thevar1able/ClickHouse
1
0
Fork 0
mirror of https://github.com/ClickHouse/ClickHouse.git synced 2024-09-23 10:10:50 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge

Browse Source
This commit is contained in:
Evgeniy Gatov 2015-03-30 18:31:23 +03:00
commit 5e012d9ac8
23 changed files with 572 additions and 875 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

18
dbms/benchmark/clickhouse/benchmark-new.sh Executable file
View File

@ -0,0 +1,18 @@
#!/bin/bash
QUERIES_FILE="queries.sql"
TABLE=$1
TRIES=3
cat "$QUERIES_FILE" | sed "s/{table}/${TABLE}/g" | while read query; do
sync
echo 3 | sudo tee /proc/sys/vm/drop_caches >/dev/null
echo -n "["
for i in $(seq 1 $TRIES); do
RES=$(clickhouse-client --time --format=Null --query="$query" 2>&1)
[[ "$?" == "0" ]] && echo -n "${RES}" || echo -n "null"
[[ "$i" != $TRIES ]] && echo -n ", "
done
echo "],"
done

366
dbms/benchmark/clickhouse/benchmark.sh
View File

@ -1,366 +0,0 @@
#!/bin/bash
test_table="hits_100m"
start_date="'2013-07-01'"
early_stop_date="'2013-07-02'"
stop_date="'2013-07-31'"
counter_id=34
function run_ck_server
{
sudo sh -c " ulimit -v 54000000; /etc/init.d/clickhouse-server restart"
}
# execute queries
function execute()
{
queries=("${@}")
queries_count=${#queries[@]}
if [ -z $TIMES ]; then
TIMES=1
fi
index=0
comment_re='\#.*'
while [ "$index" -lt "$queries_count" ]; do
query=${queries[$index]}
if [[ $query =~ $comment_re ]]; then
echo "$query"
echo
else
sync
sudo sh -c "echo 3 > /proc/sys/vm/drop_caches"
for i in $(seq $TIMES)
do
expect -f ./expect.tcl "$query"
if [ "$?" != "0" ]; then
echo "Error: $?"
#break
fi
# restart clickhouse if failed
ps aux | grep -P '\d+ clickhouse-server'
if [ "$?" != "0" ]; then
run_ck_server
fi
done
fi
let "index = $index + 1"
echo "Ran $index queries." >&2
done
}
init_queries=(
# DB structure with array arguments
#"CREATE TABLE $test_table ( WatchID UInt64, JavaEnable UInt8, Title String, GoodEvent Int16, EventTime DateTime, EventDate Date, CounterID UInt32, ClientIP UInt32, RegionID UInt32, UserID UInt64, CounterClass Int8, OS UInt8, UserAgent UInt8, URL String, Referer String, Refresh UInt8, RefererCategoryID UInt16, RefererRegionID UInt32, URLCategoryID UInt16, URLRegionID UInt32, ResolutionWidth UInt16, ResolutionHeight UInt16, ResolutionDepth UInt8, FlashMajor UInt8, FlashMinor UInt8, FlashMinor2 String, NetMajor UInt8, NetMinor UInt8, UserAgentMajor UInt16, UserAgentMinor FixedString(2), CookieEnable UInt8, JavascriptEnable UInt8, IsMobile UInt8, MobilePhone UInt8, MobilePhoneModel String, Params String, IPNetworkID UInt32, TraficSourceID Int8, SearchEngineID UInt16, SearchPhrase String, AdvEngineID UInt8, IsArtifical UInt8, WindowClientWidth UInt16, WindowClientHeight UInt16, ClientTimeZone Int16, ClientEventTime DateTime, SilverlightVersion1 UInt8, SilverlightVersion2 UInt8, SilverlightVersion3 UInt32, SilverlightVersion4 UInt16, PageCharset String, CodeVersion UInt32, IsLink UInt8, IsDownload UInt8, IsNotBounce UInt8, FUniqID UInt64, OriginalURL String, HID UInt32, IsOldCounter UInt8, IsEvent UInt8, IsParameter UInt8, DontCountHits UInt8, WithHash UInt8, HitColor FixedString(1), LocalEventTime DateTime, Age UInt8, Sex UInt8, Income UInt8, Interests UInt16, Robotness UInt8, GeneralInterests Array(UInt16), RemoteIP UInt32, WindowName Int32, OpenerName Int32, HistoryLength Int16, BrowserLanguage FixedString(2), BrowserCountry FixedString(2), SocialNetwork String, SocialAction String, HTTPError UInt16, SendTiming UInt32, DNSTiming UInt32, ConnectTiming UInt32, ResponseStartTiming UInt32, ResponseEndTiming UInt32, FetchTiming UInt32, SocialSourceNetworkID UInt8, SocialSourcePage String, ParamPrice Int64, ParamOrderID String, ParamCurrency FixedString(3), ParamCurrencyID UInt16, GoalsReached Array(UInt32), OpenstatServiceName String, OpenstatCampaignID String, OpenstatAdID String, OpenstatSourceID String, UTMSource String, UTMMedium String, UTMCampaign String, UTMContent String, UTMTerm String, FromTag String, HasGCLID UInt8, RefererHash UInt64, URLHash UInt64, CLID UInt32 ) ENGINE = MergeTree(EventDate, intHash32(UserID), tuple(CounterID, EventDate, intHash32(UserID), EventTime), 8192);"
#DB structure without array arguments
#"CREATE TABLE $test_table ( WatchID UInt64, JavaEnable UInt8, Title String, GoodEvent Int16, EventTime DateTime, EventDate Date, CounterID UInt32, ClientIP UInt32, RegionID UInt32, UserID UInt64, CounterClass Int8, OS UInt8, UserAgent UInt8, URL String, Referer String, Refresh UInt8, RefererCategoryID UInt16, RefererRegionID UInt32, URLCategoryID UInt16, URLRegionID UInt32, ResolutionWidth UInt16, ResolutionHeight UInt16, ResolutionDepth UInt8, FlashMajor UInt8, FlashMinor UInt8, FlashMinor2 String, NetMajor UInt8, NetMinor UInt8, UserAgentMajor UInt16, UserAgentMinor FixedString(2), CookieEnable UInt8, JavascriptEnable UInt8, IsMobile UInt8, MobilePhone UInt8, MobilePhoneModel String, Params String, IPNetworkID UInt32, TraficSourceID Int8, SearchEngineID UInt16, SearchPhrase String, AdvEngineID UInt8, IsArtifical UInt8, WindowClientWidth UInt16, WindowClientHeight UInt16, ClientTimeZone Int16, ClientEventTime DateTime, SilverlightVersion1 UInt8, SilverlightVersion2 UInt8, SilverlightVersion3 UInt32, SilverlightVersion4 UInt16, PageCharset String, CodeVersion UInt32, IsLink UInt8, IsDownload UInt8, IsNotBounce UInt8, FUniqID UInt64, OriginalURL String, HID UInt32, IsOldCounter UInt8, IsEvent UInt8, IsParameter UInt8, DontCountHits UInt8, WithHash UInt8, HitColor FixedString(1), LocalEventTime DateTime, Age UInt8, Sex UInt8, Income UInt8, Interests UInt16, Robotness UInt8, RemoteIP UInt32, WindowName Int32, OpenerName Int32, HistoryLength Int16, BrowserLanguage FixedString(2), BrowserCountry FixedString(2), SocialNetwork String, SocialAction String, HTTPError UInt16, SendTiming UInt32, DNSTiming UInt32, ConnectTiming UInt32, ResponseStartTiming UInt32, ResponseEndTiming UInt32, FetchTiming UInt32, SocialSourceNetworkID UInt8, SocialSourcePage String, ParamPrice Int64, ParamOrderID String, ParamCurrency FixedString(3), ParamCurrencyID UInt16, OpenstatServiceName String, OpenstatCampaignID String, OpenstatAdID String, OpenstatSourceID String, UTMSource String, UTMMedium String, UTMCampaign String, UTMContent String, UTMTerm String, FromTag String, HasGCLID UInt8, RefererHash UInt64, URLHash UInt64, CLID UInt32 ) ENGINE = MergeTree(EventDate, intHash32(UserID), tuple(CounterID, EventDate, intHash32(UserID), EventTime), 8192);"
#modified table without uint
"CREATE TABLE $test_table ( WatchID Int64, JavaEnable UInt8, Title String, GoodEvent Int16, EventTime DateTime, EventDate Date, CounterID UInt32, ClientIP UInt32, RegionID UInt32, UserID Int64, CounterClass Int8, OS UInt8, UserAgent UInt8, URL String, Referer String, Refresh UInt8, RefererCategoryID UInt16, RefererRegionID UInt32, URLCategoryID UInt16, URLRegionID UInt32, ResolutionWidth UInt16, ResolutionHeight UInt16, ResolutionDepth UInt8, FlashMajor UInt8, FlashMinor UInt8, FlashMinor2 String, NetMajor UInt8, NetMinor UInt8, UserAgentMajor UInt16, UserAgentMinor FixedString(2), CookieEnable UInt8, JavascriptEnable UInt8, IsMobile UInt8, MobilePhone UInt8, MobilePhoneModel String, Params String, IPNetworkID UInt32, TraficSourceID Int8, SearchEngineID UInt16, SearchPhrase String, AdvEngineID UInt8, IsArtifical UInt8, WindowClientWidth UInt16, WindowClientHeight UInt16, ClientTimeZone Int16, ClientEventTime DateTime, SilverlightVersion1 UInt8, SilverlightVersion2 UInt8, SilverlightVersion3 UInt32, SilverlightVersion4 UInt16, PageCharset String, CodeVersion UInt32, IsLink UInt8, IsDownload UInt8, IsNotBounce UInt8, FUniqID Int64, OriginalURL String, HID UInt32, IsOldCounter UInt8, IsEvent UInt8, IsParameter UInt8, DontCountHits UInt8, WithHash UInt8, HitColor FixedString(1), LocalEventTime DateTime, Age UInt8, Sex UInt8, Income UInt8, Interests UInt16, Robotness UInt8, RemoteIP UInt32, WindowName Int32, OpenerName Int32, HistoryLength Int16, BrowserLanguage FixedString(2), BrowserCountry FixedString(2), SocialNetwork String, SocialAction String, HTTPError UInt16, SendTiming UInt32, DNSTiming UInt32, ConnectTiming UInt32, ResponseStartTiming UInt32, ResponseEndTiming UInt32, FetchTiming UInt32, SocialSourceNetworkID UInt8, SocialSourcePage String, ParamPrice Int64, ParamOrderID String, ParamCurrency FixedString(3), ParamCurrencyID UInt16, OpenstatServiceName String, OpenstatCampaignID String, OpenstatAdID String, OpenstatSourceID String, UTMSource String, UTMMedium String, UTMCampaign String, UTMContent String, UTMTerm String, FromTag String, HasGCLID UInt8, RefererHash Int64, URLHash Int64, CLID UInt32, UserIDHash UInt64 ) ENGINE = MergeTree(EventDate, intHash32(UserID), tuple(CounterID, EventDate, intHash32(UserID), EventTime), 8192);"
)
test_queries=(
"SELECT count() FROM $test_table;"
"SELECT count() FROM $test_table WHERE AdvEngineID != 0;"
"SELECT sum(AdvEngineID), count(), avg(ResolutionWidth) FROM $test_table;"
"SELECT sum(UserID) FROM $test_table;"
"SELECT uniq(UserID) FROM $test_table;"
"SELECT uniq(SearchPhrase) FROM $test_table;"
"SELECT min(EventDate), max(EventDate) FROM $test_table;"
"SELECT AdvEngineID, count() FROM $test_table WHERE AdvEngineID != 0 GROUP BY AdvEngineID ORDER BY count() DESC;"
"#- мощная фильтрация. После фильтрации почти ничего не остаётся, но делаем ещё агрегацию.;"
"SELECT RegionID, uniq(UserID) AS u FROM $test_table GROUP BY RegionID ORDER BY u DESC LIMIT 10;"
"#- агрегация, среднее количество ключей.;"
"SELECT RegionID, sum(AdvEngineID), count() AS c, avg(ResolutionWidth), uniq(UserID) FROM $test_table GROUP BY RegionID ORDER BY c DESC LIMIT 10;"
"#- агрегация, среднее количество ключей, несколько агрегатных функций.;"
"SELECT MobilePhoneModel, uniq(UserID) AS u FROM $test_table WHERE MobilePhoneModel != '' GROUP BY MobilePhoneModel ORDER BY u DESC LIMIT 10;"
"#- мощная фильтрация по строкам, затем агрегация по строкам.;"
"SELECT MobilePhone, MobilePhoneModel, uniq(UserID) AS u FROM $test_table WHERE MobilePhoneModel != '' GROUP BY MobilePhone, MobilePhoneModel ORDER BY u DESC LIMIT 10;"
"#- мощная фильтрация по строкам, затем агрегация по паре из числа и строки.;"
"SELECT SearchPhrase, count() AS c FROM $test_table WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;"
"#- средняя фильтрация по строкам, затем агрегация по строкам, большое количество ключей.;"
"SELECT SearchPhrase, uniq(UserID) AS u FROM $test_table WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY u DESC LIMIT 10;"
"#- агрегация чуть сложнее.;"
"SELECT SearchEngineID, SearchPhrase, count() AS c FROM $test_table WHERE SearchPhrase != '' GROUP BY SearchEngineID, SearchPhrase ORDER BY c DESC LIMIT 10;"
"#- агрегация по числу и строке, большое количество ключей.;"
"SELECT UserID, count() FROM $test_table GROUP BY UserID ORDER BY count() DESC LIMIT 10;"
"#- агрегация по очень большому количеству ключей, может не хватить оперативки.;"
"SELECT UserID, SearchPhrase, count() FROM $test_table GROUP BY UserID, SearchPhrase ORDER BY count() DESC LIMIT 10;"
"#- ещё более сложная агрегация.;"
"SELECT UserID, SearchPhrase, count() FROM $test_table GROUP BY UserID, SearchPhrase LIMIT 10;"
"#- то же самое, но без сортировки.;"
"SELECT UserID, toMinute(EventTime) AS m, SearchPhrase, count() FROM $test_table GROUP BY UserID, m, SearchPhrase ORDER BY count() DESC LIMIT 10;"
"#- ещё более сложная агрегация, не стоит выполнять на больших таблицах.;"
"SELECT UserID FROM $test_table WHERE UserID = 12345678901234567890;"
"#- мощная фильтрация по столбцу типа UInt64.;"
"SELECT count() FROM $test_table WHERE URL LIKE '%metrika%';"
"#- фильтрация по поиску подстроки в строке.;"
"SELECT SearchPhrase, any(URL), count() AS c FROM $test_table WHERE URL LIKE '%metrika%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;"
"#- вынимаем большие столбцы, фильтрация по строке.;"
"SELECT SearchPhrase, any(URL), any(Title), count() AS c, uniq(UserID) FROM $test_table WHERE Title LIKE '%Яндекс%' AND URL NOT LIKE '%.yandex.%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;"
"#- чуть больше столбцы.;"
"SELECT * FROM $test_table WHERE URL LIKE '%metrika%' ORDER BY EventTime LIMIT 10;"
"#- плохой запрос - вынимаем все столбцы.;"
"SELECT SearchPhrase FROM $test_table WHERE SearchPhrase != '' ORDER BY EventTime LIMIT 10;"
"#- большая сортировка.;"
"SELECT SearchPhrase FROM $test_table WHERE SearchPhrase != '' ORDER BY SearchPhrase LIMIT 10;"
"#- большая сортировка по строкам.;"
"SELECT SearchPhrase FROM $test_table WHERE SearchPhrase != '' ORDER BY EventTime, SearchPhrase LIMIT 10;"
"#- большая сортировка по кортежу.;"
"SELECT CounterID, avg(length(URL)) AS l, count() AS c FROM $test_table WHERE URL != '' GROUP BY CounterID HAVING c > 100000 ORDER BY l DESC LIMIT 25;"
"#- считаем средние длины URL для крупных счётчиков.;"
"SELECT domainWithoutWWW(Referer) AS key, avg(length(Referer)) AS l, count() AS c, any(Referer) FROM $test_table WHERE Referer != '' GROUP BY key HAVING c > 100000 ORDER BY l DESC LIMIT 25;"
"#- то же самое, но с разбивкой по доменам.;"
"SELECT sum(ResolutionWidth), sum(ResolutionWidth + 1), sum(ResolutionWidth + 2), sum(ResolutionWidth + 3), sum(ResolutionWidth + 4), sum(ResolutionWidth + 5), sum(ResolutionWidth + 6), sum(ResolutionWidth + 7), sum(ResolutionWidth + 8), sum(ResolutionWidth + 9), sum(ResolutionWidth + 10), sum(ResolutionWidth + 11), sum(ResolutionWidth + 12), sum(ResolutionWidth + 13), sum(ResolutionWidth + 14), sum(ResolutionWidth + 15), sum(ResolutionWidth + 16), sum(ResolutionWidth + 17), sum(ResolutionWidth + 18), sum(ResolutionWidth + 19), sum(ResolutionWidth + 20), sum(ResolutionWidth + 21), sum(ResolutionWidth + 22), sum(ResolutionWidth + 23), sum(ResolutionWidth + 24), sum(ResolutionWidth + 25), sum(ResolutionWidth + 26), sum(ResolutionWidth + 27), sum(ResolutionWidth + 28), sum(ResolutionWidth + 29), sum(ResolutionWidth + 30), sum(ResolutionWidth + 31), sum(ResolutionWidth + 32), sum(ResolutionWidth + 33), sum(ResolutionWidth + 34), sum(ResolutionWidth + 35), sum(ResolutionWidth + 36), sum(ResolutionWidth + 37), sum(ResolutionWidth + 38), sum(ResolutionWidth + 39), sum(ResolutionWidth + 40), sum(ResolutionWidth + 41), sum(ResolutionWidth + 42), sum(ResolutionWidth + 43), sum(ResolutionWidth + 44), sum(ResolutionWidth + 45), sum(ResolutionWidth + 46), sum(ResolutionWidth + 47), sum(ResolutionWidth + 48), sum(ResolutionWidth + 49), sum(ResolutionWidth + 50), sum(ResolutionWidth + 51), sum(ResolutionWidth + 52), sum(ResolutionWidth + 53), sum(ResolutionWidth + 54), sum(ResolutionWidth + 55), sum(ResolutionWidth + 56), sum(ResolutionWidth + 57), sum(ResolutionWidth + 58), sum(ResolutionWidth + 59), sum(ResolutionWidth + 60), sum(ResolutionWidth + 61), sum(ResolutionWidth + 62), sum(ResolutionWidth + 63), sum(ResolutionWidth + 64), sum(ResolutionWidth + 65), sum(ResolutionWidth + 66), sum(ResolutionWidth + 67), sum(ResolutionWidth + 68), sum(ResolutionWidth + 69), sum(ResolutionWidth + 70), sum(ResolutionWidth + 71), sum(ResolutionWidth + 72), sum(ResolutionWidth + 73), sum(ResolutionWidth + 74), sum(ResolutionWidth + 75), sum(ResolutionWidth + 76), sum(ResolutionWidth + 77), sum(ResolutionWidth + 78), sum(ResolutionWidth + 79), sum(ResolutionWidth + 80), sum(ResolutionWidth + 81), sum(ResolutionWidth + 82), sum(ResolutionWidth + 83), sum(ResolutionWidth + 84), sum(ResolutionWidth + 85), sum(ResolutionWidth + 86), sum(ResolutionWidth + 87), sum(ResolutionWidth + 88), sum(ResolutionWidth + 89) FROM $test_table;"
"#- много тупых агрегатных функций.;"
"SELECT SearchEngineID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM $test_table WHERE SearchPhrase != '' GROUP BY SearchEngineID, ClientIP ORDER BY c DESC LIMIT 10;"
"#- сложная агрегация, для больших таблиц может не хватить оперативки.;"
"SELECT WatchID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM $test_table WHERE SearchPhrase != '' GROUP BY WatchID, ClientIP ORDER BY c DESC LIMIT 10;"
"#- агрегация по двум полям, которая ничего не агрегирует. Для больших таблиц выполнить не получится.;"
"SELECT WatchID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM $test_table GROUP BY WatchID, ClientIP ORDER BY c DESC LIMIT 10;"
"#- то же самое, но ещё и без фильтрации.;"
"SELECT URL, count() AS c FROM $test_table GROUP BY URL ORDER BY c DESC LIMIT 10;"
"#- агрегация по URL.;"
"SELECT 1, URL, count() AS c FROM $test_table GROUP BY 1, URL ORDER BY c DESC LIMIT 10;"
"#- агрегация по URL и числу.;"
"SELECT ClientIP AS x, x - 1, x - 2, x - 3, count() AS c FROM $test_table GROUP BY x, x - 1, x - 2, x - 3 ORDER BY c DESC LIMIT 10;"
"SELECT
URL,
count() AS PageViews
FROM $test_table
WHERE
CounterID = $counter_id
AND EventDate >= toDate($start_date)
AND EventDate <= toDate($stop_date)
AND NOT DontCountHits
AND NOT Refresh
AND notEmpty(URL)
GROUP BY URL
ORDER BY PageViews DESC
LIMIT 10;"
"SELECT
Title,
count() AS PageViews
FROM $test_table
WHERE
CounterID = $counter_id
AND EventDate >= toDate($start_date)
AND EventDate <= toDate($stop_date)
AND NOT DontCountHits
AND NOT Refresh
AND notEmpty(Title)
GROUP BY Title
ORDER BY PageViews DESC
LIMIT 10;"
"SELECT
URL,
count() AS PageViews
FROM $test_table
WHERE
CounterID = $counter_id
AND EventDate >= toDate($start_date)
AND EventDate <= toDate($stop_date)
AND NOT Refresh
AND IsLink
AND NOT IsDownload
GROUP BY URL
ORDER BY PageViews DESC
LIMIT 1000;"
"SELECT
TraficSourceID,
SearchEngineID,
AdvEngineID,
((SearchEngineID = 0 AND AdvEngineID = 0) ? Referer : '') AS Src,
URL AS Dst,
count() AS PageViews
FROM $test_table
WHERE
CounterID = $counter_id
AND EventDate >= toDate($start_date)
AND EventDate <= toDate($stop_date)
AND NOT Refresh
GROUP BY
TraficSourceID,
SearchEngineID,
AdvEngineID,
Src,
Dst
ORDER BY PageViews DESC
LIMIT 1000;"
"SELECT
URLHash,
EventDate,
count() AS PageViews
FROM $test_table
WHERE
CounterID = $counter_id
AND EventDate >= toDate($start_date)
AND EventDate <= toDate($stop_date)
AND NOT Refresh
AND TraficSourceID IN (-1, 6)
AND RefererHash = halfMD5('http://example.ru/')
GROUP BY
URLHash,
EventDate
ORDER BY PageViews DESC
LIMIT 100000;"
"SELECT
WindowClientWidth,
WindowClientHeight,
count() AS PageViews
FROM $test_table
WHERE
CounterID = $counter_id
AND EventDate >= toDate($start_date)
AND EventDate <= toDate($stop_date)
AND NOT Refresh
AND NOT DontCountHits
AND URLHash = halfMD5('http://example.ru/')
GROUP BY
WindowClientWidth,
WindowClientHeight
ORDER BY PageViews DESC
LIMIT 10000;"
"SELECT
toStartOfMinute(EventTime) AS Minute,
count() AS PageViews
FROM $test_table
WHERE
CounterID = $counter_id
AND EventDate >= toDate($start_date)
AND EventDate <= toDate($early_stop_date)
AND NOT Refresh
AND NOT DontCountHits
GROUP BY
Minute
ORDER BY Minute;"
)
function test {
TIMES=3
execute "${test_queries[@]}"
}
function init {
execute "${init_queries[@]}"
}
function debug {
TIMES=3
debug_queries=(
)
execute "${debug_queries[@]}"
}
function usage {
cat <<EOF
usage: $0 options
This script run benhmark for clickhouse
OPTIONS:
-h Show this message
-d Run debug queries
-i Init database
-p log_file Parse log file to columns with result
-t Run tests
EOF
}
function parse_log {
results=$(cat $1 | grep -P 'Elapsed: \d+.\d+ ' | awk '{print $6}')
index=1
for res in $results
do
echo -n "$res "
let "index=$index % 3"
if [ "$index" == "0" ]; then
echo
fi
let "index=$index + 1"
done
}
if [ "$#" == "0" ]; then
usage
exit 0
fi
echo "Start date" $(date)
while getopts “hitdp:” OPTION
do
case $OPTION in
h)
usage
exit 0
;;
i)
init
;;
t)
test
;;
d)
debug
;;
p)
parse_log $OPTARG
;;
?)
usage
exit 0
;;
esac
done
echo "Stop date" $(date)

4
dbms/benchmark/clickhouse/conf.sh
View File

@ -1,4 +0,0 @@
CONF_DIR=/home/kartavyy/benchmark/clickhouse
expect_file=$CONF_DIR/expect.tcl
test_file=$CONF_DIR/queries.sql
etc_init_d_service=/etc/init.d/clickhouse-server-metrika-yandex

13
dbms/benchmark/clickhouse/expect.tcl
View File

@ -1,13 +0,0 @@
#!/bin/expect
# Set timeout
set timeout 600
# Get arguments
set query [lindex $argv 0]
spawn clickhouse-client --multiline;
expect ":) "
send "$query;\r";
expect ":) "
send "quit";

152
dbms/benchmark/clickhouse/queries.sql
View File

@ -1,109 +1,43 @@
SELECT count() FROM hits_10m;
SELECT count() FROM hits_10m WHERE AdvEngineID != 0;
SELECT sum(AdvEngineID), count(), avg(ResolutionWidth) FROM hits_10m ;
SELECT sum(UserID) FROM hits_10m ;
SELECT uniq(UserID) FROM hits_10m ;
SELECT uniq(SearchPhrase) FROM hits_10m ;
SELECT min(EventDate), max(EventDate) FROM hits_10m ;
SELECT AdvEngineID, count() FROM hits_10m WHERE AdvEngineID != 0 GROUP BY AdvEngineID ORDER BY count() DESC;
-- мощная фильтрация. После фильтрации почти ничего не остаётся, но делаем ещё агрегацию.;
SELECT RegionID, uniq(UserID) AS u FROM hits_10m GROUP BY RegionID ORDER BY u DESC LIMIT 10;
-- агрегация, среднее количество ключей.;
SELECT RegionID, sum(AdvEngineID), count() AS c, avg(ResolutionWidth), uniq(UserID) FROM hits_10m GROUP BY RegionID ORDER BY c DESC LIMIT 10;
-- агрегация, среднее количество ключей, несколько агрегатных функций.;
SELECT MobilePhoneModel, uniq(UserID) AS u FROM hits_10m WHERE MobilePhoneModel != '' GROUP BY MobilePhoneModel ORDER BY u DESC LIMIT 10;
-- мощная фильтрация по строкам, затем агрегация по строкам.;
SELECT MobilePhone, MobilePhoneModel, uniq(UserID) AS u FROM hits_10m WHERE MobilePhoneModel != '' GROUP BY MobilePhone, MobilePhoneModel ORDER BY u DESC LIMIT 10;
-- мощная фильтрация по строкам, затем агрегация по паре из числа и строки.;
SELECT SearchPhrase, count() AS c FROM hits_10m WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;
-- средняя фильтрация по строкам, затем агрегация по строкам, большое количество ключей.;
SELECT SearchPhrase, uniq(UserID) AS u FROM hits_10m WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY u DESC LIMIT 10;
-- агрегация чуть сложнее.;
SELECT SearchEngineID, SearchPhrase, count() AS c FROM hits_10m WHERE SearchPhrase != '' GROUP BY SearchEngineID, SearchPhrase ORDER BY c DESC LIMIT 10;
-- агрегация по числу и строке, большое количество ключей.;
SELECT UserID, count() FROM hits_10m GROUP BY UserID ORDER BY count() DESC LIMIT 10;
-- агрегация по очень большому количеству ключей, может не хватить оперативки.;
SELECT UserID, SearchPhrase, count() FROM hits_10m GROUP BY UserID, SearchPhrase ORDER BY count() DESC LIMIT 10;
-- ещё более сложная агрегация.;
SELECT UserID, SearchPhrase, count() FROM hits_10m GROUP BY UserID, SearchPhrase LIMIT 10;
-- то же самое, но без сортировки.;
SELECT UserID, toMinute(EventTime) AS m, SearchPhrase, count() FROM hits_10m GROUP BY UserID, m, SearchPhrase ORDER BY count() DESC LIMIT 10;
-- ещё более сложная агрегация, не стоит выполнять на больших таблицах.;
SELECT UserID FROM hits_10m WHERE UserID = 12345678901234567890;
-- мощная фильтрация по столбцу типа UInt64.;
SELECT count() FROM hits_10m WHERE URL LIKE '%metrika%';
-- фильтрация по поиску подстроки в строке.;
SELECT SearchPhrase, any(URL), count() AS c FROM hits_10m WHERE URL LIKE '%metrika%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;
-- вынимаем большие столбцы, фильтрация по строке.;
SELECT SearchPhrase, any(URL), any(Title), count() AS c, uniq(UserID) FROM hits_10m WHERE Title LIKE '%Яндекс%' AND URL NOT LIKE '%.yandex.%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;
-- чуть больше столбцы.;
SELECT * FROM hits_10m WHERE URL LIKE '%metrika%' ORDER BY EventTime LIMIT 10;
-- плохой запрос - вынимаем все столбцы.;
SELECT SearchPhrase FROM hits_10m WHERE SearchPhrase != '' ORDER BY EventTime LIMIT 10;
-- большая сортировка.;
SELECT SearchPhrase FROM hits_10m WHERE SearchPhrase != '' ORDER BY SearchPhrase LIMIT 10;
-- большая сортировка по строкам.;
SELECT SearchPhrase FROM hits_10m WHERE SearchPhrase != '' ORDER BY EventTime, SearchPhrase LIMIT 10;
-- большая сортировка по кортежу.;
SELECT CounterID, avg(length(URL)) AS l, count() AS c FROM hits_10m WHERE URL != '' GROUP BY CounterID HAVING c > 100000 ORDER BY l DESC LIMIT 25;
-- считаем средние длины URL для крупных счётчиков.;
SELECT domainWithoutWWW(Referer) AS key, avg(length(Referer)) AS l, count() AS c, any(Referer) FROM hits_10m WHERE Referer != '' GROUP BY key HAVING c > 100000 ORDER BY l DESC LIMIT 25;
-- то же самое, но с разбивкой по доменам.;
SELECT sum(ResolutionWidth), sum(ResolutionWidth + 1), sum(ResolutionWidth + 2), sum(ResolutionWidth + 3), sum(ResolutionWidth + 4), sum(ResolutionWidth + 5), sum(ResolutionWidth + 6), sum(ResolutionWidth + 7), sum(ResolutionWidth + 8), sum(ResolutionWidth + 9), sum(ResolutionWidth + 10), sum(ResolutionWidth + 11), sum(ResolutionWidth + 12), sum(ResolutionWidth + 13), sum(ResolutionWidth + 14), sum(ResolutionWidth + 15), sum(ResolutionWidth + 16), sum(ResolutionWidth + 17), sum(ResolutionWidth + 18), sum(ResolutionWidth + 19), sum(ResolutionWidth + 20), sum(ResolutionWidth + 21), sum(ResolutionWidth + 22), sum(ResolutionWidth + 23), sum(ResolutionWidth + 24), sum(ResolutionWidth + 25), sum(ResolutionWidth + 26), sum(ResolutionWidth + 27), sum(ResolutionWidth + 28), sum(ResolutionWidth + 29), sum(ResolutionWidth + 30), sum(ResolutionWidth + 31), sum(ResolutionWidth + 32), sum(ResolutionWidth + 33), sum(ResolutionWidth + 34), sum(ResolutionWidth + 35), sum(ResolutionWidth + 36), sum(ResolutionWidth + 37), sum(ResolutionWidth + 38), sum(ResolutionWidth + 39), sum(ResolutionWidth + 40), sum(ResolutionWidth + 41), sum(ResolutionWidth + 42), sum(ResolutionWidth + 43), sum(ResolutionWidth + 44), sum(ResolutionWidth + 45), sum(ResolutionWidth + 46), sum(ResolutionWidth + 47), sum(ResolutionWidth + 48), sum(ResolutionWidth + 49), sum(ResolutionWidth + 50), sum(ResolutionWidth + 51), sum(ResolutionWidth + 52), sum(ResolutionWidth + 53), sum(ResolutionWidth + 54), sum(ResolutionWidth + 55), sum(ResolutionWidth + 56), sum(ResolutionWidth + 57), sum(ResolutionWidth + 58), sum(ResolutionWidth + 59), sum(ResolutionWidth + 60), sum(ResolutionWidth + 61), sum(ResolutionWidth + 62), sum(ResolutionWidth + 63), sum(ResolutionWidth + 64), sum(ResolutionWidth + 65), sum(ResolutionWidth + 66), sum(ResolutionWidth + 67), sum(ResolutionWidth + 68), sum(ResolutionWidth + 69), sum(ResolutionWidth + 70), sum(ResolutionWidth + 71), sum(ResolutionWidth + 72), sum(ResolutionWidth + 73), sum(ResolutionWidth + 74), sum(ResolutionWidth + 75), sum(ResolutionWidth + 76), sum(ResolutionWidth + 77), sum(ResolutionWidth + 78), sum(ResolutionWidth + 79), sum(ResolutionWidth + 80), sum(ResolutionWidth + 81), sum(ResolutionWidth + 82), sum(ResolutionWidth + 83), sum(ResolutionWidth + 84), sum(ResolutionWidth + 85), sum(ResolutionWidth + 86), sum(ResolutionWidth + 87), sum(ResolutionWidth + 88), sum(ResolutionWidth + 89) FROM hits_10m ;
-- много тупых агрегатных функций.;
SELECT SearchEngineID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM hits_10m WHERE SearchPhrase != '' GROUP BY SearchEngineID, ClientIP ORDER BY c DESC LIMIT 10;
-- сложная агрегация, для больших таблиц может не хватить оперативки.;
SELECT WatchID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM hits_10m WHERE SearchPhrase != '' GROUP BY WatchID, ClientIP ORDER BY c DESC LIMIT 10;
-- агрегация по двум полям, которая ничего не агрегирует. Для больших таблиц выполнить не получится.;
SELECT WatchID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM hits_10m GROUP BY WatchID, ClientIP ORDER BY c DESC LIMIT 10;
-- то же самое, но ещё и без фильтрации.;
SELECT URL, count() AS c FROM hits_10m GROUP BY URL ORDER BY c DESC LIMIT 10;
-- агрегация по URL.;
SELECT 1, URL, count() AS c FROM hits_10m GROUP BY 1, URL ORDER BY c DESC LIMIT 10;
-- агрегация по URL и числу.;
SELECT ClientIP AS x, x - 1, x - 2, x - 3, count() AS c FROM hits_10m GROUP BY x, x - 1, x - 2, x - 3 ORDER BY c DESC LIMIT 10;
SELECT URL, count() AS PageViews FROM hits_10m WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT DontCountHits AND NOT Refresh AND notEmpty(URL) GROUP BY URL ORDER BY PageViews DESC LIMIT 10;
SELECT Title, count() AS PageViews FROM hits_10m WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT DontCountHits AND NOT Refresh AND notEmpty(Title) GROUP BY Title ORDER BY PageViews DESC LIMIT 10;
SELECT URL, count() AS PageViews FROM hits_10m WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT Refresh AND IsLink AND NOT IsDownload GROUP BY URL ORDER BY PageViews DESC LIMIT 1000;
SELECT TraficSourceID, SearchEngineID, AdvEngineID, ((SearchEngineID = 0 AND AdvEngineID = 0) ? Referer : '') AS Src, URL AS Dst, count() AS PageViews FROM hits_10m WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT Refresh GROUP BY TraficSourceID, SearchEngineID, AdvEngineID, Src, Dst ORDER BY PageViews DESC LIMIT 1000;
SELECT URLHash, EventDate, count() AS PageViews FROM hits_10m WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT Refresh AND TraficSourceID IN (-1, 6) AND RefererHash = halfMD5('http://example.ru/') GROUP BY URLHash, EventDate ORDER BY PageViews DESC LIMIT 100;
SELECT WindowClientWidth, WindowClientHeight, count() AS PageViews FROM hits_10m WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT Refresh AND NOT DontCountHits AND URLHash = halfMD5('http://example.ru/') GROUP BY WindowClientWidth, WindowClientHeight ORDER BY PageViews DESC LIMIT 10000;
SELECT toStartOfMinute(EventTime) AS Minute, count() AS PageViews FROM hits_10m WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-02') AND NOT Refresh AND NOT DontCountHits GROUP BY Minute ORDER BY Minute;
SELECT count() FROM {table};
SELECT count() FROM {table} WHERE AdvEngineID != 0;
SELECT sum(AdvEngineID), count(), avg(ResolutionWidth) FROM {table} ;
SELECT sum(UserID) FROM {table} ;
SELECT uniq(UserID) FROM {table} ;
SELECT uniq(SearchPhrase) FROM {table} ;
SELECT min(EventDate), max(EventDate) FROM {table} ;
SELECT AdvEngineID, count() FROM {table} WHERE AdvEngineID != 0 GROUP BY AdvEngineID ORDER BY count() DESC;
SELECT RegionID, uniq(UserID) AS u FROM {table} GROUP BY RegionID ORDER BY u DESC LIMIT 10;
SELECT RegionID, sum(AdvEngineID), count() AS c, avg(ResolutionWidth), uniq(UserID) FROM {table} GROUP BY RegionID ORDER BY c DESC LIMIT 10;
SELECT MobilePhoneModel, uniq(UserID) AS u FROM {table} WHERE MobilePhoneModel != '' GROUP BY MobilePhoneModel ORDER BY u DESC LIMIT 10;
SELECT MobilePhone, MobilePhoneModel, uniq(UserID) AS u FROM {table} WHERE MobilePhoneModel != '' GROUP BY MobilePhone, MobilePhoneModel ORDER BY u DESC LIMIT 10;
SELECT SearchPhrase, count() AS c FROM {table} WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;
SELECT SearchPhrase, uniq(UserID) AS u FROM {table} WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY u DESC LIMIT 10;
SELECT SearchEngineID, SearchPhrase, count() AS c FROM {table} WHERE SearchPhrase != '' GROUP BY SearchEngineID, SearchPhrase ORDER BY c DESC LIMIT 10;
SELECT UserID, count() FROM {table} GROUP BY UserID ORDER BY count() DESC LIMIT 10;
SELECT UserID, SearchPhrase, count() FROM {table} GROUP BY UserID, SearchPhrase ORDER BY count() DESC LIMIT 10;
SELECT UserID, SearchPhrase, count() FROM {table} GROUP BY UserID, SearchPhrase LIMIT 10;
SELECT UserID, toMinute(EventTime) AS m, SearchPhrase, count() FROM {table} GROUP BY UserID, m, SearchPhrase ORDER BY count() DESC LIMIT 10;
SELECT UserID FROM {table} WHERE UserID = 12345678901234567890;
SELECT count() FROM {table} WHERE URL LIKE '%metrika%';
SELECT SearchPhrase, any(URL), count() AS c FROM {table} WHERE URL LIKE '%metrika%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;
SELECT SearchPhrase, any(URL), any(Title), count() AS c, uniq(UserID) FROM {table} WHERE Title LIKE '%Яндекс%' AND URL NOT LIKE '%.yandex.%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY c DESC LIMIT 10;
SELECT * FROM {table} WHERE URL LIKE '%metrika%' ORDER BY EventTime LIMIT 10;
SELECT SearchPhrase FROM {table} WHERE SearchPhrase != '' ORDER BY EventTime LIMIT 10;
SELECT SearchPhrase FROM {table} WHERE SearchPhrase != '' ORDER BY SearchPhrase LIMIT 10;
SELECT SearchPhrase FROM {table} WHERE SearchPhrase != '' ORDER BY EventTime, SearchPhrase LIMIT 10;
SELECT CounterID, avg(length(URL)) AS l, count() AS c FROM {table} WHERE URL != '' GROUP BY CounterID HAVING c > 100000 ORDER BY l DESC LIMIT 25;
SELECT domainWithoutWWW(Referer) AS key, avg(length(Referer)) AS l, count() AS c, any(Referer) FROM {table} WHERE Referer != '' GROUP BY key HAVING c > 100000 ORDER BY l DESC LIMIT 25;
SELECT sum(ResolutionWidth), sum(ResolutionWidth + 1), sum(ResolutionWidth + 2), sum(ResolutionWidth + 3), sum(ResolutionWidth + 4), sum(ResolutionWidth + 5), sum(ResolutionWidth + 6), sum(ResolutionWidth + 7), sum(ResolutionWidth + 8), sum(ResolutionWidth + 9), sum(ResolutionWidth + 10), sum(ResolutionWidth + 11), sum(ResolutionWidth + 12), sum(ResolutionWidth + 13), sum(ResolutionWidth + 14), sum(ResolutionWidth + 15), sum(ResolutionWidth + 16), sum(ResolutionWidth + 17), sum(ResolutionWidth + 18), sum(ResolutionWidth + 19), sum(ResolutionWidth + 20), sum(ResolutionWidth + 21), sum(ResolutionWidth + 22), sum(ResolutionWidth + 23), sum(ResolutionWidth + 24), sum(ResolutionWidth + 25), sum(ResolutionWidth + 26), sum(ResolutionWidth + 27), sum(ResolutionWidth + 28), sum(ResolutionWidth + 29), sum(ResolutionWidth + 30), sum(ResolutionWidth + 31), sum(ResolutionWidth + 32), sum(ResolutionWidth + 33), sum(ResolutionWidth + 34), sum(ResolutionWidth + 35), sum(ResolutionWidth + 36), sum(ResolutionWidth + 37), sum(ResolutionWidth + 38), sum(ResolutionWidth + 39), sum(ResolutionWidth + 40), sum(ResolutionWidth + 41), sum(ResolutionWidth + 42), sum(ResolutionWidth + 43), sum(ResolutionWidth + 44), sum(ResolutionWidth + 45), sum(ResolutionWidth + 46), sum(ResolutionWidth + 47), sum(ResolutionWidth + 48), sum(ResolutionWidth + 49), sum(ResolutionWidth + 50), sum(ResolutionWidth + 51), sum(ResolutionWidth + 52), sum(ResolutionWidth + 53), sum(ResolutionWidth + 54), sum(ResolutionWidth + 55), sum(ResolutionWidth + 56), sum(ResolutionWidth + 57), sum(ResolutionWidth + 58), sum(ResolutionWidth + 59), sum(ResolutionWidth + 60), sum(ResolutionWidth + 61), sum(ResolutionWidth + 62), sum(ResolutionWidth + 63), sum(ResolutionWidth + 64), sum(ResolutionWidth + 65), sum(ResolutionWidth + 66), sum(ResolutionWidth + 67), sum(ResolutionWidth + 68), sum(ResolutionWidth + 69), sum(ResolutionWidth + 70), sum(ResolutionWidth + 71), sum(ResolutionWidth + 72), sum(ResolutionWidth + 73), sum(ResolutionWidth + 74), sum(ResolutionWidth + 75), sum(ResolutionWidth + 76), sum(ResolutionWidth + 77), sum(ResolutionWidth + 78), sum(ResolutionWidth + 79), sum(ResolutionWidth + 80), sum(ResolutionWidth + 81), sum(ResolutionWidth + 82), sum(ResolutionWidth + 83), sum(ResolutionWidth + 84), sum(ResolutionWidth + 85), sum(ResolutionWidth + 86), sum(ResolutionWidth + 87), sum(ResolutionWidth + 88), sum(ResolutionWidth + 89) FROM {table};
SELECT SearchEngineID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM {table} WHERE SearchPhrase != '' GROUP BY SearchEngineID, ClientIP ORDER BY c DESC LIMIT 10;
SELECT WatchID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM {table} WHERE SearchPhrase != '' GROUP BY WatchID, ClientIP ORDER BY c DESC LIMIT 10;
SELECT WatchID, ClientIP, count() AS c, sum(Refresh), avg(ResolutionWidth) FROM {table} GROUP BY WatchID, ClientIP ORDER BY c DESC LIMIT 10;
SELECT URL, count() AS c FROM {table} GROUP BY URL ORDER BY c DESC LIMIT 10;
SELECT 1, URL, count() AS c FROM {table} GROUP BY 1, URL ORDER BY c DESC LIMIT 10;
SELECT ClientIP AS x, x - 1, x - 2, x - 3, count() AS c FROM {table} GROUP BY x, x - 1, x - 2, x - 3 ORDER BY c DESC LIMIT 10;
SELECT URL, count() AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT DontCountHits AND NOT Refresh AND notEmpty(URL) GROUP BY URL ORDER BY PageViews DESC LIMIT 10;
SELECT Title, count() AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT DontCountHits AND NOT Refresh AND notEmpty(Title) GROUP BY Title ORDER BY PageViews DESC LIMIT 10;
SELECT URL, count() AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT Refresh AND IsLink AND NOT IsDownload GROUP BY URL ORDER BY PageViews DESC LIMIT 1000;
SELECT TraficSourceID, SearchEngineID, AdvEngineID, ((SearchEngineID = 0 AND AdvEngineID = 0) ? Referer : '') AS Src, URL AS Dst, count() AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT Refresh GROUP BY TraficSourceID, SearchEngineID, AdvEngineID, Src, Dst ORDER BY PageViews DESC LIMIT 1000;
SELECT URLHash, EventDate, count() AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT Refresh AND TraficSourceID IN (-1, 6) AND RefererHash = halfMD5('http://example.ru/') GROUP BY URLHash, EventDate ORDER BY PageViews DESC LIMIT 100;
SELECT WindowClientWidth, WindowClientHeight, count() AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-31') AND NOT Refresh AND NOT DontCountHits AND URLHash = halfMD5('http://example.ru/') GROUP BY WindowClientWidth, WindowClientHeight ORDER BY PageViews DESC LIMIT 10000;
SELECT toStartOfMinute(EventTime) AS Minute, count() AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= toDate('2013-07-01') AND EventDate <= toDate('2013-07-02') AND NOT Refresh AND NOT DontCountHits GROUP BY Minute ORDER BY Minute;

19
dbms/benchmark/create_dump.sh
View File

@ -1,18 +1,3 @@
path=/opt/dump/dump_0.3
db_name=hits_1b
num=1000000000
#!/bin/bash
dump_replaced=$path/dump_"$db_name"_replaced.tsv
dump_meshed=$path/dump_"$db_name"_meshed.tsv
dump_meshed_utf8=$path/dump_"$db_name"_meshed_utf8.tsv
clickhouse-client --query="SET GLOBAL max_block_size=100000"
clickhouse-client --query="SET GLOBAL max_threads=1"
clickhouse-client --query="SELECT toInt64(WatchID), JavaEnable, Title, GoodEvent, (EventTime < toDateTime('1971-01-01 00:00:00') ? toDateTime('1971-01-01 00:00:01') : EventTime), (EventDate < toDate('1971-01-01') ? toDate('1971-01-01') : EventDate), CounterID, ClientIP, RegionID, toInt64(UserID), CounterClass, OS, UserAgent, URL, Referer, Refresh, RefererCategoryID, RefererRegionID, URLCategoryID, URLRegionID, ResolutionWidth, ResolutionHeight, ResolutionDepth, FlashMajor, FlashMinor, FlashMinor2, NetMajor, NetMinor, UserAgentMajor, UserAgentMinor, CookieEnable, JavascriptEnable, IsMobile, MobilePhone, MobilePhoneModel, Params, IPNetworkID, TraficSourceID, SearchEngineID, SearchPhrase, AdvEngineID, IsArtifical, WindowClientWidth, WindowClientHeight, ClientTimeZone, (ClientEventTime < toDateTime('1971-01-01 00:00:01') ? toDateTime('1971-01-01 00:00:01') : ClientEventTime), SilverlightVersion1, SilverlightVersion2, SilverlightVersion3, SilverlightVersion4, PageCharset, CodeVersion, IsLink, IsDownload, IsNotBounce, toInt64(FUniqID), OriginalURL, HID, IsOldCounter, IsEvent, IsParameter, DontCountHits, WithHash, HitColor, (LocalEventTime < toDateTime('1971-01-01 00:00:01') ? toDateTime('1971-01-01 00:00:01') : LocalEventTime), Age, Sex, Income, Interests, Robotness, RemoteIP, WindowName, OpenerName, HistoryLength, BrowserLanguage, BrowserCountry, SocialNetwork, SocialAction, HTTPError, SendTiming, DNSTiming, ConnectTiming, ResponseStartTiming, ResponseEndTiming, FetchTiming, SocialSourceNetworkID, SocialSourcePage, ParamPrice, ParamOrderID, ParamCurrency, ParamCurrencyID, OpenstatServiceName, OpenstatCampaignID, OpenstatAdID, OpenstatSourceID, UTMSource, UTMMedium, UTMCampaign, UTMContent, UTMTerm, FromTag, HasGCLID, toInt64(RefererHash), toInt64(URLHash), CLID, toInt64(intHash32(UserID)) FROM hits_mt_test_1b LIMIT $num FORMAT TabSeparated" > $dump_replaced
/etc/init.d/clickhouse-server-metrika-yandex-ulimit restart
sudo nsort -format=maximum_size:65535 -k1 -T /opt -o $dump_meshed $dump_replaced
cat $dump_meshed | iconv -futf8 -tutf8//IGNORE 2>/dev/null 1> $dump_meshed_utf8
table=hits_10m; time clickhouse-client --max_bytes_before_external_sort=30000000000 --query="SELECT toInt64(WatchID), JavaEnable, Title, GoodEvent, (EventTime < toDateTime('1971-01-01 00:00:00') ? toDateTime('1971-01-01 00:00:01') : EventTime), (EventDate < toDate('1971-01-01') ? toDate('1971-01-01') : EventDate), CounterID, ClientIP, RegionID, toInt64(UserID), CounterClass, OS, UserAgent, URL, Referer, Refresh, RefererCategoryID, RefererRegionID, URLCategoryID, URLRegionID, ResolutionWidth, ResolutionHeight, ResolutionDepth, FlashMajor, FlashMinor, FlashMinor2, NetMajor, NetMinor, UserAgentMajor, UserAgentMinor, CookieEnable, JavascriptEnable, IsMobile, MobilePhone, MobilePhoneModel, Params, IPNetworkID, TraficSourceID, SearchEngineID, SearchPhrase, AdvEngineID, IsArtifical, WindowClientWidth, WindowClientHeight, ClientTimeZone, (ClientEventTime < toDateTime('1971-01-01 00:00:01') ? toDateTime('1971-01-01 00:00:01') : ClientEventTime), SilverlightVersion1, SilverlightVersion2, SilverlightVersion3, SilverlightVersion4, PageCharset, CodeVersion, IsLink, IsDownload, IsNotBounce, toInt64(FUniqID), OriginalURL, HID, IsOldCounter, IsEvent, IsParameter, DontCountHits, WithHash, HitColor, (LocalEventTime < toDateTime('1971-01-01 00:00:01') ? toDateTime('1971-01-01 00:00:01') : LocalEventTime), Age, Sex, Income, Interests, Robotness, RemoteIP, WindowName, OpenerName, HistoryLength, BrowserLanguage, BrowserCountry, SocialNetwork, SocialAction, HTTPError, SendTiming, DNSTiming, ConnectTiming, ResponseStartTiming, ResponseEndTiming, FetchTiming, SocialSourceNetworkID, SocialSourcePage, ParamPrice, ParamOrderID, ParamCurrency, ParamCurrencyID, OpenstatServiceName, OpenstatCampaignID, OpenstatAdID, OpenstatSourceID, UTMSource, UTMMedium, UTMCampaign, UTMContent, UTMTerm, FromTag, HasGCLID, toInt64(RefererHash), toInt64(URLHash), CLID FROM $table ORDER BY rand()" | corrector_utf8 > /opt/dumps/${table}_corrected.tsv

26
dbms/benchmark/killif.sh
View File

@ -1,26 +0,0 @@
#!/bin/sh
if [[ $# -ne 0 ]]; then
echo "usage: if memory limit is exceeded kill process with biggest memory consumption"
exit 1
fi
while [ 1=1 ];
do
FREE_MEMORY_MB=$(free -m | sed -n '3,3p' | awk '{print $4}')
PID="$(ps -eF --sort -rss | sed -n '2,2p' | awk '{print $2}')"
NAME="$(ps -eF --sort -rss | sed -n '2,2p' | awk '{print $11}')"
SIZEGB="$(ps -eF --sort -rss | sed -n '2,2p' | awk '{print $6}')"
SIZEGB=$(($SIZEGB/1024/1024))
echo "Process id ="$PID" Size = "$SIZEGB" GB" "Free Memory = " $FREE_MEMORY_MB" MB"
if (( $FREE_MEMORY_MB < 512 ));
then echo "Killing the process with biggest memory consumption......"
sudo kill -9 $PID
echo "$(date) Killed the process with PID: $PID NAME: $NAME"
else
echo "SIZE has not yet exceeding"
fi
sleep 10
done

114
dbms/benchmark/process_log.py
View File

@ -1,114 +0,0 @@
from optparse import OptionParser
import argparse
import re
import sys
def log_to_rows(filename, pattern_select, time_pattern, pattern_ignore):
time_matcher = re.compile(time_pattern)
select_matcher = re.compile(pattern_select, re.IGNORECASE);
ignore_matcher = re.compile(pattern_ignore)
f = open(filename, 'r');
query = ''
raw_time = ''
for line in f:
if ignore_matcher.match(line):
continue
m = select_matcher.search(line)
if m :
if line != query:
query = line
sys.stdout.write("\n")
raw_time = raw_time + "\n"
m = time_matcher.search(line)
if m:
sec = 0
minute = 0
ms = 0
if 'min' in m.groupdict() and m.group('min'):
minute = float(m.group('min').replace(',','.'))
if 'sec' in m.groupdict() and m.group('sec'):
sec = float(m.group('sec').replace(',','.'))
if 'ms' in m.groupdict() and m.group('ms'):
ms = float(m.group('ms').replace(',', '.'))
sys.stdout.write( str(minute*60 + sec + ms/1000.) + " " )
raw_time = raw_time + " | " + m.group('time')
print
print " =======raw time====== \n" + raw_time
def process_log(filename, pattern_select, time_pattern, pattern_ignore, error_pattern):
time_matcher = re.compile(time_pattern)
select_matcher = re.compile(pattern_select, re.IGNORECASE);
ignore_matcher = re.compile(pattern_ignore)
error_matcher = re.compile(error_pattern, re.IGNORECASE)
f = open(filename, 'r');
query = ''
for line in f:
if error_matcher.match(line):
print line
continue
if ignore_matcher.match(line):
continue
m = select_matcher.search(line)
if m :
if line != query:
sys.stdout.flush()
query = line
print "\n\n"
print query
m = time_matcher.search(line)
if m:
sys.stdout.write(m.group('time') + " " )
def main():
parser = argparse.ArgumentParser(description="Process log files form different databases")
parser.add_argument('log_file', metavar = 'log_file', help = 'database log file')
parser.add_argument('db_name', metavar = 'db_name', help = ' database name one of clickhouse, vertica, infinidb, monetdb, infobright, hive (... more later)')
args = parser.parse_args()
log_file = args.log_file
db_name = args.db_name
time_pattern = ''
select_pattern = r'query: select '
ignore_pattern = r'#'
error_pattern = r'error .*'
if db_name == 'clickhouse':
time_pattern = r'(?P<time>(?P<sec>\d+.\d{3}) sec\.)'
select_pattern = r'query\: select '
ignore_pattern = r':\).*'
elif db_name == 'vertica' :
time_pattern = r'(?P<time>(?P<ms>\d+.\d+) ms\.)'
select_pattern = r'select '
ignore_pattern = r'(.*dbadmin=>|query:|.*Timing is on\.).*'
elif db_name == 'infinidb' :
time_pattern = r'(?P<time>(?:(?P<min>\d+) min )?(?P<sec>\d+.\d+) sec)'
ignore_pattern = r'Query OK, 0 rows affected \(0\.00 sec\)'
elif db_name == 'monetdb' :
time_pattern = r'tuples? \((?P<time>(?:(?P<min>\d+)m )?(?:(?P<sec>\d+.?\d+)s)?(?:(?P<ms>\d+.\d+)ms)?)\)'
elif db_name == 'infobright' :
time_pattern = r'(?P<time>(?:(?P<min>\d+) min ){0,1}(?P<sec>\d+.\d+) sec)'
elif db_name == 'hive':
time_pattern = r'Time taken\: (?P<time>(?:(?P<sec>\d+.?\d+) seconds))'
error_pattern = r'failed\: .*'
elif db_name == 'mysql':
time_pattern = r'(?P<time>(?:(?P<min>\d+) min )?(?P<sec>\d+.\d+) sec)'
else:
sys.exit("unknown db_name")
process_log(log_file, select_pattern, time_pattern, ignore_pattern, error_pattern )
log_to_rows(log_file, select_pattern, time_pattern, ignore_pattern )
main()

40
dbms/benchmark/vertica/README Normal file
View File

@ -0,0 +1,40 @@
Quick installation instructions
-------------------------------
Register on my.vertica.com
https://my.vertica.com/download-community-edition/
Download HP Vertica 7.1.1 Analytic Database Server, Debian or Ubuntu 14.04 version.
sudo apt-get install sysstat pstack mcelog
sudo dpkg -i vertica_7.1.1-0_amd64.deb
sudo sh -c "echo 'export TZ=Europe/Moscow' >> /home/dbadmin/.bash_profile"
sudo /opt/vertica/sbin/install_vertica --hosts=127.0.0.1 --failure-threshold=NONE
sudo mkdir /opt/vertica-data/
sudo chown dbadmin /opt/vertica-data/
sudo su dbadmin
/opt/vertica/bin/adminTools
configuration menu
create database
name: default
empty password
both directories: /opt/vertica-data/
main menu
exit
PS. Note that Vertica doesn't support IPv6.
How to prepare data
-------------------
Prepare dumps with script create_dump.sh for tables hits_10m, hits_100m, hits_1000m. It takes about 5 hours (1m41.882s, 25m11.103s, 276m36.388s).
Start vsql command line client.
Create tables with queries from hits_define_schema.sql.
Time to insert data:
hits_10m: 91 sec.
hits_100m: 774 sec.
hits_1000m:
You need to validate number of rows with SELECT count(*).

24
dbms/benchmark/vertica/benchmark.sh Normal file
View File

@ -0,0 +1,24 @@
#!/bin/bash
QUERIES_FILE="queries.sql"
TABLE=$1
TRIES=3
cat "$QUERIES_FILE" | sed "s/{table}/${TABLE}/g" | while read query; do
sync
echo 3 | sudo tee /proc/sys/vm/drop_caches >/dev/null
echo -n "["
for i in $(seq 1 $TRIES); do
RES=$((echo '\timing'; echo "$query") |
/opt/vertica/bin/vsql -U dbadmin |
grep -oP 'All rows formatted: [^ ]+ ms' |
ssed -R -e 's/^All rows formatted: ([\d,]+) ms$/\1/' |
tr ',' '.')
[[ "$?" == "0" ]] && echo -n "$(perl -e "print ${RES} / 1000")" || echo -n "null"
[[ "$i" != $TRIES ]] && echo -n ", "
done
echo "],"
done

21
dbms/benchmark/vertica/expect.tcl
View File

@ -1,21 +0,0 @@
#!/bin/bash
#!/bin/expect
# Set timeout
set timeout 600
# Get arguments
set query [lindex $argv 0]
spawn vsql -eU dbadmin
expect "dbadmin=>"
send "\\timing\r"
expect "dbadmin=>"
send "$query\r"
expect "dbadmin=>"
send "\\q\r"
expect eof

40
dbms/benchmark/vertica/hits_define_schema.sql
View File

@ -1,6 +1,6 @@
\timing
create table hits_10m_meshed
create table hits_10m
(
WatchID INTEGER,
JavaEnable INTEGER,
@ -106,17 +106,14 @@ create table hits_10m_meshed
HasGCLID INTEGER,
RefererHash INTEGER,
URLHash INTEGER,
CLID INTEGER,
UserIDHash INTEGER
) ORDER BY CounterID, EventDate, UserIDHash, EventTime;
CLID INTEGER
) ORDER BY CounterID, EventDate, UserID, EventTime;
\set input_file '''/opt/dumps/hits_10m_corrected.tsv'''
COPY hits_10m FROM :input_file DELIMITER E'\t' DIRECT;
\set input_file '''/opt/dump/dump_0.3/dump_hits_10m_meshed.tsv'''
COPY hits_10m_meshed FROM :input_file DELIMITER E'\t' DIRECT;
create table hits_100m_meshed
create table hits_100m
(
WatchID INTEGER,
JavaEnable INTEGER,
@ -222,17 +219,14 @@ create table hits_100m_meshed
HasGCLID INTEGER,
RefererHash INTEGER,
URLHash INTEGER,
CLID INTEGER,
UserIDHash INTEGER
) ORDER BY CounterID, EventDate, UserIDHash, EventTime;;
CLID INTEGER
) ORDER BY CounterID, EventDate, UserID, EventTime;
\set input_file '''/opt/dump/dump_0.3/dump_hits_100m_meshed.tsv'''
COPY hits_100m_meshed FROM :input_file DELIMITER E'\t' DIRECT;
\set input_file '''/opt/dumps/hits_100m_corrected.tsv'''
COPY hits_100m FROM :input_file DELIMITER E'\t' DIRECT;
create table hits_1b_meshed
create table hits_1000m
(
WatchID INTEGER,
JavaEnable INTEGER,
@ -338,10 +332,8 @@ create table hits_1b_meshed
HasGCLID INTEGER,
RefererHash INTEGER,
URLHash INTEGER,
CLID INTEGER,
UserIDHash INTEGER
) ORDER BY CounterID, EventDate, UserIDHash, EventTime;
CLID INTEGER
) ORDER BY CounterID, EventDate, UserID, EventTime;
\set input_file '''/opt/dump/dump_0.3/dump_hits_1b_meshed.tsv'''
COPY hits_1b_meshed FROM :input_file DELIMITER E'\t' DIRECT;
\set input_file '''/opt/dumps/hits_1000m_corrected.tsv'''
COPY hits_1000m FROM :input_file DELIMITER E'\t' DIRECT;

154
dbms/benchmark/vertica/queries_for_script.sql
View File

@ -1,111 +1,43 @@
SELECT count(*) FROM hits_100m_meshed;
SELECT count(*) FROM hits_100m_meshed WHERE AdvEngineID != 0;
SELECT sum(AdvEngineID), count(*), avg(ResolutionWidth) FROM hits_100m_meshed;
SELECT sum_float(UserID) FROM hits_100m_meshed;
SELECT count(DISTINCT UserID) FROM hits_100m_meshed;
SELECT count(DISTINCT SearchPhrase) FROM hits_100m_meshed;
SELECT min(EventDate), max(EventDate) FROM hits_100m_meshed;
SELECT AdvEngineID, count(*) FROM hits_100m_meshed WHERE AdvEngineID != 0 GROUP BY AdvEngineID ORDER BY count(*) DESC;
-- мощная фильтрация. После фильтрации почти ничего не остаётся, но делаем ещё агрегацию.;
SELECT RegionID, count(DISTINCT UserID) AS u FROM hits_100m_meshed GROUP BY RegionID ORDER BY u DESC LIMIT 10;
-- агрегация, среднее количество ключей.;
SELECT RegionID, sum(AdvEngineID), count(*) AS c, avg(ResolutionWidth), count(DISTINCT UserID) FROM hits_100m_meshed GROUP BY RegionID ORDER BY count(*) DESC LIMIT 10;
-- агрегация, среднее количество ключей, несколько агрегатных функций.;
SELECT MobilePhoneModel, count(DISTINCT UserID) AS u FROM hits_100m_meshed WHERE MobilePhoneModel != '' GROUP BY MobilePhoneModel ORDER BY u DESC LIMIT 10;
-- мощная фильтрация по строкам, затем агрегация по строкам.;
SELECT MobilePhone, MobilePhoneModel, count(DISTINCT UserID) AS u FROM hits_100m_meshed WHERE MobilePhoneModel != '' GROUP BY MobilePhone, MobilePhoneModel ORDER BY u DESC LIMIT 10;
-- мощная фильтрация по строкам, затем агрегация по паре из числа и строки.;
SELECT SearchPhrase, count(*) FROM hits_100m_meshed WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY count(*) DESC LIMIT 10;
-- средняя фильтрация по строкам, затем агрегация по строкам, большое количество ключей.;
SELECT SearchPhrase, count(DISTINCT UserID) AS u FROM hits_100m_meshed WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY u DESC LIMIT 10;
-- агрегация чуть сложнее.;
SELECT SearchEngineID, SearchPhrase, count(*) FROM hits_100m_meshed WHERE SearchPhrase != '' GROUP BY SearchEngineID, SearchPhrase ORDER BY count(*) DESC LIMIT 10;
-- агрегация по числу и строке, большое количество ключей.;
SELECT UserID, count(*) FROM hits_100m_meshed GROUP BY UserID ORDER BY count(*) DESC LIMIT 10;
-- агрегация по очень большому количеству ключей, может не хватить оперативки.;
SELECT UserID, SearchPhrase, count(*) FROM hits_100m_meshed GROUP BY UserID, SearchPhrase ORDER BY count(*) DESC LIMIT 10;
-- ещё более сложная агрегация.;
SELECT UserID, SearchPhrase, count(*) FROM hits_100m_meshed GROUP BY UserID, SearchPhrase LIMIT 10;
-- то же самое, но без сортировки.;
SELECT UserID, Minute(EventTime) AS m, SearchPhrase, count(*) FROM hits_100m_meshed GROUP BY UserID, m, SearchPhrase ORDER BY count(*) DESC LIMIT 10;
-- ещё более сложная агрегация, не стоит выполнять на больших таблицах.;
SELECT UserID FROM hits_100m_meshed WHERE UserID = 12345678901234567890;
-- мощная фильтрация по столбцу типа UInt64.;
SELECT count(*) FROM hits_100m_meshed WHERE URL LIKE '%metrika%';
-- фильтрация по поиску подстроки в строке.;
SELECT SearchPhrase, MAX(URL), count(*) FROM hits_100m_meshed WHERE URL LIKE '%metrika%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY count(*) DESC LIMIT 10;
-- вынимаем большие столбцы, фильтрация по строке.;
SELECT SearchPhrase, MAX(URL), MAX(Title), count(*) AS c, count(DISTINCT UserID) FROM hits_100m_meshed WHERE Title LIKE '%Яндекс%' AND URL NOT LIKE '%.yandex.%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY count(*) DESC LIMIT 10;
-- чуть больше столбцы.;
SELECT * FROM hits_100m_meshed WHERE URL LIKE '%metrika%' ORDER BY EventTime LIMIT 10;
-- плохой запрос - вынимаем все столбцы.;
SELECT SearchPhrase FROM hits_100m_meshed WHERE SearchPhrase != '' ORDER BY EventTime LIMIT 10;
-- большая сортировка.;
SELECT SearchPhrase FROM hits_100m_meshed WHERE SearchPhrase != '' ORDER BY SearchPhrase LIMIT 10;
-- большая сортировка по строкам.;
SELECT SearchPhrase FROM hits_100m_meshed WHERE SearchPhrase != '' ORDER BY EventTime, SearchPhrase LIMIT 10;
-- большая сортировка по кортежу.;
SELECT CounterID, avg(length(URL)) AS l, count(*) FROM hits_100m_meshed WHERE URL != '' GROUP BY CounterID HAVING count(*) > 100000 ORDER BY l DESC LIMIT 25;
-- считаем средние длины URL для крупных счётчиков.;
SELECT SUBSTRING(SUBSTRING(Referer, POSITION('//' IN Referer) + 2), 1, GREATEST(0, POSITION('/' IN SUBSTRING(Referer, POSITION('//' IN Referer) + 2)) - 1)) AS key, avg(length(Referer)) AS l, count(*) AS c, MAX(Referer) FROM hits_10m_meshed WHERE Referer != '' GROUP BY key HAVING count(*) > 100000 ORDER BY l DESC LIMIT 25;
-- то же самое, но с разбивкой по доменам.;
SELECT sum(ResolutionWidth), sum(ResolutionWidth + 1), sum(ResolutionWidth + 2), sum(ResolutionWidth + 3), sum(ResolutionWidth + 4), sum(ResolutionWidth + 5), sum(ResolutionWidth + 6), sum(ResolutionWidth + 7), sum(ResolutionWidth + 8), sum(ResolutionWidth + 9), sum(ResolutionWidth + 10), sum(ResolutionWidth + 11), sum(ResolutionWidth + 12), sum(ResolutionWidth + 13), sum(ResolutionWidth + 14), sum(ResolutionWidth + 15), sum(ResolutionWidth + 16), sum(ResolutionWidth + 17), sum(ResolutionWidth + 18), sum(ResolutionWidth + 19), sum(ResolutionWidth + 20), sum(ResolutionWidth + 21), sum(ResolutionWidth + 22), sum(ResolutionWidth + 23), sum(ResolutionWidth + 24), sum(ResolutionWidth + 25), sum(ResolutionWidth + 26), sum(ResolutionWidth + 27), sum(ResolutionWidth + 28), sum(ResolutionWidth + 29), sum(ResolutionWidth + 30), sum(ResolutionWidth + 31), sum(ResolutionWidth + 32), sum(ResolutionWidth + 33), sum(ResolutionWidth + 34), sum(ResolutionWidth + 35), sum(ResolutionWidth + 36), sum(ResolutionWidth + 37), sum(ResolutionWidth + 38), sum(ResolutionWidth + 39), sum(ResolutionWidth + 40), sum(ResolutionWidth + 41), sum(ResolutionWidth + 42), sum(ResolutionWidth + 43), sum(ResolutionWidth + 44), sum(ResolutionWidth + 45), sum(ResolutionWidth + 46), sum(ResolutionWidth + 47), sum(ResolutionWidth + 48), sum(ResolutionWidth + 49), sum(ResolutionWidth + 50), sum(ResolutionWidth + 51), sum(ResolutionWidth + 52), sum(ResolutionWidth + 53), sum(ResolutionWidth + 54), sum(ResolutionWidth + 55), sum(ResolutionWidth + 56), sum(ResolutionWidth + 57), sum(ResolutionWidth + 58), sum(ResolutionWidth + 59), sum(ResolutionWidth + 60), sum(ResolutionWidth + 61), sum(ResolutionWidth + 62), sum(ResolutionWidth + 63), sum(ResolutionWidth + 64), sum(ResolutionWidth + 65), sum(ResolutionWidth + 66), sum(ResolutionWidth + 67), sum(ResolutionWidth + 68), sum(ResolutionWidth + 69), sum(ResolutionWidth + 70), sum(ResolutionWidth + 71), sum(ResolutionWidth + 72), sum(ResolutionWidth + 73), sum(ResolutionWidth + 74), sum(ResolutionWidth + 75), sum(ResolutionWidth + 76), sum(ResolutionWidth + 77), sum(ResolutionWidth + 78), sum(ResolutionWidth + 79), sum(ResolutionWidth + 80), sum(ResolutionWidth + 81), sum(ResolutionWidth + 82), sum(ResolutionWidth + 83), sum(ResolutionWidth + 84), sum(ResolutionWidth + 85), sum(ResolutionWidth + 86), sum(ResolutionWidth + 87), sum(ResolutionWidth + 88), sum(ResolutionWidth + 89) FROM hits_100m_meshed;
-- много тупых агрегатных функций.;
SELECT SearchEngineID, ClientIP, count(*) AS c, sum(Refresh), avg(ResolutionWidth) FROM hits_100m_meshed WHERE SearchPhrase != '' GROUP BY SearchEngineID, ClientIP ORDER BY count(*) DESC LIMIT 10;
-- сложная агрегация, для больших таблиц может не хватить оперативки.;
SELECT WatchID, ClientIP, count(*) AS c, sum(Refresh), avg(ResolutionWidth) FROM hits_100m_meshed WHERE SearchPhrase != '' GROUP BY WatchID, ClientIP ORDER BY count(*) DESC LIMIT 10;
-- агрегация по двум полям, которая ничего не агрегирует. Для больших таблиц выполнить не получится.;
SELECT WatchID, ClientIP, count(*) AS c, sum(Refresh), avg(ResolutionWidth) FROM hits_100m_meshed GROUP BY WatchID, ClientIP ORDER BY count(*) DESC LIMIT 10;
-- то же самое, но ещё и без фильтрации.;
SELECT URL, count(*) FROM hits_100m_meshed GROUP BY URL ORDER BY count(*) DESC LIMIT 10;
-- агрегация по URL.;
SELECT 1, URL, count(*) FROM hits_100m_meshed GROUP BY 1, URL ORDER BY count(*) DESC LIMIT 10;
-- агрегация по URL и числу.;
SELECT ClientIP, ClientIP - 1, ClientIP - 2, ClientIP - 3, count(*) FROM hits_100m_meshed GROUP BY ClientIP, ClientIP - 1, ClientIP - 2, ClientIP - 3 ORDER BY count(*) DESC LIMIT 10;
SELECT URL, count(*) AS PageViews FROM hits_100m_meshed WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT DontCountHits AND NOT Refresh AND URL != '' GROUP BY URL ORDER BY PageViews DESC LIMIT 10;
SELECT Title, count(*) AS PageViews FROM hits_100m_meshed WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT DontCountHits AND NOT Refresh AND Title != '' GROUP BY Title ORDER BY PageViews DESC LIMIT 10;
SELECT URL, count(*) AS PageViews FROM hits_100m_meshed WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT Refresh AND IsLink AND NOT IsDownload GROUP BY URL ORDER BY PageViews DESC LIMIT 1000;
SELECT TraficSourceID, SearchEngineID, AdvEngineID, CASE WHEN SearchEngineID = 0 AND AdvEngineID = 0 THEN Referer ELSE '' END AS Src, URL AS Dst, count(*) AS PageViews FROM hits_100m_meshed WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT Refresh GROUP BY TraficSourceID, SearchEngineID, AdvEngineID, Src, Dst ORDER BY PageViews DESC LIMIT 1000;
SELECT URLHash, EventDate, count(*) AS PageViews FROM hits_100m_meshed WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT Refresh AND TraficSourceID IN (-1, 6) AND RefererHash = 6202628419148573758 GROUP BY URLHash, EventDate ORDER BY PageViews DESC LIMIT 100000;
SELECT WindowClientWidth, WindowClientHeight, count(*) AS PageViews FROM hits_100m_meshed WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT Refresh AND NOT DontCountHits AND URLHash = 6202628419148573758 GROUP BY WindowClientWidth, WindowClientHeight ORDER BY PageViews DESC LIMIT 10000;
SELECT TIME_SLICE(EventTime, 1, 'MINUTE') AS Minute, count(*) AS PageViews FROM hits_100m_meshed WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-02') AND NOT Refresh AND NOT DontCountHits GROUP BY Minute ORDER BY Minute;
SELECT count(*) FROM {table};
SELECT count(*) FROM {table} WHERE AdvEngineID != 0;
SELECT sum(AdvEngineID), count(*), avg(ResolutionWidth) FROM {table};
SELECT sum_float(UserID) FROM {table};
SELECT count(DISTINCT UserID) FROM {table};
SELECT count(DISTINCT SearchPhrase) FROM {table};
SELECT min(EventDate), max(EventDate) FROM {table};
SELECT AdvEngineID, count(*) FROM {table} WHERE AdvEngineID != 0 GROUP BY AdvEngineID ORDER BY count(*) DESC;
SELECT RegionID, count(DISTINCT UserID) AS u FROM {table} GROUP BY RegionID ORDER BY u DESC LIMIT 10;
SELECT RegionID, sum(AdvEngineID), count(*) AS c, avg(ResolutionWidth), count(DISTINCT UserID) FROM {table} GROUP BY RegionID ORDER BY count(*) DESC LIMIT 10;
SELECT MobilePhoneModel, count(DISTINCT UserID) AS u FROM {table} WHERE MobilePhoneModel != '' GROUP BY MobilePhoneModel ORDER BY u DESC LIMIT 10;
SELECT MobilePhone, MobilePhoneModel, count(DISTINCT UserID) AS u FROM {table} WHERE MobilePhoneModel != '' GROUP BY MobilePhone, MobilePhoneModel ORDER BY u DESC LIMIT 10;
SELECT SearchPhrase, count(*) FROM {table} WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY count(*) DESC LIMIT 10;
SELECT SearchPhrase, count(DISTINCT UserID) AS u FROM {table} WHERE SearchPhrase != '' GROUP BY SearchPhrase ORDER BY u DESC LIMIT 10;
SELECT SearchEngineID, SearchPhrase, count(*) FROM {table} WHERE SearchPhrase != '' GROUP BY SearchEngineID, SearchPhrase ORDER BY count(*) DESC LIMIT 10;
SELECT UserID, count(*) FROM {table} GROUP BY UserID ORDER BY count(*) DESC LIMIT 10;
SELECT UserID, SearchPhrase, count(*) FROM {table} GROUP BY UserID, SearchPhrase ORDER BY count(*) DESC LIMIT 10;
SELECT UserID, SearchPhrase, count(*) FROM {table} GROUP BY UserID, SearchPhrase LIMIT 10;
SELECT UserID, Minute(EventTime) AS m, SearchPhrase, count(*) FROM {table} GROUP BY UserID, m, SearchPhrase ORDER BY count(*) DESC LIMIT 10;
SELECT UserID FROM {table} WHERE UserID = 12345678901234567890;
SELECT count(*) FROM {table} WHERE URL LIKE '%metrika%';
SELECT SearchPhrase, MAX(URL), count(*) FROM {table} WHERE URL LIKE '%metrika%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY count(*) DESC LIMIT 10;
SELECT SearchPhrase, MAX(URL), MAX(Title), count(*) AS c, count(DISTINCT UserID) FROM {table} WHERE Title LIKE '%Яндекс%' AND URL NOT LIKE '%.yandex.%' AND SearchPhrase != '' GROUP BY SearchPhrase ORDER BY count(*) DESC LIMIT 10;
SELECT * FROM {table} WHERE URL LIKE '%metrika%' ORDER BY EventTime LIMIT 10;
SELECT SearchPhrase FROM {table} WHERE SearchPhrase != '' ORDER BY EventTime LIMIT 10;
SELECT SearchPhrase FROM {table} WHERE SearchPhrase != '' ORDER BY SearchPhrase LIMIT 10;
SELECT SearchPhrase FROM {table} WHERE SearchPhrase != '' ORDER BY EventTime, SearchPhrase LIMIT 10;
SELECT CounterID, avg(length(URL)) AS l, count(*) FROM {table} WHERE URL != '' GROUP BY CounterID HAVING count(*) > 100000 ORDER BY l DESC LIMIT 25;
SELECT SUBSTRING(SUBSTRING(Referer, POSITION('//' IN Referer) + 2), 1, GREATEST(0, POSITION('/' IN SUBSTRING(Referer, POSITION('//' IN Referer) + 2)) - 1)) AS key, avg(length(Referer)) AS l, count(*) AS c, MAX(Referer) FROM {table} WHERE Referer != '' GROUP BY key HAVING count(*) > 100000 ORDER BY l DESC LIMIT 25;
SELECT sum(ResolutionWidth), sum(ResolutionWidth + 1), sum(ResolutionWidth + 2), sum(ResolutionWidth + 3), sum(ResolutionWidth + 4), sum(ResolutionWidth + 5), sum(ResolutionWidth + 6), sum(ResolutionWidth + 7), sum(ResolutionWidth + 8), sum(ResolutionWidth + 9), sum(ResolutionWidth + 10), sum(ResolutionWidth + 11), sum(ResolutionWidth + 12), sum(ResolutionWidth + 13), sum(ResolutionWidth + 14), sum(ResolutionWidth + 15), sum(ResolutionWidth + 16), sum(ResolutionWidth + 17), sum(ResolutionWidth + 18), sum(ResolutionWidth + 19), sum(ResolutionWidth + 20), sum(ResolutionWidth + 21), sum(ResolutionWidth + 22), sum(ResolutionWidth + 23), sum(ResolutionWidth + 24), sum(ResolutionWidth + 25), sum(ResolutionWidth + 26), sum(ResolutionWidth + 27), sum(ResolutionWidth + 28), sum(ResolutionWidth + 29), sum(ResolutionWidth + 30), sum(ResolutionWidth + 31), sum(ResolutionWidth + 32), sum(ResolutionWidth + 33), sum(ResolutionWidth + 34), sum(ResolutionWidth + 35), sum(ResolutionWidth + 36), sum(ResolutionWidth + 37), sum(ResolutionWidth + 38), sum(ResolutionWidth + 39), sum(ResolutionWidth + 40), sum(ResolutionWidth + 41), sum(ResolutionWidth + 42), sum(ResolutionWidth + 43), sum(ResolutionWidth + 44), sum(ResolutionWidth + 45), sum(ResolutionWidth + 46), sum(ResolutionWidth + 47), sum(ResolutionWidth + 48), sum(ResolutionWidth + 49), sum(ResolutionWidth + 50), sum(ResolutionWidth + 51), sum(ResolutionWidth + 52), sum(ResolutionWidth + 53), sum(ResolutionWidth + 54), sum(ResolutionWidth + 55), sum(ResolutionWidth + 56), sum(ResolutionWidth + 57), sum(ResolutionWidth + 58), sum(ResolutionWidth + 59), sum(ResolutionWidth + 60), sum(ResolutionWidth + 61), sum(ResolutionWidth + 62), sum(ResolutionWidth + 63), sum(ResolutionWidth + 64), sum(ResolutionWidth + 65), sum(ResolutionWidth + 66), sum(ResolutionWidth + 67), sum(ResolutionWidth + 68), sum(ResolutionWidth + 69), sum(ResolutionWidth + 70), sum(ResolutionWidth + 71), sum(ResolutionWidth + 72), sum(ResolutionWidth + 73), sum(ResolutionWidth + 74), sum(ResolutionWidth + 75), sum(ResolutionWidth + 76), sum(ResolutionWidth + 77), sum(ResolutionWidth + 78), sum(ResolutionWidth + 79), sum(ResolutionWidth + 80), sum(ResolutionWidth + 81), sum(ResolutionWidth + 82), sum(ResolutionWidth + 83), sum(ResolutionWidth + 84), sum(ResolutionWidth + 85), sum(ResolutionWidth + 86), sum(ResolutionWidth + 87), sum(ResolutionWidth + 88), sum(ResolutionWidth + 89) FROM {table};
SELECT SearchEngineID, ClientIP, count(*) AS c, sum(Refresh), avg(ResolutionWidth) FROM {table} WHERE SearchPhrase != '' GROUP BY SearchEngineID, ClientIP ORDER BY count(*) DESC LIMIT 10;
SELECT WatchID, ClientIP, count(*) AS c, sum(Refresh), avg(ResolutionWidth) FROM {table} WHERE SearchPhrase != '' GROUP BY WatchID, ClientIP ORDER BY count(*) DESC LIMIT 10;
SELECT WatchID, ClientIP, count(*) AS c, sum(Refresh), avg(ResolutionWidth) FROM {table} GROUP BY WatchID, ClientIP ORDER BY count(*) DESC LIMIT 10;
SELECT URL, count(*) FROM {table} GROUP BY URL ORDER BY count(*) DESC LIMIT 10;
SELECT 1, URL, count(*) FROM {table} GROUP BY 1, URL ORDER BY count(*) DESC LIMIT 10;
SELECT ClientIP, ClientIP - 1, ClientIP - 2, ClientIP - 3, count(*) FROM {table} GROUP BY ClientIP, ClientIP - 1, ClientIP - 2, ClientIP - 3 ORDER BY count(*) DESC LIMIT 10;
SELECT URL, count(*) AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT DontCountHits AND NOT Refresh AND URL != '' GROUP BY URL ORDER BY PageViews DESC LIMIT 10;
SELECT Title, count(*) AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT DontCountHits AND NOT Refresh AND Title != '' GROUP BY Title ORDER BY PageViews DESC LIMIT 10;
SELECT URL, count(*) AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT Refresh AND IsLink AND NOT IsDownload GROUP BY URL ORDER BY PageViews DESC LIMIT 1000;
SELECT TraficSourceID, SearchEngineID, AdvEngineID, CASE WHEN SearchEngineID = 0 AND AdvEngineID = 0 THEN Referer ELSE '' END AS Src, URL AS Dst, count(*) AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT Refresh GROUP BY TraficSourceID, SearchEngineID, AdvEngineID, Src, Dst ORDER BY PageViews DESC LIMIT 1000;
SELECT URLHash, EventDate, count(*) AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT Refresh AND TraficSourceID IN (-1, 6) AND RefererHash = 6202628419148573758 GROUP BY URLHash, EventDate ORDER BY PageViews DESC LIMIT 100000;
SELECT WindowClientWidth, WindowClientHeight, count(*) AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-31') AND NOT Refresh AND NOT DontCountHits AND URLHash = 6202628419148573758 GROUP BY WindowClientWidth, WindowClientHeight ORDER BY PageViews DESC LIMIT 10000;
SELECT TIME_SLICE(EventTime, 1, 'MINUTE') AS Minute, count(*) AS PageViews FROM {table} WHERE CounterID = 34 AND EventDate >= DATE('2013-07-01') AND EventDate <= DATE('2013-07-02') AND NOT Refresh AND NOT DontCountHits GROUP BY Minute ORDER BY Minute;

2
dbms/include/DB/DataStreams/BlockInputStreamFromRowInputStream.h
View File

@ -32,6 +32,8 @@ public:
return res.str();
}
RowInputStreamPtr & getRowInput() { return row_input; }
protected:
Block readImpl() override;

29
dbms/include/DB/DataStreams/TabSeparatedRowInputStream.h
View File

@ -29,6 +29,35 @@ private:
bool with_names;
bool with_types;
DataTypes data_types;
/// Для удобной диагностики в случае ошибки.
size_t row_num = 0;
/// Сколько байт было считано, не считая тех, что ещё в буфере.
size_t bytes_read_at_start_of_buffer_on_current_row = 0;
size_t bytes_read_at_start_of_buffer_on_prev_row = 0;
BufferBase::Position pos_of_current_row = nullptr;
BufferBase::Position pos_of_prev_row = nullptr;
/** В случае исключения при парсинге, вызывается эта функция.
* Она выполняет заново парсинг последних двух строк и выводит подробную информацию о том, что происходит.
*/
void printDiagnosticInfo(WriteBuffer & out);
void updateDiagnosticInfo()
{
++row_num;
bytes_read_at_start_of_buffer_on_prev_row = bytes_read_at_start_of_buffer_on_current_row;
bytes_read_at_start_of_buffer_on_current_row = istr.count() - istr.offset();
pos_of_prev_row = pos_of_current_row;
pos_of_current_row = istr.position();
}
bool parseRowAndPrintDiagnosticInfo(WriteBuffer & out, size_t max_length_of_column_name, size_t max_length_of_data_type_name);
};
}

2
dbms/include/DB/Interpreters/Settings.h
View File

@ -86,7 +86,7 @@ struct Settings
/** Включена ли компиляция запросов. */ \
M(SettingBool, compile, false) \
/** Количество одинаковых по структуре запросов перед тем, как инициируется их компиляция. */ \
M(SettingUInt64, min_count_to_compile, 0) \
M(SettingUInt64, min_count_to_compile, 3) \
/** При каком количестве ключей, начинает использоваться двухуровневая агрегация. 0 - никогда не использовать. */ \
M(SettingUInt64, group_by_two_level_threshold, 100000) \
\

11
dbms/scripts/README Normal file
View File

@ -0,0 +1,11 @@
# How to create dictionaries for region* functions:
# 1. You need access to host ███████████.yandex-team.ru.
# 2. Do the following commands:
curl 'http://███████████.yandex-team.ru/?fields=id,parent_id,type,population' | tail -n+2 > regions_hierarchy.txt
curl 'http://███████████.yandex-team.ru/?fields=id,parent_id,type,population&new_parents=977:187' | tail -n+2 > regions_hierarchy_ua.txt
curl 'http://███████████.yandex-team.ru/?fields=id,ru_name' | tail -n+2 > regions_names_ru.txt
curl 'http://███████████.yandex-team.ru/?fields=id,uk_name' | tail -n+2 > regions_names_ua.txt
curl 'http://███████████.yandex-team.ru/?fields=id,by_name' | tail -n+2 > regions_names_by.txt
curl 'http://███████████.yandex-team.ru/?fields=id,kz_name' | tail -n+2 > regions_names_kz.txt
curl 'http://███████████.yandex-team.ru/?fields=id,tr_name' | tail -n+2 > regions_names_tr.txt

31
dbms/scripts/geobase_to_regions_hierarchy.pl
View File

@ -1,31 +0,0 @@
#!/usr/bin/perl -w
use strict;
use warnings;
use geobase;
sub get_population {
my $key = shift;
my $depth = shift || 0;
return 0 if ($depth > 100);
my $current = int($Region{$key}->{zip_old} || 0); # zip_old, не смотря на название, содержит население региона.
return $current if ($current);
my $sum_of_children = 0;
for my $child (@{$Region{$key}->{chld}}) {
$sum_of_children += get_population($child, $depth + 1);
}
return $sum_of_children;
}
foreach my $key (keys %Region) {
print $key . "\t"
. ($Region{$key}->{parents}[-1] || 0) . "\t"
. ($Region{$key}->{type} || 0) . "\t"
. get_population($key) . "\n";
}

25
dbms/scripts/geobase_to_regions_names.pl
View File

@ -1,25 +0,0 @@
#!/usr/bin/perl -w
use strict;
use warnings;
use geobase;
my @languages = ('ru', 'en', 'ua', 'by', 'kz', 'tr');
my @output_files = map { open(my $output, ">:encoding(UTF-8)", "regions_names_" . $_ . ".txt") || die $!; $output } @languages;
my %outputs;
@outputs{@languages} = @output_files;
foreach my $key (keys %Region) {
foreach my $lang (@languages) {
my $field = ( $lang eq 'ru' ? 'name' : $lang . '_name' );
my $name = $Region{$key}->{$field};
if ($name) {
$name =~ s/^\s+//;
$name =~ s/\s+$//;
$name =~ s/(\t|\n)/ /g;
if ($name ne '') {
print { $outputs{$lang} } $key . "\t" . $name . "\n";
}
}
}
}

32
dbms/src/Client/Client.cpp
View File

@ -38,6 +38,8 @@
#include <DB/IO/ReadBufferFromIStream.h>
#include <DB/DataStreams/AsynchronousBlockInputStream.h>
#include <DB/DataStreams/BlockInputStreamFromRowInputStream.h>
#include <DB/DataStreams/TabSeparatedRowInputStream.h>
#include <DB/Parsers/ParserQuery.h>
#include <DB/Parsers/ASTSetQuery.h>
@ -57,6 +59,7 @@
#include <DB/Common/ExternalTable.h>
#include <DB/Common/UnicodeBar.h>
#include <DB/Common/formatReadable.h>
#include <DB/Columns/ColumnString.h>
/// http://en.wikipedia.org/wiki/ANSI_escape_code
@ -93,6 +96,7 @@ private:
};
bool is_interactive = true; /// Использовать readline интерфейс или batch режим.
bool print_time_to_stderr = false; /// В неинтерактивном режиме, выводить время выполнения в stderr.
bool stdin_is_not_tty = false; /// stdin - не терминал.
winsize terminal_size {}; /// Размер терминала - для вывода прогресс-бара.
@ -257,6 +261,9 @@ private:
if (is_interactive)
{
if (print_time_to_stderr)
throw Exception("time option could be specified only in non-interactive mode", ErrorCodes::BAD_ARGUMENTS);
/// Отключаем tab completion.
rl_bind_key('\t', rl_insert);
@ -557,6 +564,10 @@ private:
std::cout << std::endl << std::endl;
}
else if (print_time_to_stderr)
{
std::cerr << watch.elapsedSeconds() << "\n";
}
return true;
}
@ -679,13 +690,16 @@ private:
if (!insert->format.empty())
current_format = insert->format;
BlockInputStreamPtr block_std_in = new AsynchronousBlockInputStream(context.getFormatFactory().getInput(
current_format, buf, sample, insert_format_max_block_size, context.getDataTypeFactory()));
block_std_in->readPrefix();
BlockInputStreamPtr block_input = context.getFormatFactory().getInput(
current_format, buf, sample, insert_format_max_block_size, context.getDataTypeFactory());
BlockInputStreamPtr async_block_input = new AsynchronousBlockInputStream(block_input);
async_block_input->readPrefix();
while (true)
{
Block block = block_std_in->read();
Block block = async_block_input->read();
connection->sendData(block);
processed_rows += block.rows();
@ -693,7 +707,7 @@ private:
break;
}
block_std_in->readSuffix();
async_block_input->readSuffix();
}
@ -1022,7 +1036,9 @@ public:
("database,d", boost::program_options::value<std::string>(), "database")
("multiline,m", "multiline")
("multiquery,n", "multiquery")
("vertical,E", "vertical")
("format,f", boost::program_options::value<std::string>(), "default output format")
("vertical,E", "vertical output format, same as --format=Vertical or FORMAT Vertical or \\G at end of command")
("time,t", "print query execution time to stderr in non-interactive mode (for benchmarks)")
APPLY_FOR_SETTINGS(DECLARE_SETTING)
APPLY_FOR_LIMITS(DECLARE_LIMIT)
;
@ -1135,8 +1151,12 @@ public:
config().setBool("multiline", true);
if (options.count("multiquery"))
config().setBool("multiquery", true);
if (options.count("format"))
config().setString("format", options["format"].as<std::string>());
if (options.count("vertical"))
config().setBool("vertical", true);
if (options.count("time"))
print_time_to_stderr = true;
}
};

314
dbms/src/DataStreams/TabSeparatedRowInputStream.cpp
View File

@ -1,6 +1,8 @@
#include <DB/IO/ReadHelpers.h>
#include <DB/IO/Operators.h>
#include <DB/DataStreams/TabSeparatedRowInputStream.h>
#include <DB/DataTypes/DataTypesNumberFixed.h>
namespace DB
@ -43,29 +45,327 @@ void TabSeparatedRowInputStream::readPrefix()
}
/** Проверка на распространённый случай ошибки - использование Windows перевода строки.
*/
static void checkForCarriageReturn(ReadBuffer & istr)
{
if (istr.position()[0] == '\r' || (istr.position() != istr.buffer().begin() && istr.position()[-1] == '\r'))
throw Exception("\nYou have carriage return (\\r, 0x0D, ASCII 13) at end of first row."
"\nIt's like your input data has DOS/Windows style line separators, that are illegal in TabSeparated format."
" You must transform your file to Unix format."
"\nBut if you really need carriage return at end of string value of last column, you need to escape it as \\r.",
ErrorCodes::INCORRECT_DATA);
}
bool TabSeparatedRowInputStream::read(Row & row)
{
updateDiagnosticInfo();
size_t size = data_types.size();
row.resize(size);
try
{
for (size_t i = 0; i < size; ++i)
{
if (i == 0 && istr.eof())
{
row.clear();
return false;
}
data_types[i]->deserializeTextEscaped(row[i], istr);
/// пропускаем разделители
if (i + 1 == size)
{
if (!istr.eof())
{
if (unlikely(row_num == 1))
checkForCarriageReturn(istr);
assertString("\n", istr);
}
}
else
assertString("\t", istr);
}
}
catch (Exception & e)
{
String verbose_diagnostic;
{
WriteBufferFromString diagnostic_out(verbose_diagnostic);
printDiagnosticInfo(diagnostic_out);
}
e.addMessage("\n" + verbose_diagnostic);
throw;
}
return true;
}
void TabSeparatedRowInputStream::printDiagnosticInfo(WriteBuffer & out)
{
/// Вывести подробную диагностику возможно лишь если последняя и предпоследняя строка ещё находятся в буфере для чтения.
size_t bytes_read_at_start_of_buffer = istr.count() - istr.offset();
if (bytes_read_at_start_of_buffer != bytes_read_at_start_of_buffer_on_prev_row)
{
out << "Could not print diagnostic info because two last rows aren't in buffer (rare case)\n";
return;
}
size_t max_length_of_column_name = 0;
for (size_t i = 0; i < sample.columns(); ++i)
if (sample.getByPosition(i).name.size() > max_length_of_column_name)
max_length_of_column_name = sample.getByPosition(i).name.size();
size_t max_length_of_data_type_name = 0;
for (size_t i = 0; i < sample.columns(); ++i)
if (sample.getByPosition(i).type->getName().size() > max_length_of_data_type_name)
max_length_of_data_type_name = sample.getByPosition(i).type->getName().size();
/// Откатываем курсор для чтения на начало предыдущей или текущей строки и парсим всё заново. Но теперь выводим подробную информацию.
if (pos_of_prev_row)
{
istr.position() = pos_of_prev_row;
out << "\nRow " << (row_num - 1) << ":\n";
if (!parseRowAndPrintDiagnosticInfo(out, max_length_of_column_name, max_length_of_data_type_name))
return;
}
else
{
if (!pos_of_current_row)
{
out << "Could not print diagnostic info because parsing of data hasn't started.\n";
return;
}
istr.position() = pos_of_current_row;
}
out << "\nRow " << row_num << ":\n";
parseRowAndPrintDiagnosticInfo(out, max_length_of_column_name, max_length_of_data_type_name);
out << "\n";
}
static void verbosePrintString(BufferBase::Position begin, BufferBase::Position end, WriteBuffer & out)
{
if (end == begin)
{
out << "<EMPTY>";
return;
}
out << "\"";
for (auto pos = begin; pos < end; ++pos)
{
switch (*pos)
{
case '\0':
out << "<ASCII NUL>";
break;
case '\b':
out << "<BACKSPACE>";
break;
case '\f':
out << "<FORM FEED>";
break;
case '\n':
out << "<LINE FEED>";
break;
case '\r':
out << "<CARRIAGE RETURN>";
break;
case '\t':
out << "<TAB>";
break;
case '\\':
out << "<BACKSLASH>";
break;
case '"':
out << "<DOUBLE QUOTE>";
break;
case '\'':
out << "<SINGLE QUOTE>";
break;
default:
{
if (*pos >= 0 && *pos < 32)
{
static const char * hex = "0123456789ABCDEF";
out << "<0x" << hex[*pos / 16] << hex[*pos % 16] << ">";
}
else
out << *pos;
}
}
}
out << "\"";
}
bool TabSeparatedRowInputStream::parseRowAndPrintDiagnosticInfo(
WriteBuffer & out, size_t max_length_of_column_name, size_t max_length_of_data_type_name)
{
size_t size = data_types.size();
for (size_t i = 0; i < size; ++i)
{
if (i == 0 && istr.eof())
{
row.clear();
out << "<End of stream>\n";
return false;
}
data_types[i]->deserializeTextEscaped(row[i], istr);
/// пропускаем разделители
out << "Column " << i << ", " << std::string((i < 10 ? 2 : i < 100 ? 1 : 0), ' ')
<< "name: " << sample.getByPosition(i).name << ", " << std::string(max_length_of_column_name - sample.getByPosition(i).name.size(), ' ')
<< "type: " << data_types[i]->getName() << ", " << std::string(max_length_of_data_type_name - data_types[i]->getName().size(), ' ');
auto prev_position = istr.position();
std::exception_ptr exception;
Field field;
try
{
data_types[i]->deserializeTextEscaped(field, istr);
}
catch (...)
{
exception = std::current_exception();
}
auto curr_position = istr.position();
if (curr_position < prev_position)
throw Exception("Logical error: parsing is non-deterministic.", ErrorCodes::LOGICAL_ERROR);
if (data_types[i]->isNumeric())
{
/// Пустая строка вместо числа.
if (curr_position == prev_position)
{
out << "ERROR: text ";
verbosePrintString(prev_position, std::min(prev_position + 10, istr.buffer().end()), out);
out << " is not like " << data_types[i]->getName() << "\n";
return false;
}
}
out << "parsed text: ";
verbosePrintString(prev_position, curr_position, out);
if (exception)
{
if (data_types[i]->getName() == "DateTime")
out << "ERROR: DateTime must be in YYYY-MM-DD hh:mm:ss format.\n";
else if (data_types[i]->getName() == "Date")
out << "ERROR: Date must be in YYYY-MM-DD format.\n";
else
out << "ERROR\n";
return false;
}
out << "\n";
if (data_types[i]->isNumeric())
{
if (*curr_position != '\n' && *curr_position != '\t')
{
out << "ERROR: garbage after " << data_types[i]->getName() << ": ";
verbosePrintString(curr_position, std::min(curr_position + 10, istr.buffer().end()), out);
out << "\n";
if (data_types[i]->getName() == "DateTime")
out << "ERROR: DateTime must be in YYYY-MM-DD hh:mm:ss format.\n";
else if (data_types[i]->getName() == "Date")
out << "ERROR: Date must be in YYYY-MM-DD format.\n";
return false;
}
}
if ( (typeid_cast<const DataTypeUInt8 *>(data_types[i].get()) && field.get<UInt64>() > std::numeric_limits<UInt8>::max())
|| (typeid_cast<const DataTypeUInt16 *>(data_types[i].get()) && field.get<UInt64>() > std::numeric_limits<UInt16>::max())
|| (typeid_cast<const DataTypeUInt32 *>(data_types[i].get()) && field.get<UInt64>() > std::numeric_limits<UInt32>::max())
|| (typeid_cast<const DataTypeInt8 *>(data_types[i].get())
&& (field.get<Int64>() > std::numeric_limits<Int8>::max() || field.get<Int64>() < std::numeric_limits<Int8>::min()))
|| (typeid_cast<const DataTypeInt16 *>(data_types[i].get())
&& (field.get<Int64>() > std::numeric_limits<Int16>::max() || field.get<Int64>() < std::numeric_limits<Int16>::min()))
|| (typeid_cast<const DataTypeInt32 *>(data_types[i].get())
&& (field.get<Int64>() > std::numeric_limits<Int32>::max() || field.get<Int64>() < std::numeric_limits<Int32>::min())))
{
out << "ERROR: parsed number is out of range of data type.\n";
return false;
}
/// Разделители
if (i + 1 == size)
{
if (!istr.eof())
assertString("\n", istr);
{
try
{
assertString("\n", istr);
}
catch (const DB::Exception &)
{
if (*istr.position() == '\t')
{
out << "ERROR: Tab found where line feed is expected."
" It's like your file has more columns than expected.\n"
"And if your file have right number of columns, maybe it have unescaped tab in value.\n";
}
else if (*istr.position() == '\r')
{
out << "ERROR: Carriage return found where line feed is expected."
" It's like your file has DOS/Windows style line separators, that is illegal in TabSeparated format.\n";
}
else
{
out << "ERROR: There is no line feed. ";
verbosePrintString(istr.position(), istr.position() + 1, out);
out << " found instead.\n";
}
return false;
}
}
}
else
assertString("\t", istr);
{
try
{
assertString("\t", istr);
}
catch (const DB::Exception &)
{
if (*istr.position() == '\n')
{
out << "ERROR: Line feed found where tab is expected."
" It's like your file has less columns than expected.\n"
"And if your file have right number of columns, maybe it have unescaped backslash in value before tab, which cause tab has escaped.\n";
}
else if (*istr.position() == '\r')
{
out << "ERROR: Carriage return found where tab is expected.\n";
}
else
{
out << "ERROR: There is no tab. ";
verbosePrintString(istr.position(), istr.position() + 1, out);
out << " found instead.\n";
}
return false;
}
}
}
return true;

1
dbms/src/Interpreters/Compiler.cpp
View File

@ -218,6 +218,7 @@ void Compiler::compile(
" -I /usr/share/clickhouse/headers/libs/libcityhash/"
" -I /usr/share/clickhouse/headers/libs/libcommon/include/"
" -I /usr/share/clickhouse/headers/libs/libdouble-conversion/"
" -I /usr/share/clickhouse/headers/libs/libcpuid/include/"
" -I /usr/share/clickhouse/headers/libs/libmysqlxx/include/"
" -I /usr/share/clickhouse/headers/libs/libstatdaemons/include/"
" -I /usr/share/clickhouse/headers/libs/libstats/include/"

9
dbms/src/Parsers/ParserInsertQuery.cpp
View File

@ -111,6 +111,15 @@ bool ParserInsertQuery::parseImpl(Pos & pos, Pos end, ASTPtr & node, Expected &
ParserWhiteSpaceOrComments ws_without_nl(false);
ws_without_nl.ignore(pos, end);
if (pos != end && *pos == ';')
throw Exception("You have excessive ';' symbol before data for INSERT.\n"
"Example:\n\n"
"INSERT INTO t (x, y) FORMAT TabSeparated\n"
"1\tHello\n"
"2\tWorld\n"
"\n"
"Note that there is no ';' in first line.", ErrorCodes::SYNTAX_ERROR);
if (pos != end && *pos == '\n')
++pos;