||||
* Do not build helper_container image inside integrational tests. Build docker container in CI and use pre-built helper_container in integration tests. [#12953](https://github.com/ClickHouse/ClickHouse/pull/12953) ([Ilya Yatsishin](https://github.com/qoega)).
|
||||
* Add a test for `ALTER TABLE CLEAR COLUMN` query for primary key columns. [#12951](https://github.com/ClickHouse/ClickHouse/pull/12951) ([alesapin](https://github.com/alesapin)).
|
||||
* Increased timeouts in testflows tests. [#12949](https://github.com/ClickHouse/ClickHouse/pull/12949) ([vzakaznikov](https://github.com/vzakaznikov)).
|
||||
* Fix build of test under Mac OS X. This closes [#12767](https://github.com/ClickHouse/ClickHouse/issues/12767). [#12772](https://github.com/ClickHouse/ClickHouse/pull/12772) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Connector-ODBC updated to mysql-connector-odbc-8.0.21. [#12739](https://github.com/ClickHouse/ClickHouse/pull/12739) ([Ilya Yatsishin](https://github.com/qoega)).
|
||||
* Adding RBAC syntax tests in TestFlows. [#12642](https://github.com/ClickHouse/ClickHouse/pull/12642) ([vzakaznikov](https://github.com/vzakaznikov)).
|
||||
* Improve performance of TestKeeper. This will speedup tests with heavy usage of Replicated tables. [#12505](https://github.com/ClickHouse/ClickHouse/pull/12505) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Now we check that server is able to start after stress tests run. This fixes [#12473](https://github.com/ClickHouse/ClickHouse/issues/12473). [#12496](https://github.com/ClickHouse/ClickHouse/pull/12496) ([alesapin](https://github.com/alesapin)).
|
||||
* Update fmtlib to master (7.0.1). [#12446](https://github.com/ClickHouse/ClickHouse/pull/12446) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Add docker image for fast tests. [#12294](https://github.com/ClickHouse/ClickHouse/pull/12294) ([alesapin](https://github.com/alesapin)).
|
||||
* Rework configuration paths for integration tests. [#12285](https://github.com/ClickHouse/ClickHouse/pull/12285) ([Ilya Yatsishin](https://github.com/qoega)).
|
||||
* Add compiler option to control that stack frames are not too large. This will help to run the code in fibers with small stack size. [#11524](https://github.com/ClickHouse/ClickHouse/pull/11524) ([alexey-milovidov](https://github.com/alexey-milovidov)).
|
||||
* Update gitignore-files. [#13447](https://github.com/ClickHouse/ClickHouse/pull/13447) ([vladimir-golovchenko](https://github.com/vladimir-golovchenko)).
|
||||
|
||||
|
||||
## ClickHouse release 20.6
|
||||
|
||||
### ClickHouse release v20.6.3.28-stable
|
||||
|
||||
@ -17,5 +17,4 @@ ClickHouse is an open-source column-oriented database management system that all
|
||||
|
||||
## Upcoming Events
|
||||
|
||||
* [ClickHouse at ByteDance (in Chinese)](https://mp.weixin.qq.com/s/Em-HjPylO8D7WPui4RREAQ) on August 28, 2020.
|
||||
* [ClickHouse Data Integration Virtual Meetup](https://www.eventbrite.com/e/clickhouse-september-virtual-meetup-data-integration-tickets-117421895049) on September 10, 2020.
|
||||
|
||||
@ -78,8 +78,9 @@ if (COMPILER_GCC)
|
||||
-Wno-deprecated-declarations -Wno-class-memaccess)
|
||||
elseif (COMPILER_CLANG)
|
||||
set (SUPPRESS_WARNINGS -Wno-non-virtual-dtor -Wno-sign-compare -Wno-strict-aliasing -Wno-deprecated-declarations)
|
||||
set (CAPNP_PRIVATE_CXX_FLAGS -fno-char8_t)
|
||||
endif ()
|
||||
|
||||
target_compile_options(kj PRIVATE ${SUPPRESS_WARNINGS})
|
||||
target_compile_options(capnp PRIVATE ${SUPPRESS_WARNINGS})
|
||||
target_compile_options(capnpc PRIVATE ${SUPPRESS_WARNINGS})
|
||||
target_compile_options(kj PRIVATE ${SUPPRESS_WARNINGS} ${CAPNP_PRIVATE_CXX_FLAGS})
|
||||
target_compile_options(capnp PRIVATE ${SUPPRESS_WARNINGS} ${CAPNP_PRIVATE_CXX_FLAGS})
|
||||
target_compile_options(capnpc PRIVATE ${SUPPRESS_WARNINGS} ${CAPNP_PRIVATE_CXX_FLAGS})
|
||||
|
||||
23
debian/clickhouse-server.init
vendored
23
debian/clickhouse-server.init
vendored
@ -67,13 +67,6 @@ if uname -mpi | grep -q 'x86_64'; then
|
||||
fi
|
||||
|
||||
|
||||
SUPPORTED_COMMANDS="{start|stop|status|restart|forcestop|forcerestart|reload|condstart|condstop|condrestart|condreload|initdb}"
|
||||
is_supported_command()
|
||||
{
|
||||
echo "$SUPPORTED_COMMANDS" | grep -E "(\{|\|)$1(\||})" &> /dev/null
|
||||
}
|
||||
|
||||
|
||||
is_running()
|
||||
{
|
||||
pgrep --pidfile "$CLICKHOUSE_PIDFILE" $(echo "${PROGRAM}" | cut -c1-15) 1> /dev/null 2> /dev/null
|
||||
@ -283,13 +276,12 @@ use_cron()
|
||||
fi
|
||||
return 0
|
||||
}
|
||||
|
||||
# returns false if cron disabled (with systemd)
|
||||
enable_cron()
|
||||
{
|
||||
use_cron && sed -i 's/^#*//' "$CLICKHOUSE_CRONFILE"
|
||||
}
|
||||
|
||||
|
||||
# returns false if cron disabled (with systemd)
|
||||
disable_cron()
|
||||
{
|
||||
use_cron && sed -i 's/^#*/#/' "$CLICKHOUSE_CRONFILE"
|
||||
@ -312,15 +304,14 @@ main()
|
||||
EXIT_STATUS=0
|
||||
case "$1" in
|
||||
start)
|
||||
start && enable_cron
|
||||
service_or_func start && enable_cron
|
||||
;;
|
||||
stop)
|
||||
# disable_cron returns false if cron disabled (with systemd) - not checking return status
|
||||
disable_cron
|
||||
stop
|
||||
service_or_func stop
|
||||
;;
|
||||
restart)
|
||||
restart && enable_cron
|
||||
service_or_func restart && enable_cron
|
||||
;;
|
||||
forcestop)
|
||||
disable_cron
|
||||
@ -330,7 +321,7 @@ main()
|
||||
forcerestart && enable_cron
|
||||
;;
|
||||
reload)
|
||||
restart
|
||||
service_or_func restart
|
||||
;;
|
||||
condstart)
|
||||
is_running || service_or_func start
|
||||
@ -354,7 +345,7 @@ main()
|
||||
disable_cron
|
||||
;;
|
||||
*)
|
||||
echo "Usage: $0 $SUPPORTED_COMMANDS"
|
||||
echo "Usage: $0 {start|stop|status|restart|forcestop|forcerestart|reload|condstart|condstop|condrestart|condreload|initdb}"
|
||||
exit 2
|
||||
;;
|
||||
esac
|
||||
|
||||
2
debian/rules
vendored
2
debian/rules
vendored
@ -18,7 +18,7 @@ ifeq ($(CCACHE_PREFIX),distcc)
|
||||
THREADS_COUNT=$(shell distcc -j)
|
||||
endif
|
||||
ifeq ($(THREADS_COUNT),)
|
||||
THREADS_COUNT=$(shell nproc || grep -c ^processor /proc/cpuinfo || sysctl -n hw.ncpu || echo 4)
|
||||
THREADS_COUNT=$(shell echo $$(( $$(nproc || grep -c ^processor /proc/cpuinfo || sysctl -n hw.ncpu || echo 8) / 2 )) )
|
||||
endif
|
||||
DEB_BUILD_OPTIONS+=parallel=$(THREADS_COUNT)
|
||||
|
||||
|
||||
@ -6,7 +6,7 @@ RUN apt-get update \
|
||||
&& apt-get install ca-certificates lsb-release wget gnupg apt-transport-https \
|
||||
--yes --no-install-recommends --verbose-versions \
|
||||
&& export LLVM_PUBKEY_HASH="bda960a8da687a275a2078d43c111d66b1c6a893a3275271beedf266c1ff4a0cdecb429c7a5cccf9f486ea7aa43fd27f" \
|
||||
&& wget -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& wget -nv -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& echo "${LLVM_PUBKEY_HASH} /tmp/llvm-snapshot.gpg.key" | sha384sum -c \
|
||||
&& apt-key add /tmp/llvm-snapshot.gpg.key \
|
||||
&& export CODENAME="$(lsb_release --codename --short | tr 'A-Z' 'a-z')" \
|
||||
|
||||
@ -7,7 +7,7 @@ RUN apt-get update \
|
||||
&& apt-get install ca-certificates lsb-release wget gnupg apt-transport-https \
|
||||
--yes --no-install-recommends --verbose-versions \
|
||||
&& export LLVM_PUBKEY_HASH="bda960a8da687a275a2078d43c111d66b1c6a893a3275271beedf266c1ff4a0cdecb429c7a5cccf9f486ea7aa43fd27f" \
|
||||
&& wget -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& wget -nv -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& echo "${LLVM_PUBKEY_HASH} /tmp/llvm-snapshot.gpg.key" | sha384sum -c \
|
||||
&& apt-key add /tmp/llvm-snapshot.gpg.key \
|
||||
&& export CODENAME="$(lsb_release --codename --short | tr 'A-Z' 'a-z')" \
|
||||
@ -55,7 +55,6 @@ RUN apt-get update \
|
||||
cmake \
|
||||
gdb \
|
||||
rename \
|
||||
wget \
|
||||
build-essential \
|
||||
--yes --no-install-recommends
|
||||
|
||||
@ -83,14 +82,14 @@ RUN git clone https://github.com/tpoechtrager/cctools-port.git \
|
||||
&& rm -rf cctools-port
|
||||
|
||||
# Download toolchain for Darwin
|
||||
RUN wget https://github.com/phracker/MacOSX-SDKs/releases/download/10.14-beta4/MacOSX10.14.sdk.tar.xz
|
||||
RUN wget -nv https://github.com/phracker/MacOSX-SDKs/releases/download/10.14-beta4/MacOSX10.14.sdk.tar.xz
|
||||
|
||||
# Download toolchain for ARM
|
||||
# It contains all required headers and libraries. Note that it's named as "gcc" but actually we are using clang for cross compiling.
|
||||
RUN wget "https://developer.arm.com/-/media/Files/downloads/gnu-a/8.3-2019.03/binrel/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu.tar.xz?revision=2e88a73f-d233-4f96-b1f4-d8b36e9bb0b9&la=en" -O gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu.tar.xz
|
||||
RUN wget -nv "https://developer.arm.com/-/media/Files/downloads/gnu-a/8.3-2019.03/binrel/gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu.tar.xz?revision=2e88a73f-d233-4f96-b1f4-d8b36e9bb0b9&la=en" -O gcc-arm-8.3-2019.03-x86_64-aarch64-linux-gnu.tar.xz
|
||||
|
||||
# Download toolchain for FreeBSD 11.3
|
||||
RUN wget https://clickhouse-datasets.s3.yandex.net/toolchains/toolchains/freebsd-11.3-toolchain.tar.xz
|
||||
RUN wget -nv https://clickhouse-datasets.s3.yandex.net/toolchains/toolchains/freebsd-11.3-toolchain.tar.xz
|
||||
|
||||
COPY build.sh /
|
||||
CMD ["/bin/bash", "/build.sh"]
|
||||
|
||||
@ -7,7 +7,7 @@ RUN apt-get update \
|
||||
&& apt-get install ca-certificates lsb-release wget gnupg apt-transport-https \
|
||||
--yes --no-install-recommends --verbose-versions \
|
||||
&& export LLVM_PUBKEY_HASH="bda960a8da687a275a2078d43c111d66b1c6a893a3275271beedf266c1ff4a0cdecb429c7a5cccf9f486ea7aa43fd27f" \
|
||||
&& wget -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& wget -nv -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& echo "${LLVM_PUBKEY_HASH} /tmp/llvm-snapshot.gpg.key" | sha384sum -c \
|
||||
&& apt-key add /tmp/llvm-snapshot.gpg.key \
|
||||
&& export CODENAME="$(lsb_release --codename --short | tr 'A-Z' 'a-z')" \
|
||||
@ -34,7 +34,7 @@ RUN curl -O https://clickhouse-builds.s3.yandex.net/utils/1/dpkg-deb \
|
||||
ENV APACHE_PUBKEY_HASH="bba6987b63c63f710fd4ed476121c588bc3812e99659d27a855f8c4d312783ee66ad6adfce238765691b04d62fa3688f"
|
||||
|
||||
RUN export CODENAME="$(lsb_release --codename --short | tr 'A-Z' 'a-z')" \
|
||||
&& wget -O /tmp/arrow-keyring.deb "https://apache.bintray.com/arrow/ubuntu/apache-arrow-archive-keyring-latest-${CODENAME}.deb" \
|
||||
&& wget -nv -O /tmp/arrow-keyring.deb "https://apache.bintray.com/arrow/ubuntu/apache-arrow-archive-keyring-latest-${CODENAME}.deb" \
|
||||
&& echo "${APACHE_PUBKEY_HASH} /tmp/arrow-keyring.deb" | sha384sum -c \
|
||||
&& dpkg -i /tmp/arrow-keyring.deb
|
||||
|
||||
|
||||
@ -7,7 +7,7 @@ RUN apt-get update \
|
||||
&& apt-get install ca-certificates lsb-release wget gnupg apt-transport-https \
|
||||
--yes --no-install-recommends --verbose-versions \
|
||||
&& export LLVM_PUBKEY_HASH="bda960a8da687a275a2078d43c111d66b1c6a893a3275271beedf266c1ff4a0cdecb429c7a5cccf9f486ea7aa43fd27f" \
|
||||
&& wget -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& wget -nv -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& echo "${LLVM_PUBKEY_HASH} /tmp/llvm-snapshot.gpg.key" | sha384sum -c \
|
||||
&& apt-key add /tmp/llvm-snapshot.gpg.key \
|
||||
&& export CODENAME="$(lsb_release --codename --short | tr 'A-Z' 'a-z')" \
|
||||
|
||||
@ -15,7 +15,7 @@ RUN apt-get --allow-unauthenticated update -y \
|
||||
gpg-agent \
|
||||
git
|
||||
|
||||
RUN wget -O - https://apt.kitware.com/keys/kitware-archive-latest.asc 2>/dev/null | sudo apt-key add -
|
||||
RUN wget -nv -O - https://apt.kitware.com/keys/kitware-archive-latest.asc | sudo apt-key add -
|
||||
RUN sudo apt-add-repository 'deb https://apt.kitware.com/ubuntu/ bionic main'
|
||||
RUN sudo echo "deb [trusted=yes] http://apt.llvm.org/bionic/ llvm-toolchain-bionic-8 main" >> /etc/apt/sources.list
|
||||
|
||||
|
||||
@ -7,7 +7,7 @@ RUN apt-get update \
|
||||
&& apt-get install ca-certificates lsb-release wget gnupg apt-transport-https \
|
||||
--yes --no-install-recommends --verbose-versions \
|
||||
&& export LLVM_PUBKEY_HASH="bda960a8da687a275a2078d43c111d66b1c6a893a3275271beedf266c1ff4a0cdecb429c7a5cccf9f486ea7aa43fd27f" \
|
||||
&& wget -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& wget -nv -O /tmp/llvm-snapshot.gpg.key https://apt.llvm.org/llvm-snapshot.gpg.key \
|
||||
&& echo "${LLVM_PUBKEY_HASH} /tmp/llvm-snapshot.gpg.key" | sha384sum -c \
|
||||
&& apt-key add /tmp/llvm-snapshot.gpg.key \
|
||||
&& export CODENAME="$(lsb_release --codename --short | tr 'A-Z' 'a-z')" \
|
||||
@ -61,7 +61,6 @@ RUN apt-get update \
|
||||
software-properties-common \
|
||||
tzdata \
|
||||
unixodbc \
|
||||
wget \
|
||||
--yes --no-install-recommends
|
||||
|
||||
# This symlink required by gcc to find lld compiler
|
||||
@ -70,7 +69,7 @@ RUN ln -s /usr/bin/lld-${LLVM_VERSION} /usr/bin/ld.lld
|
||||
ARG odbc_driver_url="https://github.com/ClickHouse/clickhouse-odbc/releases/download/v1.1.4.20200302/clickhouse-odbc-1.1.4-Linux.tar.gz"
|
||||
|
||||
RUN mkdir -p /tmp/clickhouse-odbc-tmp \
|
||||
&& wget --quiet -O - ${odbc_driver_url} | tar --strip-components=1 -xz -C /tmp/clickhouse-odbc-tmp \
|
||||
&& wget -nv -O - ${odbc_driver_url} | tar --strip-components=1 -xz -C /tmp/clickhouse-odbc-tmp \
|
||||
&& cp /tmp/clickhouse-odbc-tmp/lib64/*.so /usr/local/lib/ \
|
||||
&& odbcinst -i -d -f /tmp/clickhouse-odbc-tmp/share/doc/clickhouse-odbc/config/odbcinst.ini.sample \
|
||||
&& odbcinst -i -s -l -f /tmp/clickhouse-odbc-tmp/share/doc/clickhouse-odbc/config/odbc.ini.sample \
|
||||
|
||||
@ -2,6 +2,15 @@
|
||||
<profiles>
|
||||
<default>
|
||||
<max_execution_time>10</max_execution_time>
|
||||
<!--
|
||||
Don't let the fuzzer change this setting (I've actually seen it
|
||||
do this before).
|
||||
-->
|
||||
<constraints>
|
||||
<max_execution_time>
|
||||
<max>10</max>
|
||||
</max_execution_time>
|
||||
</constraints>
|
||||
</default>
|
||||
</profiles>
|
||||
</yandex>
|
||||
|
||||
@ -91,7 +91,7 @@ function fuzz
|
||||
|
||||
./clickhouse-client --query "select elapsed, query from system.processes" ||:
|
||||
killall clickhouse-server ||:
|
||||
for x in {1..10}
|
||||
for _ in {1..10}
|
||||
do
|
||||
if ! pgrep -f clickhouse-server
|
||||
then
|
||||
@ -172,8 +172,59 @@ case "$stage" in
|
||||
echo "failure" > status.txt
|
||||
echo "Fuzzer failed ($fuzzer_exit_code). See the logs" > description.txt
|
||||
fi
|
||||
;&
|
||||
"report")
|
||||
cat > report.html <<EOF ||:
|
||||
<!DOCTYPE html>
|
||||
<html lang="en">
|
||||
<link rel="preload" as="font" href="https://yastatic.net/adv-www/_/sUYVCPUAQE7ExrvMS7FoISoO83s.woff2" type="font/woff2" crossorigin="anonymous"/>
|
||||
<style>
|
||||
@font-face {
|
||||
font-family:'Yandex Sans Display Web';
|
||||
src:url(https://yastatic.net/adv-www/_/H63jN0veW07XQUIA2317lr9UIm8.eot);
|
||||
src:url(https://yastatic.net/adv-www/_/H63jN0veW07XQUIA2317lr9UIm8.eot?#iefix) format('embedded-opentype'),
|
||||
url(https://yastatic.net/adv-www/_/sUYVCPUAQE7ExrvMS7FoISoO83s.woff2) format('woff2'),
|
||||
url(https://yastatic.net/adv-www/_/v2Sve_obH3rKm6rKrtSQpf-eB7U.woff) format('woff'),
|
||||
url(https://yastatic.net/adv-www/_/PzD8hWLMunow5i3RfJ6WQJAL7aI.ttf) format('truetype'),
|
||||
url(https://yastatic.net/adv-www/_/lF_KG5g4tpQNlYIgA0e77fBSZ5s.svg#YandexSansDisplayWeb-Regular) format('svg');
|
||||
font-weight:400;
|
||||
font-style:normal;
|
||||
font-stretch:normal
|
||||
}
|
||||
|
||||
exit $task_exit_code
|
||||
body { font-family: "Yandex Sans Display Web", Arial, sans-serif; background: #EEE; }
|
||||
h1 { margin-left: 10px; }
|
||||
th, td { border: 0; padding: 5px 10px 5px 10px; text-align: left; vertical-align: top; line-height: 1.5; background-color: #FFF;
|
||||
td { white-space: pre; font-family: Monospace, Courier New; }
|
||||
border: 0; box-shadow: 0 0 0 1px rgba(0, 0, 0, 0.05), 0 8px 25px -5px rgba(0, 0, 0, 0.1); }
|
||||
a { color: #06F; text-decoration: none; }
|
||||
a:hover, a:active { color: #F40; text-decoration: underline; }
|
||||
table { border: 0; }
|
||||
.main { margin-left: 10%; }
|
||||
p.links a { padding: 5px; margin: 3px; background: #FFF; line-height: 2; white-space: nowrap; box-shadow: 0 0 0 1px rgba(0, 0, 0, 0.05), 0 8px 25px -5px rgba(0, 0, 0, 0.1); }
|
||||
th { cursor: pointer; }
|
||||
|
||||
</style>
|
||||
<title>AST Fuzzer for PR #${PR_TO_TEST} @ ${SHA_TO_TEST}</title>
|
||||
</head>
|
||||
<body>
|
||||
<div class="main">
|
||||
|
||||
<h1>AST Fuzzer for PR #${PR_TO_TEST} @ ${SHA_TO_TEST}</h1>
|
||||
<p class="links">
|
||||
<a href="fuzzer.log">fuzzer.log</a>
|
||||
<a href="server.log">server.log</a>
|
||||
<a href="main.log">main.log</a>
|
||||
</p>
|
||||
<table>
|
||||
<tr><th>Test name</th><th>Test status</th><th>Description</th></tr>
|
||||
<tr><td>AST Fuzzer</td><td>$(cat status.txt)</td><td>$(cat description.txt)</td></tr>
|
||||
</table>
|
||||
</body>
|
||||
</html>
|
||||
|
||||
EOF
|
||||
;&
|
||||
esac
|
||||
|
||||
exit $task_exit_code
|
||||
@ -46,7 +46,7 @@ RUN set -eux; \
|
||||
\
|
||||
# this "case" statement is generated via "update.sh"
|
||||
\
|
||||
if ! wget -O docker.tgz "https://download.docker.com/linux/static/${DOCKER_CHANNEL}/x86_64/docker-${DOCKER_VERSION}.tgz"; then \
|
||||
if ! wget -nv -O docker.tgz "https://download.docker.com/linux/static/${DOCKER_CHANNEL}/x86_64/docker-${DOCKER_VERSION}.tgz"; then \
|
||||
echo >&2 "error: failed to download 'docker-${DOCKER_VERSION}' from '${DOCKER_CHANNEL}' for '${x86_64}'"; \
|
||||
exit 1; \
|
||||
fi; \
|
||||
|
||||
@ -5,3 +5,4 @@ services:
|
||||
restart: always
|
||||
ports:
|
||||
- 6380:6379
|
||||
command: redis-server --requirepass "clickhouse"
|
||||
|
||||
@ -536,7 +536,9 @@ create table queries engine File(TSVWithNamesAndTypes, 'report/queries.tsv')
|
||||
left join query_display_names
|
||||
on query_metric_stats.test = query_display_names.test
|
||||
and query_metric_stats.query_index = query_display_names.query_index
|
||||
where metric_name = 'server_time'
|
||||
-- 'server_time' is rounded down to ms, which might be bad for very short queries.
|
||||
-- Use 'client_time' instead.
|
||||
where metric_name = 'client_time'
|
||||
order by test, query_index, metric_name
|
||||
;
|
||||
|
||||
@ -563,40 +565,54 @@ create table unstable_queries_report engine File(TSV, 'report/unstable-queries.t
|
||||
toDecimal64(stat_threshold, 3), unstable_fail, test, query_index, query_display_name
|
||||
from queries where unstable_show order by stat_threshold desc;
|
||||
|
||||
create table test_time_changes engine File(TSV, 'report/test-time-changes.tsv') as
|
||||
select test, queries, average_time_change from (
|
||||
select test, count(*) queries,
|
||||
sum(left) as left, sum(right) as right,
|
||||
(right - left) / right average_time_change
|
||||
|
||||
create view test_speedup as
|
||||
select
|
||||
test,
|
||||
exp2(avg(log2(left / right))) times_speedup,
|
||||
count(*) queries,
|
||||
unstable + changed bad,
|
||||
sum(changed_show) changed,
|
||||
sum(unstable_show) unstable
|
||||
from queries
|
||||
group by test
|
||||
order by abs(average_time_change) desc
|
||||
)
|
||||
order by times_speedup desc
|
||||
;
|
||||
|
||||
create table unstable_tests engine File(TSV, 'report/unstable-tests.tsv') as
|
||||
select test, sum(unstable_show) total_unstable, sum(changed_show) total_changed
|
||||
from queries
|
||||
group by test
|
||||
order by total_unstable + total_changed desc
|
||||
create view total_speedup as
|
||||
select
|
||||
'Total' test,
|
||||
exp2(avg(log2(times_speedup))) times_speedup,
|
||||
sum(queries) queries,
|
||||
unstable + changed bad,
|
||||
sum(changed) changed,
|
||||
sum(unstable) unstable
|
||||
from test_speedup
|
||||
;
|
||||
|
||||
create table test_perf_changes_report engine File(TSV, 'report/test-perf-changes.tsv') as
|
||||
select test,
|
||||
queries,
|
||||
coalesce(total_unstable, 0) total_unstable,
|
||||
coalesce(total_changed, 0) total_changed,
|
||||
total_unstable + total_changed total_bad,
|
||||
coalesce(toString(toDecimal64(average_time_change, 3)), '??') average_time_change_str
|
||||
from test_time_changes
|
||||
full join unstable_tests
|
||||
using test
|
||||
where (abs(average_time_change) > 0.05 and queries > 5)
|
||||
or (total_bad > 0)
|
||||
order by total_bad desc, average_time_change desc
|
||||
settings join_use_nulls = 1
|
||||
with
|
||||
(times_speedup >= 1
|
||||
? '-' || toString(toDecimal64(times_speedup, 3)) || 'x'
|
||||
: '+' || toString(toDecimal64(1 / times_speedup, 3)) || 'x')
|
||||
as times_speedup_str
|
||||
select test, times_speedup_str, queries, bad, changed, unstable
|
||||
-- Not sure what's the precedence of UNION ALL vs WHERE & ORDER BY, hence all
|
||||
-- the braces.
|
||||
from (
|
||||
(
|
||||
select * from total_speedup
|
||||
) union all (
|
||||
select * from test_speedup
|
||||
where
|
||||
(times_speedup >= 1 ? times_speedup : (1 / times_speedup)) >= 1.005
|
||||
or bad
|
||||
)
|
||||
)
|
||||
order by test = 'Total' desc
|
||||
;
|
||||
|
||||
|
||||
create view total_client_time_per_query as select *
|
||||
from file('analyze/client-times.tsv', TSV,
|
||||
'test text, query_index int, client float, server float');
|
||||
@ -888,7 +904,10 @@ for log in *-err.log
|
||||
do
|
||||
test=$(basename "$log" "-err.log")
|
||||
{
|
||||
grep -H -m2 -i '\(Exception\|Error\):[^:]' "$log" \
|
||||
# The second grep is a heuristic for error messages like
|
||||
# "socket.timeout: timed out".
|
||||
grep -h -m2 -i '\(Exception\|Error\):[^:]' "$log" \
|
||||
|| grep -h -m2 -i '^[^ ]\+: ' "$log" \
|
||||
|| head -2 "$log"
|
||||
} | sed "s/^/$test\t/" >> run-errors.tsv ||:
|
||||
done
|
||||
|
||||
@ -168,12 +168,6 @@ def nextRowAnchor():
|
||||
global table_anchor
|
||||
return f'{table_anchor}.{row_anchor + 1}'
|
||||
|
||||
def setRowAnchor(anchor_row_part):
|
||||
global row_anchor
|
||||
global table_anchor
|
||||
row_anchor = anchor_row_part
|
||||
return currentRowAnchor()
|
||||
|
||||
def advanceRowAnchor():
|
||||
global row_anchor
|
||||
global table_anchor
|
||||
@ -376,7 +370,7 @@ if args.report == 'main':
|
||||
columns = [
|
||||
'Old, s', # 0
|
||||
'New, s', # 1
|
||||
'Times speedup / slowdown', # 2
|
||||
'Ratio of speedup (-) or slowdown (+)', # 2
|
||||
'Relative difference (new − old) / old', # 3
|
||||
'p < 0.001 threshold', # 4
|
||||
# Failed # 5
|
||||
@ -453,7 +447,7 @@ if args.report == 'main':
|
||||
addSimpleTable('Skipped tests', ['Test', 'Reason'], skipped_tests_rows)
|
||||
|
||||
addSimpleTable('Test performance changes',
|
||||
['Test', 'Queries', 'Unstable', 'Changed perf', 'Total not OK', 'Avg relative time diff'],
|
||||
['Test', 'Ratio of speedup (-) or slowdown (+)', 'Queries', 'Total not OK', 'Changed perf', 'Unstable'],
|
||||
tsvRows('report/test-perf-changes.tsv'))
|
||||
|
||||
def add_test_times():
|
||||
@ -480,11 +474,12 @@ if args.report == 'main':
|
||||
total_runs = (nominal_runs + 1) * 2 # one prewarm run, two servers
|
||||
attrs = ['' for c in columns]
|
||||
for r in rows:
|
||||
anchor = f'{currentTableAnchor()}.{r[0]}'
|
||||
if float(r[6]) > 1.5 * total_runs:
|
||||
# FIXME should be 15s max -- investigate parallel_insert
|
||||
slow_average_tests += 1
|
||||
attrs[6] = f'style="background: {color_bad}"'
|
||||
errors_explained.append([f'<a href="./all-queries.html#all-query-times.{r[0]}.0">The test \'{r[0]}\' is too slow to run as a whole. Investigate whether the create and fill queries can be sped up'])
|
||||
errors_explained.append([f'<a href="#{anchor}">The test \'{r[0]}\' is too slow to run as a whole. Investigate whether the create and fill queries can be sped up'])
|
||||
else:
|
||||
attrs[6] = ''
|
||||
|
||||
@ -495,7 +490,7 @@ if args.report == 'main':
|
||||
else:
|
||||
attrs[5] = ''
|
||||
|
||||
text += tableRow(r, attrs)
|
||||
text += tableRow(r, attrs, anchor)
|
||||
|
||||
text += tableEnd()
|
||||
tables.append(text)
|
||||
@ -652,7 +647,7 @@ elif args.report == 'all-queries':
|
||||
# Unstable #1
|
||||
'Old, s', #2
|
||||
'New, s', #3
|
||||
'Times speedup / slowdown', #4
|
||||
'Ratio of speedup (-) or slowdown (+)', #4
|
||||
'Relative difference (new − old) / old', #5
|
||||
'p < 0.001 threshold', #6
|
||||
'Test', #7
|
||||
|
||||
@ -12,8 +12,8 @@ RUN apt-get update --yes \
|
||||
strace \
|
||||
--yes --no-install-recommends
|
||||
|
||||
#RUN wget -q -O - http://files.viva64.com/etc/pubkey.txt | sudo apt-key add -
|
||||
#RUN sudo wget -O /etc/apt/sources.list.d/viva64.list http://files.viva64.com/etc/viva64.list
|
||||
#RUN wget -nv -O - http://files.viva64.com/etc/pubkey.txt | sudo apt-key add -
|
||||
#RUN sudo wget -nv -O /etc/apt/sources.list.d/viva64.list http://files.viva64.com/etc/viva64.list
|
||||
#
|
||||
#RUN apt-get --allow-unauthenticated update -y \
|
||||
# && env DEBIAN_FRONTEND=noninteractive \
|
||||
@ -24,10 +24,10 @@ ENV PKG_VERSION="pvs-studio-latest"
|
||||
|
||||
RUN set -x \
|
||||
&& export PUBKEY_HASHSUM="486a0694c7f92e96190bbfac01c3b5ac2cb7823981db510a28f744c99eabbbf17a7bcee53ca42dc6d84d4323c2742761" \
|
||||
&& wget https://files.viva64.com/etc/pubkey.txt -O /tmp/pubkey.txt \
|
||||
&& wget -nv https://files.viva64.com/etc/pubkey.txt -O /tmp/pubkey.txt \
|
||||
&& echo "${PUBKEY_HASHSUM} /tmp/pubkey.txt" | sha384sum -c \
|
||||
&& apt-key add /tmp/pubkey.txt \
|
||||
&& wget "https://files.viva64.com/${PKG_VERSION}.deb" \
|
||||
&& wget -nv "https://files.viva64.com/${PKG_VERSION}.deb" \
|
||||
&& { debsig-verify ${PKG_VERSION}.deb \
|
||||
|| echo "WARNING: Some file was just downloaded from the internet without any validation and we are installing it into the system"; } \
|
||||
&& dpkg -i "${PKG_VERSION}.deb"
|
||||
|
||||
@ -26,7 +26,7 @@ RUN apt-get update -y \
|
||||
zookeeperd
|
||||
|
||||
RUN mkdir -p /tmp/clickhouse-odbc-tmp \
|
||||
&& wget --quiet -O - ${odbc_driver_url} | tar --strip-components=1 -xz -C /tmp/clickhouse-odbc-tmp \
|
||||
&& wget -nv -O - ${odbc_driver_url} | tar --strip-components=1 -xz -C /tmp/clickhouse-odbc-tmp \
|
||||
&& cp /tmp/clickhouse-odbc-tmp/lib64/*.so /usr/local/lib/ \
|
||||
&& odbcinst -i -d -f /tmp/clickhouse-odbc-tmp/share/doc/clickhouse-odbc/config/odbcinst.ini.sample \
|
||||
&& odbcinst -i -s -l -f /tmp/clickhouse-odbc-tmp/share/doc/clickhouse-odbc/config/odbc.ini.sample \
|
||||
|
||||
@ -71,7 +71,7 @@ RUN apt-get --allow-unauthenticated update -y \
|
||||
zookeeperd
|
||||
|
||||
RUN mkdir -p /tmp/clickhouse-odbc-tmp \
|
||||
&& wget --quiet -O - ${odbc_driver_url} | tar --strip-components=1 -xz -C /tmp/clickhouse-odbc-tmp \
|
||||
&& wget -nv -O - ${odbc_driver_url} | tar --strip-components=1 -xz -C /tmp/clickhouse-odbc-tmp \
|
||||
&& cp /tmp/clickhouse-odbc-tmp/lib64/*.so /usr/local/lib/ \
|
||||
&& odbcinst -i -d -f /tmp/clickhouse-odbc-tmp/share/doc/clickhouse-odbc/config/odbcinst.ini.sample \
|
||||
&& odbcinst -i -s -l -f /tmp/clickhouse-odbc-tmp/share/doc/clickhouse-odbc/config/odbc.ini.sample \
|
||||
|
||||
@ -33,7 +33,7 @@ RUN apt-get update -y \
|
||||
qemu-user-static
|
||||
|
||||
RUN mkdir -p /tmp/clickhouse-odbc-tmp \
|
||||
&& wget --quiet -O - ${odbc_driver_url} | tar --strip-components=1 -xz -C /tmp/clickhouse-odbc-tmp \
|
||||
&& wget -nv -O - ${odbc_driver_url} | tar --strip-components=1 -xz -C /tmp/clickhouse-odbc-tmp \
|
||||
&& cp /tmp/clickhouse-odbc-tmp/lib64/*.so /usr/local/lib/ \
|
||||
&& odbcinst -i -d -f /tmp/clickhouse-odbc-tmp/share/doc/clickhouse-odbc/config/odbcinst.ini.sample \
|
||||
&& odbcinst -i -s -l -f /tmp/clickhouse-odbc-tmp/share/doc/clickhouse-odbc/config/odbc.ini.sample \
|
||||
|
||||
@ -44,7 +44,7 @@ RUN set -eux; \
|
||||
\
|
||||
# this "case" statement is generated via "update.sh"
|
||||
\
|
||||
if ! wget -O docker.tgz "https://download.docker.com/linux/static/${DOCKER_CHANNEL}/x86_64/docker-${DOCKER_VERSION}.tgz"; then \
|
||||
if ! wget -nv -O docker.tgz "https://download.docker.com/linux/static/${DOCKER_CHANNEL}/x86_64/docker-${DOCKER_VERSION}.tgz"; then \
|
||||
echo >&2 "error: failed to download 'docker-${DOCKER_VERSION}' from '${DOCKER_CHANNEL}' for '${x86_64}'"; \
|
||||
exit 1; \
|
||||
fi; \
|
||||
|
||||
@ -7,7 +7,7 @@ toc_title: RabbitMQ
|
||||
|
||||
This engine allows integrating ClickHouse with [RabbitMQ](https://www.rabbitmq.com).
|
||||
|
||||
RabbitMQ lets you:
|
||||
`RabbitMQ` lets you:
|
||||
|
||||
- Publish or subscribe to data flows.
|
||||
- Process streams as they become available.
|
||||
@ -44,8 +44,8 @@ Optional parameters:
|
||||
- `rabbitmq_routing_key_list` – A comma-separated list of routing keys.
|
||||
- `rabbitmq_row_delimiter` – Delimiter character, which ends the message.
|
||||
- `rabbitmq_num_consumers` – The number of consumers per table. Default: `1`. Specify more consumers if the throughput of one consumer is insufficient.
|
||||
- `rabbitmq_num_queues` – The number of queues per consumer. Default: `1`. Specify more queues if the capacity of one queue per consumer is insufficient. Single queue can contain up to 50K messages at the same time.
|
||||
- `rabbitmq_transactional_channel` – Wrap insert queries in transactions. Default: `0`.
|
||||
- `rabbitmq_num_queues` – The number of queues per consumer. Default: `1`. Specify more queues if the capacity of one queue per consumer is insufficient. A single queue can contain up to 50K messages at the same time.
|
||||
- `rabbitmq_transactional_channel` – Wrap `INSERT` queries in transactions. Default: `0`.
|
||||
|
||||
Required configuration:
|
||||
|
||||
@ -72,7 +72,7 @@ Example:
|
||||
|
||||
## Description {#description}
|
||||
|
||||
`SELECT` is not particularly useful for reading messages (except for debugging), because each message can be read only once. It is more practical to create real-time threads using materialized views. To do this:
|
||||
`SELECT` is not particularly useful for reading messages (except for debugging), because each message can be read only once. It is more practical to create real-time threads using [materialized views](../../../sql-reference/statements/create/view.md). To do this:
|
||||
|
||||
1. Use the engine to create a RabbitMQ consumer and consider it a data stream.
|
||||
2. Create a table with the desired structure.
|
||||
@ -86,13 +86,13 @@ There can be no more than one exchange per table. One exchange can be shared bet
|
||||
|
||||
Exchange type options:
|
||||
|
||||
- `direct` - Routing is based on exact matching of keys. Example table key list: `key1,key2,key3,key4,key5`, message key can eqaul any of them.
|
||||
- `direct` - Routing is based on the exact matching of keys. Example table key list: `key1,key2,key3,key4,key5`, message key can equal any of them.
|
||||
- `fanout` - Routing to all tables (where exchange name is the same) regardless of the keys.
|
||||
- `topic` - Routing is based on patterns with dot-separated keys. Examples: `*.logs`, `records.*.*.2020`, `*.2018,*.2019,*.2020`.
|
||||
- `headers` - Routing is based on `key=value` matches with a setting `x-match=all` or `x-match=any`. Example table key list: `x-match=all,format=logs,type=report,year=2020`.
|
||||
- `consistent-hash` - Data is evenly distributed between all bound tables (where exchange name is the same). Note that this exchange type must be enabled with RabbitMQ plugin: `rabbitmq-plugins enable rabbitmq_consistent_hash_exchange`.
|
||||
- `consistent-hash` - Data is evenly distributed between all bound tables (where the exchange name is the same). Note that this exchange type must be enabled with RabbitMQ plugin: `rabbitmq-plugins enable rabbitmq_consistent_hash_exchange`.
|
||||
|
||||
If exchange type is not specified, then default is `fanout` and routing keys for data publishing must be randomized in range `[1, num_consumers]` for every message/batch (or in range `[1, num_consumers * num_queues]` if `rabbitmq_num_queues` is set). This table configuration works quicker then any other, especially when `rabbitmq_num_consumers` and/or `rabbitmq_num_queues` parameters are set.
|
||||
If exchange type is not specified, then default is `fanout` and routing keys for data publishing must be randomized in range `[1, num_consumers]` for every message/batch (or in range `[1, num_consumers * num_queues]` if `rabbitmq_num_queues` is set). This table configuration works quicker than any other, especially when `rabbitmq_num_consumers` and/or `rabbitmq_num_queues` parameters are set.
|
||||
|
||||
If `rabbitmq_num_consumers` and/or `rabbitmq_num_queues` parameters are specified along with `rabbitmq_exchange_type`, then:
|
||||
|
||||
|
||||
@ -31,7 +31,7 @@ For a description of request parameters, see [statement description](../../../sq
|
||||
|
||||
**ReplacingMergeTree Parameters**
|
||||
|
||||
- `ver` — column with version. Type `UInt*`, `Date`, `DateTime` or `DateTime64`. Optional parameter.
|
||||
- `ver` — column with version. Type `UInt*`, `Date` or `DateTime`. Optional parameter.
|
||||
|
||||
When merging, `ReplacingMergeTree` from all the rows with the same sorting key leaves only one:
|
||||
|
||||
|
||||
@ -121,7 +121,7 @@ To find out why we need two rows for each change, see [Algorithm](#table_engines
|
||||
|
||||
**Notes on Usage**
|
||||
|
||||
1. The program that writes the data should remember the state of an object in order to cancel it. The “cancel” string should be a copy of the “state” string with the opposite `Sign`. This increases the initial size of storage but allows to write the data quickly.
|
||||
1. The program that writes the data should remember the state of an object to be able to cancel it. “Cancel” string should contain copies of the primary key fields and the version of the “state” string and the opposite `Sign`. It increases the initial size of storage but allows to write the data quickly.
|
||||
2. Long growing arrays in columns reduce the efficiency of the engine due to the load for writing. The more straightforward the data, the better the efficiency.
|
||||
3. `SELECT` results depend strongly on the consistency of the history of object changes. Be accurate when preparing data for inserting. You can get unpredictable results with inconsistent data, such as negative values for non-negative metrics like session depth.
|
||||
|
||||
|
||||
@ -36,7 +36,7 @@ Examples:
|
||||
$ curl 'http://localhost:8123/?query=SELECT%201'
|
||||
1
|
||||
|
||||
$ wget -O- -q 'http://localhost:8123/?query=SELECT 1'
|
||||
$ wget -nv -O- 'http://localhost:8123/?query=SELECT 1'
|
||||
1
|
||||
|
||||
$ echo -ne 'GET /?query=SELECT%201 HTTP/1.0\r\n\r\n' | nc localhost 8123
|
||||
|
||||
@ -1290,6 +1290,47 @@ Possible values:
|
||||
|
||||
Default value: 0.
|
||||
|
||||
## distributed\_group\_by\_no\_merge {#distributed-group-by-no-merge}
|
||||
|
||||
Do not merge aggregation states from different servers for distributed query processing, you can use this in case it is for certain that there are different keys on different shards
|
||||
|
||||
Possible values:
|
||||
|
||||
- 0 — Disabled (final query processing is done on the initiator node).
|
||||
- 1 - Do not merge aggregation states from different servers for distributed query processing (query completelly processed on the shard, initiator only proxy the data).
|
||||
- 2 - Same as 1 but apply `ORDER BY` and `LIMIT` on the initiator (can be used for queries with `ORDER BY` and/or `LIMIT`).
|
||||
|
||||
**Example**
|
||||
|
||||
```sql
|
||||
SELECT *
|
||||
FROM remote('127.0.0.{2,3}', system.one)
|
||||
GROUP BY dummy
|
||||
LIMIT 1
|
||||
SETTINGS distributed_group_by_no_merge = 1
|
||||
FORMAT PrettyCompactMonoBlock
|
||||
|
||||
┌─dummy─┐
|
||||
│ 0 │
|
||||
│ 0 │
|
||||
└───────┘
|
||||
```
|
||||
|
||||
```sql
|
||||
SELECT *
|
||||
FROM remote('127.0.0.{2,3}', system.one)
|
||||
GROUP BY dummy
|
||||
LIMIT 1
|
||||
SETTINGS distributed_group_by_no_merge = 2
|
||||
FORMAT PrettyCompactMonoBlock
|
||||
|
||||
┌─dummy─┐
|
||||
│ 0 │
|
||||
└───────┘
|
||||
```
|
||||
|
||||
Default value: 0
|
||||
|
||||
## optimize\_skip\_unused\_shards {#optimize-skip-unused-shards}
|
||||
|
||||
Enables or disables skipping of unused shards for [SELECT](../../sql-reference/statements/select/index.md) queries that have sharding key condition in `WHERE/PREWHERE` (assuming that the data is distributed by sharding key, otherwise does nothing).
|
||||
@ -1337,6 +1378,40 @@ Possible values:
|
||||
|
||||
Default value: 0
|
||||
|
||||
## optimize\_distributed\_group\_by\_sharding\_key {#optimize-distributed-group-by-sharding-key}
|
||||
|
||||
Optimize `GROUP BY sharding_key` queries, by avoiding costly aggregation on the initiator server (which will reduce memory usage for the query on the initiator server).
|
||||
|
||||
The following types of queries are supported (and all combinations of them):
|
||||
|
||||
- `SELECT DISTINCT [..., ]sharding_key[, ...] FROM dist`
|
||||
- `SELECT ... FROM dist GROUP BY sharding_key[, ...]`
|
||||
- `SELECT ... FROM dist GROUP BY sharding_key[, ...] ORDER BY x`
|
||||
- `SELECT ... FROM dist GROUP BY sharding_key[, ...] LIMIT 1`
|
||||
- `SELECT ... FROM dist GROUP BY sharding_key[, ...] LIMIT 1 BY x`
|
||||
|
||||
The following types of queries are not supported (support for some of them may be added later):
|
||||
|
||||
- `SELECT ... GROUP BY sharding_key[, ...] WITH TOTALS`
|
||||
- `SELECT ... GROUP BY sharding_key[, ...] WITH ROLLUP`
|
||||
- `SELECT ... GROUP BY sharding_key[, ...] WITH CUBE`
|
||||
- `SELECT ... GROUP BY sharding_key[, ...] SETTINGS extremes=1`
|
||||
|
||||
Possible values:
|
||||
|
||||
- 0 — Disabled.
|
||||
- 1 — Enabled.
|
||||
|
||||
Default value: 0
|
||||
|
||||
See also:
|
||||
|
||||
- [distributed\_group\_by\_no\_merge](#distributed-group-by-no-merge)
|
||||
- [optimize\_skip\_unused\_shards](#optimize-skip-unused-shards)
|
||||
|
||||
!!! note "Note"
|
||||
Right now it requires `optimize_skip_unused_shards` (the reason behind this is that one day it may be enabled by default, and it will work correctly only if data was inserted via Distributed table, i.e. data is distributed according to sharding_key).
|
||||
|
||||
## optimize\_throw\_if\_noop {#setting-optimize_throw_if_noop}
|
||||
|
||||
Enables or disables throwing an exception if an [OPTIMIZE](../../sql-reference/statements/misc.md#misc_operations-optimize) query didn’t perform a merge.
|
||||
@ -1894,9 +1969,9 @@ Locking timeout is used to protect from deadlocks while executing read/write ope
|
||||
|
||||
Possible values:
|
||||
|
||||
- Positive integer.
|
||||
- Positive integer (in seconds).
|
||||
- 0 — No locking timeout.
|
||||
|
||||
Default value: `120`.
|
||||
Default value: `120` seconds.
|
||||
|
||||
[Original article](https://clickhouse.tech/docs/en/operations/settings/settings/) <!-- hide -->
|
||||
|
||||
@ -33,7 +33,7 @@ Para obtener una descripción de los parámetros de solicitud, consulte [descrip
|
||||
|
||||
**ReplacingMergeTree Parámetros**
|
||||
|
||||
- `ver` — column with version. Type `UInt*`, `Date`, `DateTime` o `DateTime64`. Parámetro opcional.
|
||||
- `ver` — column with version. Type `UInt*`, `Date` o `DateTime`. Parámetro opcional.
|
||||
|
||||
Al fusionar, `ReplacingMergeTree` de todas las filas con la misma clave primaria deja solo una:
|
||||
|
||||
|
||||
@ -38,7 +38,7 @@ Ejemplos:
|
||||
$ curl 'http://localhost:8123/?query=SELECT%201'
|
||||
1
|
||||
|
||||
$ wget -O- -q 'http://localhost:8123/?query=SELECT 1'
|
||||
$ wget -nv -O- 'http://localhost:8123/?query=SELECT 1'
|
||||
1
|
||||
|
||||
$ echo -ne 'GET /?query=SELECT%201 HTTP/1.0\r\n\r\n' | nc localhost 8123
|
||||
|
||||
@ -33,7 +33,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
|
||||
|
||||
**پارامترهای جایگزین**
|
||||
|
||||
- `ver` — column with version. Type `UInt*`, `Date`, `DateTime` یا `DateTime64`. پارامتر اختیاری.
|
||||
- `ver` — column with version. Type `UInt*`, `Date` یا `DateTime`. پارامتر اختیاری.
|
||||
|
||||
هنگام ادغام, `ReplacingMergeTree` از تمام ردیف ها با همان کلید اصلی تنها یک برگ دارد:
|
||||
|
||||
|
||||
@ -38,7 +38,7 @@ Ok.
|
||||
$ curl 'http://localhost:8123/?query=SELECT%201'
|
||||
1
|
||||
|
||||
$ wget -O- -q 'http://localhost:8123/?query=SELECT 1'
|
||||
$ wget -nv -O- 'http://localhost:8123/?query=SELECT 1'
|
||||
1
|
||||
|
||||
$ echo -ne 'GET /?query=SELECT%201 HTTP/1.0\r\n\r\n' | nc localhost 8123
|
||||
|
||||
@ -33,7 +33,7 @@ Pour une description des paramètres de requête, voir [demande de description](
|
||||
|
||||
**ReplacingMergeTree Paramètres**
|
||||
|
||||
- `ver` — column with version. Type `UInt*`, `Date`, `DateTime` ou `DateTime64`. Paramètre facultatif.
|
||||
- `ver` — column with version. Type `UInt*`, `Date` ou `DateTime`. Paramètre facultatif.
|
||||
|
||||
Lors de la fusion, `ReplacingMergeTree` de toutes les lignes avec la même clé primaire ne laisse qu'un:
|
||||
|
||||
|
||||
@ -38,7 +38,7 @@ Exemple:
|
||||
$ curl 'http://localhost:8123/?query=SELECT%201'
|
||||
1
|
||||
|
||||
$ wget -O- -q 'http://localhost:8123/?query=SELECT 1'
|
||||
$ wget -nv -O- 'http://localhost:8123/?query=SELECT 1'
|
||||
1
|
||||
|
||||
$ echo -ne 'GET /?query=SELECT%201 HTTP/1.0\r\n\r\n' | nc localhost 8123
|
||||
|
||||
@ -33,7 +33,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
|
||||
|
||||
**ReplacingMergeTreeパラメータ**
|
||||
|
||||
- `ver` — column with version. Type `UInt*`, `Date`, `DateTime` または `DateTime64`. 任意パラメータ。
|
||||
- `ver` — column with version. Type `UInt*`, `Date` または `DateTime`. 任意パラメータ。
|
||||
|
||||
マージ時, `ReplacingMergeTree` 同じ主キーを持つすべての行から、一つだけを残します:
|
||||
|
||||
|
||||
@ -38,7 +38,7 @@ GETメソッドを使用する場合, ‘readonly’ 設定されています。
|
||||
$ curl 'http://localhost:8123/?query=SELECT%201'
|
||||
1
|
||||
|
||||
$ wget -O- -q 'http://localhost:8123/?query=SELECT 1'
|
||||
$ wget -nv -O- 'http://localhost:8123/?query=SELECT 1'
|
||||
1
|
||||
|
||||
$ echo -ne 'GET /?query=SELECT%201 HTTP/1.0\r\n\r\n' | nc localhost 8123
|
||||
|
||||
@ -1,14 +1,11 @@
|
||||
---
|
||||
machine_translated: true
|
||||
machine_translated_rev: 72537a2d527c63c07aa5d2361a8829f3895cf2bd
|
||||
toc_priority: 47
|
||||
toc_title: "\u65E5\u4ED8"
|
||||
---
|
||||
|
||||
# 日付 {#date}
|
||||
|
||||
デートだ 1970-01-01(符号なし)以降の日数として二バイト単位で格納されます。 Unixエポックの開始直後から、コンパイル段階で定数によって定義される上限しきい値までの値を格納できます（現在は2106年までですが、完全にサポート
|
||||
最小値は1970-01-01として出力されます。
|
||||
日付型です。 1970-01-01 からの日数が2バイトの符号なし整数として格納されます。 UNIX時間の開始直後から、変換段階で定数として定義される上限しきい値までの値を格納できます（現在は2106年までですが、一年分を完全にサポートしているのは2105年までです）。
|
||||
|
||||
日付値は、タイムゾーンなしで格納されます。
|
||||
|
||||
|
||||
122
docs/ru/engines/table-engines/integrations/rabbitmq.md
Normal file
122
docs/ru/engines/table-engines/integrations/rabbitmq.md
Normal file
@ -0,0 +1,122 @@
|
||||
---
|
||||
toc_priority: 6
|
||||
toc_title: RabbitMQ
|
||||
---
|
||||
|
||||
# RabbitMQ {#rabbitmq-engine}
|
||||
|
||||
Движок работает с [RabbitMQ](https://www.rabbitmq.com).
|
||||
|
||||
`RabbitMQ` позволяет:
|
||||
|
||||
- Публиковать/подписываться на потоки данных.
|
||||
- Обрабатывать потоки по мере их появления.
|
||||
|
||||
## Создание таблицы {#table_engine-rabbitmq-creating-a-table}
|
||||
|
||||
``` sql
|
||||
CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
|
||||
(
|
||||
name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1],
|
||||
name2 [type2] [DEFAULT|MATERIALIZED|ALIAS expr2],
|
||||
...
|
||||
) ENGINE = RabbitMQ SETTINGS
|
||||
rabbitmq_host_port = 'host:port',
|
||||
rabbitmq_exchange_name = 'exchange_name',
|
||||
rabbitmq_format = 'data_format'[,]
|
||||
[rabbitmq_exchange_type = 'exchange_type',]
|
||||
[rabbitmq_routing_key_list = 'key1,key2,...',]
|
||||
[rabbitmq_row_delimiter = 'delimiter_symbol',]
|
||||
[rabbitmq_num_consumers = N,]
|
||||
[rabbitmq_num_queues = N,]
|
||||
[rabbitmq_transactional_channel = 0]
|
||||
```
|
||||
|
||||
Обязательные параметры:
|
||||
|
||||
- `rabbitmq_host_port` – адрес сервера (`хост:порт`). Например: `localhost:5672`.
|
||||
- `rabbitmq_exchange_name` – имя точки обмена в RabbitMQ.
|
||||
- `rabbitmq_format` – формат сообщения. Используется такое же обозначение, как и в функции `FORMAT` в SQL, например, `JSONEachRow`. Подробнее см. в разделе [Форматы входных и выходных данных](../../../interfaces/formats.md).
|
||||
|
||||
Дополнительные параметры:
|
||||
|
||||
- `rabbitmq_exchange_type` – тип точки обмена в RabbitMQ: `direct`, `fanout`, `topic`, `headers`, `consistent-hash`. По умолчанию: `fanout`.
|
||||
- `rabbitmq_routing_key_list` – список ключей маршрутизации, через запятую.
|
||||
- `rabbitmq_row_delimiter` – символ-разделитель, который завершает сообщение.
|
||||
- `rabbitmq_num_consumers` – количество потребителей на таблицу. По умолчанию: `1`. Укажите больше потребителей, если пропускная способность одного потребителя недостаточна.
|
||||
- `rabbitmq_num_queues` – количество очередей на потребителя. По умолчанию: `1`. Укажите больше потребителей, если пропускная способность одной очереди на потребителя недостаточна. Одна очередь поддерживает до 50 тысяч сообщений одновременно.
|
||||
- `rabbitmq_transactional_channel` – обернутые запросы `INSERT` в транзакциях. По умолчанию: `0`.
|
||||
|
||||
Требуемая конфигурация:
|
||||
|
||||
Конфигурация сервера RabbitMQ добавляется с помощью конфигурационного файла ClickHouse.
|
||||
|
||||
``` xml
|
||||
<rabbitmq>
|
||||
<username>root</username>
|
||||
<password>clickhouse</password>
|
||||
</rabbitmq>
|
||||
```
|
||||
|
||||
Example:
|
||||
|
||||
``` sql
|
||||
CREATE TABLE queue (
|
||||
key UInt64,
|
||||
value UInt64
|
||||
) ENGINE = RabbitMQ SETTINGS rabbitmq_host_port = 'localhost:5672',
|
||||
rabbitmq_exchange_name = 'exchange1',
|
||||
rabbitmq_format = 'JSONEachRow',
|
||||
rabbitmq_num_consumers = 5;
|
||||
```
|
||||
|
||||
## Описание {#description}
|
||||
|
||||
Запрос `SELECT` не очень полезен для чтения сообщений (за исключением отладки), поскольку каждое сообщение может быть прочитано только один раз. Практичнее создавать потоки реального времени с помощью [материализованных преставлений](../../../sql-reference/statements/create/view.md). Для этого:
|
||||
|
||||
1. Создайте потребителя RabbitMQ с помощью движка и рассматривайте его как поток данных.
|
||||
2. Создайте таблицу с необходимой структурой.
|
||||
3. Создайте материализованное представление, которое преобразует данные от движка и помещает их в ранее созданную таблицу.
|
||||
|
||||
Когда к движку присоединяется материализованное представление, оно начинает в фоновом режиме собирать данные. Это позволяет непрерывно получать сообщения от RabbitMQ и преобразовывать их в необходимый формат с помощью `SELECT`.
|
||||
У одной таблицы RabbitMQ может быть неограниченное количество материализованных представлений.
|
||||
|
||||
Данные передаются с помощью параметров `rabbitmq_exchange_type` и `rabbitmq_routing_key_list`.
|
||||
Может быть не более одной точки обмена на таблицу. Одна точка обмена может использоваться несколькими таблицами: это позволяет выполнять маршрутизацию по нескольким таблицам одновременно.
|
||||
|
||||
Параметры точек обмена:
|
||||
|
||||
- `direct` - маршрутизация основана на точном совпадении ключей. Пример списка ключей: `key1,key2,key3,key4,key5`. Ключ сообщения может совпадать с одним из них.
|
||||
- `fanout` - маршрутизация по всем таблицам, где имя точки обмена совпадает, независимо от ключей.
|
||||
- `topic` - маршрутизация основана на правилах с ключами, разделенными точками. Например: `*.logs`, `records.*.*.2020`, `*.2018,*.2019,*.2020`.
|
||||
- `headers` - маршрутизация основана на совпадении `key=value` с настройкой `x-match=all` или `x-match=any`. Пример списка ключей таблицы: `x-match=all,format=logs,type=report,year=2020`.
|
||||
- `consistent-hash` - данные равномерно распределяются между всеми связанными таблицами, где имя точки обмена совпадает. Обратите внимание, что этот тип обмена должен быть включен с помощью плагина RabbitMQ: `rabbitmq-plugins enable rabbitmq_consistent_hash_exchange`.
|
||||
|
||||
Если тип точки обмена не задан, по умолчанию используется `fanout`. В таком случае ключи маршрутизации для публикации данных должны быть рандомизированы в диапазоне `[1, num_consumers]` за каждое сообщение/пакет (или в диапазоне `[1, num_consumers * num_queues]`, если `rabbitmq_num_queues` задано). Эта конфигурация таблицы работает быстрее, чем любая другая, особенно когда заданы параметры `rabbitmq_num_consumers` и/или `rabbitmq_num_queues`.
|
||||
|
||||
Если параметры`rabbitmq_num_consumers` и/или `rabbitmq_num_queues` заданы вместе с параметром `rabbitmq_exchange_type`:
|
||||
|
||||
- плагин `rabbitmq-consistent-hash-exchange` должен быть включен.
|
||||
- свойство `message_id` должно быть определено (уникальное для каждого сообщения/пакета).
|
||||
|
||||
Пример:
|
||||
|
||||
``` sql
|
||||
CREATE TABLE queue (
|
||||
key UInt64,
|
||||
value UInt64
|
||||
) ENGINE = RabbitMQ SETTINGS rabbitmq_host_port = 'localhost:5672',
|
||||
rabbitmq_exchange_name = 'exchange1',
|
||||
rabbitmq_exchange_type = 'headers',
|
||||
rabbitmq_routing_key_list = 'format=logs,type=report,year=2020',
|
||||
rabbitmq_format = 'JSONEachRow',
|
||||
rabbitmq_num_consumers = 5;
|
||||
|
||||
CREATE TABLE daily (key UInt64, value UInt64)
|
||||
ENGINE = MergeTree();
|
||||
|
||||
CREATE MATERIALIZED VIEW consumer TO daily
|
||||
AS SELECT key, value FROM queue;
|
||||
|
||||
SELECT key, value FROM daily ORDER BY key;
|
||||
```
|
||||
@ -25,7 +25,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
|
||||
|
||||
**Параметры ReplacingMergeTree**
|
||||
|
||||
- `ver` — столбец с версией, тип `UInt*`, `Date`, `DateTime` или `DateTime64`. Необязательный параметр.
|
||||
- `ver` — столбец с версией, тип `UInt*`, `Date` или `DateTime`. Необязательный параметр.
|
||||
|
||||
При слиянии, из всех строк с одинаковым значением ключа сортировки `ReplacingMergeTree` оставляет только одну:
|
||||
|
||||
|
||||
@ -116,7 +116,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
|
||||
|
||||
**Примечания по использованию**
|
||||
|
||||
1. Программа, которая записывает данные, должна помнить состояние объекта, чтобы иметь возможность отменить его. Строка отмены состояния должна быть копией предыдущей строки состояния с противоположным значением `Sign`. Это увеличивает начальный размер хранилища, но позволяет быстро записывать данные.
|
||||
1. Программа, которая записывает данные, должна помнить состояние объекта, чтобы иметь возможность отменить его. Строка отмены состояния должна содержать копии полей первичного ключа и копию версии строки состояния и противоположное значение `Sign`. Это увеличивает начальный размер хранилища, но позволяет быстро записывать данные.
|
||||
2. Длинные растущие массивы в столбцах снижают эффективность работы движка за счёт нагрузки на запись. Чем проще данные, тем выше эффективность.
|
||||
3. `SELECT` результаты сильно зависят от согласованности истории изменений объекта. Будьте точны при подготовке данных для вставки. Вы можете получить непредсказуемые результаты с несогласованными данными, такими как отрицательные значения для неотрицательных метрик, таких как глубина сеанса.
|
||||
|
||||
|
||||
@ -31,7 +31,7 @@ Ok.
|
||||
$ curl 'http://localhost:8123/?query=SELECT%201'
|
||||
1
|
||||
|
||||
$ wget -O- -q 'http://localhost:8123/?query=SELECT 1'
|
||||
$ wget -nv -O- 'http://localhost:8123/?query=SELECT 1'
|
||||
1
|
||||
|
||||
$ echo -ne 'GET /?query=SELECT%201 HTTP/1.0\r\n\r\n' | nc localhost 8123
|
||||
|
||||
@ -401,13 +401,91 @@ INSERT INTO test VALUES (lower('Hello')), (lower('world')), (lower('INSERT')), (
|
||||
|
||||
Устанавливает тип поведения [JOIN](../../sql-reference/statements/select/join.md). При объединении таблиц могут появиться пустые ячейки. ClickHouse заполняет их по-разному в зависимости от настроек.
|
||||
|
||||
Возможные значения
|
||||
Возможные значения:
|
||||
|
||||
- 0 — пустые ячейки заполняются значением по умолчанию соответствующего типа поля.
|
||||
- 1 — `JOIN` ведёт себя как в стандартном SQL. Тип соответствующего поля преобразуется в [Nullable](../../sql-reference/data-types/nullable.md#data_type-nullable), а пустые ячейки заполняются значениями [NULL](../../sql-reference/syntax.md).
|
||||
|
||||
## partial_merge_join_optimizations {#partial_merge_join_optimizations}
|
||||
|
||||
Отключает все оптимизации для запросов [JOIN](../../sql-reference/statements/select/join.md) с частичным MergeJoin алгоритмом.
|
||||
|
||||
По умолчанию оптимизации включены, что может привести к неправильным результатам. Если вы видите подозрительные результаты в своих запросах, отключите оптимизацию с помощью этого параметра. В различных версиях сервера ClickHouse, оптимизация может отличаться.
|
||||
|
||||
Возможные значения:
|
||||
|
||||
- 0 — Оптимизация отключена.
|
||||
- 1 — Оптимизация включена.
|
||||
|
||||
Значение по умолчанию: 1.
|
||||
|
||||
## partial_merge_join_rows_in_right_blocks {#partial_merge_join_rows_in_right_blocks}
|
||||
|
||||
Устанавливает предельные размеры блоков данных «правого» соединения, для запросов [JOIN](../../sql-reference/statements/select/join.md) с частичным MergeJoin алгоритмом.
|
||||
|
||||
Сервер ClickHouse:
|
||||
|
||||
1. Разделяет данные правого соединения на блоки с заданным числом строк.
|
||||
2. Индексирует для каждого блока минимальное и максимальное значение.
|
||||
3. Выгружает подготовленные блоки на диск, если это возможно.
|
||||
|
||||
Возможные значения:
|
||||
|
||||
- Положительное целое число. Рекомендуемый диапазон значений [1000, 100000].
|
||||
|
||||
Значение по умолчанию: 65536.
|
||||
|
||||
## join_on_disk_max_files_to_merge {#join_on_disk_max_files_to_merge}
|
||||
|
||||
Устанавливет количество файлов, разрешенных для параллельной сортировки, при выполнении операций MergeJoin на диске.
|
||||
|
||||
Чем больше значение параметра, тем больше оперативной памяти используется и тем меньше используется диск (I/O).
|
||||
|
||||
Возможные значения:
|
||||
|
||||
- Положительное целое число, больше 2.
|
||||
|
||||
Значение по умолчанию: 64.
|
||||
|
||||
## temporary_files_codec {#temporary_files_codec}
|
||||
|
||||
Устанавливает метод сжатия для временных файлов на диске, используемых при сортировки и объединения.
|
||||
|
||||
Возможные значения:
|
||||
|
||||
- LZ4 — применять сжатие, используя алгоритм [LZ4](https://ru.wikipedia.org/wiki/LZ4)
|
||||
- NONE — не применять сжатие.
|
||||
|
||||
Значение по умолчанию: LZ4.
|
||||
|
||||
## any_join_distinct_right_table_keys {#any_join_distinct_right_table_keys}
|
||||
|
||||
Включает устаревшее поведение сервера ClickHouse при выполнении операций `ANY INNER|LEFT JOIN`.
|
||||
|
||||
!!! note "Внимание"
|
||||
Используйте этот параметр только в целях обратной совместимости, если ваши варианты использования требуют устаревшего поведения `JOIN`.
|
||||
|
||||
Когда включено устаревшее поведение:
|
||||
|
||||
- Результаты операций "t1 ANY LEFT JOIN t2" и "t2 ANY RIGHT JOIN t1" не равны, поскольку ClickHouse использует логику с сопоставлением ключей таблицы "многие к одному слева направо".
|
||||
- Результаты операций `ANY INNER JOIN` содержат все строки из левой таблицы, аналогично операции `SEMI LEFT JOIN`.
|
||||
|
||||
Когда устаревшее поведение отключено:
|
||||
|
||||
- Результаты операций `t1 ANY LEFT JOIN t2` и `t2 ANY RIGHT JOIN t1` равно, потому что ClickHouse использует логику сопоставления ключей один-ко-многим в операциях `ANY RIGHT JOIN`.
|
||||
- Результаты операций `ANY INNER JOIN` содержат по одной строке на ключ из левой и правой таблиц.
|
||||
|
||||
Возможные значения:
|
||||
|
||||
- 0 — Устаревшее поведение отключено.
|
||||
- 1 — Устаревшее поведение включено.
|
||||
|
||||
Значение по умолчанию: 0.
|
||||
|
||||
См. также:
|
||||
|
||||
- [JOIN strictness](../../sql-reference/statements/select/join.md#select-join-strictness)
|
||||
|
||||
## max\_block\_size {#setting-max_block_size}
|
||||
|
||||
Данные в ClickHouse обрабатываются по блокам (наборам кусочков столбцов). Внутренние циклы обработки для одного блока достаточно эффективны, но есть заметные издержки на каждый блок. Настройка `max_block_size` — это рекомендация, какой размер блока (в количестве строк) загружать из таблиц. Размер блока не должен быть слишком маленьким, чтобы затраты на каждый блок были заметны, но не слишком велики, чтобы запрос с LIMIT, который завершается после первого блока, обрабатывался быстро. Цель состоит в том, чтобы не использовалось слишком много оперативки при вынимании большого количества столбцов в несколько потоков; чтобы оставалась хоть какая-нибудь кэш-локальность.
|
||||
@ -1678,4 +1756,17 @@ SELECT idx, i FROM null_in WHERE i IN (1, NULL) SETTINGS transform_null_in = 1;
|
||||
- [Секции и настройки запроса CREATE TABLE](../../engines/table-engines/mergetree-family/mergetree.md#mergetree-query-clauses) (настройка `merge_with_ttl_timeout`)
|
||||
- [Table TTL](../../engines/table-engines/mergetree-family/mergetree.md#mergetree-table-ttl)
|
||||
|
||||
## lock_acquire_timeout {#lock_acquire_timeout}
|
||||
|
||||
Устанавливает, сколько секунд сервер ожидает возможности выполнить блокировку таблицы.
|
||||
|
||||
Таймаут устанавливается для защиты от взаимоблокировки при выполнении операций чтения или записи. Если время ожидания истекло, а блокировку выполнить не удалось, сервер возвращает исключение с кодом `DEADLOCK_AVOIDED` и сообщением "Locking attempt timed out! Possible deadlock avoided. Client should retry." ("Время ожидания блокировки истекло! Возможная взаимоблокировка предотвращена. Повторите запрос.").
|
||||
|
||||
Возможные значения:
|
||||
|
||||
- Положительное целое число (в секундах).
|
||||
- 0 — таймаут не устанавливается.
|
||||
|
||||
Значение по умолчанию: `120` секунд.
|
||||
|
||||
[Оригинальная статья](https://clickhouse.tech/docs/ru/operations/settings/settings/) <!--hide-->
|
||||
|
||||
@ -24,7 +24,7 @@
|
||||
Утилиту следует запускать вручную следующим образом:
|
||||
|
||||
``` bash
|
||||
$ clickhouse-copier copier --daemon --config zookeeper.xml --task-path /task/path --base-dir /path/to/dir
|
||||
$ clickhouse-copier --daemon --config zookeeper.xml --task-path /task/path --base-dir /path/to/dir
|
||||
```
|
||||
|
||||
Параметры запуска:
|
||||
|
||||
@ -37,6 +37,8 @@ FROM <left_table>
|
||||
!!! note "Примечание"
|
||||
Значение строгости по умолчанию может быть переопределено с помощью настройки [join\_default\_strictness](../../../operations/settings/settings.md#settings-join_default_strictness).
|
||||
|
||||
Поведение сервера ClickHouse для операций `ANY JOIN` зависит от параметра [any_join_distinct_right_table_keys](../../../operations/settings/settings.md#any_join_distinct_right_table_keys).
|
||||
|
||||
### Использование ASOF JOIN {#asof-join-usage}
|
||||
|
||||
`ASOF JOIN` применим в том случае, когда необходимо объединять записи, которые не имеют точного совпадения.
|
||||
|
||||
@ -180,10 +180,11 @@ def build(args):
|
||||
if not args.skip_website:
|
||||
website.build_website(args)
|
||||
|
||||
if not args.skip_test_templates:
|
||||
test.test_templates(args.website_dir)
|
||||
|
||||
if not args.skip_docs:
|
||||
build_docs(args)
|
||||
|
||||
from github import build_releases
|
||||
build_releases(args, build_docs)
|
||||
|
||||
@ -220,6 +221,8 @@ if __name__ == '__main__':
|
||||
arg_parser.add_argument('--skip-website', action='store_true')
|
||||
arg_parser.add_argument('--skip-blog', action='store_true')
|
||||
arg_parser.add_argument('--skip-git-log', action='store_true')
|
||||
arg_parser.add_argument('--skip-docs', action='store_true')
|
||||
arg_parser.add_argument('--skip-test-templates', action='store_true')
|
||||
arg_parser.add_argument('--test-only', action='store_true')
|
||||
arg_parser.add_argument('--minify', action='store_true')
|
||||
arg_parser.add_argument('--htmlproofer', action='store_true')
|
||||
|
||||
@ -33,7 +33,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
|
||||
|
||||
**ReplacingMergeTree Parametreleri**
|
||||
|
||||
- `ver` — column with version. Type `UInt*`, `Date`, `DateTime` veya `DateTime64`. İsteğe bağlı parametre.
|
||||
- `ver` — column with version. Type `UInt*`, `Date` veya `DateTime`. İsteğe bağlı parametre.
|
||||
|
||||
Birleş whenirken, `ReplacingMergeTree` aynı birincil anahtara sahip tüm satırlardan sadece bir tane bırakır:
|
||||
|
||||
|
||||
@ -38,7 +38,7 @@ GET yöntemini kullanırken, ‘readonly’ ayar .lanmıştır. Başka bir deyi
|
||||
$ curl 'http://localhost:8123/?query=SELECT%201'
|
||||
1
|
||||
|
||||
$ wget -O- -q 'http://localhost:8123/?query=SELECT 1'
|
||||
$ wget -nv -O- 'http://localhost:8123/?query=SELECT 1'
|
||||
1
|
||||
|
||||
$ echo -ne 'GET /?query=SELECT%201 HTTP/1.0\r\n\r\n' | nc localhost 8123
|
||||
|
||||
@ -2,44 +2,47 @@
|
||||
|
||||
Clickhouse 中最强大的表引擎当属 `MergeTree` （合并树）引擎及该系列（`*MergeTree`）中的其他引擎。
|
||||
|
||||
`MergeTree` 引擎系列的基本理念如下。当你有巨量数据要插入到表中，你要高效地一批批写入数据片段，并希望这些数据片段在后台按照一定规则合并。相比在插入时不断修改（重写）数据进存储，这种策略会高效很多。
|
||||
`MergeTree` 系列的引擎被设计用于插入极大量的数据到一张表当中。数据可以以数据片段的形式一个接着一个的快速写入，数据片段在后台按照一定的规则进行合并。相比在插入时不断修改（重写）已存储的数据，这种策略会高效很多。
|
||||
|
||||
主要特点:
|
||||
|
||||
- 存储的数据按主键排序。
|
||||
|
||||
这让你可以创建一个用于快速检索数据的小稀疏索引。
|
||||
这使得你能够创建一个小型的稀疏索引来加快数据检索。
|
||||
|
||||
- 允许使用分区，如果指定了 [分区键](custom-partitioning-key.md) 的话。
|
||||
- 支持数据分区，如果指定了 [分区键](custom-partitioning-key.md) 的话。
|
||||
|
||||
在相同数据集和相同结果集的情况下 ClickHouse 中某些带分区的操作会比普通操作更快。查询中指定了分区键时 ClickHouse 会自动截取分区数据。这也有效增加了查询性能。
|
||||
|
||||
- 支持数据副本。
|
||||
|
||||
`ReplicatedMergeTree` 系列的表便是用于此。更多信息，请参阅 [数据副本](replication.md) 一节。
|
||||
`ReplicatedMergeTree` 系列的表提供了数据副本功能。更多信息，请参阅 [数据副本](replication.md) 一节。
|
||||
|
||||
- 支持数据采样。
|
||||
|
||||
需要的话，你可以给表设置一个采样方法。
|
||||
|
||||
!!! 注意 "注意"
|
||||
!!! note "注意"
|
||||
[合并](../special/merge.md#merge) 引擎并不属于 `*MergeTree` 系列。
|
||||
|
||||
## 建表 {#table_engine-mergetree-creating-a-table}
|
||||
|
||||
``` sql
|
||||
CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
|
||||
(
|
||||
name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1],
|
||||
name2 [type2] [DEFAULT|MATERIALIZED|ALIAS expr2],
|
||||
name1 [type1] [DEFAULT|MATERIALIZED|ALIAS expr1] [TTL expr1],
|
||||
name2 [type2] [DEFAULT|MATERIALIZED|ALIAS expr2] [TTL expr2],
|
||||
...
|
||||
INDEX index_name1 expr1 TYPE type1(...) GRANULARITY value1,
|
||||
INDEX index_name2 expr2 TYPE type2(...) GRANULARITY value2
|
||||
) ENGINE = MergeTree()
|
||||
ORDER BY expr
|
||||
[PARTITION BY expr]
|
||||
[ORDER BY expr]
|
||||
[PRIMARY KEY expr]
|
||||
[SAMPLE BY expr]
|
||||
[TTL expr [DELETE|TO DISK 'xxx'|TO VOLUME 'xxx'], ...]
|
||||
[SETTINGS name=value, ...]
|
||||
```
|
||||
|
||||
对于以上参数的描述，可参考 [CREATE 语句 的描述](../../../engines/table-engines/mergetree-family/mergetree.md) 。
|
||||
|
||||
@ -62,7 +65,7 @@ Clickhouse 中最强大的表引擎当属 `MergeTree` （合并树）引擎及
|
||||
|
||||
要按月分区，可以使用表达式 `toYYYYMM(date_column)` ，这里的 `date_column` 是一个 [Date](../../../engines/table-engines/mergetree-family/mergetree.md) 类型的列。分区名的格式会是 `"YYYYMM"` 。
|
||||
|
||||
- `PRIMARY KEY` - 主键，如果要设成 [跟排序键不相同](#xuan-ze-gen-pai-xu-jian-bu-yi-yang-zhu-jian)，可选。
|
||||
- `PRIMARY KEY` - 主键，如果要 [选择与排序键不同的主键](#choosing-a-primary-key-that-differs-from-the-sorting-key)，可选。
|
||||
|
||||
默认情况下主键跟排序键（由 `ORDER BY` 子句指定）相同。
|
||||
因此，大部分情况下不需要再专门指定一个 `PRIMARY KEY` 子句。
|
||||
@ -72,17 +75,19 @@ Clickhouse 中最强大的表引擎当属 `MergeTree` （合并树）引擎及
|
||||
如果要用抽样表达式，主键中必须包含这个表达式。例如：
|
||||
`SAMPLE BY intHash32(UserID) ORDER BY (CounterID, EventDate, intHash32(UserID))` 。
|
||||
|
||||
- TTL 指定行存储的持续时间并定义 PART 在硬盘和卷上的移动逻辑的规则列表，可选。
|
||||
- TTL 指定行存储的持续时间并定义数据片段在硬盘和卷上的移动逻辑的规则列表，可选。
|
||||
|
||||
表达式中必须存在至少一个 `Date` 或 `DateTime` 类型的列，比如：
|
||||
|
||||
`TTL date + INTERVAl 1 DAY`
|
||||
|
||||
规则的类型 `DELETE|TO DISK 'xxx'|TO VOLUME 'xxx'`指定了当满足条件（到达当前时间）时所要执行的动作：移除过期的行，还是将 PART （如果PART中的所有行都满足表达式的话）移动到指定的磁盘（`TO DISK 'xxx'`) 或 卷（`TO VOLUME 'xxx'`）。默认的规则是移除（`DELETE`）。可以在列表中指定多个规则，但最多只能有一个`DELETE`的规则。
|
||||
规则的类型 `DELETE|TO DISK 'xxx'|TO VOLUME 'xxx'`指定了当满足条件（到达指定时间）时所要执行的动作：移除过期的行，还是将数据片段（如果数据片段中的所有行都满足表达式的话）移动到指定的磁盘（`TO DISK 'xxx'`) 或 卷（`TO VOLUME 'xxx'`）。默认的规则是移除（`DELETE`）。可以在列表中指定多个规则，但最多只能有一个`DELETE`的规则。
|
||||
|
||||
- `SETTINGS` — 影响 `MergeTree` 性能的额外参数：
|
||||
更多细节，请查看 [表和列的 TTL](#table_engine-mergetree-ttl)
|
||||
|
||||
- `index_granularity` — 索引粒度。索引中相邻的『标记』间的数据行数。默认值，8192 。参考[Data Storage](#mergetree-data-storage)。
|
||||
- `SETTINGS` — 控制 `MergeTree` 行为的额外参数：
|
||||
|
||||
- `index_granularity` — 索引粒度。索引中相邻的『标记』间的数据行数。默认值，8192 。参考[数据存储](#mergetree-data-storage)。
|
||||
- `index_granularity_bytes` — 索引粒度，以字节为单位，默认值: 10Mb。如果想要仅按数据行数限制索引粒度, 请设置为0(不建议)。
|
||||
- `enable_mixed_granularity_parts` — 是否启用通过 `index_granularity_bytes` 控制索引粒度的大小。在19.11版本之前, 只有 `index_granularity` 配置能够用于限制索引粒度的大小。当从具有很大的行（几十上百兆字节）的表中查询数据时候，`index_granularity_bytes` 配置能够提升ClickHouse的性能。如果你的表里有很大的行，可以开启这项配置来提升`SELECT` 查询的性能。
|
||||
- `use_minimalistic_part_header_in_zookeeper` — 是否在 ZooKeeper 中启用最小的数据片段头 。如果设置了 `use_minimalistic_part_header_in_zookeeper=1` ，ZooKeeper 会存储更少的数据。更多信息参考『服务配置参数』这章中的 [设置描述](../../../operations/server-configuration-parameters/settings.md#server-settings-use_minimalistic_part_header_in_zookeeper) 。
|
||||
@ -90,18 +95,21 @@ Clickhouse 中最强大的表引擎当属 `MergeTree` （合并树）引擎及
|
||||
<a name="mergetree_setting-merge_with_ttl_timeout"></a>
|
||||
- `merge_with_ttl_timeout` — TTL合并频率的最小间隔时间，单位：秒。默认值: 86400 (1 天)。
|
||||
- `write_final_mark` — 是否启用在数据片段尾部写入最终索引标记。默认值: 1（不建议更改）。
|
||||
- `storage_policy` — 存储策略。 参见 [使用多个区块装置进行数据存储](#table_engine-mergetree-multiple-volumes).
|
||||
- `min_bytes_for_wide_part`,`min_rows_for_wide_part` 在数据分段中可以使用`Wide`格式进行存储的最小字节数/行数。你可以不设置、只设置一个，或全都设置。参考：[Data Storage](#mergetree-data-storage)
|
||||
- `merge_max_block_size` — 在块中进行合并操作时的最大行数限制。默认值：8192
|
||||
- `storage_policy` — 存储策略。 参见 [使用具有多个块的设备进行数据存储](#table_engine-mergetree-multiple-volumes).
|
||||
- `min_bytes_for_wide_part`,`min_rows_for_wide_part` 在数据片段中可以使用`Wide`格式进行存储的最小字节数/行数。你可以不设置、只设置一个，或全都设置。参考：[数据存储](#mergetree-data-storage)
|
||||
|
||||
**示例配置**
|
||||
|
||||
``` sql
|
||||
ENGINE MergeTree() PARTITION BY toYYYYMM(EventDate) ORDER BY (CounterID, EventDate, intHash32(UserID)) SAMPLE BY intHash32(UserID) SETTINGS index_granularity=8192
|
||||
```
|
||||
|
||||
示例中，我们设为按月分区。
|
||||
在这个例子中，我们设置了按月进行分区。
|
||||
|
||||
同时我们设置了一个按用户ID哈希的抽样表达式。这让你可以有该表中每个 `CounterID` 和 `EventDate` 下面的数据的伪随机分布。如果你在查询时指定了 [SAMPLE](../../../engines/table-engines/mergetree-family/mergetree.md#select-sample-clause) 子句。 ClickHouse会返回对于用户子集的一个均匀的伪随机数据采样。
|
||||
同时我们设置了一个按用户 ID 哈希的抽样表达式。这使得你可以对该表中每个 `CounterID` 和 `EventDate` 的数据伪随机分布。如果你在查询时指定了 [SAMPLE](../../../engines/table-engines/mergetree-family/mergetree.md#select-sample-clause) 子句。 ClickHouse会返回对于用户子集的一个均匀的伪随机数据采样。
|
||||
|
||||
`index_granularity` 可省略，默认值为 8192 。
|
||||
`index_granularity` 可省略因为 8192 是默认设置 。
|
||||
|
||||
<details markdown="1">
|
||||
|
||||
@ -133,15 +141,20 @@ Clickhouse 中最强大的表引擎当属 `MergeTree` （合并树）引擎及
|
||||
|
||||
## 数据存储 {#mergetree-data-storage}
|
||||
|
||||
表由按主键排序的数据 *片段* 组成。
|
||||
表由按主键排序的数据片段（DATA PART）组成。
|
||||
|
||||
当数据被插入到表中时，会分成数据片段并按主键的字典序排序。例如，主键是 `(CounterID, Date)` 时，片段中数据按 `CounterID` 排序，具有相同 `CounterID` 的部分按 `Date` 排序。
|
||||
当数据被插入到表中时，会创建多个数据片段并按主键的字典序排序。例如，主键是 `(CounterID, Date)` 时，片段中数据首先按 `CounterID` 排序，具有相同 `CounterID` 的部分按 `Date` 排序。
|
||||
|
||||
不同分区的数据会被分成不同的片段，ClickHouse 在后台合并数据片段以便更高效存储。不会合并来自不同分区的数据片段。这个合并机制并不保证相同主键的所有行都会合并到同一个数据片段中。
|
||||
不同分区的数据会被分成不同的片段，ClickHouse 在后台合并数据片段以便更高效存储。不同分区的数据片段不会进行合并。合并机制并不保证具有相同主键的行全都合并到同一个数据片段中。
|
||||
|
||||
ClickHouse 会为每个数据片段创建一个索引文件，索引文件包含每个索引行（『标记』）的主键值。索引行号定义为 `n * index_granularity` 。最大的 `n` 等于总行数除以 `index_granularity` 的值的整数部分。对于每列，跟主键相同的索引行处也会写入『标记』。这些『标记』让你可以直接找到数据所在的列。
|
||||
数据片段可以以 `Wide` 或 `Compact` 格式存储。在 `Wide` 格式下，每一列都会在文件系统中存储为单独的文件，在 `Compact` 格式下所有列都存储在一个文件中。`Compact` 格式可以提高插入量少插入频率频繁时的性能。
|
||||
|
||||
你可以只用一单一大表并不断地一块块往里面加入数据 – `MergeTree` 引擎的就是为了这样的场景。
|
||||
数据存储格式由 `min_bytes_for_wide_part` 和 `min_rows_for_wide_part` 表引擎参数控制。如果数据片段中的字节数或行数少于相应的设置值，数据片段会以 `Compact` 格式存储，否则会以 `Wide` 格式存储。
|
||||
|
||||
每个数据片段被逻辑的分割成颗粒（granules）。颗粒是 ClickHouse 中进行数据查询时的最小不可分割数据集。ClickHouse 不会对行或值进行拆分，所以每个颗粒总是包含整数个行。每个颗粒的第一行通过该行的主键值进行标记，
|
||||
ClickHouse 会为每个数据片段创建一个索引文件来存储这些标记。对于每列，无论它是否包含在主键当中，ClickHouse 都会存储类似标记。这些标记让你可以在列文件中直接找到数据。
|
||||
|
||||
颗粒的大小通过表引擎参数 `index_granularity` 和 `index_granularity_bytes` 控制。取决于行的大小，颗粒的行数的在 `[1, index_granularity]` 范围中。如果单行的大小超过了 `index_granularity_bytes` 设置的值，那么一个颗粒的大小会超过 `index_granularity_bytes`。在这种情况下，颗粒的大小等于该行的大小。
|
||||
|
||||
## 主键和索引在查询中的表现 {#primary-keys-and-indexes-in-queries}
|
||||
|
||||
@ -162,56 +175,53 @@ ClickHouse 会为每个数据片段创建一个索引文件，索引文件包含
|
||||
|
||||
上面例子可以看出使用索引通常会比全表描述要高效。
|
||||
|
||||
稀疏索引会引起额外的数据读取。当读取主键单个区间范围的数据时，每个数据块中最多会多读 `index_granularity * 2` 行额外的数据。大部分情况下，当 `index_granularity = 8192` 时，ClickHouse的性能并不会降级。
|
||||
稀疏索引会引起额外的数据读取。当读取主键单个区间范围的数据时，每个数据块中最多会多读 `index_granularity * 2` 行额外的数据。
|
||||
|
||||
稀疏索引让你能操作有巨量行的表。因为这些索引是常驻内存（RAM）的。
|
||||
稀疏索引使得你可以处理极大量的行，因为大多数情况下，这些索引常驻与内存（RAM）中。
|
||||
|
||||
ClickHouse 不要求主键惟一。所以，你可以插入多条具有相同主键的行。
|
||||
ClickHouse 不要求主键惟一，所以你可以插入多条具有相同主键的行。
|
||||
|
||||
### 主键的选择 {#zhu-jian-de-xuan-ze}
|
||||
|
||||
主键中列的数量并没有明确的限制。依据数据结构，你应该让主键包含多些或少些列。这样可以：
|
||||
主键中列的数量并没有明确的限制。依据数据结构，你可以在主键包含多些或少些列。这样可以：
|
||||
|
||||
- 改善索引的性能。
|
||||
|
||||
如果当前主键是 `(a, b)` ，然后加入另一个 `c` 列，满足下面条件时，则可以改善性能：
|
||||
- 有带有 `c` 列条件的查询。
|
||||
- 很长的数据范围（ `index_granularity` 的数倍）里 `(a, b)` 都是相同的值，并且这种的情况很普遍。换言之，就是加入另一列后，可以让你的查询略过很长的数据范围。
|
||||
如果当前主键是 `(a, b)` ，在下列情况下添加另一个 `c` 列会提升性能：
|
||||
|
||||
- 查询会使用 `c` 列作为条件
|
||||
- 很长的数据范围（ `index_granularity` 的数倍）里 `(a, b)` 都是相同的值，并且这样的情况很普遍。换言之，就是加入另一列后，可以让你的查询略过很长的数据范围。
|
||||
|
||||
- 改善数据压缩。
|
||||
|
||||
ClickHouse 以主键排序片段数据，所以，数据的一致性越高，压缩越好。
|
||||
|
||||
- [折叠树](collapsingmergetree.md#table_engine-collapsingmergetree) 和 [SummingMergeTree](summingmergetree.md) 引擎里，数据合并时，会有额外的处理逻辑。
|
||||
- 在[CollapsingMergeTree](collapsingmergetree.md#table_engine-collapsingmergetree) 和 [SummingMergeTree](summingmergetree.md) 引擎里进行数据合并时会提供额外的处理逻辑。
|
||||
|
||||
在这种情况下，指定一个跟主键不同的 *排序键* 也是有意义的。
|
||||
在这种情况下，指定与主键不同的 *排序键* 也是有意义的。
|
||||
|
||||
长的主键会对插入性能和内存消耗有负面影响，但主键中额外的列并不影响 `SELECT` 查询的性能。
|
||||
|
||||
### 选择跟排序键不一样主键 {#xuan-ze-gen-pai-xu-jian-bu-yi-yang-zhu-jian}
|
||||
可以使用 `ORDER BY tuple()` 语法创建没有主键的表。在这种情况下 ClickHouse 根据数据插入的顺序存储。如果在使用 `INSERT ... SELECT` 时希望保持数据的排序，请设置 [max\_insert\_threads = 1](../../../operations/settings/settings.md#settings-max-insert-threads)。
|
||||
|
||||
指定一个跟排序键（用于排序数据片段中行的表达式）
|
||||
不一样的主键（用于计算写到索引文件的每个标记值的表达式）是可以的。
|
||||
这种情况下，主键表达式元组必须是排序键表达式元组的一个前缀。
|
||||
想要根据初始顺序进行数据查询，使用 [单线程查询](../../../operations/settings/settings.md#settings-max_threads)
|
||||
|
||||
当使用 [SummingMergeTree](summingmergetree.md) 和
|
||||
[AggregatingMergeTree](aggregatingmergetree.md) 引擎时，这个特性非常有用。
|
||||
通常，使用这类引擎时，表里列分两种：*维度* 和 *度量* 。
|
||||
典型的查询是在 `GROUP BY` 并过虑维度的情况下统计度量列的值。
|
||||
像 SummingMergeTree 和 AggregatingMergeTree ，用相同的排序键值统计行时，
|
||||
通常会加上所有的维度。结果就是，这键的表达式会是一长串的列组成，
|
||||
并且这组列还会因为新加维度必须频繁更新。
|
||||
### 选择与排序键不同主键 {#choosing-a-primary-key-that-differs-from-the-sorting-key}
|
||||
|
||||
这种情况下，主键中仅预留少量列保证高效范围扫描，
|
||||
剩下的维度列放到排序键元组里。这样是合理的。
|
||||
指定一个跟排序键不一样的主键是可以的，此时排序键用于在数据片段中进行排序，主键用于在索引文件中进行标记的写入。这种情况下，主键表达式元组必须是排序键表达式元组的前缀。
|
||||
|
||||
[排序键的修改](../../../engines/table-engines/mergetree-family/mergetree.md) 是轻量级的操作，因为一个新列同时被加入到表里和排序键后时，已存在的数据片段并不需要修改。由于旧的排序键是新排序键的前缀，并且刚刚添加的列中没有数据，因此在表修改时的数据对于新旧的排序键来说都是有序的。
|
||||
当使用 [SummingMergeTree](summingmergetree.md) 和 [AggregatingMergeTree](aggregatingmergetree.md) 引擎时，这个特性非常有用。通常在使用这类引擎时，表里的列分两种：*维度* 和 *度量* 。典型的查询会通过任意的 `GROUP BY` 对度量列进行聚合并通过维度列进行过滤。由于 SummingMergeTree 和 AggregatingMergeTree 会对排序键相同的行进行聚合，所以把所有的维度放进排序键是很自然的做法。但这将导致排序键中包含大量的列，并且排序键会伴随着新添加的维度不断的更新。
|
||||
|
||||
### 索引和分区在查询中的应用 {#suo-yin-he-fen-qu-zai-cha-xun-zhong-de-ying-yong}
|
||||
在这种情况下合理的做法是，只保留少量的列在主键当中用于提升扫描效率，将维度列添加到排序键中。
|
||||
|
||||
对于 `SELECT` 查询，ClickHouse 分析是否可以使用索引。如果 `WHERE/PREWHERE` 子句具有下面这些表达式（作为谓词链接一子项或整个）则可以使用索引：基于主键或分区键的列或表达式的部分的等式或比较运算表达式；基于主键或分区键的列或表达式的固定前缀的 `IN` 或 `LIKE` 表达式；基于主键或分区键的列的某些函数；基于主键或分区键的表达式的逻辑表达式。 <!-- It is too hard for me to translate this section as the original text completely. So I did it with my own understanding. If you have good idea, please help me. -->
|
||||
对排序键进行 [ALTER](../../../sql-reference/statements/alter.md) 是轻量级的操作，因为当一个新列同时被加入到表里和排序键里时，已存在的数据片段并不需要修改。由于旧的排序键是新排序键的前缀，并且新添加的列中没有数据，因此在表修改时的数据对于新旧的排序键来说都是有序的。
|
||||
|
||||
因此，在索引键的一个或多个区间上快速地跑查询都是可能的。下面例子中，指定标签；指定标签和日期范围；指定标签和日期；指定多个标签和日期范围等运行查询，都会非常快。
|
||||
### 索引和分区在查询中的应用 {#use-of-indexes-and-partitions-in-queries}
|
||||
|
||||
对于 `SELECT` 查询，ClickHouse 分析是否可以使用索引。如果 `WHERE/PREWHERE` 子句具有下面这些表达式（作为谓词链接一子项或整个）则可以使用索引：包含一个表示与主键/分区键中的部分字段或全部字段相等/不等的比较表达式；基于主键/分区键的字段上的 `IN` 或 固定前缀的`LIKE` 表达式；基于主键/分区键的字段上的某些函数；基于主键/分区键的表达式的逻辑表达式。 <!-- It is too hard for me to translate this section as the original text completely. So I did it with my own understanding. If you have good idea, please help me. -->
|
||||
<!-- It is hard for me to translate this section too, but I think change the sentence struct is helpful for understanding. So I change the phraseology-->
|
||||
|
||||
因此，在索引键的一个或多个区间上快速地执行查询都是可能的。下面例子中，指定标签；指定标签和日期范围；指定标签和日期；指定多个标签和日期范围等执行查询，都会非常快。
|
||||
|
||||
当引擎配置如下时：
|
||||
|
||||
@ -237,11 +247,18 @@ SELECT count() FROM table WHERE CounterID = 34 OR URL LIKE '%upyachka%'
|
||||
|
||||
要检查 ClickHouse 执行一个查询时能否使用索引，可设置 [force\_index\_by\_date](../../../operations/settings/settings.md#settings-force_index_by_date) 和 [force\_primary\_key](../../../operations/settings/settings.md) 。
|
||||
|
||||
按月分区的分区键是只能读取包含适当范围日期的数据块。这种情况下，数据块会包含很多天（最多整月）的数据。在块中，数据按主键排序，主键第一列可能不包含日期。因此，仅使用日期而没有带主键前缀条件的查询将会导致读取超过这个日期范围。
|
||||
按月分区的分区键是只能读取包含适当范围日期的数据块。这种情况下，数据块会包含很多天（最多整月）的数据。在块中，数据按主键排序，主键第一列可能不包含日期。因此，仅使用日期而没有带主键前几个字段作为条件的查询将会导致需要读取超过这个指定日期以外的数据。
|
||||
|
||||
### 跳数索引（分段汇总索引，实验性的） {#tiao-shu-suo-yin-fen-duan-hui-zong-suo-yin-shi-yan-xing-de}
|
||||
### 部分单调主键的使用
|
||||
|
||||
需要设置 `allow_experimental_data_skipping_indices` 为 1 才能使用此索引。（执行 `SET allow_experimental_data_skipping_indices = 1`）。
|
||||
考虑这样的场景，比如一个月中的几天。它们在一个月的范围内形成一个[单调序列](https://zh.wikipedia.org/wiki/单调函数) ，但如果扩展到更大的时间范围它们就不再单调了。这就是一个部分单调序列。如果用户使用部分单调的主键创建表，ClickHouse同样会创建一个稀疏索引。当用户从这类表中查询数据时，ClickHouse 会对查询条件进行分析。如果用户希望获取两个索引标记之间的数据并且这两个标记在一个月以内，ClickHouse 可以在这种特殊情况下使用到索引，因为它可以计算出查询参数与索引标记之间的距离。
|
||||
|
||||
如果查询参数范围内的主键不是单调序列，那么 ClickHouse 无法使用索引。在这种情况下，ClickHouse 会进行全表扫描。
|
||||
|
||||
ClickHouse 在任何主键代表一个部分单调序列的情况下都会使用这个逻辑。
|
||||
|
||||
|
||||
### 跳数索引 {#tiao-shu-suo-yin-fen-duan-hui-zong-suo-yin-shi-yan-xing-de}
|
||||
|
||||
此索引在 `CREATE` 语句的列部分里定义。
|
||||
|
||||
@ -249,12 +266,14 @@ SELECT count() FROM table WHERE CounterID = 34 OR URL LIKE '%upyachka%'
|
||||
INDEX index_name expr TYPE type(...) GRANULARITY granularity_value
|
||||
```
|
||||
|
||||
`*MergeTree` 系列的表都能指定跳数索引。
|
||||
`*MergeTree` 系列的表可以指定跳数索引。
|
||||
|
||||
这些索引是由数据块按粒度分割后的每部分在指定表达式上汇总信息 `granularity_value` 组成（粒度大小用表引擎里 `index_granularity` 的指定）。
|
||||
这些汇总信息有助于用 `where` 语句跳过大片不满足的数据，从而减少 `SELECT` 查询从磁盘读取的数据量，
|
||||
|
||||
示例
|
||||
这些索引会在数据块上聚合指定表达式的信息，这些信息以 granularity_value 指定的粒度组成 （粒度的大小通过在表引擎中定义 index_granularity 定义）。这些汇总信息有助于跳过大片不满足 `where` 条件的数据，从而减少 `SELECT` 查询从磁盘读取的数据量。
|
||||
|
||||
**示例**
|
||||
|
||||
``` sql
|
||||
CREATE TABLE table_name
|
||||
@ -282,18 +301,26 @@ SELECT count() FROM table WHERE u64 * i32 == 10 AND u64 * length(s) >= 1234
|
||||
存储指定表达式的极值（如果表达式是 `tuple` ，则存储 `tuple` 中每个元素的极值），这些信息用于跳过数据块，类似主键。
|
||||
|
||||
- `set(max_rows)`
|
||||
存储指定表达式的惟一值（不超过 `max_rows` 个，`max_rows=0` 则表示『无限制』）。这些信息可用于检查 `WHERE` 表达式是否满足某个数据块。
|
||||
存储指定表达式的不重复值（不超过 `max_rows` 个，`max_rows=0` 则表示『无限制』）。这些信息可用于检查 数据块是否满足 `WHERE` 条件。
|
||||
|
||||
- `ngrambf_v1(n, size_of_bloom_filter_in_bytes, number_of_hash_functions, random_seed)`
|
||||
存储包含数据块中所有 n 元短语的 [布隆过滤器](https://en.wikipedia.org/wiki/Bloom_filter) 。只可用在字符串上。
|
||||
存储一个包含数据块中所有 n元短语（ngram） 的 [布隆过滤器](https://en.wikipedia.org/wiki/Bloom_filter) 。只可用在字符串上。
|
||||
可用于优化 `equals` ， `like` 和 `in` 表达式的性能。
|
||||
`n` – 短语长度。
|
||||
`size_of_bloom_filter_in_bytes` – 布隆过滤器大小，单位字节。（因为压缩得好，可以指定比较大的值，如 256 或 512）。
|
||||
`number_of_hash_functions` – 布隆过滤器中使用的 hash 函数的个数。
|
||||
`random_seed` – hash 函数的随机种子。
|
||||
`number_of_hash_functions` – 布隆过滤器中使用的哈希函数的个数。
|
||||
`random_seed` – 哈希函数的随机种子。
|
||||
|
||||
- `tokenbf_v1(size_of_bloom_filter_in_bytes, number_of_hash_functions, random_seed)`
|
||||
跟 `ngrambf_v1` 类似，不同于 ngrams 存储字符串指定长度的所有片段。它只存储被非字母数据字符分割的片段。
|
||||
跟 `ngrambf_v1` 类似，不同于 ngrams 存储字符串指定长度的所有片段。它只存储被非字母数字字符分割的片段。
|
||||
|
||||
- `bloom_filter(bloom_filter([false_positive])` – 为指定的列存储布隆过滤器
|
||||
|
||||
可选的参数 false_positive 用来指定从布隆过滤器收到错误响应的几率。取值范围是 (0,1)，默认值：0.025
|
||||
|
||||
支持的数据类型：`Int*`, `UInt*`, `Float*`, `Enum`, `Date`, `DateTime`, `String`, `FixedString`, `Array`, `LowCardinality`, `Nullable`。
|
||||
|
||||
以下函数会用到这个索引： [equals](../../../sql-reference/functions/comparison-functions.md), [notEquals](../../../sql-reference/functions/comparison-functions.md), [in](../../../sql-reference/functions/in-functions.md), [notIn](../../../sql-reference/functions/in-functions.md), [has](../../../sql-reference/functions/array-functions.md)
|
||||
|
||||
<!-- -->
|
||||
|
||||
@ -303,7 +330,50 @@ INDEX sample_index2 (u64 * length(str), i32 + f64 * 100, date, str) TYPE set(100
|
||||
INDEX sample_index3 (lower(str), str) TYPE ngrambf_v1(3, 256, 2, 0) GRANULARITY 4
|
||||
```
|
||||
|
||||
## 并发数据访问 {#bing-fa-shu-ju-fang-wen}
|
||||
#### 函数支持 {#functions-support}
|
||||
|
||||
WHERE 子句中的条件包含对列的函数调用，如果列是索引的一部分，ClickHouse 会在执行函数时尝试使用索引。不同的函数对索引的支持是不同的。
|
||||
|
||||
`set` 索引会对所有函数生效，其他索引对函数的生效情况见下表
|
||||
|
||||
| 函数 (操作符) / 索引 | primary key | minmax | ngrambf\_v1 | tokenbf\_v1 | bloom\_filter |
|
||||
|------------------------------------------------------------------------------------------------------------|-------------|--------|-------------|-------------|---------------|
|
||||
| [equals (=, ==)](../../../sql-reference/functions/comparison-functions.md#function-equals) | ✔ | ✔ | ✔ | ✔ | ✔ |
|
||||
| [notEquals(!=, \<\>)](../../../sql-reference/functions/comparison-functions.md#function-notequals) | ✔ | ✔ | ✔ | ✔ | ✔ |
|
||||
| [like](../../../sql-reference/functions/string-search-functions.md#function-like) | ✔ | ✔ | ✔ | ✔ | ✔ |
|
||||
| [notLike](../../../sql-reference/functions/string-search-functions.md#function-notlike) | ✔ | ✔ | ✗ | ✗ | ✗ |
|
||||
| [startsWith](../../../sql-reference/functions/string-functions.md#startswith) | ✔ | ✔ | ✔ | ✔ | ✗ |
|
||||
| [endsWith](../../../sql-reference/functions/string-functions.md#endswith) | ✗ | ✗ | ✔ | ✔ | ✗ |
|
||||
| [multiSearchAny](../../../sql-reference/functions/string-search-functions.md#function-multisearchany) | ✗ | ✗ | ✔ | ✗ | ✗ |
|
||||
| [in](../../../sql-reference/functions/in-functions.md#in-functions) | ✔ | ✔ | ✔ | ✔ | ✔ |
|
||||
| [notIn](../../../sql-reference/functions/in-functions.md#in-functions) | ✔ | ✔ | ✔ | ✔ | ✔ |
|
||||
| [less (\<)](../../../sql-reference/functions/comparison-functions.md#function-less) | ✔ | ✔ | ✗ | ✗ | ✗ |
|
||||
| [greater (\>)](../../../sql-reference/functions/comparison-functions.md#function-greater) | ✔ | ✔ | ✗ | ✗ | ✗ |
|
||||
| [lessOrEquals (\<=)](../../../sql-reference/functions/comparison-functions.md#function-lessorequals) | ✔ | ✔ | ✗ | ✗ | ✗ |
|
||||
| [greaterOrEquals (\>=)](../../../sql-reference/functions/comparison-functions.md#function-greaterorequals) | ✔ | ✔ | ✗ | ✗ | ✗ |
|
||||
| [empty](../../../sql-reference/functions/array-functions.md#function-empty) | ✔ | ✔ | ✗ | ✗ | ✗ |
|
||||
| [notEmpty](../../../sql-reference/functions/array-functions.md#function-notempty) | ✔ | ✔ | ✗ | ✗ | ✗ |
|
||||
| hasToken | ✗ | ✗ | ✗ | ✔ | ✗ |
|
||||
|
||||
常量参数小于 ngram 大小的函数不能使用 `ngrambf_v1` 进行查询优化。
|
||||
|
||||
!!! note "注意"
|
||||
布隆过滤器可能会包含不符合条件的匹配，所以 `ngrambf_v1`, `tokenbf_v1` 和 `bloom_filter` 索引不能用于负向的函数，例如：
|
||||
|
||||
- 可以用来优化的场景
|
||||
- `s LIKE '%test%'`
|
||||
- `NOT s NOT LIKE '%test%'`
|
||||
- `s = 1`
|
||||
- `NOT s != 1`
|
||||
- `startsWith(s, 'test')`
|
||||
- 不能用来优化的场景
|
||||
- `NOT s LIKE '%test%'`
|
||||
- `s NOT LIKE '%test%'`
|
||||
- `NOT s = 1`
|
||||
- `s != 1`
|
||||
- `NOT startsWith(s, 'test')`
|
||||
|
||||
## 并发数据访问 {#concurrent-data-access}
|
||||
|
||||
应对表的并发访问，我们使用多版本机制。换言之，当同时读和更新表时，数据从当前查询到的一组片段中读取。没有冗长的的锁。插入不会阻碍读取。
|
||||
|
||||
@ -311,9 +381,11 @@ INDEX sample_index3 (lower(str), str) TYPE ngrambf_v1(3, 256, 2, 0) GRANULARITY
|
||||
|
||||
## 列和表的 TTL {#table_engine-mergetree-ttl}
|
||||
|
||||
TTL可以设置值的生命周期，它既可以为整张表设置，也可以为每个列字段单独设置。如果`TTL`同时作用于表和字段，ClickHouse会使用先到期的那个。
|
||||
TTL 可以设置值的生命周期，它既可以为整张表设置，也可以为每个列字段单独设置。表级别的 TTL 还会指定数据在磁盘和卷上自动转移的逻辑。
|
||||
|
||||
被设置TTL的表，必须拥有[日期](../../../engines/table-engines/mergetree-family/mergetree.md) 或 [日期时间](../../../engines/table-engines/mergetree-family/mergetree.md) 类型的字段。要定义数据的生命周期，需要在这个日期字段上使用操作符，例如:
|
||||
TTL 表达式的计算结果必须是 [日期](../../../engines/table-engines/mergetree-family/mergetree.md) 或 [日期时间](../../../engines/table-engines/mergetree-family/mergetree.md) 类型的字段。
|
||||
|
||||
示例：
|
||||
|
||||
``` sql
|
||||
TTL time_column
|
||||
@ -327,15 +399,15 @@ TTL date_time + INTERVAL 1 MONTH
|
||||
TTL date_time + INTERVAL 15 HOUR
|
||||
```
|
||||
|
||||
### 列字段 TTL {#mergetree-column-ttl}
|
||||
### 列 TTL {#mergetree-column-ttl}
|
||||
|
||||
当列字段中的值过期时, ClickHouse会将它们替换成数据类型的默认值。如果分区内，某一列的所有值均已过期，则ClickHouse会从文件系统中删除这个分区目录下的列文件。
|
||||
当列中的值过期时, ClickHouse会将它们替换成该列数据类型的默认值。如果数据片段中列的所有值均已过期，则ClickHouse 会从文件系统中的数据片段中此列。
|
||||
|
||||
`TTL`子句不能被用于主键字段。
|
||||
|
||||
示例说明:
|
||||
示例:
|
||||
|
||||
创建一张包含 `TTL` 的表
|
||||
创建表时指定 `TTL`
|
||||
|
||||
``` sql
|
||||
CREATE TABLE example_table
|
||||
@ -368,11 +440,21 @@ ALTER TABLE example_table
|
||||
|
||||
### 表 TTL {#mergetree-table-ttl}
|
||||
|
||||
当表内的数据过期时, ClickHouse会删除所有对应的行。
|
||||
表可以设置一个用于移除过期行的表达式，以及多个用于在磁盘或卷上自动转移数据片段的表达式。当表中的行过期时，ClickHouse 会删除所有对应的行。对于数据片段的转移特性，必须所有的行都满足转移条件。
|
||||
|
||||
举例说明:
|
||||
``` sql
|
||||
TTL expr [DELETE|TO DISK 'aaa'|TO VOLUME 'bbb'], ...
|
||||
```
|
||||
|
||||
创建一张包含 `TTL` 的表
|
||||
TTL 规则的类型紧跟在每个 TTL 表达式后面，它会影响满足表达式时（到达指定时间时）应当执行的操作：
|
||||
|
||||
- `DELETE` - 删除过期的行（默认操作）;
|
||||
- `TO DISK 'aaa'` - 将数据片段移动到磁盘 `aaa`;
|
||||
- `TO VOLUME 'bbb'` - 将数据片段移动到卷 `bbb`.
|
||||
|
||||
示例:
|
||||
|
||||
创建时指定 TTL
|
||||
|
||||
``` sql
|
||||
CREATE TABLE example_table
|
||||
@ -383,7 +465,9 @@ CREATE TABLE example_table
|
||||
ENGINE = MergeTree
|
||||
PARTITION BY toYYYYMM(d)
|
||||
ORDER BY d
|
||||
TTL d + INTERVAL 1 MONTH;
|
||||
TTL d + INTERVAL 1 MONTH [DELETE],
|
||||
d + INTERVAL 1 WEEK TO VOLUME 'aaa',
|
||||
d + INTERVAL 2 WEEK TO DISK 'bbb';
|
||||
```
|
||||
|
||||
修改表的 `TTL`
|
||||
@ -395,14 +479,179 @@ ALTER TABLE example_table
|
||||
|
||||
**删除数据**
|
||||
|
||||
当ClickHouse合并数据分区时, 会删除TTL过期的数据。
|
||||
ClickHouse 在数据片段合并时会删除掉过期的数据。
|
||||
|
||||
当ClickHouse发现数据过期时, 它将会执行一个计划外的合并。要控制这类合并的频率, 你可以设置 `merge_with_ttl_timeout`。如果该值被设置的太低, 它将导致执行许多的计划外合并，这可能会消耗大量资源。
|
||||
当ClickHouse发现数据过期时, 它将会执行一个计划外的合并。要控制这类合并的频率, 你可以设置 `merge_with_ttl_timeout`。如果该值被设置的太低, 它将引发大量计划外的合并，这可能会消耗大量资源。
|
||||
|
||||
如果在合并的时候执行`SELECT` 查询, 则可能会得到过期的数据。为了避免这种情况，可以在`SELECT`之前使用 [OPTIMIZE](../../../engines/table-engines/mergetree-family/mergetree.md#misc_operations-optimize) 查询。
|
||||
如果在合并的过程中执行 `SELECT` 查询, 则可能会得到过期的数据。为了避免这种情况，可以在 `SELECT` 之前使用 [OPTIMIZE](../../../engines/table-engines/mergetree-family/mergetree.md#misc_operations-optimize) 查询。
|
||||
|
||||
## 使用多个块设备进行数据存储 {#table_engine-mergetree-multiple-volumes}
|
||||
## 使用具有多个块的设备进行数据存储 {#table_engine-mergetree-multiple-volumes}
|
||||
|
||||
### 介绍 {#introduction}
|
||||
|
||||
MergeTree 系列表引擎可以将数据存储在多块设备上。这对某些可以潜在被划分为“冷”“热”的表来说是很有用的。近期数据被定期的查询但只需要很小的空间。相反，详尽的历史数据很少被用到。如果有多块磁盘可用，那么“热”的数据可以放置在快速的磁盘上（比如 NVMe 固态硬盘或内存），“冷”的数据可以放在相对较慢的磁盘上（比如机械硬盘）。
|
||||
|
||||
数据片段是 `MergeTree` 引擎表的最小可移动单元。属于同一个数据片段的数据被存储在同一块磁盘上。数据片段会在后台自动的在磁盘间移动，也可以通过 [ALTER](../../../sql-reference/statements/alter.md#alter_move-partition) 查询来移动。
|
||||
|
||||
### 术语 {#terms}
|
||||
|
||||
- 磁盘 — 挂载到文件系统的块设备
|
||||
- 默认磁盘 — 在服务器设置中通过 [path](../../../operations/server-configuration-parameters/settings.md#server_configuration_parameters-path) 参数指定的数据存储
|
||||
- 卷 — 磁盘的等效有序集合 （类似于 [JBOD](https://en.wikipedia.org/wiki/Non-RAID_drive_architectures)）
|
||||
- 存储策略 — 卷的集合及他们之间的数据移动规则
|
||||
|
||||
### 配置 {#table_engine-mergetree-multiple-volumes_configure}
|
||||
|
||||
[来源文章](https://clickhouse.tech/docs/en/operations/table_engines/mergetree/) <!--hide-->
|
||||
磁盘、卷和存储策略应当在主文件 `config.xml` 或 `config.d` 目录中的独立文件中的 `<storage_configuration>` 标签内定义。
|
||||
|
||||
配置结构：
|
||||
|
||||
``` xml
|
||||
<storage_configuration>
|
||||
<disks>
|
||||
<disk_name_1> <!-- disk name -->
|
||||
<path>/mnt/fast_ssd/clickhouse/</path>
|
||||
</disk_name_1>
|
||||
<disk_name_2>
|
||||
<path>/mnt/hdd1/clickhouse/</path>
|
||||
<keep_free_space_bytes>10485760</keep_free_space_bytes>
|
||||
</disk_name_2>
|
||||
<disk_name_3>
|
||||
<path>/mnt/hdd2/clickhouse/</path>
|
||||
<keep_free_space_bytes>10485760</keep_free_space_bytes>
|
||||
</disk_name_3>
|
||||
|
||||
...
|
||||
</disks>
|
||||
|
||||
...
|
||||
</storage_configuration>
|
||||
```
|
||||
|
||||
标签：
|
||||
|
||||
- `<disk_name_N>` — 磁盘名，名称必须与其他磁盘不同.
|
||||
- `path` — 服务器将用来存储数据 (`data` 和 `shadow` 目录) 的路径, 应当以 ‘/’ 结尾.
|
||||
- `keep_free_space_bytes` — 需要保留的剩余磁盘空间.
|
||||
|
||||
磁盘定义的顺序无关紧要。
|
||||
|
||||
存储策略配置：
|
||||
|
||||
``` xml
|
||||
<storage_configuration>
|
||||
...
|
||||
<policies>
|
||||
<policy_name_1>
|
||||
<volumes>
|
||||
<volume_name_1>
|
||||
<disk>disk_name_from_disks_configuration</disk>
|
||||
<max_data_part_size_bytes>1073741824</max_data_part_size_bytes>
|
||||
</volume_name_1>
|
||||
<volume_name_2>
|
||||
<!-- configuration -->
|
||||
</volume_name_2>
|
||||
<!-- more volumes -->
|
||||
</volumes>
|
||||
<move_factor>0.2</move_factor>
|
||||
</policy_name_1>
|
||||
<policy_name_2>
|
||||
<!-- configuration -->
|
||||
</policy_name_2>
|
||||
|
||||
<!-- more policies -->
|
||||
</policies>
|
||||
...
|
||||
</storage_configuration>
|
||||
```
|
||||
|
||||
标签：
|
||||
|
||||
- `policy_name_N` — 策略名称，不能重复。
|
||||
- `volume_name_N` — 卷名称，不能重复。
|
||||
- `disk` — 卷中的磁盘。
|
||||
- `max_data_part_size_bytes` — 任意卷上的磁盘可以存储的数据片段的最大大小。
|
||||
- `move_factor` — 当可用空间少于这个因子时，数据将自动的向下一个卷（如果有的话）移动 (默认值为 0.1)。
|
||||
|
||||
配置示例：
|
||||
|
||||
``` xml
|
||||
<storage_configuration>
|
||||
...
|
||||
<policies>
|
||||
<hdd_in_order> <!-- policy name -->
|
||||
<volumes>
|
||||
<single> <!-- volume name -->
|
||||
<disk>disk1</disk>
|
||||
<disk>disk2</disk>
|
||||
</single>
|
||||
</volumes>
|
||||
</hdd_in_order>
|
||||
|
||||
<moving_from_ssd_to_hdd>
|
||||
<volumes>
|
||||
<hot>
|
||||
<disk>fast_ssd</disk>
|
||||
<max_data_part_size_bytes>1073741824</max_data_part_size_bytes>
|
||||
</hot>
|
||||
<cold>
|
||||
<disk>disk1</disk>
|
||||
</cold>
|
||||
</volumes>
|
||||
<move_factor>0.2</move_factor>
|
||||
</moving_from_ssd_to_hdd>
|
||||
</policies>
|
||||
...
|
||||
</storage_configuration>
|
||||
```
|
||||
|
||||
在给出的例子中， `hdd_in_order` 策略实现了 [循环制](https://zh.wikipedia.org/wiki/循环制) 方法。因此这个策略只定义了一个卷（`single`），数据片段会以循环的顺序全部存储到它的磁盘上。当有多个类似的磁盘挂载到系统上，但没有配置 RAID 时，这种策略非常有用。请注意一个每个独立的磁盘驱动都并不可靠，你可能需要用 3 或更大的复制因此来补偿它。
|
||||
|
||||
如果在系统中有不同类型的磁盘可用，可以使用 `moving_from_ssd_to_hdd`。`hot` 卷由 SSD 磁盘（`fast_ssd`）组成，这个卷上可以存储的数据片段的最大大小为 1GB。所有大于 1GB 的数据片段都会被直接存储到 `cold` 卷上，`cold` 卷包含一个名为 `disk1` 的 HDD 磁盘。
|
||||
同样，一旦 `fast_ssd` 被填充超过 80%，数据会通过后台进程向 `disk1` 进行转移。
|
||||
|
||||
存储策略中卷的枚举顺序是很重要的。因为当一个卷被充满时，数据会向下一个卷转移。磁盘的枚举顺序同样重要，因为数据是依次存储在磁盘上的。
|
||||
|
||||
在创建表时，可以将一个配置好的策略应用到表：
|
||||
|
||||
``` sql
|
||||
CREATE TABLE table_with_non_default_policy (
|
||||
EventDate Date,
|
||||
OrderID UInt64,
|
||||
BannerID UInt64,
|
||||
SearchPhrase String
|
||||
) ENGINE = MergeTree
|
||||
ORDER BY (OrderID, BannerID)
|
||||
PARTITION BY toYYYYMM(EventDate)
|
||||
SETTINGS storage_policy = 'moving_from_ssd_to_hdd'
|
||||
```
|
||||
|
||||
`default` 存储策略意味着只使用一个卷，这个卷只包含一个在 `<path>` 中定义的磁盘。表创建后，它的存储策略就不能改变了。
|
||||
|
||||
可以通过 [background\_move\_pool\_size](../../../operations/settings/settings.md#background_move_pool_size) 设置调整执行后台任务的线程数。
|
||||
|
||||
### 详细说明 {#details}
|
||||
|
||||
对于 `MergeTree` 表，数据通过以下不同的方式写入到磁盘当中：
|
||||
|
||||
- 作为插入（`INSERT`查询）的结果
|
||||
- 在后台合并和[数据变异](../../../sql-reference/statements/alter.md#alter-mutations)期间
|
||||
- 当从另一个副本下载时
|
||||
- 作为 [ALTER TABLE … FREEZE PARTITION](../../../sql-reference/statements/alter.md#alter_freeze-partition) 冻结分区的结果
|
||||
|
||||
除了数据变异和冻结分区以外的情况下，数据按照以下逻辑存储到卷或磁盘上：
|
||||
|
||||
1. 首个卷（按定义顺序）拥有足够的磁盘空间存储数据片段（`unreserved_space > current_part_size`）并且允许存储给定数据片段的大小（`max_data_part_size_bytes > current_part_size`）
|
||||
2. 在这个数据卷内，紧挨着先前存储数据的那块磁盘之后的磁盘，拥有比数据片段大的剩余空间。（`unreserved_space - keep_free_space_bytes > current_part_size`）
|
||||
|
||||
更进一步，数据变异和分区冻结使用的是 [硬链接](https://en.wikipedia.org/wiki/Hard_link)。不同磁盘之间的硬链接是不支持的，所以在这种情况下数据片段都会被存储到初始化的那一块磁盘上。
|
||||
|
||||
在后台，数据片段基于剩余空间（`move_factor`参数）根据卷在配置文件中定义的顺序进行转移。数据永远不会从最后一个移出也不会从第一个移入。可以通过系统表 [system.part\_log](../../../operations/system-tables/part_log.md#system_tables-part-log) (字段 `type = MOVE_PART`) 和 [system.parts](../../../operations/system-tables/parts.md#system_tables-parts) (字段 `path` 和 `disk`) 来监控后台的移动情况。同时，具体细节可以通过服务器日志查看。
|
||||
|
||||
用户可以通过 [ALTER TABLE … MOVE PART\|PARTITION … TO VOLUME\|DISK …](../../../sql-reference/statements/alter.md#alter_move-partition) 强制移动一个数据片段或分区到另外一个卷，所有后台移动的限制都会被考虑在内。这个查询会自行启动，无需等待后台操作完成。如果没有足够的可用空间或任何必须条件没有被满足，用户会收到报错信息。
|
||||
|
||||
数据移动不会妨碍到数据复制。也就是说，同一张表的不同副本可以指定不同的存储策略。
|
||||
|
||||
在后台合并和数据变异之后，就的数据片段会在一定时间后被移除 (`old_parts_lifetime`)。在这期间，他们不能被移动到其他的卷或磁盘。也就是说，直到数据片段被完全移除，它们仍然会被磁盘占用空间计算在内。
|
||||
|
||||
[原始文章](https://clickhouse.tech/docs/en/operations/table_engines/mergetree/) <!--hide-->
|
||||
|
||||
@ -25,7 +25,7 @@ CREATE TABLE [IF NOT EXISTS] [db.]table_name [ON CLUSTER cluster]
|
||||
|
||||
**参数**
|
||||
|
||||
- `ver` — 版本列。类型为 `UInt*`, `Date`, `DateTime` 或 `DateTime64`。可选参数。
|
||||
- `ver` — 版本列。类型为 `UInt*`, `Date` 或 `DateTime`。可选参数。
|
||||
|
||||
合并的时候，`ReplacingMergeTree` 从所有具有相同主键的行中选择一行留下：
|
||||
- 如果 `ver` 列未指定，选择最后一条。
|
||||
|
||||
@ -23,7 +23,7 @@ Ok.
|
||||
$ curl 'http://localhost:8123/?query=SELECT%201'
|
||||
1
|
||||
|
||||
$ wget -O- -q 'http://localhost:8123/?query=SELECT 1'
|
||||
$ wget -nv -O- 'http://localhost:8123/?query=SELECT 1'
|
||||
1
|
||||
|
||||
$ GET 'http://localhost:8123/?query=SELECT 1'
|
||||
|
||||
@ -104,6 +104,8 @@ public:
|
||||
query_processing_stage = QueryProcessingStage::FetchColumns;
|
||||
else if (stage == "with_mergeable_state")
|
||||
query_processing_stage = QueryProcessingStage::WithMergeableState;
|
||||
else if (stage == "with_mergeable_state_after_aggregation")
|
||||
query_processing_stage = QueryProcessingStage::WithMergeableStateAfterAggregation;
|
||||
else
|
||||
throw Exception("Unknown query processing stage: " + stage, ErrorCodes::BAD_ARGUMENTS);
|
||||
|
||||
@ -565,7 +567,7 @@ int mainEntryClickHouseBenchmark(int argc, char ** argv)
|
||||
("help", "produce help message")
|
||||
("concurrency,c", value<unsigned>()->default_value(1), "number of parallel queries")
|
||||
("delay,d", value<double>()->default_value(1), "delay between intermediate reports in seconds (set 0 to disable reports)")
|
||||
("stage", value<std::string>()->default_value("complete"), "request query processing up to specified stage: complete,fetch_columns,with_mergeable_state")
|
||||
("stage", value<std::string>()->default_value("complete"), "request query processing up to specified stage: complete,fetch_columns,with_mergeable_state,with_mergeable_state_after_aggregation")
|
||||
("iterations,i", value<size_t>()->default_value(0), "amount of queries to be executed")
|
||||
("timelimit,t", value<double>()->default_value(0.), "stop launch of queries after specified time limit")
|
||||
("randomize,r", value<bool>()->default_value(false), "randomize order of execution")
|
||||
|
||||
@ -847,32 +847,26 @@ private:
|
||||
}
|
||||
|
||||
// Parse and execute what we've read.
|
||||
fprintf(stderr, "will now parse '%s'\n", text.c_str());
|
||||
|
||||
const auto * new_end = processWithFuzzing(text);
|
||||
|
||||
if (new_end > &text[0])
|
||||
{
|
||||
const auto rest_size = text.size() - (new_end - &text[0]);
|
||||
|
||||
fprintf(stderr, "total %zd, rest %zd\n", text.size(), rest_size);
|
||||
|
||||
memcpy(&text[0], new_end, rest_size);
|
||||
text.resize(rest_size);
|
||||
}
|
||||
else
|
||||
{
|
||||
fprintf(stderr, "total %zd, can't parse\n", text.size());
|
||||
// We didn't read enough text to parse a query. Will read more.
|
||||
}
|
||||
|
||||
if (!connection->isConnected())
|
||||
{
|
||||
// Uh-oh...
|
||||
std::cerr << "Lost connection to the server." << std::endl;
|
||||
last_exception_received_from_server
|
||||
= std::make_unique<Exception>(210, "~");
|
||||
return;
|
||||
}
|
||||
// Ensure that we're still connected to the server. If the server died,
|
||||
// the reconnect is going to fail with an exception, and the fuzzer
|
||||
// will exit. The ping() would be the best match here, but it's
|
||||
// private, probably for a good reason that the protocol doesn't allow
|
||||
// pings at any possible moment.
|
||||
connection->forceConnected(connection_parameters.timeouts);
|
||||
|
||||
if (text.size() > 4 * 1024)
|
||||
{
|
||||
@ -880,9 +874,6 @@ private:
|
||||
// and we still cannot parse a single query in it. Abort.
|
||||
std::cerr << "Read too much text and still can't parse a query."
|
||||
" Aborting." << std::endl;
|
||||
last_exception_received_from_server
|
||||
= std::make_unique<Exception>(1, "~");
|
||||
// return;
|
||||
exit(1);
|
||||
}
|
||||
}
|
||||
|
||||
@ -716,6 +716,7 @@ int Server::main(const std::vector<std::string> & /*args*/)
|
||||
{
|
||||
/// Disable DNS caching at all
|
||||
DNSResolver::instance().setDisableCacheFlag();
|
||||
LOG_DEBUG(log, "DNS caching disabled");
|
||||
}
|
||||
else
|
||||
{
|
||||
|
||||
@ -281,41 +281,33 @@ void AccessControlManager::addStoragesFromMainConfig(
|
||||
String config_dir = std::filesystem::path{config_path}.remove_filename().string();
|
||||
String dbms_dir = config.getString("path", DBMS_DEFAULT_PATH);
|
||||
String include_from_path = config.getString("include_from", "/etc/metrika.xml");
|
||||
bool has_user_directories = config.has("user_directories");
|
||||
|
||||
if (config.has("user_directories"))
|
||||
{
|
||||
if (config.has("users_config"))
|
||||
LOG_WARNING(getLogger(), "<user_directories> is specified, the path from <users_config> won't be used: " + config.getString("users_config"));
|
||||
if (config.has("access_control_path"))
|
||||
LOG_WARNING(getLogger(), "<access_control_path> is specified, the path from <access_control_path> won't be used: " + config.getString("access_control_path"));
|
||||
|
||||
addStoragesFromUserDirectoriesConfig(
|
||||
config,
|
||||
"user_directories",
|
||||
config_dir,
|
||||
dbms_dir,
|
||||
include_from_path,
|
||||
get_zookeeper_function);
|
||||
}
|
||||
else
|
||||
{
|
||||
/// If path to users' config isn't absolute, try guess its root (current) dir.
|
||||
/// At first, try to find it in dir of main config, after will use current dir.
|
||||
String users_config_path = config.getString("users_config", "");
|
||||
if (users_config_path.empty())
|
||||
{
|
||||
if (!has_user_directories)
|
||||
users_config_path = config_path;
|
||||
}
|
||||
else if (std::filesystem::path{users_config_path}.is_relative() && std::filesystem::exists(config_dir + users_config_path))
|
||||
users_config_path = config_dir + users_config_path;
|
||||
|
||||
if (!users_config_path.empty())
|
||||
{
|
||||
if (users_config_path != config_path)
|
||||
checkForUsersNotInMainConfig(config, config_path, users_config_path, getLogger());
|
||||
|
||||
addUsersConfigStorage(users_config_path, include_from_path, dbms_dir, get_zookeeper_function);
|
||||
}
|
||||
|
||||
String disk_storage_dir = config.getString("access_control_path", "");
|
||||
if (!disk_storage_dir.empty())
|
||||
addDiskStorage(disk_storage_dir);
|
||||
}
|
||||
|
||||
if (has_user_directories)
|
||||
addStoragesFromUserDirectoriesConfig(config, "user_directories", config_dir, dbms_dir, include_from_path, get_zookeeper_function);
|
||||
}
|
||||
|
||||
|
||||
|
||||
@ -47,7 +47,7 @@ class AccessControlManager : public MultipleAccessStorage
|
||||
{
|
||||
public:
|
||||
AccessControlManager();
|
||||
~AccessControlManager();
|
||||
~AccessControlManager() override;
|
||||
|
||||
/// Parses access entities from a configuration loaded from users.xml.
|
||||
/// This function add UsersConfigAccessStorage if it wasn't added before.
|
||||
|
||||
@ -96,6 +96,22 @@ public:
|
||||
/// Returns all the flags related to a dictionary.
|
||||
static AccessFlags allDictionaryFlags();
|
||||
|
||||
/// Returns all the flags which could be granted on the global level.
|
||||
/// The same as allFlags().
|
||||
static AccessFlags allFlagsGrantableOnGlobalLevel();
|
||||
|
||||
/// Returns all the flags which could be granted on the database level.
|
||||
/// Returns allDatabaseFlags() | allTableFlags() | allDictionaryFlags() | allColumnFlags().
|
||||
static AccessFlags allFlagsGrantableOnDatabaseLevel();
|
||||
|
||||
/// Returns all the flags which could be granted on the table level.
|
||||
/// Returns allTableFlags() | allDictionaryFlags() | allColumnFlags().
|
||||
static AccessFlags allFlagsGrantableOnTableLevel();
|
||||
|
||||
/// Returns all the flags which could be granted on the global level.
|
||||
/// The same as allColumnFlags().
|
||||
static AccessFlags allFlagsGrantableOnColumnLevel();
|
||||
|
||||
private:
|
||||
static constexpr size_t NUM_FLAGS = 128;
|
||||
using Flags = std::bitset<NUM_FLAGS>;
|
||||
@ -193,6 +209,10 @@ public:
|
||||
const Flags & getTableFlags() const { return all_flags_for_target[TABLE]; }
|
||||
const Flags & getColumnFlags() const { return all_flags_for_target[COLUMN]; }
|
||||
const Flags & getDictionaryFlags() const { return all_flags_for_target[DICTIONARY]; }
|
||||
const Flags & getAllFlagsGrantableOnGlobalLevel() const { return getAllFlags(); }
|
||||
const Flags & getAllFlagsGrantableOnDatabaseLevel() const { return all_flags_grantable_on_database_level; }
|
||||
const Flags & getAllFlagsGrantableOnTableLevel() const { return all_flags_grantable_on_table_level; }
|
||||
const Flags & getAllFlagsGrantableOnColumnLevel() const { return getColumnFlags(); }
|
||||
|
||||
private:
|
||||
enum NodeType
|
||||
@ -381,6 +401,9 @@ private:
|
||||
}
|
||||
for (const auto & child : start_node->children)
|
||||
collectAllFlags(child.get());
|
||||
|
||||
all_flags_grantable_on_table_level = all_flags_for_target[TABLE] | all_flags_for_target[DICTIONARY] | all_flags_for_target[COLUMN];
|
||||
all_flags_grantable_on_database_level = all_flags_for_target[DATABASE] | all_flags_grantable_on_table_level;
|
||||
}
|
||||
|
||||
Impl()
|
||||
@ -431,6 +454,8 @@ private:
|
||||
std::vector<Flags> access_type_to_flags_mapping;
|
||||
Flags all_flags;
|
||||
Flags all_flags_for_target[static_cast<size_t>(DICTIONARY) + 1];
|
||||
Flags all_flags_grantable_on_database_level;
|
||||
Flags all_flags_grantable_on_table_level;
|
||||
};
|
||||
|
||||
|
||||
@ -447,6 +472,10 @@ inline AccessFlags AccessFlags::allDatabaseFlags() { return Impl<>::instance().g
|
||||
inline AccessFlags AccessFlags::allTableFlags() { return Impl<>::instance().getTableFlags(); }
|
||||
inline AccessFlags AccessFlags::allColumnFlags() { return Impl<>::instance().getColumnFlags(); }
|
||||
inline AccessFlags AccessFlags::allDictionaryFlags() { return Impl<>::instance().getDictionaryFlags(); }
|
||||
inline AccessFlags AccessFlags::allFlagsGrantableOnGlobalLevel() { return Impl<>::instance().getAllFlagsGrantableOnGlobalLevel(); }
|
||||
inline AccessFlags AccessFlags::allFlagsGrantableOnDatabaseLevel() { return Impl<>::instance().getAllFlagsGrantableOnDatabaseLevel(); }
|
||||
inline AccessFlags AccessFlags::allFlagsGrantableOnTableLevel() { return Impl<>::instance().getAllFlagsGrantableOnTableLevel(); }
|
||||
inline AccessFlags AccessFlags::allFlagsGrantableOnColumnLevel() { return Impl<>::instance().getAllFlagsGrantableOnColumnLevel(); }
|
||||
|
||||
inline AccessFlags operator |(AccessType left, AccessType right) { return AccessFlags(left) | right; }
|
||||
inline AccessFlags operator &(AccessType left, AccessType right) { return AccessFlags(left) & right; }
|
||||
|
||||
@ -1,5 +1,4 @@
|
||||
#include <Access/AccessRights.h>
|
||||
#include <Common/Exception.h>
|
||||
#include <common/logger_useful.h>
|
||||
#include <boost/container/small_vector.hpp>
|
||||
#include <boost/range/adaptor/map.hpp>
|
||||
@ -8,12 +7,6 @@
|
||||
|
||||
namespace DB
|
||||
{
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int INVALID_GRANT;
|
||||
}
|
||||
|
||||
|
||||
namespace
|
||||
{
|
||||
using Kind = AccessRightsElementWithOptions::Kind;
|
||||
@ -214,30 +207,14 @@ namespace
|
||||
COLUMN_LEVEL,
|
||||
};
|
||||
|
||||
AccessFlags getAcceptableFlags(Level level)
|
||||
AccessFlags getAllGrantableFlags(Level level)
|
||||
{
|
||||
switch (level)
|
||||
{
|
||||
case GLOBAL_LEVEL:
|
||||
{
|
||||
static const AccessFlags res = AccessFlags::allFlags();
|
||||
return res;
|
||||
}
|
||||
case DATABASE_LEVEL:
|
||||
{
|
||||
static const AccessFlags res = AccessFlags::allDatabaseFlags() | AccessFlags::allTableFlags() | AccessFlags::allDictionaryFlags() | AccessFlags::allColumnFlags();
|
||||
return res;
|
||||
}
|
||||
case TABLE_LEVEL:
|
||||
{
|
||||
static const AccessFlags res = AccessFlags::allTableFlags() | AccessFlags::allDictionaryFlags() | AccessFlags::allColumnFlags();
|
||||
return res;
|
||||
}
|
||||
case COLUMN_LEVEL:
|
||||
{
|
||||
static const AccessFlags res = AccessFlags::allColumnFlags();
|
||||
return res;
|
||||
}
|
||||
case GLOBAL_LEVEL: return AccessFlags::allFlagsGrantableOnGlobalLevel();
|
||||
case DATABASE_LEVEL: return AccessFlags::allFlagsGrantableOnDatabaseLevel();
|
||||
case TABLE_LEVEL: return AccessFlags::allFlagsGrantableOnTableLevel();
|
||||
case COLUMN_LEVEL: return AccessFlags::allFlagsGrantableOnColumnLevel();
|
||||
}
|
||||
__builtin_unreachable();
|
||||
}
|
||||
@ -276,21 +253,7 @@ public:
|
||||
|
||||
void grant(const AccessFlags & flags_)
|
||||
{
|
||||
if (!flags_)
|
||||
return;
|
||||
|
||||
AccessFlags flags_to_add = flags_ & getAcceptableFlags();
|
||||
|
||||
if (!flags_to_add)
|
||||
{
|
||||
if (level == DATABASE_LEVEL)
|
||||
throw Exception(flags_.toString() + " cannot be granted on the database level", ErrorCodes::INVALID_GRANT);
|
||||
else if (level == TABLE_LEVEL)
|
||||
throw Exception(flags_.toString() + " cannot be granted on the table level", ErrorCodes::INVALID_GRANT);
|
||||
else if (level == COLUMN_LEVEL)
|
||||
throw Exception(flags_.toString() + " cannot be granted on the column level", ErrorCodes::INVALID_GRANT);
|
||||
}
|
||||
|
||||
AccessFlags flags_to_add = flags_ & getAllGrantableFlags();
|
||||
addGrantsRec(flags_to_add);
|
||||
optimizeTree();
|
||||
}
|
||||
@ -456,8 +419,8 @@ public:
|
||||
}
|
||||
|
||||
private:
|
||||
AccessFlags getAcceptableFlags() const { return ::DB::getAcceptableFlags(level); }
|
||||
AccessFlags getChildAcceptableFlags() const { return ::DB::getAcceptableFlags(static_cast<Level>(level + 1)); }
|
||||
AccessFlags getAllGrantableFlags() const { return ::DB::getAllGrantableFlags(level); }
|
||||
AccessFlags getChildAllGrantableFlags() const { return ::DB::getAllGrantableFlags(static_cast<Level>(level + 1)); }
|
||||
|
||||
Node * tryGetChild(const std::string_view & name) const
|
||||
{
|
||||
@ -480,7 +443,7 @@ private:
|
||||
Node & new_child = (*children)[*new_child_name];
|
||||
new_child.node_name = std::move(new_child_name);
|
||||
new_child.level = static_cast<Level>(level + 1);
|
||||
new_child.flags = flags & new_child.getAcceptableFlags();
|
||||
new_child.flags = flags & new_child.getAllGrantableFlags();
|
||||
return new_child;
|
||||
}
|
||||
|
||||
@ -496,12 +459,12 @@ private:
|
||||
|
||||
bool canEraseChild(const Node & child) const
|
||||
{
|
||||
return ((flags & child.getAcceptableFlags()) == child.flags) && !child.children;
|
||||
return ((flags & child.getAllGrantableFlags()) == child.flags) && !child.children;
|
||||
}
|
||||
|
||||
void addGrantsRec(const AccessFlags & flags_)
|
||||
{
|
||||
if (auto flags_to_add = flags_ & getAcceptableFlags())
|
||||
if (auto flags_to_add = flags_ & getAllGrantableFlags())
|
||||
{
|
||||
flags |= flags_to_add;
|
||||
if (children)
|
||||
@ -547,7 +510,7 @@ private:
|
||||
const AccessFlags & parent_flags)
|
||||
{
|
||||
auto flags = node.flags;
|
||||
auto parent_fl = parent_flags & node.getAcceptableFlags();
|
||||
auto parent_fl = parent_flags & node.getAllGrantableFlags();
|
||||
auto revokes = parent_fl - flags;
|
||||
auto grants = flags - parent_fl;
|
||||
|
||||
@ -576,9 +539,9 @@ private:
|
||||
const Node * node_go,
|
||||
const AccessFlags & parent_flags_go)
|
||||
{
|
||||
auto acceptable_flags = ::DB::getAcceptableFlags(static_cast<Level>(full_name.size()));
|
||||
auto parent_fl = parent_flags & acceptable_flags;
|
||||
auto parent_fl_go = parent_flags_go & acceptable_flags;
|
||||
auto grantable_flags = ::DB::getAllGrantableFlags(static_cast<Level>(full_name.size()));
|
||||
auto parent_fl = parent_flags & grantable_flags;
|
||||
auto parent_fl_go = parent_flags_go & grantable_flags;
|
||||
auto flags = node ? node->flags : parent_fl;
|
||||
auto flags_go = node_go ? node_go->flags : parent_fl_go;
|
||||
auto revokes = parent_fl - flags;
|
||||
@ -672,8 +635,8 @@ private:
|
||||
}
|
||||
|
||||
max_flags_with_children |= max_among_children;
|
||||
AccessFlags add_acceptable_flags = getAcceptableFlags() - getChildAcceptableFlags();
|
||||
min_flags_with_children &= min_among_children | add_acceptable_flags;
|
||||
AccessFlags add_flags = getAllGrantableFlags() - getChildAllGrantableFlags();
|
||||
min_flags_with_children &= min_among_children | add_flags;
|
||||
}
|
||||
|
||||
void makeUnionRec(const Node & rhs)
|
||||
@ -689,7 +652,7 @@ private:
|
||||
for (auto & [lhs_childname, lhs_child] : *children)
|
||||
{
|
||||
if (!rhs.tryGetChild(lhs_childname))
|
||||
lhs_child.flags |= rhs.flags & lhs_child.getAcceptableFlags();
|
||||
lhs_child.flags |= rhs.flags & lhs_child.getAllGrantableFlags();
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -738,7 +701,7 @@ private:
|
||||
|
||||
if (new_flags != flags)
|
||||
{
|
||||
new_flags &= getAcceptableFlags();
|
||||
new_flags &= getAllGrantableFlags();
|
||||
flags_added |= static_cast<bool>(new_flags - flags);
|
||||
flags_removed |= static_cast<bool>(flags - new_flags);
|
||||
flags = new_flags;
|
||||
|
||||
@ -71,6 +71,8 @@ struct AccessRightsElement
|
||||
{
|
||||
}
|
||||
|
||||
bool empty() const { return !access_flags || (!any_column && columns.empty()); }
|
||||
|
||||
auto toTuple() const { return std::tie(access_flags, any_database, database, any_table, table, any_column, columns); }
|
||||
friend bool operator==(const AccessRightsElement & left, const AccessRightsElement & right) { return left.toTuple() == right.toTuple(); }
|
||||
friend bool operator!=(const AccessRightsElement & left, const AccessRightsElement & right) { return !(left == right); }
|
||||
@ -86,6 +88,9 @@ struct AccessRightsElement
|
||||
/// If the database is empty, replaces it with `new_database`. Otherwise does nothing.
|
||||
void replaceEmptyDatabase(const String & new_database);
|
||||
|
||||
/// Resets flags which cannot be granted.
|
||||
void removeNonGrantableFlags();
|
||||
|
||||
/// Returns a human-readable representation like "SELECT, UPDATE(x, y) ON db.table".
|
||||
String toString() const;
|
||||
};
|
||||
@ -111,6 +116,9 @@ struct AccessRightsElementWithOptions : public AccessRightsElement
|
||||
friend bool operator==(const AccessRightsElementWithOptions & left, const AccessRightsElementWithOptions & right) { return left.toTuple() == right.toTuple(); }
|
||||
friend bool operator!=(const AccessRightsElementWithOptions & left, const AccessRightsElementWithOptions & right) { return !(left == right); }
|
||||
|
||||
/// Resets flags which cannot be granted.
|
||||
void removeNonGrantableFlags();
|
||||
|
||||
/// Returns a human-readable representation like "GRANT SELECT, UPDATE(x, y) ON db.table".
|
||||
String toString() const;
|
||||
};
|
||||
@ -120,9 +128,14 @@ struct AccessRightsElementWithOptions : public AccessRightsElement
|
||||
class AccessRightsElements : public std::vector<AccessRightsElement>
|
||||
{
|
||||
public:
|
||||
bool empty() const { return std::all_of(begin(), end(), [](const AccessRightsElement & e) { return e.empty(); }); }
|
||||
|
||||
/// Replaces the empty database with `new_database`.
|
||||
void replaceEmptyDatabase(const String & new_database);
|
||||
|
||||
/// Resets flags which cannot be granted.
|
||||
void removeNonGrantableFlags();
|
||||
|
||||
/// Returns a human-readable representation like "GRANT SELECT, UPDATE(x, y) ON db.table".
|
||||
String toString() const;
|
||||
};
|
||||
@ -134,6 +147,9 @@ public:
|
||||
/// Replaces the empty database with `new_database`.
|
||||
void replaceEmptyDatabase(const String & new_database);
|
||||
|
||||
/// Resets flags which cannot be granted.
|
||||
void removeNonGrantableFlags();
|
||||
|
||||
/// Returns a human-readable representation like "GRANT SELECT, UPDATE(x, y) ON db.table".
|
||||
String toString() const;
|
||||
};
|
||||
@ -157,4 +173,34 @@ inline void AccessRightsElementsWithOptions::replaceEmptyDatabase(const String &
|
||||
element.replaceEmptyDatabase(new_database);
|
||||
}
|
||||
|
||||
inline void AccessRightsElement::removeNonGrantableFlags()
|
||||
{
|
||||
if (!any_column)
|
||||
access_flags &= AccessFlags::allFlagsGrantableOnColumnLevel();
|
||||
else if (!any_table)
|
||||
access_flags &= AccessFlags::allFlagsGrantableOnTableLevel();
|
||||
else if (!any_database)
|
||||
access_flags &= AccessFlags::allFlagsGrantableOnDatabaseLevel();
|
||||
else
|
||||
access_flags &= AccessFlags::allFlagsGrantableOnGlobalLevel();
|
||||
}
|
||||
|
||||
inline void AccessRightsElementWithOptions::removeNonGrantableFlags()
|
||||
{
|
||||
if (kind == Kind::GRANT)
|
||||
AccessRightsElement::removeNonGrantableFlags();
|
||||
}
|
||||
|
||||
inline void AccessRightsElements::removeNonGrantableFlags()
|
||||
{
|
||||
for (auto & element : *this)
|
||||
element.removeNonGrantableFlags();
|
||||
}
|
||||
|
||||
inline void AccessRightsElementsWithOptions::removeNonGrantableFlags()
|
||||
{
|
||||
for (auto & element : *this)
|
||||
element.removeNonGrantableFlags();
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -85,12 +85,12 @@ AggregateFunctionPtr createAggregateFunctionTopK(const std::string & name, const
|
||||
load_factor = applyVisitor(FieldVisitorConvertToNumber<UInt64>(), params[1]);
|
||||
|
||||
if (load_factor < 1)
|
||||
throw Exception("Too small parameter for aggregate function " + name + ". Minimum: 1",
|
||||
throw Exception("Too small parameter 'load_factor' for aggregate function " + name + ". Minimum: 1",
|
||||
ErrorCodes::ARGUMENT_OUT_OF_BOUND);
|
||||
}
|
||||
|
||||
if (k > TOP_K_MAX_SIZE)
|
||||
throw Exception("Too large parameter for aggregate function " + name + ". Maximum: " + toString(TOP_K_MAX_SIZE),
|
||||
if (k > TOP_K_MAX_SIZE || load_factor > TOP_K_MAX_SIZE || k * load_factor > TOP_K_MAX_SIZE)
|
||||
throw Exception("Too large parameter(s) for aggregate function " + name + ". Maximum: " + toString(TOP_K_MAX_SIZE),
|
||||
ErrorCodes::ARGUMENT_OUT_OF_BOUND);
|
||||
|
||||
if (k == 0)
|
||||
|
||||
@ -147,16 +147,17 @@ public:
|
||||
{
|
||||
// Increase weight of a key that already exists
|
||||
auto hash = counter_map.hash(key);
|
||||
auto counter = findCounter(key, hash);
|
||||
if (counter)
|
||||
|
||||
if (auto counter = findCounter(key, hash); counter)
|
||||
{
|
||||
counter->count += increment;
|
||||
counter->error += error;
|
||||
percolate(counter);
|
||||
return;
|
||||
}
|
||||
|
||||
// Key doesn't exist, but can fit in the top K
|
||||
else if (unlikely(size() < capacity()))
|
||||
if (unlikely(size() < capacity()))
|
||||
{
|
||||
auto c = new Counter(arena.emplace(key), increment, error, hash);
|
||||
push(c);
|
||||
|
||||
@ -254,7 +254,7 @@ public:
|
||||
const auto offset = __builtin_ctz(mask);
|
||||
haystack += offset;
|
||||
|
||||
if (haystack < haystack_end && haystack + n <= haystack_end && pageSafe(haystack))
|
||||
if (haystack + n <= haystack_end && pageSafe(haystack))
|
||||
{
|
||||
const auto v_haystack_offset = _mm_loadu_si128(reinterpret_cast<const __m128i *>(haystack));
|
||||
const auto v_against_l_offset = _mm_cmpeq_epi8(v_haystack_offset, cachel);
|
||||
@ -463,7 +463,7 @@ public:
|
||||
const auto offset = __builtin_ctz(mask);
|
||||
haystack += offset;
|
||||
|
||||
if (haystack < haystack_end && haystack + n <= haystack_end && pageSafe(haystack))
|
||||
if (haystack + n <= haystack_end && pageSafe(haystack))
|
||||
{
|
||||
const auto v_haystack_offset = _mm_loadu_si128(reinterpret_cast<const __m128i *>(haystack));
|
||||
const auto v_against_l_offset = _mm_cmpeq_epi8(v_haystack_offset, cachel);
|
||||
@ -652,7 +652,7 @@ public:
|
||||
const auto offset = __builtin_ctz(mask);
|
||||
haystack += offset;
|
||||
|
||||
if (haystack < haystack_end && haystack + n <= haystack_end && pageSafe(haystack))
|
||||
if (haystack + n <= haystack_end && pageSafe(haystack))
|
||||
{
|
||||
/// check for first 16 octets
|
||||
const auto v_haystack_offset = _mm_loadu_si128(reinterpret_cast<const __m128i *>(haystack));
|
||||
|
||||
@ -13,6 +13,7 @@ namespace DB
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int UNKNOWN_EXCEPTION;
|
||||
extern const int LOGICAL_ERROR;
|
||||
}
|
||||
|
||||
namespace MySQLReplication
|
||||
@ -103,19 +104,19 @@ namespace MySQLReplication
|
||||
= header.event_size - EVENT_HEADER_LENGTH - 4 - 4 - 1 - 2 - 2 - status_len - schema_len - 1 - CHECKSUM_CRC32_SIGNATURE_LENGTH;
|
||||
query.resize(len);
|
||||
payload.readStrict(reinterpret_cast<char *>(query.data()), len);
|
||||
if (query.rfind("BEGIN", 0) == 0 || query.rfind("COMMIT") == 0)
|
||||
if (query.starts_with("BEGIN") || query.starts_with("COMMIT"))
|
||||
{
|
||||
typ = QUERY_EVENT_MULTI_TXN_FLAG;
|
||||
}
|
||||
else if (query.rfind("XA", 0) == 0)
|
||||
else if (query.starts_with("XA"))
|
||||
{
|
||||
if (query.rfind("XA ROLLBACK", 0) == 0)
|
||||
throw ReplicationError("ParseQueryEvent: Unsupported query event:" + query, ErrorCodes::UNKNOWN_EXCEPTION);
|
||||
if (query.starts_with("XA ROLLBACK"))
|
||||
throw ReplicationError("ParseQueryEvent: Unsupported query event:" + query, ErrorCodes::LOGICAL_ERROR);
|
||||
typ = QUERY_EVENT_XA;
|
||||
}
|
||||
else if (query.rfind("SAVEPOINT", 0) == 0)
|
||||
else if (query.starts_with("SAVEPOINT"))
|
||||
{
|
||||
throw ReplicationError("ParseQueryEvent: Unsupported query event:" + query, ErrorCodes::UNKNOWN_EXCEPTION);
|
||||
throw ReplicationError("ParseQueryEvent: Unsupported query event:" + query, ErrorCodes::LOGICAL_ERROR);
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@ -10,17 +10,36 @@ namespace DB
|
||||
namespace QueryProcessingStage
|
||||
{
|
||||
/// Numbers matter - the later stage has a larger number.
|
||||
///
|
||||
/// It is part of Protocol ABI, add values only to the end.
|
||||
/// Also keep in mind that the code may depends on the order of fields, so be double aware when you will add new values.
|
||||
enum Enum
|
||||
{
|
||||
FetchColumns = 0, /// Only read/have been read the columns specified in the query.
|
||||
WithMergeableState = 1, /// Until the stage where the results of processing on different servers can be combined.
|
||||
Complete = 2, /// Completely.
|
||||
/// Only read/have been read the columns specified in the query.
|
||||
FetchColumns = 0,
|
||||
/// Until the stage where the results of processing on different servers can be combined.
|
||||
WithMergeableState = 1,
|
||||
/// Completely.
|
||||
Complete = 2,
|
||||
/// Until the stage where the aggregate functions were calculated and finalized.
|
||||
///
|
||||
/// It is used for auto distributed_group_by_no_merge optimization for distributed engine.
|
||||
/// (See comments in StorageDistributed).
|
||||
WithMergeableStateAfterAggregation = 3,
|
||||
|
||||
MAX = 4,
|
||||
};
|
||||
|
||||
inline const char * toString(UInt64 stage)
|
||||
{
|
||||
static const char * data[] = { "FetchColumns", "WithMergeableState", "Complete" };
|
||||
return stage < 3
|
||||
static const char * data[] =
|
||||
{
|
||||
"FetchColumns",
|
||||
"WithMergeableState",
|
||||
"Complete",
|
||||
"WithMergeableStateAfterAggregation",
|
||||
};
|
||||
return stage < MAX
|
||||
? data[stage]
|
||||
: "Unknown stage";
|
||||
}
|
||||
|
||||
@ -107,8 +107,8 @@ class IColumn;
|
||||
\
|
||||
M(Bool, skip_unavailable_shards, false, "If 1, ClickHouse silently skips unavailable shards and nodes unresolvable through DNS. Shard is marked as unavailable when none of the replicas can be reached.", 0) \
|
||||
\
|
||||
M(Bool, distributed_group_by_no_merge, false, "Do not merge aggregation states from different servers for distributed query processing - in case it is for certain that there are different keys on different shards.", 0) \
|
||||
M(UInt64, parallel_distributed_insert_select, 0, "Process distributed INSERT SELECT query in the same cluster on local tables on every shard, if 1 SELECT is executed on each shard, if 2 SELECT and INSERT is executed on each shard", 0) \
|
||||
M(UInt64, distributed_group_by_no_merge, 0, "If 1, Do not merge aggregation states from different servers for distributed query processing - in case it is for certain that there are different keys on different shards. If 2 - same as 1 but also apply ORDER BY and LIMIT stages", 0) \
|
||||
M(Bool, optimize_distributed_group_by_sharding_key, false, "Optimize GROUP BY sharding_key queries (by avodiing costly aggregation on the initiator server).", 0) \
|
||||
M(Bool, optimize_skip_unused_shards, false, "Assumes that data is distributed by sharding_key. Optimization to skip unused shards if SELECT query filters by sharding_key.", 0) \
|
||||
M(UInt64, force_optimize_skip_unused_shards, 0, "Throw an exception if unused shards cannot be skipped (1 - throw only if the table has the sharding key, 2 - always throw.", 0) \
|
||||
|
||||
@ -185,31 +185,4 @@ bool DataTypeDateTime::equals(const IDataType & rhs) const
|
||||
return typeid(rhs) == typeid(*this);
|
||||
}
|
||||
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
|
||||
extern const int ILLEGAL_TYPE_OF_ARGUMENT;
|
||||
}
|
||||
|
||||
static DataTypePtr create(const ASTPtr & arguments)
|
||||
{
|
||||
if (!arguments)
|
||||
return std::make_shared<DataTypeDateTime>();
|
||||
|
||||
if (arguments->children.size() != 1)
|
||||
throw Exception("DateTime data type can optionally have only one argument - time zone name", ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
|
||||
|
||||
const auto * arg = arguments->children[0]->as<ASTLiteral>();
|
||||
if (!arg || arg->value.getType() != Field::Types::String)
|
||||
throw Exception("Parameter for DateTime data type must be string literal", ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
|
||||
|
||||
return std::make_shared<DataTypeDateTime>(arg->value.get<String>());
|
||||
}
|
||||
|
||||
void registerDataTypeDateTime(DataTypeFactory & factory)
|
||||
{
|
||||
factory.registerDataType("DateTime", create, DataTypeFactory::CaseInsensitive);
|
||||
factory.registerAlias("TIMESTAMP", "DateTime", DataTypeFactory::CaseInsensitive);
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -201,65 +201,4 @@ bool DataTypeDateTime64::equals(const IDataType & rhs) const
|
||||
return false;
|
||||
}
|
||||
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
|
||||
extern const int ILLEGAL_TYPE_OF_ARGUMENT;
|
||||
}
|
||||
|
||||
enum class ArgumentKind
|
||||
{
|
||||
Optional,
|
||||
Mandatory
|
||||
};
|
||||
|
||||
template <typename T, ArgumentKind Kind>
|
||||
std::conditional_t<Kind == ArgumentKind::Optional, std::optional<T>, T>
|
||||
getArgument(const ASTPtr & arguments, size_t argument_index, const char * argument_name, const std::string context_data_type_name)
|
||||
{
|
||||
using NearestResultType = NearestFieldType<T>;
|
||||
const auto field_type = Field::TypeToEnum<NearestResultType>::value;
|
||||
const ASTLiteral * argument = nullptr;
|
||||
|
||||
auto exception_message = [=](const String & message)
|
||||
{
|
||||
return std::string("Parameter #") + std::to_string(argument_index) + " '"
|
||||
+ argument_name + "' for " + context_data_type_name
|
||||
+ message
|
||||
+ ", expected: " + Field::Types::toString(field_type) + " literal.";
|
||||
};
|
||||
|
||||
if (!arguments || arguments->children.size() <= argument_index
|
||||
|| !(argument = arguments->children[argument_index]->as<ASTLiteral>()))
|
||||
{
|
||||
if constexpr (Kind == ArgumentKind::Optional)
|
||||
return {};
|
||||
else
|
||||
throw Exception(exception_message(" is missing"),
|
||||
ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
|
||||
}
|
||||
|
||||
if (argument->value.getType() != field_type)
|
||||
throw Exception(exception_message(String(" has wrong type: ") + argument->value.getTypeName()),
|
||||
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
|
||||
|
||||
return argument->value.get<NearestResultType>();
|
||||
}
|
||||
|
||||
static DataTypePtr create64(const ASTPtr & arguments)
|
||||
{
|
||||
if (!arguments || arguments->size() == 0)
|
||||
return std::make_shared<DataTypeDateTime64>(DataTypeDateTime64::default_scale);
|
||||
|
||||
const auto scale = getArgument<UInt64, ArgumentKind::Optional>(arguments, 0, "scale", "DateType64");
|
||||
const auto timezone = getArgument<String, ArgumentKind::Optional>(arguments, !!scale, "timezone", "DateType64");
|
||||
|
||||
return std::make_shared<DataTypeDateTime64>(scale.value_or(DataTypeDateTime64::default_scale), timezone.value_or(String{}));
|
||||
}
|
||||
|
||||
void registerDataTypeDateTime64(DataTypeFactory & factory)
|
||||
{
|
||||
factory.registerDataType("DateTime64", create64, DataTypeFactory::CaseInsensitive);
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -165,7 +165,6 @@ DataTypeFactory::DataTypeFactory()
|
||||
registerDataTypeDecimal(*this);
|
||||
registerDataTypeDate(*this);
|
||||
registerDataTypeDateTime(*this);
|
||||
registerDataTypeDateTime64(*this);
|
||||
registerDataTypeString(*this);
|
||||
registerDataTypeFixedString(*this);
|
||||
registerDataTypeEnum(*this);
|
||||
|
||||
@ -82,7 +82,6 @@ void registerDataTypeInterval(DataTypeFactory & factory);
|
||||
void registerDataTypeLowCardinality(DataTypeFactory & factory);
|
||||
void registerDataTypeDomainIPv4AndIPv6(DataTypeFactory & factory);
|
||||
void registerDataTypeDomainSimpleAggregateFunction(DataTypeFactory & factory);
|
||||
void registerDataTypeDateTime64(DataTypeFactory & factory);
|
||||
void registerDataTypeDomainGeo(DataTypeFactory & factory);
|
||||
|
||||
}
|
||||
|
||||
118
src/DataTypes/registerDataTypeDateTime.cpp
Normal file
118
src/DataTypes/registerDataTypeDateTime.cpp
Normal file
@ -0,0 +1,118 @@
|
||||
|
||||
#include <Core/Field.h>
|
||||
#include <Parsers/IAST.h>
|
||||
#include <Parsers/ASTLiteral.h>
|
||||
#include <DataTypes/IDataType.h>
|
||||
#include <DataTypes/DataTypeDateTime.h>
|
||||
#include <DataTypes/DataTypeDateTime64.h>
|
||||
#include <DataTypes/DataTypeFactory.h>
|
||||
|
||||
namespace DB
|
||||
{
|
||||
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int NUMBER_OF_ARGUMENTS_DOESNT_MATCH;
|
||||
extern const int ILLEGAL_TYPE_OF_ARGUMENT;
|
||||
}
|
||||
|
||||
enum class ArgumentKind
|
||||
{
|
||||
Optional,
|
||||
Mandatory
|
||||
};
|
||||
|
||||
String getExceptionMessage(
|
||||
const String & message, size_t argument_index, const char * argument_name,
|
||||
const std::string & context_data_type_name, Field::Types::Which field_type)
|
||||
{
|
||||
return std::string("Parameter #") + std::to_string(argument_index) + " '"
|
||||
+ argument_name + "' for " + context_data_type_name
|
||||
+ message
|
||||
+ ", expected: " + Field::Types::toString(field_type) + " literal.";
|
||||
}
|
||||
|
||||
template <typename T, ArgumentKind Kind>
|
||||
std::conditional_t<Kind == ArgumentKind::Optional, std::optional<T>, T>
|
||||
getArgument(const ASTPtr & arguments, size_t argument_index, const char * argument_name [[maybe_unused]], const std::string context_data_type_name)
|
||||
{
|
||||
using NearestResultType = NearestFieldType<T>;
|
||||
const auto field_type = Field::TypeToEnum<NearestResultType>::value;
|
||||
const ASTLiteral * argument = nullptr;
|
||||
|
||||
if (!arguments || arguments->children.size() <= argument_index
|
||||
|| !(argument = arguments->children[argument_index]->as<ASTLiteral>())
|
||||
|| argument->value.getType() != field_type)
|
||||
{
|
||||
if constexpr (Kind == ArgumentKind::Optional)
|
||||
return {};
|
||||
else
|
||||
{
|
||||
if (argument && argument->value.getType() != field_type)
|
||||
throw Exception(getExceptionMessage(String(" has wrong type: ") + argument->value.getTypeName(),
|
||||
argument_index, argument_name, context_data_type_name, field_type), ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
|
||||
else
|
||||
throw Exception(getExceptionMessage(" is missing", argument_index, argument_name, context_data_type_name, field_type),
|
||||
ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
|
||||
}
|
||||
}
|
||||
|
||||
return argument->value.get<NearestResultType>();
|
||||
}
|
||||
|
||||
static DataTypePtr create(const ASTPtr & arguments)
|
||||
{
|
||||
if (!arguments || arguments->children.empty())
|
||||
return std::make_shared<DataTypeDateTime>();
|
||||
|
||||
const auto scale = getArgument<UInt64, ArgumentKind::Optional>(arguments, 0, "scale", "DateTime");
|
||||
const auto timezone = getArgument<String, ArgumentKind::Optional>(arguments, !!scale, "timezone", "DateTime");
|
||||
|
||||
if (!scale && !timezone)
|
||||
throw Exception(getExceptionMessage(" has wrong type: ", 0, "scale", "DateTime", Field::Types::Which::UInt64),
|
||||
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
|
||||
|
||||
/// If scale is defined, the data type is DateTime when scale = 0 otherwise the data type is DateTime64
|
||||
if (scale && scale.value() != 0)
|
||||
return std::make_shared<DataTypeDateTime64>(scale.value(), timezone.value_or(String{}));
|
||||
|
||||
return std::make_shared<DataTypeDateTime>(timezone.value_or(String{}));
|
||||
}
|
||||
|
||||
static DataTypePtr create32(const ASTPtr & arguments)
|
||||
{
|
||||
if (!arguments || arguments->children.empty())
|
||||
return std::make_shared<DataTypeDateTime>();
|
||||
|
||||
if (arguments->children.size() != 1)
|
||||
throw Exception("DateTime32 data type can optionally have only one argument - time zone name", ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
|
||||
|
||||
const auto timezone = getArgument<String, ArgumentKind::Mandatory>(arguments, 0, "timezone", "DateTime32");
|
||||
|
||||
return std::make_shared<DataTypeDateTime>(timezone);
|
||||
}
|
||||
|
||||
static DataTypePtr create64(const ASTPtr & arguments)
|
||||
{
|
||||
if (!arguments || arguments->children.empty())
|
||||
return std::make_shared<DataTypeDateTime64>(DataTypeDateTime64::default_scale);
|
||||
|
||||
if (arguments->children.size() > 2)
|
||||
throw Exception("DateTime64 data type can optionally have two argument - scale and time zone name", ErrorCodes::NUMBER_OF_ARGUMENTS_DOESNT_MATCH);
|
||||
|
||||
const auto scale = getArgument<UInt64, ArgumentKind::Mandatory>(arguments, 0, "scale", "DateTime64");
|
||||
const auto timezone = getArgument<String, ArgumentKind::Optional>(arguments, 1, "timezone", "DateTime64");
|
||||
|
||||
return std::make_shared<DataTypeDateTime64>(scale, timezone.value_or(String{}));
|
||||
}
|
||||
|
||||
void registerDataTypeDateTime(DataTypeFactory & factory)
|
||||
{
|
||||
factory.registerDataType("DateTime", create, DataTypeFactory::CaseInsensitive);
|
||||
factory.registerDataType("DateTime32", create32, DataTypeFactory::CaseInsensitive);
|
||||
factory.registerDataType("DateTime64", create64, DataTypeFactory::CaseInsensitive);
|
||||
|
||||
factory.registerAlias("TIMESTAMP", "DateTime", DataTypeFactory::CaseInsensitive);
|
||||
}
|
||||
|
||||
}
|
||||
@ -38,6 +38,7 @@ SRCS(
|
||||
getMostSubtype.cpp
|
||||
IDataType.cpp
|
||||
NestedUtils.cpp
|
||||
registerDataTypeDateTime.cpp
|
||||
|
||||
)
|
||||
|
||||
|
||||
@ -221,6 +221,11 @@ void DatabaseOnDisk::dropTable(const Context & context, const String & table_nam
|
||||
throw Exception(ErrorCodes::LOGICAL_ERROR, "Path is empty");
|
||||
|
||||
StoragePtr table = detachTable(table_name);
|
||||
|
||||
/// This is possible for Lazy database.
|
||||
if (!table)
|
||||
return;
|
||||
|
||||
bool renamed = false;
|
||||
try
|
||||
{
|
||||
|
||||
@ -33,6 +33,11 @@ static constexpr size_t PRINT_MESSAGE_EACH_N_OBJECTS = 256;
|
||||
static constexpr size_t PRINT_MESSAGE_EACH_N_SECONDS = 5;
|
||||
static constexpr size_t METADATA_FILE_BUFFER_SIZE = 32768;
|
||||
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int NOT_IMPLEMENTED;
|
||||
}
|
||||
|
||||
namespace
|
||||
{
|
||||
void tryAttachTable(
|
||||
@ -249,6 +254,9 @@ void DatabaseOrdinary::alterTable(const Context & context, const StorageID & tab
|
||||
|
||||
auto & ast_create_query = ast->as<ASTCreateQuery &>();
|
||||
|
||||
if (ast_create_query.as_table_function)
|
||||
throw Exception(ErrorCodes::NOT_IMPLEMENTED, "Cannot alter table {} because it was created AS table function", backQuote(table_name));
|
||||
|
||||
ASTPtr new_columns = InterpreterCreateQuery::formatColumns(metadata.columns);
|
||||
ASTPtr new_indices = InterpreterCreateQuery::formatIndices(metadata.secondary_indices);
|
||||
ASTPtr new_constraints = InterpreterCreateQuery::formatConstraints(metadata.constraints);
|
||||
|
||||
@ -13,6 +13,7 @@ namespace DB
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int UNSUPPORTED_METHOD;
|
||||
extern const int LOGICAL_ERROR;
|
||||
}
|
||||
|
||||
|
||||
@ -239,12 +240,15 @@ std::string ExternalQueryBuilder::composeLoadIdsQuery(const std::vector<UInt64>
|
||||
}
|
||||
|
||||
|
||||
std::string
|
||||
ExternalQueryBuilder::composeLoadKeysQuery(const Columns & key_columns, const std::vector<size_t> & requested_rows, LoadKeysMethod method, size_t partition_key_prefix)
|
||||
std::string ExternalQueryBuilder::composeLoadKeysQuery(
|
||||
const Columns & key_columns, const std::vector<size_t> & requested_rows, LoadKeysMethod method, size_t partition_key_prefix)
|
||||
{
|
||||
if (!dict_struct.key)
|
||||
throw Exception{"Composite key required for method", ErrorCodes::UNSUPPORTED_METHOD};
|
||||
|
||||
if (key_columns.size() != dict_struct.key->size())
|
||||
throw Exception{"The size of key_columns does not equal to the size of dictionary key", ErrorCodes::LOGICAL_ERROR};
|
||||
|
||||
WriteBufferFromOwnString out;
|
||||
writeString("SELECT ", out);
|
||||
|
||||
|
||||
@ -52,12 +52,14 @@ namespace DB
|
||||
const String & host_,
|
||||
UInt16 port_,
|
||||
UInt8 db_index_,
|
||||
const String & password_,
|
||||
RedisStorageType storage_type_,
|
||||
const Block & sample_block_)
|
||||
: dict_struct{dict_struct_}
|
||||
, host{host_}
|
||||
, port{port_}
|
||||
, db_index{db_index_}
|
||||
, password{password_}
|
||||
, storage_type{storage_type_}
|
||||
, sample_block{sample_block_}
|
||||
, client{std::make_shared<Poco::Redis::Client>(host, port)}
|
||||
@ -78,16 +80,22 @@ namespace DB
|
||||
ErrorCodes::INVALID_CONFIG_PARAMETER};
|
||||
// suppose key[0] is primary key, key[1] is secondary key
|
||||
}
|
||||
if (!password.empty())
|
||||
{
|
||||
RedisCommand command("AUTH");
|
||||
command << password;
|
||||
String reply = client->execute<String>(command);
|
||||
if (reply != "OK")
|
||||
throw Exception{"Authentication failed with reason "
|
||||
+ reply, ErrorCodes::INTERNAL_REDIS_ERROR};
|
||||
}
|
||||
|
||||
if (db_index != 0)
|
||||
{
|
||||
RedisCommand command("SELECT");
|
||||
// Use poco's Int64, because it is defined as long long, and on
|
||||
// MacOS, for the purposes of template instantiation, this type is
|
||||
// distinct from int64_t, which is our Int64.
|
||||
command << static_cast<Poco::Int64>(db_index);
|
||||
command << std::to_string(db_index);
|
||||
String reply = client->execute<String>(command);
|
||||
if (reply != "+OK\r\n")
|
||||
if (reply != "OK")
|
||||
throw Exception{"Selecting database with index " + DB::toString(db_index)
|
||||
+ " failed with reason " + reply, ErrorCodes::INTERNAL_REDIS_ERROR};
|
||||
}
|
||||
@ -104,6 +112,7 @@ namespace DB
|
||||
config_.getString(config_prefix_ + ".host"),
|
||||
config_.getUInt(config_prefix_ + ".port"),
|
||||
config_.getUInt(config_prefix_ + ".db_index", 0),
|
||||
config_.getString(config_prefix_ + ".password",""),
|
||||
parseStorageType(config_.getString(config_prefix_ + ".storage_type", "")),
|
||||
sample_block_)
|
||||
{
|
||||
@ -115,6 +124,7 @@ namespace DB
|
||||
other.host,
|
||||
other.port,
|
||||
other.db_index,
|
||||
other.password,
|
||||
other.storage_type,
|
||||
other.sample_block}
|
||||
{
|
||||
|
||||
@ -41,6 +41,7 @@ namespace ErrorCodes
|
||||
const std::string & host,
|
||||
UInt16 port,
|
||||
UInt8 db_index,
|
||||
const std::string & password,
|
||||
RedisStorageType storage_type,
|
||||
const Block & sample_block);
|
||||
|
||||
@ -91,6 +92,7 @@ namespace ErrorCodes
|
||||
const std::string host;
|
||||
const UInt16 port;
|
||||
const UInt8 db_index;
|
||||
const std::string password;
|
||||
const RedisStorageType storage_type;
|
||||
Block sample_block;
|
||||
|
||||
|
||||
@ -1120,6 +1120,8 @@ void SSDComplexKeyCacheStorage::update(
|
||||
AbsentIdHandler && on_key_not_found,
|
||||
const DictionaryLifetime lifetime)
|
||||
{
|
||||
assert(key_columns.size() == key_types.size());
|
||||
|
||||
auto append_block = [&key_types, this](
|
||||
const Columns & new_keys,
|
||||
const SSDComplexKeyCachePartition::Attributes & new_attributes,
|
||||
@ -1447,6 +1449,12 @@ void SSDComplexKeyCacheDictionary::getItemsNumberImpl(
|
||||
const Columns & key_columns, const DataTypes & key_types,
|
||||
ResultArrayType<OutputType> & out, DefaultGetter && get_default) const
|
||||
{
|
||||
assert(dict_struct.key);
|
||||
assert(key_columns.size() == key_types.size());
|
||||
assert(key_columns.size() == dict_struct.key->size());
|
||||
|
||||
dict_struct.validateKeyTypes(key_types);
|
||||
|
||||
const auto now = std::chrono::system_clock::now();
|
||||
|
||||
TemporalComplexKeysPool not_found_pool;
|
||||
@ -1527,6 +1535,8 @@ void SSDComplexKeyCacheDictionary::getItemsStringImpl(
|
||||
ColumnString * out,
|
||||
DefaultGetter && get_default) const
|
||||
{
|
||||
dict_struct.validateKeyTypes(key_types);
|
||||
|
||||
const auto now = std::chrono::system_clock::now();
|
||||
|
||||
TemporalComplexKeysPool not_found_pool;
|
||||
|
||||
@ -427,8 +427,7 @@ private:
|
||||
using SSDComplexKeyCachePartitionPtr = std::shared_ptr<SSDComplexKeyCachePartition>;
|
||||
|
||||
|
||||
/*
|
||||
Class for managing SSDCachePartition and getting data from source.
|
||||
/** Class for managing SSDCachePartition and getting data from source.
|
||||
*/
|
||||
class SSDComplexKeyCacheStorage
|
||||
{
|
||||
@ -515,8 +514,7 @@ private:
|
||||
};
|
||||
|
||||
|
||||
/*
|
||||
Dictionary interface
|
||||
/** Dictionary interface
|
||||
*/
|
||||
class SSDComplexKeyCacheDictionary final : public IDictionaryBase
|
||||
{
|
||||
|
||||
@ -36,6 +36,7 @@ void registerFunctionsConversion(FunctionFactory & factory)
|
||||
|
||||
factory.registerFunction<FunctionToDate>();
|
||||
factory.registerFunction<FunctionToDateTime>();
|
||||
factory.registerFunction<FunctionToDateTime32>();
|
||||
factory.registerFunction<FunctionToDateTime64>();
|
||||
factory.registerFunction<FunctionToUUID>();
|
||||
factory.registerFunction<FunctionToString>();
|
||||
@ -93,6 +94,9 @@ void registerFunctionsConversion(FunctionFactory & factory)
|
||||
factory.registerFunction<FunctionParseDateTimeBestEffortUS>();
|
||||
factory.registerFunction<FunctionParseDateTimeBestEffortOrZero>();
|
||||
factory.registerFunction<FunctionParseDateTimeBestEffortOrNull>();
|
||||
factory.registerFunction<FunctionParseDateTime32BestEffort>();
|
||||
factory.registerFunction<FunctionParseDateTime32BestEffortOrZero>();
|
||||
factory.registerFunction<FunctionParseDateTime32BestEffortOrNull>();
|
||||
factory.registerFunction<FunctionParseDateTime64BestEffort>();
|
||||
factory.registerFunction<FunctionParseDateTime64BestEffortOrZero>();
|
||||
factory.registerFunction<FunctionParseDateTime64BestEffortOrNull>();
|
||||
|
||||
@ -977,6 +977,7 @@ struct ConvertImpl<DataTypeFixedString, DataTypeString, Name>
|
||||
/// Declared early because used below.
|
||||
struct NameToDate { static constexpr auto name = "toDate"; };
|
||||
struct NameToDateTime { static constexpr auto name = "toDateTime"; };
|
||||
struct NameToDateTime32 { static constexpr auto name = "toDateTime32"; };
|
||||
struct NameToDateTime64 { static constexpr auto name = "toDateTime64"; };
|
||||
struct NameToString { static constexpr auto name = "toString"; };
|
||||
struct NameToDecimal32 { static constexpr auto name = "toDecimal32"; };
|
||||
@ -1003,6 +1004,26 @@ DEFINE_NAME_TO_INTERVAL(Year)
|
||||
|
||||
#undef DEFINE_NAME_TO_INTERVAL
|
||||
|
||||
struct NameParseDateTimeBestEffort;
|
||||
struct NameParseDateTimeBestEffortOrZero;
|
||||
struct NameParseDateTimeBestEffortOrNull;
|
||||
|
||||
template<typename Name, typename ToDataType>
|
||||
static inline bool isDateTime64(const ColumnsWithTypeAndName & arguments, const ColumnNumbers & arguments_index = {})
|
||||
{
|
||||
if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
|
||||
return true;
|
||||
else if constexpr (std::is_same_v<Name, NameToDateTime> || std::is_same_v<Name, NameParseDateTimeBestEffort>
|
||||
|| std::is_same_v<Name, NameParseDateTimeBestEffortOrZero> || std::is_same_v<Name, NameParseDateTimeBestEffortOrNull>)
|
||||
{
|
||||
if (arguments_index.empty())
|
||||
return (arguments.size() == 2 && isUnsignedInteger(arguments[1].type)) || arguments.size() == 3;
|
||||
else
|
||||
return (arguments_index.size() == 2 && isUnsignedInteger(arguments[arguments_index[1]].type)) || arguments_index.size() == 3;
|
||||
}
|
||||
|
||||
return false;
|
||||
}
|
||||
|
||||
template <typename ToDataType, typename Name, typename MonotonicityImpl>
|
||||
class FunctionConvert : public IFunction
|
||||
@ -1034,10 +1055,16 @@ public:
|
||||
FunctionArgumentDescriptors mandatory_args = {{"Value", nullptr, nullptr, nullptr}};
|
||||
FunctionArgumentDescriptors optional_args;
|
||||
|
||||
if constexpr (to_decimal || to_datetime64)
|
||||
if constexpr (to_decimal)
|
||||
{
|
||||
mandatory_args.push_back({"scale", &isNativeInteger, &isColumnConst, "const Integer"});
|
||||
}
|
||||
|
||||
if (!to_decimal && isDateTime64<Name, ToDataType>(arguments))
|
||||
{
|
||||
mandatory_args.push_back({"scale", &isNativeInteger, &isColumnConst, "const Integer"});
|
||||
}
|
||||
|
||||
// toString(DateTime or DateTime64, [timezone: String])
|
||||
if ((std::is_same_v<Name, NameToString> && arguments.size() > 0 && (isDateTime64(arguments[0].type) || isDateTime(arguments[0].type)))
|
||||
// toUnixTimestamp(value[, timezone : String])
|
||||
@ -1080,16 +1107,22 @@ public:
|
||||
UInt32 scale [[maybe_unused]] = DataTypeDateTime64::default_scale;
|
||||
|
||||
// DateTime64 requires more arguments: scale and timezone. Since timezone is optional, scale should be first.
|
||||
if constexpr (to_datetime64)
|
||||
if (isDateTime64<Name, ToDataType>(arguments))
|
||||
{
|
||||
timezone_arg_position += 1;
|
||||
scale = static_cast<UInt32>(arguments[1].column->get64(0));
|
||||
|
||||
if (to_datetime64 || scale != 0) /// toDateTime('xxxx-xx-xx xx:xx:xx', 0) return DateTime
|
||||
return std::make_shared<DataTypeDateTime64>(scale,
|
||||
extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0));
|
||||
|
||||
return std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0));
|
||||
}
|
||||
|
||||
if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
|
||||
return std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0));
|
||||
else if constexpr (to_datetime64)
|
||||
return std::make_shared<DataTypeDateTime64>(scale, extractTimeZoneNameFromFunctionArguments(arguments, timezone_arg_position, 0));
|
||||
else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
|
||||
throw Exception("LOGICAL ERROR: It is a bug.", ErrorCodes::LOGICAL_ERROR);
|
||||
else
|
||||
return std::make_shared<ToDataType>();
|
||||
}
|
||||
@ -1208,6 +1241,22 @@ private:
|
||||
return true;
|
||||
};
|
||||
|
||||
if (isDateTime64<Name, ToDataType>(block.getColumnsWithTypeAndName(), arguments))
|
||||
{
|
||||
/// For toDateTime('xxxx-xx-xx xx:xx:xx.00', 2[, 'timezone']) we need to it convert to DateTime64
|
||||
const ColumnWithTypeAndName & scale_column = block.getByPosition(arguments[1]);
|
||||
UInt32 scale = extractToDecimalScale(scale_column);
|
||||
|
||||
if (to_datetime64 || scale != 0) /// When scale = 0, the data type is DateTime otherwise the data type is DateTime64
|
||||
{
|
||||
if (!callOnIndexAndDataType<DataTypeDateTime64>(from_type->getTypeId(), call))
|
||||
throw Exception("Illegal type " + block.getByPosition(arguments[0]).type->getName() + " of argument of function " + getName(),
|
||||
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
|
||||
|
||||
return;
|
||||
}
|
||||
}
|
||||
|
||||
bool done = callOnIndexAndDataType<ToDataType>(from_type->getTypeId(), call);
|
||||
if (!done)
|
||||
{
|
||||
@ -1262,7 +1311,8 @@ public:
|
||||
DataTypePtr getReturnTypeImpl(const ColumnsWithTypeAndName & arguments) const override
|
||||
{
|
||||
DataTypePtr res;
|
||||
if constexpr (to_datetime64)
|
||||
|
||||
if (isDateTime64<Name, ToDataType>(arguments))
|
||||
{
|
||||
validateFunctionArgumentTypes(*this, arguments,
|
||||
FunctionArgumentDescriptors{{"string", isStringOrFixedString, nullptr, "String or FixedString"}},
|
||||
@ -1272,11 +1322,12 @@ public:
|
||||
{"timezone", isStringOrFixedString, isColumnConst, "const String or FixedString"},
|
||||
});
|
||||
|
||||
UInt64 scale = DataTypeDateTime64::default_scale;
|
||||
UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
|
||||
if (arguments.size() > 1)
|
||||
scale = extractToDecimalScale(arguments[1]);
|
||||
const auto timezone = extractTimeZoneNameFromFunctionArguments(arguments, 2, 0);
|
||||
res = std::make_shared<DataTypeDateTime64>(scale, timezone);
|
||||
|
||||
res = scale == 0 ? res = std::make_shared<DataTypeDateTime>(timezone) : std::make_shared<DataTypeDateTime64>(scale, timezone);
|
||||
}
|
||||
else
|
||||
{
|
||||
@ -1323,6 +1374,8 @@ public:
|
||||
|
||||
if constexpr (std::is_same_v<ToDataType, DataTypeDateTime>)
|
||||
res = std::make_shared<DataTypeDateTime>(extractTimeZoneNameFromFunctionArguments(arguments, 1, 0));
|
||||
else if constexpr (std::is_same_v<ToDataType, DataTypeDateTime64>)
|
||||
throw Exception("LOGICAL ERROR: It is a bug.", ErrorCodes::LOGICAL_ERROR);
|
||||
else if constexpr (to_decimal)
|
||||
{
|
||||
UInt64 scale = extractToDecimalScale(arguments[1]);
|
||||
@ -1340,42 +1393,53 @@ public:
|
||||
return res;
|
||||
}
|
||||
|
||||
void executeImpl(Block & block, const ColumnNumbers & arguments, size_t result, size_t input_rows_count) const override
|
||||
template <typename ConvertToDataType>
|
||||
bool executeInternal(Block & block, const ColumnNumbers & arguments, size_t result, size_t input_rows_count, UInt32 scale = 0) const
|
||||
{
|
||||
const IDataType * from_type = block.getByPosition(arguments[0]).type.get();
|
||||
|
||||
bool ok = true;
|
||||
if constexpr (to_decimal || to_datetime64)
|
||||
if (checkAndGetDataType<DataTypeString>(from_type))
|
||||
{
|
||||
const UInt32 scale = assert_cast<const ToDataType &>(*removeNullable(block.getByPosition(result).type)).getScale();
|
||||
ConvertThroughParsing<DataTypeString, ConvertToDataType, Name, exception_mode, parsing_mode>::execute(
|
||||
block, arguments, result, input_rows_count, scale);
|
||||
return true;
|
||||
}
|
||||
else if (checkAndGetDataType<DataTypeFixedString>(from_type))
|
||||
{
|
||||
ConvertThroughParsing<DataTypeFixedString, ConvertToDataType, Name, exception_mode, parsing_mode>::execute(
|
||||
block, arguments, result, input_rows_count, scale);
|
||||
return true;
|
||||
}
|
||||
|
||||
if (checkAndGetDataType<DataTypeString>(from_type))
|
||||
{
|
||||
ConvertThroughParsing<DataTypeString, ToDataType, Name, exception_mode, parsing_mode>::execute(
|
||||
block, arguments, result, input_rows_count, scale);
|
||||
return false;
|
||||
}
|
||||
else if (checkAndGetDataType<DataTypeFixedString>(from_type))
|
||||
|
||||
void executeImpl(Block & block, const ColumnNumbers & arguments, size_t result, size_t input_rows_count) const override
|
||||
{
|
||||
ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, exception_mode, parsing_mode>::execute(
|
||||
block, arguments, result, input_rows_count, scale);
|
||||
}
|
||||
bool ok = true;
|
||||
|
||||
if constexpr (to_decimal)
|
||||
ok = executeInternal<ToDataType>(block, arguments, result, input_rows_count,
|
||||
assert_cast<const ToDataType &>(*removeNullable(block.getByPosition(result).type)).getScale());
|
||||
else
|
||||
ok = false;
|
||||
{
|
||||
if (isDateTime64<Name, ToDataType>(block.getColumnsWithTypeAndName(), arguments))
|
||||
{
|
||||
UInt64 scale = to_datetime64 ? DataTypeDateTime64::default_scale : 0;
|
||||
if (arguments.size() > 1)
|
||||
scale = extractToDecimalScale(block.getColumnsWithTypeAndName()[arguments[1]]);
|
||||
|
||||
if (scale == 0)
|
||||
ok = executeInternal<DataTypeDateTime>(block, arguments, result, input_rows_count);
|
||||
else
|
||||
{
|
||||
ok = executeInternal<DataTypeDateTime64>(block, arguments, result, input_rows_count, static_cast<UInt32>(scale));
|
||||
}
|
||||
}
|
||||
else
|
||||
{
|
||||
if (checkAndGetDataType<DataTypeString>(from_type))
|
||||
{
|
||||
ConvertThroughParsing<DataTypeString, ToDataType, Name, exception_mode, parsing_mode>::execute(
|
||||
block, arguments, result, input_rows_count);
|
||||
ok = executeInternal<ToDataType>(block, arguments, result, input_rows_count);
|
||||
}
|
||||
else if (checkAndGetDataType<DataTypeFixedString>(from_type))
|
||||
{
|
||||
ConvertThroughParsing<DataTypeFixedString, ToDataType, Name, exception_mode, parsing_mode>::execute(
|
||||
block, arguments, result, input_rows_count);
|
||||
}
|
||||
else
|
||||
ok = false;
|
||||
}
|
||||
|
||||
if (!ok)
|
||||
@ -1638,6 +1702,7 @@ using FunctionToFloat32 = FunctionConvert<DataTypeFloat32, NameToFloat32, ToNumb
|
||||
using FunctionToFloat64 = FunctionConvert<DataTypeFloat64, NameToFloat64, ToNumberMonotonicity<Float64>>;
|
||||
using FunctionToDate = FunctionConvert<DataTypeDate, NameToDate, ToDateMonotonicity>;
|
||||
using FunctionToDateTime = FunctionConvert<DataTypeDateTime, NameToDateTime, ToDateTimeMonotonicity>;
|
||||
using FunctionToDateTime32 = FunctionConvert<DataTypeDateTime, NameToDateTime32, ToDateTimeMonotonicity>;
|
||||
using FunctionToDateTime64 = FunctionConvert<DataTypeDateTime64, NameToDateTime64, UnknownMonotonicity>;
|
||||
using FunctionToUUID = FunctionConvert<DataTypeUUID, NameToUUID, ToNumberMonotonicity<UInt128>>;
|
||||
using FunctionToString = FunctionConvert<DataTypeString, NameToString, ToStringMonotonicity>;
|
||||
@ -1767,6 +1832,9 @@ struct NameParseDateTimeBestEffort { static constexpr auto name = "parseDateTime
|
||||
struct NameParseDateTimeBestEffortUS { static constexpr auto name = "parseDateTimeBestEffortUS"; };
|
||||
struct NameParseDateTimeBestEffortOrZero { static constexpr auto name = "parseDateTimeBestEffortOrZero"; };
|
||||
struct NameParseDateTimeBestEffortOrNull { static constexpr auto name = "parseDateTimeBestEffortOrNull"; };
|
||||
struct NameParseDateTime32BestEffort { static constexpr auto name = "parseDateTime32BestEffort"; };
|
||||
struct NameParseDateTime32BestEffortOrZero { static constexpr auto name = "parseDateTime32BestEffortOrZero"; };
|
||||
struct NameParseDateTime32BestEffortOrNull { static constexpr auto name = "parseDateTime32BestEffortOrNull"; };
|
||||
struct NameParseDateTime64BestEffort { static constexpr auto name = "parseDateTime64BestEffort"; };
|
||||
struct NameParseDateTime64BestEffortOrZero { static constexpr auto name = "parseDateTime64BestEffortOrZero"; };
|
||||
struct NameParseDateTime64BestEffortOrNull { static constexpr auto name = "parseDateTime64BestEffortOrNull"; };
|
||||
@ -1781,6 +1849,13 @@ using FunctionParseDateTimeBestEffortOrZero = FunctionConvertFromString<
|
||||
using FunctionParseDateTimeBestEffortOrNull = FunctionConvertFromString<
|
||||
DataTypeDateTime, NameParseDateTimeBestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
|
||||
|
||||
using FunctionParseDateTime32BestEffort = FunctionConvertFromString<
|
||||
DataTypeDateTime, NameParseDateTime32BestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
|
||||
using FunctionParseDateTime32BestEffortOrZero = FunctionConvertFromString<
|
||||
DataTypeDateTime, NameParseDateTime32BestEffortOrZero, ConvertFromStringExceptionMode::Zero, ConvertFromStringParsingMode::BestEffort>;
|
||||
using FunctionParseDateTime32BestEffortOrNull = FunctionConvertFromString<
|
||||
DataTypeDateTime, NameParseDateTime32BestEffortOrNull, ConvertFromStringExceptionMode::Null, ConvertFromStringParsingMode::BestEffort>;
|
||||
|
||||
using FunctionParseDateTime64BestEffort = FunctionConvertFromString<
|
||||
DataTypeDateTime64, NameParseDateTime64BestEffort, ConvertFromStringExceptionMode::Throw, ConvertFromStringParsingMode::BestEffort>;
|
||||
using FunctionParseDateTime64BestEffortOrZero = FunctionConvertFromString<
|
||||
|
||||
@ -1,6 +1,6 @@
|
||||
#!/usr/bin/env bash
|
||||
|
||||
[ ! -f public_suffix_list.dat ] && wget -O public_suffix_list.dat https://publicsuffix.org/list/public_suffix_list.dat
|
||||
[ ! -f public_suffix_list.dat ] && wget -nv -O public_suffix_list.dat https://publicsuffix.org/list/public_suffix_list.dat
|
||||
|
||||
echo '%language=C++
|
||||
%define lookup-function-name is_valid
|
||||
|
||||
@ -447,6 +447,19 @@ void ScopeStack::addAction(const ExpressionAction & action)
|
||||
}
|
||||
}
|
||||
|
||||
void ScopeStack::addActionNoInput(const ExpressionAction & action)
|
||||
{
|
||||
size_t level = 0;
|
||||
Names required = action.getNeededColumns();
|
||||
for (const auto & elem : required)
|
||||
level = std::max(level, getColumnLevel(elem));
|
||||
|
||||
Names added;
|
||||
stack[level].actions->add(action, added);
|
||||
|
||||
stack[level].new_columns.insert(added.begin(), added.end());
|
||||
}
|
||||
|
||||
ExpressionActionsPtr ScopeStack::popLevel()
|
||||
{
|
||||
ExpressionActionsPtr res = stack.back().actions;
|
||||
@ -549,7 +562,7 @@ void ActionsMatcher::visit(const ASTFunction & node, const ASTPtr & ast, Data &
|
||||
/// It could have been possible to implement arrayJoin which keeps source column,
|
||||
/// but in this case it will always be replicated (as many arrays), which is expensive.
|
||||
String tmp_name = data.getUniqueName("_array_join_" + arg->getColumnName());
|
||||
data.addAction(ExpressionAction::copyColumn(arg->getColumnName(), tmp_name));
|
||||
data.addActionNoInput(ExpressionAction::copyColumn(arg->getColumnName(), tmp_name));
|
||||
data.addAction(ExpressionAction::arrayJoin(tmp_name, result_name));
|
||||
}
|
||||
|
||||
|
||||
@ -12,6 +12,7 @@ namespace DB
|
||||
class Context;
|
||||
class ASTFunction;
|
||||
|
||||
struct ExpressionAction;
|
||||
class ExpressionActions;
|
||||
using ExpressionActionsPtr = std::shared_ptr<ExpressionActions>;
|
||||
|
||||
@ -49,6 +50,8 @@ struct ScopeStack
|
||||
size_t getColumnLevel(const std::string & name);
|
||||
|
||||
void addAction(const ExpressionAction & action);
|
||||
/// For arrayJoin() to avoid double columns in the input.
|
||||
void addActionNoInput(const ExpressionAction & action);
|
||||
|
||||
ExpressionActionsPtr popLevel();
|
||||
|
||||
@ -115,6 +118,10 @@ public:
|
||||
{
|
||||
actions_stack.addAction(action);
|
||||
}
|
||||
void addActionNoInput(const ExpressionAction & action)
|
||||
{
|
||||
actions_stack.addActionNoInput(action);
|
||||
}
|
||||
|
||||
const Block & getSampleBlock() const
|
||||
{
|
||||
|
||||
@ -13,6 +13,7 @@
|
||||
#include <Processors/Transforms/ConvertingTransform.h>
|
||||
#include <Processors/Sources/RemoteSource.h>
|
||||
#include <Processors/Sources/DelayedSource.h>
|
||||
#include <Processors/QueryPlan/QueryPlan.h>
|
||||
|
||||
namespace ProfileEvents
|
||||
{
|
||||
@ -68,14 +69,19 @@ SelectStreamFactory::SelectStreamFactory(
|
||||
namespace
|
||||
{
|
||||
|
||||
QueryPipeline createLocalStream(
|
||||
auto createLocalPipe(
|
||||
const ASTPtr & query_ast, const Block & header, const Context & context, QueryProcessingStage::Enum processed_stage)
|
||||
{
|
||||
checkStackSize();
|
||||
|
||||
InterpreterSelectQuery interpreter{query_ast, context, SelectQueryOptions(processed_stage)};
|
||||
InterpreterSelectQuery interpreter(query_ast, context, SelectQueryOptions(processed_stage));
|
||||
auto query_plan = std::make_unique<QueryPlan>();
|
||||
|
||||
auto pipeline = interpreter.execute().pipeline;
|
||||
interpreter.buildQueryPlan(*query_plan);
|
||||
auto pipeline = std::move(*query_plan->buildQueryPipeline());
|
||||
|
||||
/// Avoid going it out-of-scope for EXPLAIN
|
||||
pipeline.addQueryPlan(std::move(query_plan));
|
||||
|
||||
pipeline.addSimpleTransform([&](const Block & source_header)
|
||||
{
|
||||
@ -94,7 +100,7 @@ QueryPipeline createLocalStream(
|
||||
/// return std::make_shared<MaterializingBlockInputStream>(stream);
|
||||
|
||||
pipeline.setMaxThreads(1);
|
||||
return pipeline;
|
||||
return QueryPipeline::getPipe(std::move(pipeline));
|
||||
}
|
||||
|
||||
String formattedAST(const ASTPtr & ast)
|
||||
@ -130,7 +136,7 @@ void SelectStreamFactory::createForShard(
|
||||
|
||||
auto emplace_local_stream = [&]()
|
||||
{
|
||||
pipes.emplace_back(QueryPipeline::getPipe(createLocalStream(modified_query_ast, header, context, processed_stage)));
|
||||
pipes.emplace_back(createLocalPipe(modified_query_ast, header, context, processed_stage));
|
||||
};
|
||||
|
||||
String modified_query = formattedAST(modified_query_ast);
|
||||
@ -270,7 +276,7 @@ void SelectStreamFactory::createForShard(
|
||||
}
|
||||
|
||||
if (try_results.empty() || local_delay < max_remote_delay)
|
||||
return QueryPipeline::getPipe(createLocalStream(modified_query_ast, header, context, stage));
|
||||
return createLocalPipe(modified_query_ast, header, context, stage);
|
||||
else
|
||||
{
|
||||
std::vector<IConnectionPool::Entry> connections;
|
||||
|
||||
@ -42,6 +42,7 @@ void DNSCacheUpdater::run()
|
||||
|
||||
void DNSCacheUpdater::start()
|
||||
{
|
||||
LOG_INFO(&Poco::Logger::get("DNSCacheUpdater"), "Update period {} seconds", update_period_seconds);
|
||||
task_handle->activateAndSchedule();
|
||||
}
|
||||
|
||||
|
||||
@ -101,7 +101,7 @@ BlockIO InterpreterAlterQuery::execute()
|
||||
switch (command.type)
|
||||
{
|
||||
case LiveViewCommand::REFRESH:
|
||||
live_view->refresh(context);
|
||||
live_view->refresh();
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
@ -269,7 +269,9 @@ BlockInputStreamPtr InterpreterExplainQuery::executeImpl()
|
||||
|
||||
if (settings.graph)
|
||||
{
|
||||
auto processors = Pipe::detachProcessors(QueryPipeline::getPipe(std::move(*pipeline)));
|
||||
/// Pipe holds QueryPlan, should not go out-of-scope
|
||||
auto pipe = QueryPipeline::getPipe(std::move(*pipeline));
|
||||
const auto & processors = pipe.getProcessors();
|
||||
|
||||
if (settings.compact)
|
||||
printPipelineCompact(processors, buffer, settings.query_pipeline_options.header);
|
||||
|
||||
@ -29,7 +29,6 @@ namespace
|
||||
current_access.grant(access_to_grant);
|
||||
}
|
||||
|
||||
|
||||
AccessRightsElements getFilteredAccessRightsElementsToRevoke(
|
||||
const AccessRights & current_access, const AccessRightsElements & access_to_revoke, bool grant_option)
|
||||
{
|
||||
@ -214,6 +213,7 @@ BlockIO InterpreterGrantQuery::execute()
|
||||
auto access = context.getAccess();
|
||||
auto & access_control = context.getAccessControlManager();
|
||||
query.replaceEmptyDatabaseWithCurrent(context.getCurrentDatabase());
|
||||
query.removeNonGrantableFlags();
|
||||
|
||||
RolesOrUsersSet roles_from_query;
|
||||
if (query.roles)
|
||||
|
||||
@ -551,6 +551,11 @@ Block InterpreterSelectQuery::getSampleBlockImpl()
|
||||
return res;
|
||||
}
|
||||
|
||||
if (options.to_stage == QueryProcessingStage::Enum::WithMergeableStateAfterAggregation)
|
||||
{
|
||||
return analysis_result.before_order_and_select->getSampleBlock();
|
||||
}
|
||||
|
||||
return analysis_result.final_projection->getSampleBlock();
|
||||
}
|
||||
|
||||
@ -738,6 +743,8 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, const BlockInpu
|
||||
auto & expressions = analysis_result;
|
||||
auto & subqueries_for_sets = query_analyzer->getSubqueriesForSets();
|
||||
bool intermediate_stage = false;
|
||||
bool to_aggregation_stage = false;
|
||||
bool from_aggregation_stage = false;
|
||||
|
||||
if (options.only_analyze)
|
||||
{
|
||||
@ -786,6 +793,14 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, const BlockInpu
|
||||
options.to_stage == QueryProcessingStage::WithMergeableState)
|
||||
intermediate_stage = true;
|
||||
|
||||
/// Support optimize_distributed_group_by_sharding_key
|
||||
/// Is running on the initiating server during distributed processing?
|
||||
if (from_stage == QueryProcessingStage::WithMergeableStateAfterAggregation)
|
||||
from_aggregation_stage = true;
|
||||
/// Is running on remote servers during distributed processing?
|
||||
if (options.to_stage == QueryProcessingStage::WithMergeableStateAfterAggregation)
|
||||
to_aggregation_stage = true;
|
||||
|
||||
if (storage && expressions.filter_info && expressions.prewhere_info)
|
||||
throw Exception("PREWHERE is not supported if the table is filtered by row-level security expression", ErrorCodes::ILLEGAL_PREWHERE);
|
||||
|
||||
@ -846,6 +861,12 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, const BlockInpu
|
||||
if (expressions.need_aggregate)
|
||||
executeMergeAggregated(query_plan, aggregate_overflow_row, aggregate_final);
|
||||
}
|
||||
if (from_aggregation_stage)
|
||||
{
|
||||
if (intermediate_stage || expressions.first_stage || expressions.second_stage)
|
||||
throw Exception("Query with after aggregation stage cannot have any other stages", ErrorCodes::LOGICAL_ERROR);
|
||||
}
|
||||
|
||||
|
||||
if (expressions.first_stage)
|
||||
{
|
||||
@ -937,9 +958,13 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, const BlockInpu
|
||||
executeSubqueriesInSetsAndJoins(query_plan, subqueries_for_sets);
|
||||
}
|
||||
|
||||
if (expressions.second_stage)
|
||||
if (expressions.second_stage || from_aggregation_stage)
|
||||
{
|
||||
if (expressions.need_aggregate)
|
||||
if (from_aggregation_stage)
|
||||
{
|
||||
/// No need to aggregate anything, since this was done on remote shards.
|
||||
}
|
||||
else if (expressions.need_aggregate)
|
||||
{
|
||||
/// If you need to combine aggregated results from multiple servers
|
||||
if (!expressions.first_stage)
|
||||
@ -992,7 +1017,8 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, const BlockInpu
|
||||
* limiting the number of rows in each up to `offset + limit`.
|
||||
*/
|
||||
bool has_prelimit = false;
|
||||
if (query.limitLength() && !query.limit_with_ties && !hasWithTotalsInAnySubqueryInFromClause(query) &&
|
||||
if (!to_aggregation_stage &&
|
||||
query.limitLength() && !query.limit_with_ties && !hasWithTotalsInAnySubqueryInFromClause(query) &&
|
||||
!query.arrayJoinExpressionList() && !query.distinct && !expressions.hasLimitBy() && !settings.extremes)
|
||||
{
|
||||
executePreLimit(query_plan, false);
|
||||
@ -1021,17 +1047,22 @@ void InterpreterSelectQuery::executeImpl(QueryPlan & query_plan, const BlockInpu
|
||||
has_prelimit = true;
|
||||
}
|
||||
|
||||
/** We must do projection after DISTINCT because projection may remove some columns.
|
||||
*/
|
||||
/// Projection not be done on the shards, since then initiator will not find column in blocks.
|
||||
/// (significant only for WithMergeableStateAfterAggregation).
|
||||
if (!to_aggregation_stage)
|
||||
{
|
||||
/// We must do projection after DISTINCT because projection may remove some columns.
|
||||
executeProjection(query_plan, expressions.final_projection);
|
||||
}
|
||||
|
||||
/** Extremes are calculated before LIMIT, but after LIMIT BY. This is Ok.
|
||||
*/
|
||||
/// Extremes are calculated before LIMIT, but after LIMIT BY. This is Ok.
|
||||
executeExtremes(query_plan);
|
||||
|
||||
if (!has_prelimit) /// Limit is no longer needed if there is prelimit.
|
||||
/// Limit is no longer needed if there is prelimit.
|
||||
if (!to_aggregation_stage && !has_prelimit)
|
||||
executeLimit(query_plan);
|
||||
|
||||
if (!to_aggregation_stage)
|
||||
executeOffset(query_plan);
|
||||
}
|
||||
}
|
||||
|
||||
@ -18,6 +18,7 @@
|
||||
#include <DataTypes/DataTypeNullable.h>
|
||||
#include <Parsers/MySQL/ASTDeclareIndex.h>
|
||||
#include <Common/quoteString.h>
|
||||
#include <Common/assert_cast.h>
|
||||
#include <Interpreters/Context.h>
|
||||
#include <Interpreters/InterpreterCreateQuery.h>
|
||||
#include <Storages/IStorage.h>
|
||||
@ -124,8 +125,37 @@ static NamesAndTypesList getNames(const ASTFunction & expr, const Context & cont
|
||||
return expression->getRequiredColumnsWithTypes();
|
||||
}
|
||||
|
||||
static NamesAndTypesList modifyPrimaryKeysToNonNullable(const NamesAndTypesList & primary_keys, NamesAndTypesList & columns)
|
||||
{
|
||||
/// https://dev.mysql.com/doc/refman/5.7/en/create-table.html#create-table-indexes-keys
|
||||
/// PRIMARY KEY:
|
||||
/// A unique index where all key columns must be defined as NOT NULL.
|
||||
/// If they are not explicitly declared as NOT NULL, MySQL declares them so implicitly (and silently).
|
||||
/// A table can have only one PRIMARY KEY. The name of a PRIMARY KEY is always PRIMARY,
|
||||
/// which thus cannot be used as the name for any other kind of index.
|
||||
NamesAndTypesList non_nullable_primary_keys;
|
||||
for (const auto & primary_key : primary_keys)
|
||||
{
|
||||
if (!primary_key.type->isNullable())
|
||||
non_nullable_primary_keys.emplace_back(primary_key);
|
||||
else
|
||||
{
|
||||
non_nullable_primary_keys.emplace_back(
|
||||
NameAndTypePair(primary_key.name, assert_cast<const DataTypeNullable *>(primary_key.type.get())->getNestedType()));
|
||||
|
||||
for (auto & column : columns)
|
||||
{
|
||||
if (column.name == primary_key.name)
|
||||
column.type = assert_cast<const DataTypeNullable *>(column.type.get())->getNestedType();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
return non_nullable_primary_keys;
|
||||
}
|
||||
|
||||
static inline std::tuple<NamesAndTypesList, NamesAndTypesList, NamesAndTypesList, NameSet> getKeys(
|
||||
ASTExpressionList * columns_define, ASTExpressionList * indices_define, const Context & context, const NamesAndTypesList & columns)
|
||||
ASTExpressionList * columns_define, ASTExpressionList * indices_define, const Context & context, NamesAndTypesList & columns)
|
||||
{
|
||||
NameSet increment_columns;
|
||||
auto keys = makeASTFunction("tuple");
|
||||
@ -171,8 +201,9 @@ static inline std::tuple<NamesAndTypesList, NamesAndTypesList, NamesAndTypesList
|
||||
}
|
||||
}
|
||||
|
||||
return std::make_tuple(
|
||||
getNames(*primary_keys, context, columns), getNames(*unique_keys, context, columns), getNames(*keys, context, columns), increment_columns);
|
||||
const auto & primary_keys_names_and_types = getNames(*primary_keys, context, columns);
|
||||
const auto & non_nullable_primary_keys_names_and_types = modifyPrimaryKeysToNonNullable(primary_keys_names_and_types, columns);
|
||||
return std::make_tuple(non_nullable_primary_keys_names_and_types, getNames(*unique_keys, context, columns), getNames(*keys, context, columns), increment_columns);
|
||||
}
|
||||
|
||||
static String getUniqueColumnName(NamesAndTypesList columns_name_and_type, const String & prefix)
|
||||
@ -201,14 +232,13 @@ static String getUniqueColumnName(NamesAndTypesList columns_name_and_type, const
|
||||
|
||||
static ASTPtr getPartitionPolicy(const NamesAndTypesList & primary_keys)
|
||||
{
|
||||
const auto & numbers_partition = [&](const String & column_name, bool is_nullable, size_t type_max_size)
|
||||
const auto & numbers_partition = [&](const String & column_name, size_t type_max_size) -> ASTPtr
|
||||
{
|
||||
ASTPtr column = std::make_shared<ASTIdentifier>(column_name);
|
||||
if (type_max_size <= 1000)
|
||||
return std::make_shared<ASTIdentifier>(column_name);
|
||||
|
||||
if (is_nullable)
|
||||
column = makeASTFunction("assumeNotNull", column);
|
||||
|
||||
return makeASTFunction("intDiv", column, std::make_shared<ASTLiteral>(UInt64(type_max_size / 1000)));
|
||||
return makeASTFunction("intDiv", std::make_shared<ASTIdentifier>(column_name),
|
||||
std::make_shared<ASTLiteral>(UInt64(type_max_size / 1000)));
|
||||
};
|
||||
|
||||
ASTPtr best_partition;
|
||||
@ -219,16 +249,12 @@ static ASTPtr getPartitionPolicy(const NamesAndTypesList & primary_keys)
|
||||
WhichDataType which(type);
|
||||
|
||||
if (which.isNullable())
|
||||
{
|
||||
type = (static_cast<const DataTypeNullable &>(*type)).getNestedType();
|
||||
which = WhichDataType(type);
|
||||
}
|
||||
throw Exception("LOGICAL ERROR: MySQL primary key must be not null, it is a bug.", ErrorCodes::LOGICAL_ERROR);
|
||||
|
||||
if (which.isDateOrDateTime())
|
||||
{
|
||||
/// In any case, date or datetime is always the best partitioning key
|
||||
ASTPtr res = std::make_shared<ASTIdentifier>(primary_key.name);
|
||||
return makeASTFunction("toYYYYMM", primary_key.type->isNullable() ? makeASTFunction("assumeNotNull", res) : res);
|
||||
return makeASTFunction("toYYYYMM", std::make_shared<ASTIdentifier>(primary_key.name));
|
||||
}
|
||||
|
||||
if (type->haveMaximumSizeOfValue() && (!best_size || type->getSizeOfValueInMemory() < best_size))
|
||||
@ -236,25 +262,22 @@ static ASTPtr getPartitionPolicy(const NamesAndTypesList & primary_keys)
|
||||
if (which.isInt8() || which.isUInt8())
|
||||
{
|
||||
best_size = type->getSizeOfValueInMemory();
|
||||
best_partition = std::make_shared<ASTIdentifier>(primary_key.name);
|
||||
|
||||
if (primary_key.type->isNullable())
|
||||
best_partition = makeASTFunction("assumeNotNull", best_partition);
|
||||
best_partition = numbers_partition(primary_key.name, std::numeric_limits<UInt8>::max());
|
||||
}
|
||||
else if (which.isInt16() || which.isUInt16())
|
||||
{
|
||||
best_size = type->getSizeOfValueInMemory();
|
||||
best_partition = numbers_partition(primary_key.name, primary_key.type->isNullable(), std::numeric_limits<UInt16>::max());
|
||||
best_partition = numbers_partition(primary_key.name, std::numeric_limits<UInt16>::max());
|
||||
}
|
||||
else if (which.isInt32() || which.isUInt32())
|
||||
{
|
||||
best_size = type->getSizeOfValueInMemory();
|
||||
best_partition = numbers_partition(primary_key.name, primary_key.type->isNullable(), std::numeric_limits<UInt32>::max());
|
||||
best_partition = numbers_partition(primary_key.name, std::numeric_limits<UInt32>::max());
|
||||
}
|
||||
else if (which.isInt64() || which.isUInt64())
|
||||
{
|
||||
best_size = type->getSizeOfValueInMemory();
|
||||
best_partition = numbers_partition(primary_key.name, primary_key.type->isNullable(), std::numeric_limits<UInt64>::max());
|
||||
best_partition = numbers_partition(primary_key.name, std::numeric_limits<UInt64>::max());
|
||||
}
|
||||
}
|
||||
}
|
||||
@ -266,12 +289,12 @@ static ASTPtr getOrderByPolicy(
|
||||
const NamesAndTypesList & primary_keys, const NamesAndTypesList & unique_keys, const NamesAndTypesList & keys, const NameSet & increment_columns)
|
||||
{
|
||||
NameSet order_by_columns_set;
|
||||
std::deque<std::vector<String>> order_by_columns_list;
|
||||
std::deque<NamesAndTypesList> order_by_columns_list;
|
||||
|
||||
const auto & add_order_by_expression = [&](const NamesAndTypesList & names_and_types)
|
||||
{
|
||||
std::vector<String> increment_keys;
|
||||
std::vector<String> non_increment_keys;
|
||||
NamesAndTypesList increment_keys;
|
||||
NamesAndTypesList non_increment_keys;
|
||||
|
||||
for (const auto & [name, type] : names_and_types)
|
||||
{
|
||||
@ -280,13 +303,13 @@ static ASTPtr getOrderByPolicy(
|
||||
|
||||
if (increment_columns.count(name))
|
||||
{
|
||||
increment_keys.emplace_back(name);
|
||||
order_by_columns_set.emplace(name);
|
||||
increment_keys.emplace_back(NameAndTypePair(name, type));
|
||||
}
|
||||
else
|
||||
{
|
||||
order_by_columns_set.emplace(name);
|
||||
non_increment_keys.emplace_back(name);
|
||||
non_increment_keys.emplace_back(NameAndTypePair(name, type));
|
||||
}
|
||||
}
|
||||
|
||||
@ -305,8 +328,13 @@ static ASTPtr getOrderByPolicy(
|
||||
|
||||
for (const auto & order_by_columns : order_by_columns_list)
|
||||
{
|
||||
for (const auto & order_by_column : order_by_columns)
|
||||
order_by_expression->arguments->children.emplace_back(std::make_shared<ASTIdentifier>(order_by_column));
|
||||
for (const auto & [name, type] : order_by_columns)
|
||||
{
|
||||
order_by_expression->arguments->children.emplace_back(std::make_shared<ASTIdentifier>(name));
|
||||
|
||||
if (type->isNullable())
|
||||
order_by_expression->arguments->children.back() = makeASTFunction("assumeNotNull", order_by_expression->arguments->children.back());
|
||||
}
|
||||
}
|
||||
|
||||
return order_by_expression;
|
||||
|
||||
@ -103,21 +103,12 @@ TEST(MySQLCreateRewritten, PartitionPolicy)
|
||||
{"TIMESTAMP", "DateTime", " PARTITION BY toYYYYMM(key)"}, {"BOOLEAN", "Int8", " PARTITION BY key"}
|
||||
};
|
||||
|
||||
const auto & replace_string = [](const String & str, const String & old_str, const String & new_str)
|
||||
{
|
||||
String new_string = str;
|
||||
size_t pos = new_string.find(old_str);
|
||||
if (pos != std::string::npos)
|
||||
new_string = new_string.replace(pos, old_str.size(), new_str);
|
||||
return new_string;
|
||||
};
|
||||
|
||||
for (const auto & [test_type, mapped_type, partition_policy] : test_types)
|
||||
{
|
||||
EXPECT_EQ(queryToString(tryRewrittenCreateQuery(
|
||||
"CREATE TABLE `test_database`.`test_table_1` (`key` " + test_type + " PRIMARY KEY)", context_holder.context)),
|
||||
"CREATE TABLE test_database.test_table_1 (`key` Nullable(" + mapped_type + "), `_sign` Int8() MATERIALIZED 1, "
|
||||
"`_version` UInt64() MATERIALIZED 1) ENGINE = ReplacingMergeTree(_version)" + replace_string(partition_policy, "key", "assumeNotNull(key)") + " ORDER BY tuple(key)");
|
||||
"CREATE TABLE test_database.test_table_1 (`key` " + mapped_type + ", `_sign` Int8() MATERIALIZED 1, "
|
||||
"`_version` UInt64() MATERIALIZED 1) ENGINE = ReplacingMergeTree(_version)" + partition_policy + " ORDER BY tuple(key)");
|
||||
|
||||
EXPECT_EQ(queryToString(tryRewrittenCreateQuery(
|
||||
"CREATE TABLE `test_database`.`test_table_1` (`key` " + test_type + " NOT NULL PRIMARY KEY)", context_holder.context)),
|
||||
@ -126,6 +117,45 @@ TEST(MySQLCreateRewritten, PartitionPolicy)
|
||||
}
|
||||
}
|
||||
|
||||
TEST(MySQLCreateRewritten, OrderbyPolicy)
|
||||
{
|
||||
tryRegisterFunctions();
|
||||
const auto & context_holder = getContext();
|
||||
|
||||
std::vector<std::tuple<String, String, String>> test_types
|
||||
{
|
||||
{"TINYINT", "Int8", " PARTITION BY key"}, {"SMALLINT", "Int16", " PARTITION BY intDiv(key, 65)"},
|
||||
{"MEDIUMINT", "Int32", " PARTITION BY intDiv(key, 4294967)"}, {"INT", "Int32", " PARTITION BY intDiv(key, 4294967)"},
|
||||
{"INTEGER", "Int32", " PARTITION BY intDiv(key, 4294967)"}, {"BIGINT", "Int64", " PARTITION BY intDiv(key, 18446744073709551)"},
|
||||
{"FLOAT", "Float32", ""}, {"DOUBLE", "Float64", ""}, {"VARCHAR(10)", "String", ""}, {"CHAR(10)", "String", ""},
|
||||
{"Date", "Date", " PARTITION BY toYYYYMM(key)"}, {"DateTime", "DateTime", " PARTITION BY toYYYYMM(key)"},
|
||||
{"TIMESTAMP", "DateTime", " PARTITION BY toYYYYMM(key)"}, {"BOOLEAN", "Int8", " PARTITION BY key"}
|
||||
};
|
||||
|
||||
for (const auto & [test_type, mapped_type, partition_policy] : test_types)
|
||||
{
|
||||
EXPECT_EQ(queryToString(tryRewrittenCreateQuery(
|
||||
"CREATE TABLE `test_database`.`test_table_1` (`key` " + test_type + " PRIMARY KEY, `key2` " + test_type + " UNIQUE KEY)", context_holder.context)),
|
||||
"CREATE TABLE test_database.test_table_1 (`key` " + mapped_type + ", `key2` Nullable(" + mapped_type + "), `_sign` Int8() MATERIALIZED 1, "
|
||||
"`_version` UInt64() MATERIALIZED 1) ENGINE = ReplacingMergeTree(_version)" + partition_policy + " ORDER BY (key, assumeNotNull(key2))");
|
||||
|
||||
EXPECT_EQ(queryToString(tryRewrittenCreateQuery(
|
||||
"CREATE TABLE `test_database`.`test_table_1` (`key` " + test_type + " NOT NULL PRIMARY KEY, `key2` " + test_type + " NOT NULL UNIQUE KEY)", context_holder.context)),
|
||||
"CREATE TABLE test_database.test_table_1 (`key` " + mapped_type + ", `key2` " + mapped_type + ", `_sign` Int8() MATERIALIZED 1, "
|
||||
"`_version` UInt64() MATERIALIZED 1) ENGINE = ReplacingMergeTree(_version)" + partition_policy + " ORDER BY (key, key2)");
|
||||
|
||||
EXPECT_EQ(queryToString(tryRewrittenCreateQuery(
|
||||
"CREATE TABLE `test_database`.`test_table_1` (`key` " + test_type + " KEY UNIQUE KEY)", context_holder.context)),
|
||||
"CREATE TABLE test_database.test_table_1 (`key` " + mapped_type + ", `_sign` Int8() MATERIALIZED 1, "
|
||||
"`_version` UInt64() MATERIALIZED 1) ENGINE = ReplacingMergeTree(_version)" + partition_policy + " ORDER BY tuple(key)");
|
||||
|
||||
EXPECT_EQ(queryToString(tryRewrittenCreateQuery(
|
||||
"CREATE TABLE `test_database`.`test_table_1` (`key` " + test_type + ", `key2` " + test_type + " UNIQUE KEY, PRIMARY KEY(`key`, `key2`))", context_holder.context)),
|
||||
"CREATE TABLE test_database.test_table_1 (`key` " + mapped_type + ", `key2` " + mapped_type + ", `_sign` Int8() MATERIALIZED 1, "
|
||||
"`_version` UInt64() MATERIALIZED 1) ENGINE = ReplacingMergeTree(_version)" + partition_policy + " ORDER BY (key, key2)");
|
||||
}
|
||||
}
|
||||
|
||||
TEST(MySQLCreateRewritten, RewrittenQueryWithPrimaryKey)
|
||||
{
|
||||
tryRegisterFunctions();
|
||||
|
||||
@ -20,6 +20,7 @@ namespace ErrorCodes
|
||||
extern const int TOO_DEEP_AST;
|
||||
extern const int CYCLIC_ALIASES;
|
||||
extern const int UNKNOWN_QUERY_PARAMETER;
|
||||
extern const int BAD_ARGUMENTS;
|
||||
}
|
||||
|
||||
|
||||
@ -151,6 +152,13 @@ void QueryNormalizer::visitChildren(const ASTPtr & node, Data & data)
|
||||
{
|
||||
if (const auto * func_node = node->as<ASTFunction>())
|
||||
{
|
||||
if (func_node->query)
|
||||
{
|
||||
if (func_node->name != "view")
|
||||
throw Exception("Query argument can only be used in the `view` TableFunction", ErrorCodes::BAD_ARGUMENTS);
|
||||
/// Don't go into query argument.
|
||||
return;
|
||||
}
|
||||
/// We skip the first argument. We also assume that the lambda function can not have parameters.
|
||||
size_t first_pos = 0;
|
||||
if (func_node->name == "lambda")
|
||||
|
||||
@ -18,6 +18,7 @@
|
||||
#include <Parsers/ASTLiteral.h>
|
||||
#include <Parsers/ASTFunction.h>
|
||||
#include <Parsers/ASTColumnsMatcher.h>
|
||||
#include <Parsers/ASTColumnsTransformers.h>
|
||||
|
||||
|
||||
namespace DB
|
||||
@ -135,8 +136,8 @@ void TranslateQualifiedNamesMatcher::visit(ASTFunction & node, const ASTPtr &, D
|
||||
|
||||
void TranslateQualifiedNamesMatcher::visit(const ASTQualifiedAsterisk &, const ASTPtr & ast, Data & data)
|
||||
{
|
||||
if (ast->children.size() != 1)
|
||||
throw Exception("Logical error: qualified asterisk must have exactly one child", ErrorCodes::LOGICAL_ERROR);
|
||||
if (ast->children.empty())
|
||||
throw Exception("Logical error: qualified asterisk must have children", ErrorCodes::LOGICAL_ERROR);
|
||||
|
||||
auto & ident = ast->children[0];
|
||||
|
||||
@ -242,6 +243,10 @@ void TranslateQualifiedNamesMatcher::visit(ASTExpressionList & node, const ASTPt
|
||||
|
||||
first_table = false;
|
||||
}
|
||||
for (const auto & transformer : asterisk->children)
|
||||
{
|
||||
IASTColumnsTransformer::transform(transformer, node.children);
|
||||
}
|
||||
}
|
||||
else if (const auto * asterisk_pattern = child->as<ASTColumnsMatcher>())
|
||||
{
|
||||
@ -258,6 +263,11 @@ void TranslateQualifiedNamesMatcher::visit(ASTExpressionList & node, const ASTPt
|
||||
|
||||
first_table = false;
|
||||
}
|
||||
// ColumnsMatcher's transformers start to appear at child 1
|
||||
for (auto it = asterisk_pattern->children.begin() + 1; it != asterisk_pattern->children.end(); ++it)
|
||||
{
|
||||
IASTColumnsTransformer::transform(*it, node.children);
|
||||
}
|
||||
}
|
||||
else if (const auto * qualified_asterisk = child->as<ASTQualifiedAsterisk>())
|
||||
{
|
||||
@ -274,6 +284,11 @@ void TranslateQualifiedNamesMatcher::visit(ASTExpressionList & node, const ASTPt
|
||||
break;
|
||||
}
|
||||
}
|
||||
// QualifiedAsterisk's transformers start to appear at child 1
|
||||
for (auto it = qualified_asterisk->children.begin() + 1; it != qualified_asterisk->children.end(); ++it)
|
||||
{
|
||||
IASTColumnsTransformer::transform(*it, node.children);
|
||||
}
|
||||
}
|
||||
else
|
||||
node.children.emplace_back(child);
|
||||
|
||||
@ -13,9 +13,14 @@ ASTPtr ASTAsterisk::clone() const
|
||||
|
||||
void ASTAsterisk::appendColumnName(WriteBuffer & ostr) const { ostr.write('*'); }
|
||||
|
||||
void ASTAsterisk::formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const
|
||||
void ASTAsterisk::formatImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const
|
||||
{
|
||||
settings.ostr << "*";
|
||||
for (const auto & child : children)
|
||||
{
|
||||
settings.ostr << ' ';
|
||||
child->formatImpl(settings, state, frame);
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
@ -9,6 +9,9 @@ namespace DB
|
||||
struct AsteriskSemantic;
|
||||
struct AsteriskSemanticImpl;
|
||||
|
||||
/** SELECT * is expanded to all visible columns of the source table.
|
||||
* Optional transformers can be attached to further manipulate these expanded columns.
|
||||
*/
|
||||
class ASTAsterisk : public IAST
|
||||
{
|
||||
public:
|
||||
|
||||
@ -28,10 +28,15 @@ void ASTColumnsMatcher::updateTreeHashImpl(SipHash & hash_state) const
|
||||
IAST::updateTreeHashImpl(hash_state);
|
||||
}
|
||||
|
||||
void ASTColumnsMatcher::formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const
|
||||
void ASTColumnsMatcher::formatImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const
|
||||
{
|
||||
settings.ostr << (settings.hilite ? hilite_keyword : "") << "COLUMNS" << (settings.hilite ? hilite_none : "") << "("
|
||||
<< quoteString(original_pattern) << ")";
|
||||
for (ASTs::const_iterator it = children.begin() + 1; it != children.end(); ++it)
|
||||
{
|
||||
settings.ostr << ' ';
|
||||
(*it)->formatImpl(settings, state, frame);
|
||||
}
|
||||
}
|
||||
|
||||
void ASTColumnsMatcher::setPattern(String pattern)
|
||||
|
||||
@ -23,6 +23,7 @@ struct AsteriskSemanticImpl;
|
||||
|
||||
|
||||
/** SELECT COLUMNS('regexp') is expanded to multiple columns like * (asterisk).
|
||||
* Optional transformers can be attached to further manipulate these expanded columns.
|
||||
*/
|
||||
class ASTColumnsMatcher : public IAST
|
||||
{
|
||||
|
||||
159
src/Parsers/ASTColumnsTransformers.cpp
Normal file
159
src/Parsers/ASTColumnsTransformers.cpp
Normal file
@ -0,0 +1,159 @@
|
||||
#include <map>
|
||||
#include "ASTColumnsTransformers.h"
|
||||
#include <IO/WriteHelpers.h>
|
||||
#include <Parsers/ASTFunction.h>
|
||||
#include <Parsers/ASTIdentifier.h>
|
||||
#include <Common/SipHash.h>
|
||||
#include <Common/quoteString.h>
|
||||
|
||||
|
||||
namespace DB
|
||||
{
|
||||
namespace ErrorCodes
|
||||
{
|
||||
extern const int ILLEGAL_TYPE_OF_ARGUMENT;
|
||||
}
|
||||
|
||||
void IASTColumnsTransformer::transform(const ASTPtr & transformer, ASTs & nodes)
|
||||
{
|
||||
if (const auto * apply = transformer->as<ASTColumnsApplyTransformer>())
|
||||
{
|
||||
apply->transform(nodes);
|
||||
}
|
||||
else if (const auto * except = transformer->as<ASTColumnsExceptTransformer>())
|
||||
{
|
||||
except->transform(nodes);
|
||||
}
|
||||
else if (const auto * replace = transformer->as<ASTColumnsReplaceTransformer>())
|
||||
{
|
||||
replace->transform(nodes);
|
||||
}
|
||||
}
|
||||
|
||||
void ASTColumnsApplyTransformer::formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const
|
||||
{
|
||||
settings.ostr << (settings.hilite ? hilite_keyword : "") << "APPLY" << (settings.hilite ? hilite_none : "") << "(" << func_name << ")";
|
||||
}
|
||||
|
||||
void ASTColumnsApplyTransformer::transform(ASTs & nodes) const
|
||||
{
|
||||
for (auto & column : nodes)
|
||||
{
|
||||
column = makeASTFunction(func_name, column);
|
||||
}
|
||||
}
|
||||
|
||||
void ASTColumnsExceptTransformer::formatImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const
|
||||
{
|
||||
settings.ostr << (settings.hilite ? hilite_keyword : "") << "EXCEPT" << (settings.hilite ? hilite_none : "") << "(";
|
||||
|
||||
for (ASTs::const_iterator it = children.begin(); it != children.end(); ++it)
|
||||
{
|
||||
if (it != children.begin())
|
||||
{
|
||||
settings.ostr << ", ";
|
||||
}
|
||||
(*it)->formatImpl(settings, state, frame);
|
||||
}
|
||||
|
||||
settings.ostr << ")";
|
||||
}
|
||||
|
||||
void ASTColumnsExceptTransformer::transform(ASTs & nodes) const
|
||||
{
|
||||
nodes.erase(
|
||||
std::remove_if(
|
||||
nodes.begin(),
|
||||
nodes.end(),
|
||||
[this](const ASTPtr & node_child)
|
||||
{
|
||||
if (const auto * id = node_child->as<ASTIdentifier>())
|
||||
{
|
||||
for (const auto & except_child : children)
|
||||
{
|
||||
if (except_child->as<const ASTIdentifier &>().name == id->shortName())
|
||||
return true;
|
||||
}
|
||||
}
|
||||
return false;
|
||||
}),
|
||||
nodes.end());
|
||||
}
|
||||
|
||||
void ASTColumnsReplaceTransformer::Replacement::formatImpl(
|
||||
const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const
|
||||
{
|
||||
expr->formatImpl(settings, state, frame);
|
||||
settings.ostr << (settings.hilite ? hilite_keyword : "") << " AS " << (settings.hilite ? hilite_none : "") << name;
|
||||
}
|
||||
|
||||
void ASTColumnsReplaceTransformer::formatImpl(const FormatSettings & settings, FormatState & state, FormatStateStacked frame) const
|
||||
{
|
||||
settings.ostr << (settings.hilite ? hilite_keyword : "") << "REPLACE" << (settings.hilite ? hilite_none : "") << "(";
|
||||
|
||||
for (ASTs::const_iterator it = children.begin(); it != children.end(); ++it)
|
||||
{
|
||||
if (it != children.begin())
|
||||
{
|
||||
settings.ostr << ", ";
|
||||
}
|
||||
(*it)->formatImpl(settings, state, frame);
|
||||
}
|
||||
|
||||
settings.ostr << ")";
|
||||
}
|
||||
|
||||
void ASTColumnsReplaceTransformer::replaceChildren(ASTPtr & node, const ASTPtr & replacement, const String & name)
|
||||
{
|
||||
for (auto & child : node->children)
|
||||
{
|
||||
if (const auto * id = child->as<ASTIdentifier>())
|
||||
{
|
||||
if (id->shortName() == name)
|
||||
child = replacement;
|
||||
}
|
||||
else
|
||||
replaceChildren(child, replacement, name);
|
||||
}
|
||||
}
|
||||
|
||||
void ASTColumnsReplaceTransformer::transform(ASTs & nodes) const
|
||||
{
|
||||
std::map<String, ASTPtr> replace_map;
|
||||
for (const auto & replace_child : children)
|
||||
{
|
||||
auto & replacement = replace_child->as<Replacement &>();
|
||||
if (replace_map.find(replacement.name) != replace_map.end())
|
||||
throw Exception(
|
||||
"Expressions in columns transformer REPLACE should not contain the same replacement more than once",
|
||||
ErrorCodes::ILLEGAL_TYPE_OF_ARGUMENT);
|
||||
replace_map.emplace(replacement.name, replacement.expr);
|
||||
}
|
||||
|
||||
for (auto & column : nodes)
|
||||
{
|
||||
if (const auto * id = column->as<ASTIdentifier>())
|
||||
{
|
||||
auto replace_it = replace_map.find(id->shortName());
|
||||
if (replace_it != replace_map.end())
|
||||
{
|
||||
column = replace_it->second;
|
||||
column->setAlias(replace_it->first);
|
||||
}
|
||||
}
|
||||
else if (auto * ast_with_alias = dynamic_cast<ASTWithAlias *>(column.get()))
|
||||
{
|
||||
auto replace_it = replace_map.find(ast_with_alias->alias);
|
||||
if (replace_it != replace_map.end())
|
||||
{
|
||||
auto new_ast = replace_it->second->clone();
|
||||
ast_with_alias->alias = ""; // remove the old alias as it's useless after replace transformation
|
||||
replaceChildren(new_ast, column, replace_it->first);
|
||||
column = new_ast;
|
||||
column->setAlias(replace_it->first);
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
}
|
||||
85
src/Parsers/ASTColumnsTransformers.h
Normal file
85
src/Parsers/ASTColumnsTransformers.h
Normal file
@ -0,0 +1,85 @@
|
||||
#pragma once
|
||||
|
||||
#include <Parsers/IAST.h>
|
||||
|
||||
namespace DB
|
||||
{
|
||||
class IASTColumnsTransformer : public IAST
|
||||
{
|
||||
public:
|
||||
virtual void transform(ASTs & nodes) const = 0;
|
||||
static void transform(const ASTPtr & transformer, ASTs & nodes);
|
||||
};
|
||||
|
||||
class ASTColumnsApplyTransformer : public IASTColumnsTransformer
|
||||
{
|
||||
public:
|
||||
String getID(char) const override { return "ColumnsApplyTransformer"; }
|
||||
ASTPtr clone() const override
|
||||
{
|
||||
auto res = std::make_shared<ASTColumnsApplyTransformer>(*this);
|
||||
return res;
|
||||
}
|
||||
void transform(ASTs & nodes) const override;
|
||||
String func_name;
|
||||
|
||||
protected:
|
||||
void formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const override;
|
||||
};
|
||||
|
||||
class ASTColumnsExceptTransformer : public IASTColumnsTransformer
|
||||
{
|
||||
public:
|
||||
String getID(char) const override { return "ColumnsExceptTransformer"; }
|
||||
ASTPtr clone() const override
|
||||
{
|
||||
auto clone = std::make_shared<ASTColumnsExceptTransformer>(*this);
|
||||
clone->cloneChildren();
|
||||
return clone;
|
||||
}
|
||||
void transform(ASTs & nodes) const override;
|
||||
|
||||
protected:
|
||||
void formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const override;
|
||||
};
|
||||
|
||||
class ASTColumnsReplaceTransformer : public IASTColumnsTransformer
|
||||
{
|
||||
public:
|
||||
class Replacement : public IAST
|
||||
{
|
||||
public:
|
||||
String getID(char) const override { return "ColumnsReplaceTransformer::Replacement"; }
|
||||
ASTPtr clone() const override
|
||||
{
|
||||
auto replacement = std::make_shared<Replacement>(*this);
|
||||
replacement->name = name;
|
||||
replacement->expr = expr->clone();
|
||||
replacement->children.push_back(replacement->expr);
|
||||
return replacement;
|
||||
}
|
||||
|
||||
String name;
|
||||
ASTPtr expr;
|
||||
|
||||
protected:
|
||||
void formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const override;
|
||||
};
|
||||
|
||||
String getID(char) const override { return "ColumnsReplaceTransformer"; }
|
||||
ASTPtr clone() const override
|
||||
{
|
||||
auto clone = std::make_shared<ASTColumnsReplaceTransformer>(*this);
|
||||
clone->cloneChildren();
|
||||
return clone;
|
||||
}
|
||||
void transform(ASTs & nodes) const override;
|
||||
|
||||
protected:
|
||||
void formatImpl(const FormatSettings & settings, FormatState &, FormatStateStacked) const override;
|
||||
|
||||
private:
|
||||
static void replaceChildren(ASTPtr & node, const ASTPtr & replacement, const String & name);
|
||||
};
|
||||
|
||||
}
|
||||
@ -48,6 +48,7 @@ ASTPtr ASTFunction::clone() const
|
||||
auto res = std::make_shared<ASTFunction>(*this);
|
||||
res->children.clear();
|
||||
|
||||
if (query) { res->query = query->clone(); res->children.push_back(res->query); }
|
||||
if (arguments) { res->arguments = arguments->clone(); res->children.push_back(res->arguments); }
|
||||
if (parameters) { res->parameters = parameters->clone(); res->children.push_back(res->parameters); }
|
||||
|
||||
@ -118,6 +119,18 @@ void ASTFunction::formatImplWithoutAlias(const FormatSettings & settings, Format
|
||||
nested_need_parens.need_parens = true;
|
||||
nested_dont_need_parens.need_parens = false;
|
||||
|
||||
if (query)
|
||||
{
|
||||
std::string nl_or_nothing = settings.one_line ? "" : "\n";
|
||||
std::string indent_str = settings.one_line ? "" : std::string(4u * frame.indent, ' ');
|
||||
settings.ostr << (settings.hilite ? hilite_function : "") << name << "(" << nl_or_nothing;
|
||||
FormatStateStacked frame_nested = frame;
|
||||
frame_nested.need_parens = false;
|
||||
++frame_nested.indent;
|
||||
query->formatImpl(settings, state, frame_nested);
|
||||
settings.ostr << nl_or_nothing << indent_str << ")";
|
||||
return;
|
||||
}
|
||||
/// Should this function to be written as operator?
|
||||
bool written = false;
|
||||
if (arguments && !parameters)
|
||||
|
||||
@ -13,6 +13,7 @@ class ASTFunction : public ASTWithAlias
|
||||
{
|
||||
public:
|
||||
String name;
|
||||
ASTPtr query; // It's possible for a function to accept a query as its only argument.
|
||||
ASTPtr arguments;
|
||||
/// parameters - for parametric aggregate function. Example: quantile(0.9)(x) - what in first parens are 'parameters'.
|
||||
ASTPtr parameters;
|
||||
|
||||
@ -144,4 +144,12 @@ void ASTGrantQuery::replaceCurrentUserTagWithName(const String & current_user_na
|
||||
if (to_roles)
|
||||
to_roles->replaceCurrentUserTagWithName(current_user_name);
|
||||
}
|
||||
|
||||
|
||||
void ASTGrantQuery::removeNonGrantableFlags()
|
||||
{
|
||||
if (kind == Kind::GRANT)
|
||||
access_rights_elements.removeNonGrantableFlags();
|
||||
}
|
||||
|
||||
}
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Loading…
Reference in New Issue
Block a user