Commit Graph

766 Commits

Author SHA1 Message Date
proller
300bbdfb43 Fix cyclic lib depend ppart 2: clean 2018-12-28 21:15:26 +03:00
proller
0c78ef8573 Fix cyclic lib depend (make clickhouse_compression lib) 2018-12-28 20:27:53 +03:00
alexey-milovidov
d16447c3f4
Merge pull request #3899 from yandex/custom_compression_codecs
Custom compression codecs
2018-12-28 08:29:09 +03:00
proller
a75d8794ef Merge remote-tracking branch 'upstream/master' into fix10 2018-12-27 21:21:20 +03:00
proller
1a0a879a51 Build fixes 2018-12-27 18:23:37 +03:00
Alexey Milovidov
2e323192a2 Addition to prev. revision #3926 2018-12-27 03:02:11 +03:00
Alexey Milovidov
15a218f951 Fixed UB #3926 2018-12-26 09:03:18 +03:00
Alexey Milovidov
3d3385139b Fixed error #3926 2018-12-26 07:09:24 +03:00
Alexey Milovidov
ee953b4653 Fixed UB (requires performance testing) #3569 2018-12-26 03:31:51 +03:00
alesapin
6e8ed0c942 Merge branch 'master' into custom_compression_codecs 2018-12-25 14:34:19 +03:00
alesapin
c91687f9d7 Small refactoring 2018-12-25 13:35:46 +03:00
Alexey Milovidov
66a0c1691d Added more formats to "parseDateTimeBestEffort" function [#CLICKHOUSE-2] 2018-12-24 21:28:41 +03:00
alesapin
bbdfc77f21
Update CachedCompressedReadBuffer.h 2018-12-24 19:12:25 +03:00
alesapin
6f7333840a Move header with compression info to another place 2018-12-21 16:25:39 +03:00
alesapin
7c11455bd3 Remove compression settings 2018-12-21 15:17:30 +03:00
alesapin
5b23210062 Merge branch 'master' into custom_compression_codecs 2018-12-21 11:57:21 +03:00
Alexey Milovidov
980d6d9e46 Fixed error in "regexpQuoteMeta" function #3826 2018-12-20 21:16:06 +03:00
alesapin
aee6c0a606 Remove dump code 2018-12-20 20:37:02 +03:00
alesapin
512fe3c854 Refactoring (sanitizer failure) 2018-12-19 20:20:18 +03:00
alesapin
5c30bd2384 Compilable code 2018-12-13 13:25:11 +03:00
zhang2014
6d0c4eaf89 ISSUES-838 add lz4、none、zstd codec 2018-12-13 12:10:34 +08:00
alesapin
80b49e4c0a Add integration test, put dependent libs to contrib, slightly refine
code.
2018-12-07 14:41:24 +03:00
alesapin
03ce4c5bc5
Merge branch 'master' into master 2018-12-03 13:02:34 +03:00
Alexey Milovidov
2cf627457a Fixed error #3720 2018-12-01 05:38:54 +03:00
Alexey Milovidov
be80b7b188 Added suggestion from Nikolay Kochetov #3720 2018-11-30 22:47:47 +03:00
Alexey Milovidov
35f6fb5ae6 Avoid memory over-allocation when large value of "max_query_size" is specified [#CLICKHOUSE-4169] 2018-11-30 18:36:41 +03:00
proller
dd2371e071 CLICKHOUSE-4137 DictionaryFactory, DictionarySourceFactory (#3653)
* Split ComplexKeyCacheDictionary to faster compile (part2)

* Dictionaries as lib WIP

* wip

* clean

* Fix build with old capnp

* fix

* wip

* fixes

* fix

* clean

* clean

* clean

* wip

* wip

* wip

* flat

* wip

* cache

* clean

* wip

* faster

* fix style

* fixes

* clean

* clean

* Split CacheDictionary.cpp for faster compile

* fix

* fix

* Less memory usage while compiling

* missing file

* format

* Update registerDictionaries.h

* clean
2018-11-28 14:37:12 +03:00
Alexey Milovidov
2115c31dda Fixed wrong exception message [#CLICKHOUSE-2] 2018-11-26 20:11:04 +03:00
Alexey Milovidov
c77930fe30 Whitespaces [#CLICKHOUSE-2] 2018-11-26 03:56:50 +03:00
chenxing.xc
ff6e849c0f address review comments | 2018-11-25 12:22:01 +08:00
Alexey Milovidov
141e9799e4 Fixed error with "extractURLParameter" function (read after buffer); improved performance; added support for zero bytes in URLs; renamed Chars_t type [#CLICKHOUSE-2] 2018-11-25 03:08:50 +03:00
Alexey Milovidov
7d6fffff48 More style checks [#CLICKHOUSE-3] 2018-11-24 04:48:06 +03:00
alexey-milovidov
507990d5f2
Merge branch 'master' into master 2018-11-23 23:41:38 +03:00
Alexey Milovidov
dbbacf4f87 Added more style checks [#CLICKHOUSE-3] 2018-11-23 21:52:00 +03:00
alesapin
7f40494330 Enable keep alive only with patched poco 2018-11-23 12:24:59 +03:00
alexey-milovidov
3729c4e532
Update ReadBufferFromHDFS.h 2018-11-23 03:20:33 +03:00
alexey-milovidov
2c2d75cceb
Don't allow exceptions without code (#3645)
* Don't allow exceptions without code #3633

* Don't allow exceptions without code #3633
2018-11-23 00:19:58 +03:00
Alexey Milovidov
a5d1fc0ca7 Removed unused constructor #3633 2018-11-22 21:09:17 +03:00
proller
1d40c2c871 Better link and deps (#3631)
* try fix 1

* Better deps

* fix

* fix

* fix

* Update CMakeLists.txt

* Update CMakeLists.txt

* fix

* fix

* fix

* disable old

* Update find_odbc.cmake
2018-11-21 22:20:27 +03:00
alexey-milovidov
7a98d55415
Update ReadBufferFromHDFS.h 2018-11-20 21:31:21 +03:00
alesapin
f08e090bda Better uri scheme check and fix in poco submodule 2018-11-20 20:21:32 +03:00
alesapin
01e2137364 CLICKHOUSE-4112: Add keepalive for http connections 2018-11-20 16:15:44 +03:00
chenxing.xc
76843c2dc0 hdfs read support 2018-11-19 16:17:09 +08:00
alexey-milovidov
e7fb653d0c
Update HTTPCommon.cpp 2018-11-19 01:20:36 +03:00
alexey-milovidov
f35d300d33
Update HTTPCommon.cpp 2018-11-19 01:15:42 +03:00
alesapin
2a4a8d180b Try to change stupid naming 2018-11-16 16:33:43 +03:00
alesapin
d4ce674b93 Add http connection pool between replicas 2018-11-16 16:15:17 +03:00
alesapin
bdb473b425 Fix AIO unknown compression method when merge tree has mark to last byte of file 2018-11-06 20:29:31 +03:00
CurtizJ
7963e952f1 rename setting to tcp_keep_alive 2018-10-23 02:02:57 +03:00
CurtizJ
a05aa5448d add setting keep_alive 2018-10-22 21:09:55 +03:00
Alexey Milovidov
681929cbda Attempt to fix performance regression while parsing JSONEachRow #3223 2018-10-10 00:32:15 +03:00
proller
ec11bef633
Build fixes (#3181) 2018-09-28 17:53:20 +03:00
Alexey Milovidov
599ec4c577 Merge branch 'veloman-yunkan-jsoneachrow_import_nested' 2018-09-17 23:36:01 +03:00
Alexey Milovidov
8852660b24 Merge branch 'jsoneachrow_import_nested' of https://github.com/veloman-yunkan/ClickHouse into veloman-yunkan-jsoneachrow_import_nested 2018-09-17 23:26:43 +03:00
alexey-milovidov
006119b81d
Merge pull request #3155 from yandex/fix-float-parsing-with-custom-csv-delimiter
Fix float parsing with custom csv delimiter
2018-09-17 22:39:45 +03:00
Alexey Milovidov
c00147f37d Fixed parsing of floats in CSV file with custom delimiter #3142 2018-09-17 21:17:30 +03:00
chertus
7db097a49a support e-notation for decimal input [issue-3135] 2018-09-17 20:07:38 +03:00
proller
d752d5d62d Prepare to new poco, build fixes (#3072)
* WIp

* fix

* fix

* test

* fix poco netssl find

* clean

* fixes

* wip

* wip

* new poco

* fixes

* Fixes

* wip

*  clean

* clean

*     clean

* clean

* Update HTTPHandlerFactory.h
2018-09-14 22:48:51 +03:00
Veloman Yunkan
47eb0e28b3 Nested JSON data is mapped to nested table data 2018-09-14 12:15:32 +00:00
chertus
dd5c55df2c decimal sum [CLICKHOUSE-3765] 2018-09-11 21:42:06 +03:00
proller
db13ef96d6 Fix ssl init (#3056)
* Try to fix ../libs/libglibc-compatibility/glibc-compatibility.c:151: multiple definition of `explicit_bzero'
contrib/ssl/crypto/libcrypto.a(explicit_bzero.c.o):../contrib/ssl/crypto/compat/explicit_bzero.c:16: first defined here

* Fix ssl init

* fix

* noncopyable

* clean
2018-09-06 21:05:33 +03:00
Alexey Milovidov
2332bf1a50 Code consistency [#CLICKHOUSE-3943] 2018-09-04 22:34:34 +03:00
Alexey Milovidov
d10f9200b9 Consistency of FileOpen event [#CLICKHOUSE-3943] 2018-09-04 22:24:45 +03:00
alexey-milovidov
72c799ad76
Merge pull request #3008 from 4ertus2/decimal
even more Decimal
2018-09-03 07:19:33 +03:00
Alexey Milovidov
538edbb8e2 Addition to prev. revision [#CLICKHOUSE-2] 2018-09-02 06:33:48 +03:00
Alexey Milovidov
fc27be314b Avoid temporary nullptr references [#CLICKHOUSE-2] 2018-09-02 06:00:04 +03:00
Alexey Milovidov
f7107f12b4 Merge branch 'master' into extra-warnings 2018-09-02 03:13:36 +03:00
Alexey Milovidov
8eb746b216 Documentation for ProfileEvents (draft) #3010 2018-09-02 02:13:17 +03:00
chertus
fe1e3300b7 remove dirty code [CLICKHOUSE-3765] 2018-08-31 20:36:27 +03:00
alexey-milovidov
a05cdfbb45
Merge branch 'master' into extra-warnings 2018-08-31 03:03:44 +03:00
Alexey Milovidov
5df5df8971 utils/check-style/check-style -n 2018-08-31 00:13:34 +03:00
Alexey Milovidov
06053d9c1f Added comments #2980 2018-08-30 23:12:15 +03:00
alexey-milovidov
e0b1b5ff78
Merge pull request #2980 from amosbird/memcpySmall
Padding for IO buffers.
2018-08-30 22:53:47 +03:00
chertus
7bac9698f6 fix build 2018-08-29 16:24:33 +03:00
chertus
a69c936b44 improve field visitors for UUID CLICKHOUSE-3249 2018-08-29 14:58:40 +03:00
Amos Bird
8851fbcab2 Padding for IO buffers.
Testing data

```
select 'aaaaaaaa','bbbbbbbb','cccccccc','dddddddd','eeeeeeee','ffffffff','gggg','hhh' from numbers(3000000) into outfile '/tmp/test.tsv'
```

Testing command
```
echo "select count() from file('/tmp/test.tsv', CSV, 'a String, b String, c String, d String, e String, f String, g String, h String') where not ignore(e)" | clickhouse-benchmark
```

TSV parser has less overhead than CSV, using it would better unveil the benefits of memcpySmall.

Before
```
QPS: 1.662, RPS: 4985463.906, MiB/s: 603.823, result RPS: 1.662, result MiB/s: 0.000.
0.000%  0.559 sec.
10.000% 0.564 sec.
20.000% 0.568 sec.
30.000% 0.572 sec.
40.000% 0.575 sec.
50.000% 0.581 sec.
60.000% 0.592 sec.
70.000% 0.624 sec.
80.000% 0.639 sec.
90.000% 0.664 sec.
95.000% 0.686 sec.
99.000% 0.711 sec.
99.900% 0.715 sec.
99.990% 0.716 sec.
```

After
```
QPS: 1.861, RPS: 5582303.107, MiB/s: 676.110, result RPS: 1.861, result MiB/s: 0.000.
0.000%  0.510 sec.
10.000% 0.514 sec.
20.000% 0.517 sec.
30.000% 0.521 sec.
40.000% 0.523 sec.
50.000% 0.527 sec.
60.000% 0.530 sec.
70.000% 0.539 sec.
80.000% 0.558 sec.
90.000% 0.584 sec.
95.000% 0.589 sec.
99.000% 0.608 sec.
99.900% 0.655 sec.
99.990% 0.663 sec.
```
2018-08-29 09:39:44 +08:00
alexey-milovidov
a5d7097c08
Merge pull request #2977 from amosbird/sse2csv
Enable sse2 for CSV parsing.
2018-08-29 03:04:17 +03:00
chertus
6a02b18b93 field visitors for UUID CLICKHOUSE-3249 2018-08-28 21:36:41 +03:00
Amos Bird
8d6a26b8df Enable sse2 for CSV parsing.
Testing data

```
select 'aaaaaaaa,bbbbbbbb,cccccccc,dddddddd,eeeeeeee,ffffffff,gggg,hhh' from numbers(3000000) into outfile '/tmp/test.csv'
```

Testing command
```
echo "select count() from file('/tmp/test.csv', CSV, 'a String, b String, c String, d String, e String, f String, g String, h String') where not ignore(e)" | clickhouse-benchmark

```

Before
```
QPS: 1.317, RPS: 3949749.687, MiB/s: 478.380, result RPS: 1.317, result MiB/s: 0.000.
0.000%  0.704 sec.
10.000% 0.712 sec.
20.000% 0.718 sec.
30.000% 0.726 sec.
40.000% 0.739 sec.
50.000% 0.754 sec.
60.000% 0.770 sec.
70.000% 0.788 sec.
80.000% 0.798 sec.
90.000% 0.815 sec.
95.000% 0.826 sec.
99.000% 0.850 sec.
99.900% 0.857 sec.
99.990% 0.858 sec.
```

After
```
QPS: 1.533, RPS: 4598308.336, MiB/s: 556.932, result RPS: 1.533, result MiB/s: 0.000.
0.000%  0.626 sec.
10.000% 0.635 sec.
20.000% 0.639 sec.
30.000% 0.642 sec.
40.000% 0.643 sec.
50.000% 0.645 sec.
60.000% 0.649 sec.
70.000% 0.652 sec.
80.000% 0.658 sec.
90.000% 0.682 sec.
95.000% 0.710 sec.
99.000% 0.727 sec.
99.900% 0.733 sec.
99.990% 0.734 sec.
```
2018-08-28 19:46:11 +08:00
Alexey Milovidov
24bd803c07 Whitespaces #2758 2018-08-27 22:06:32 +03:00
Alexey Milovidov
d4ea8bdb88 Merge remote-tracking branch 'origin/master' into extra-warnings 2018-08-27 20:18:39 +03:00
alexey-milovidov
98148d914c
Merge pull request #2958 from blahgeek/skip-json-object
Skip object fields on json input
2018-08-27 20:07:32 +03:00
Alexey Milovidov
486d87ebcd Merge remote-tracking branch 'origin/master' into extra-warnings 2018-08-27 20:01:33 +03:00
chertus
ee873c8e9f Merge branch 'master' into decimal 2018-08-27 19:18:48 +03:00
alexey-milovidov
7de0549879
Update ReadHelpers.h 2018-08-27 18:00:08 +03:00
alexey-milovidov
01f01b1ee1
Update ReadHelpers.cpp 2018-08-27 17:54:58 +03:00
BlahGeek
739119d355 skip object fields on json input 2018-08-27 12:01:47 +08:00
Alexey Milovidov
eed804ba94 Enabling extra warnings [#CLICKHOUSE-2] 2018-08-26 15:38:45 +03:00
Alexey Milovidov
281faa1686 Enabling -Wshadow [#CLICKHOUSE-2] 2018-08-26 05:13:41 +03:00
Alexey Milovidov
8a0239b6ad Enabling -Wshadow [#CLICKHOUSE-2] 2018-08-26 05:08:35 +03:00
Alexey Milovidov
a1a70d4a65 Enabling -Wzero-as-null-pointer-constant [#CLICKHOUSE-2] 2018-08-26 04:56:34 +03:00
Alexey Milovidov
44e848856e Enabling -Wunused-exception-parameter [#CLICKHOUSE-2] 2018-08-26 04:33:20 +03:00
Alexey Milovidov
2ce783e016 Merge branch 'read' of https://github.com/amosbird/ClickHouse into amosbird-read 2018-08-26 02:55:22 +03:00
alexey-milovidov
00404a911c
Update VarInt.h 2018-08-26 02:46:18 +03:00
Amos Bird
a589230249 Faster path for varint read.
clickhouse-benchmark <<< 'select count() from s where not ignore(s);'

before this patch:

```

QPS: 0.732, RPS: 2346562049.608, MiB/s: 22378.560, result RPS: 0.732, result MiB/s: 0.000.
0.000%  1.310 sec.
10.000% 1.321 sec.
20.000% 1.327 sec.
30.000% 1.337 sec.
40.000% 1.343 sec.
50.000% 1.359 sec.
60.000% 1.366 sec.
70.000% 1.381 sec.
80.000% 1.400 sec.
90.000% 1.434 sec.
95.000% 1.448 sec.
99.000% 1.489 sec.
99.900% 1.499 sec.
99.990% 1.500 sec.

```

after this patch:

```
QPS: 0.787, RPS: 2524560389.064, MiB/s: 24076.084, result RPS: 0.787, result MiB/s: 0.000.
0.000%  1.228 sec.
10.000% 1.232 sec.
20.000% 1.235 sec.
30.000% 1.241 sec.
40.000% 1.246 sec.
50.000% 1.256 sec.
60.000% 1.265 sec.
70.000% 1.278 sec.
80.000% 1.296 sec.
90.000% 1.321 sec.
95.000% 1.354 sec.
99.000% 1.421 sec.
99.900% 1.453 sec.
99.990% 1.456 sec.
```

I also tried a SSE2 implementation and it's much slower (50%)
2018-08-26 02:23:26 +08:00
Amos Bird
9623e59870 Simplify read helpers.
Use similar APIs like https://en.cppreference.com/w/cpp/string/byte/strstr
2018-08-26 01:27:28 +08:00
Amos Bird
5f03082af9 Make sure BufferBase is fully swapped. 2018-08-25 19:58:16 +08:00
chertus
54a3c028c5 Merge branch 'master' into decimal 2018-08-24 14:19:41 +03:00
Alexey Milovidov
080b0f7f1f Speed up fuzz tests, part 2 #2924 2018-08-24 10:35:53 +03:00
Alexey Milovidov
a2fefa1617 Speed up fuzz tests, part 2 #2924 2018-08-24 10:30:53 +03:00