You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
24 KiB
24 KiB
sergeant : Tools to Transform and Query Data with 'Apache' 'Drill'
Drill + sergeant is (IMO) a nice alternative to Spark + sparklyr if you don't need the ML components of Spark (i.e. just need to query "big data" sources, need to interface with parquet, need to combine disparate data source types — json, csv, parquet, rdbms - for aggregation, etc). Drill also has support for spatial queries.
I find writing SQL queries to parquet files with Drill on a local linux or macOS workstation to be more performant than doing the data ingestion work with R (especially for large or disperate data sets). I also work with many tiny JSON files on a daily basis and Drill makes it much easier to do so. YMMV.
You can download Drill from https://drill.apache.org/download/ (use "Direct File Download"). I use /usr/local/drill as the install directory. drill-embedded is a super-easy way to get started playing with Drill on a single workstation and most of my workflows can get by using Drill this way. If there is sufficient desire for an automated downloader and a way to start the drill-embedded server from within R, please file an issue.
There are a few convenience wrappers for various informational SQL queries (like drill_version()). Please file an PR if you add more.
The package has been written with retrieval of rectangular data sources in mind. If you need/want a version of drill_query() that will enable returning of non-rectangular data (which is possible with Drill) then please file an issue.
Some of the more "controlling vs data ops" REST API functions aren't implemented. Please file a PR if you need those.
Finally, I run most of this locally and at home, so it's all been coded with no authentication or encryption in mind. If you want/need support for that, please file an issue. If there is demand for this, it will change the R API a bit (I've already thought out what to do but have no need for it right now).
The following functions are implemented:
DBI
- A "just enough" feature complete R
DBIdriver has been implemented using the Drill REST API, mostly to facilitate thedplyrinterface. Use theRJDBCdriver interface if you need moreDBIfunctionality. - This also means that SQL functions unique to Drill have also been "implemented" (i.e. made accessible to the
dplyrinterface). If you have custom Drill SQL functions that need to be implemented please file an issue on GitHub. Many should work without it, but some may require a custom interface.
RJDBC
drill_jdbc: Connect to Drill using JDBC, enabling use of said idioms. SeeRJDBCfor more info.- NOTE: The DRILL JDBC driver fully-qualified path must be placed in the
DRILL_JDBC_JARenvironment variable. This is best done via~/.Renvironfor interactive work. i.e.DRILL_JDBC_JAR=/usr/local/drill/jars/drill-jdbc-all-1.10.0.jar
dplyr:
src_drill: Connect to Drill (using dplyr) + supporting functions
See dplyr for the dplyr operations (light testing shows they work in basic SQL use-cases but Drill's SQL engine has issues with more complex queries).
Drill APIs:
drill_connection: Setup parameters for a Drill server/cluster connectiondrill_active: Test whether Drill HTTP REST API server is updrill_cancel: Cancel the query that has the given queryiddrill_jdbc: Connect to Drill using JDBCdrill_metrics: Get the current memory metricsdrill_options: List the name, default, and data type of the system and session optionsdrill_profile: Get the profile of the query that has the given query iddrill_profiles: Get the profiles of running and completed queriesdrill_query: Submit a query and return resultsdrill_set: Set Drill SYSTEM or SESSION optionsdrill_settings_reset: Changes (optionally, all) session settings back to system defaultsdrill_show_files: Show files in a file system schema.drill_show_schemas: Returns a list of available schemas.drill_stats: Get Drillbit information, such as ports numbersdrill_status: Get the status of Drilldrill_storage: Get the list of storage plugin names and configurationsdrill_system_reset: Changes (optionally, all) system settings back to system defaultsdrill_threads: Get information about threadsdrill_uplift: Turn a columnar query results into a type-converted tbldrill_use: Change to a particular schema.drill_version: Identify the version of Drill running
Installation
devtools::install_github("hrbrmstr/sergeant")
Experimental dplyr interface
library(sergeant)
ds <- src_drill("localhost") # use localhost if running standalone on same system otherwise the host or IP of your Drill server
ds
#> src: DrillConnection
#> tbls: INFORMATION_SCHEMA, cp.default, dfs.d, dfs.default, dfs.h, dfs.natexp, dfs.p, dfs.root, dfs.tmp, sys
db <- tbl(ds, "cp.`employee.json`")
# without `collect()`:
count(db, gender, marital_status)
#> # Source: lazy query [?? x 3]
#> # Database: DrillConnection
#> # Groups: gender
#> marital_status gender n
#> <chr> <chr> <int>
#> 1 S F 297
#> 2 M M 278
#> 3 S M 276
#> 4 M F 304
# ^^ gets translated to:
#
# SELECT *
# FROM (SELECT gender , marital_status , COUNT(*) AS n
# FROM cp.`employee.json`
# GROUP BY gender , marital_status ) govketbhqb
# LIMIT 1000
count(db, gender, marital_status) %>% collect()
#> # A tibble: 4 x 3
#> # Groups: gender [2]
#> marital_status gender n
#> * <chr> <chr> <int>
#> 1 S F 297
#> 2 M M 278
#> 3 S M 276
#> 4 M F 304
# ^^ gets translated to:
#
# SELECT gender , marital_status , COUNT(*) AS n
# FROM cp.`employee.json`
# GROUP BY gender , marital_status
group_by(db, position_title) %>%
count(gender) -> tmp2
group_by(db, position_title) %>%
count(gender) %>%
ungroup() %>%
mutate(full_desc=ifelse(gender=="F", "Female", "Male")) %>%
collect() %>%
select(Title=position_title, Gender=full_desc, Count=n)
#> # A tibble: 30 x 3
#> Title Gender Count
#> * <chr> <chr> <int>
#> 1 President Female 1
#> 2 VP Country Manager Male 3
#> 3 VP Country Manager Female 3
#> 4 VP Information Systems Female 1
#> 5 VP Human Resources Female 1
#> 6 Store Manager Female 13
#> 7 VP Finance Male 1
#> 8 Store Manager Male 11
#> 9 HQ Marketing Female 2
#> 10 HQ Information Systems Female 4
#> # ... with 20 more rows
# ^^ gets translated to:
#
# SELECT position_title , gender , n ,
# CASE WHEN ( gender = 'F') THEN ('Female') ELSE ('Male') END AS full_desc
# FROM (SELECT position_title , gender , COUNT(*) AS n
# FROM cp.`employee.json`
# GROUP BY position_title , gender ) dcyuypuypb
arrange(db, desc(employee_id)) %>% print(n=20)
#> # Source: table<cp.`employee.json`> [?? x 16]
#> # Database: DrillConnection
#> # Ordered by: desc(employee_id)
#> store_id gender department_id birth_date supervisor_id last_name position_title hire_date
#> <int> <chr> <int> <date> <int> <chr> <chr> <dttm>
#> 1 18 F 18 1914-02-02 1140 Stand Store Temporary Stocker 1998-01-01
#> 2 18 M 18 1914-02-02 1140 Burnham Store Temporary Stocker 1998-01-01
#> 3 18 F 18 1914-02-02 1139 Doolittle Store Temporary Stocker 1998-01-01
#> 4 18 M 18 1914-02-02 1139 Pirnie Store Temporary Stocker 1998-01-01
#> 5 18 M 17 1914-02-02 1140 Younce Store Permanent Stocker 1998-01-01
#> 6 18 F 17 1914-02-02 1140 Biltoft Store Permanent Stocker 1998-01-01
#> 7 18 M 17 1914-02-02 1139 Detwiler Store Permanent Stocker 1998-01-01
#> 8 18 F 17 1914-02-02 1139 Ciruli Store Permanent Stocker 1998-01-01
#> 9 18 F 16 1914-02-02 1140 Bishop Store Temporary Checker 1998-01-01
#> 10 18 F 16 1914-02-02 1140 Cutwright Store Temporary Checker 1998-01-01
#> 11 18 F 16 1914-02-02 1139 Anderson Store Temporary Checker 1998-01-01
#> 12 18 F 16 1914-02-02 1139 Swartwood Store Temporary Checker 1998-01-01
#> 13 18 M 15 1914-02-02 1140 Curtsinger Store Permanent Checker 1998-01-01
#> 14 18 F 15 1914-02-02 1140 Quick Store Permanent Checker 1998-01-01
#> 15 18 M 15 1914-02-02 1139 Souza Store Permanent Checker 1998-01-01
#> 16 18 M 15 1914-02-02 1139 Compagno Store Permanent Checker 1998-01-01
#> 17 18 M 11 1961-09-24 1139 Jaramillo Store Shift Supervisor 1998-01-01
#> 18 18 M 11 1972-05-12 17 Belsey Store Assistant Manager 1998-01-01
#> 19 12 M 18 1914-02-02 1069 Eichorn Store Temporary Stocker 1998-01-01
#> 20 12 F 18 1914-02-02 1069 Geiermann Store Temporary Stocker 1998-01-01
#> # ... with more rows, and 8 more variables: management_role <chr>, salary <dbl>, marital_status <chr>, full_name <chr>,
#> # employee_id <int>, education_level <chr>, first_name <chr>, position_id <int>
# ^^ gets translated to:
#
# SELECT *
# FROM (SELECT *
# FROM cp.`employee.json`
# ORDER BY employee_id DESC) lvpxoaejbc
# LIMIT 5
mutate(db, position_title=tolower(position_title)) %>%
mutate(salary=as.numeric(salary)) %>%
mutate(gender=ifelse(gender=="F", "Female", "Male")) %>%
mutate(marital_status=ifelse(marital_status=="S", "Single", "Married")) %>%
group_by(supervisor_id) %>%
summarise(underlings_count=n()) %>%
collect()
#> # A tibble: 112 x 2
#> supervisor_id underlings_count
#> * <int> <int>
#> 1 0 1
#> 2 1 7
#> 3 5 9
#> 4 4 2
#> 5 2 3
#> 6 20 2
#> 7 21 4
#> 8 22 7
#> 9 6 4
#> 10 36 2
#> # ... with 102 more rows
# ^^ gets translated to:
#
# SELECT supervisor_id , COUNT(*) AS underlings_count
# FROM (SELECT employee_id , full_name , first_name , last_name , position_id , position_title , store_id , department_id , birth_date , hire_date , salary , supervisor_id , education_level , gender , management_role , CASE WHEN ( marital_status = 'S') THEN ('Single') ELSE ('Married') END AS marital_status
# FROM (SELECT employee_id , full_name , first_name , last_name , position_id , position_title , store_id , department_id , birth_date , hire_date , salary , supervisor_id , education_level , marital_status , management_role , CASE WHEN ( gender = 'F') THEN ('Female') ELSE ('Male') END AS gender
# FROM (SELECT employee_id , full_name , first_name , last_name , position_id , position_title , store_id , department_id , birth_date , hire_date , supervisor_id , education_level , marital_status , gender , management_role , CAST( salary AS DOUBLE) AS salary
# FROM (SELECT employee_id , full_name , first_name , last_name , position_id , store_id , department_id , birth_date , hire_date , salary , supervisor_id , education_level , marital_status , gender , management_role , LOWER( position_title ) AS position_title
# FROM cp.`employee.json` ) cnjsqxeick ) bnbnjrubna ) wavfmhkczv ) zaxeyyicxo
# GROUP BY supervisor_id
Usage
library(sergeant)
# current verison
packageVersion("sergeant")
#> [1] '0.5.2'
dc <- drill_connection("localhost")
drill_active(dc)
#> [1] TRUE
drill_version(dc)
#> [1] "1.10.0"
drill_storage(dc)$name
#> [1] "cp" "dfs" "hbase" "hive" "kudu" "mongo" "s3"
Working with the built-in JSON data sets:
drill_query(dc, "SELECT * FROM cp.`employee.json` limit 100")
#> Parsed with column specification:
#> cols(
#> store_id = col_integer(),
#> gender = col_character(),
#> department_id = col_integer(),
#> birth_date = col_date(format = ""),
#> supervisor_id = col_integer(),
#> last_name = col_character(),
#> position_title = col_character(),
#> hire_date = col_datetime(format = ""),
#> management_role = col_character(),
#> salary = col_double(),
#> marital_status = col_character(),
#> full_name = col_character(),
#> employee_id = col_integer(),
#> education_level = col_character(),
#> first_name = col_character(),
#> position_id = col_integer()
#> )
#> # A tibble: 100 x 16
#> store_id gender department_id birth_date supervisor_id last_name position_title hire_date management_role
#> * <int> <chr> <int> <date> <int> <chr> <chr> <dttm> <chr>
#> 1 0 F 1 1961-08-26 0 Nowmer President 1994-12-01 Senior Management
#> 2 0 M 1 1915-07-03 1 Whelply VP Country Manager 1994-12-01 Senior Management
#> 3 0 M 1 1969-06-20 1 Spence VP Country Manager 1998-01-01 Senior Management
#> 4 0 F 1 1951-05-10 1 Gutierrez VP Country Manager 1998-01-01 Senior Management
#> 5 0 F 2 1942-10-08 1 Damstra VP Information Systems 1994-12-01 Senior Management
#> 6 0 F 3 1949-03-27 1 Kanagaki VP Human Resources 1994-12-01 Senior Management
#> 7 9 F 11 1922-08-10 5 Brunner Store Manager 1998-01-01 Store Management
#> 8 21 F 11 1979-06-23 5 Blumberg Store Manager 1998-01-01 Store Management
#> 9 0 M 5 1949-08-26 1 Stanz VP Finance 1994-12-01 Senior Management
#> 10 1 M 11 1967-06-20 5 Murraiin Store Manager 1998-01-01 Store Management
#> # ... with 90 more rows, and 7 more variables: salary <dbl>, marital_status <chr>, full_name <chr>, employee_id <int>,
#> # education_level <chr>, first_name <chr>, position_id <int>
drill_query(dc, "SELECT COUNT(gender) AS gender FROM cp.`employee.json` GROUP BY gender")
#> Parsed with column specification:
#> cols(
#> gender = col_integer()
#> )
#> # A tibble: 2 x 1
#> gender
#> * <int>
#> 1 601
#> 2 554
drill_options(dc)
#> # A tibble: 113 x 4
#> name value type kind
#> * <chr> <chr> <chr> <chr>
#> 1 planner.enable_hash_single_key TRUE SYSTEM BOOLEAN
#> 2 store.parquet.reader.pagereader.queuesize 2 SYSTEM LONG
#> 3 planner.enable_limit0_optimization FALSE SYSTEM BOOLEAN
#> 4 store.json.read_numbers_as_double FALSE SYSTEM BOOLEAN
#> 5 planner.enable_constant_folding TRUE SYSTEM BOOLEAN
#> 6 store.json.extended_types FALSE SYSTEM BOOLEAN
#> 7 planner.memory.non_blocking_operators_memory 64 SYSTEM LONG
#> 8 planner.enable_multiphase_agg TRUE SYSTEM BOOLEAN
#> 9 exec.query_profile.debug_mode FALSE SYSTEM BOOLEAN
#> 10 planner.filter.max_selectivity_estimate_factor 1 SYSTEM DOUBLE
#> # ... with 103 more rows
drill_options(dc, "json")
#> # A tibble: 7 x 4
#> name value type kind
#> <chr> <chr> <chr> <chr>
#> 1 store.json.read_numbers_as_double FALSE SYSTEM BOOLEAN
#> 2 store.json.extended_types FALSE SYSTEM BOOLEAN
#> 3 store.json.writer.uglify FALSE SYSTEM BOOLEAN
#> 4 store.json.reader.skip_invalid_records FALSE SYSTEM BOOLEAN
#> 5 store.json.reader.print_skipped_invalid_record_number FALSE SYSTEM BOOLEAN
#> 6 store.json.all_text_mode FALSE SYSTEM BOOLEAN
#> 7 store.json.writer.skip_null_fields TRUE SYSTEM BOOLEAN
Working with parquet files
drill_query(dc, "SELECT * FROM dfs.`/usr/local/drill/sample-data/nation.parquet` LIMIT 5")
#> Parsed with column specification:
#> cols(
#> N_COMMENT = col_character(),
#> N_NAME = col_character(),
#> N_NATIONKEY = col_integer(),
#> N_REGIONKEY = col_integer()
#> )
#> # A tibble: 5 x 4
#> N_COMMENT N_NAME N_NATIONKEY N_REGIONKEY
#> * <chr> <chr> <int> <int>
#> 1 haggle. carefully f ALGERIA 0 0
#> 2 al foxes promise sly ARGENTINA 1 1
#> 3 y alongside of the p BRAZIL 2 1
#> 4 eas hang ironic, sil CANADA 3 1
#> 5 y above the carefull EGYPT 4 4
Including multiple parquet files in different directories (note the wildcard support):
drill_query(dc, "SELECT * FROM dfs.`/usr/local/drill/sample-data/nations*/nations*.parquet` LIMIT 5")
#> Parsed with column specification:
#> cols(
#> N_COMMENT = col_character(),
#> N_NAME = col_character(),
#> N_NATIONKEY = col_integer(),
#> N_REGIONKEY = col_integer(),
#> dir0 = col_character()
#> )
#> # A tibble: 5 x 5
#> N_COMMENT N_NAME N_NATIONKEY N_REGIONKEY dir0
#> * <chr> <chr> <int> <int> <chr>
#> 1 haggle. carefully f ALGERIA 0 0 nationsSF
#> 2 al foxes promise sly ARGENTINA 1 1 nationsSF
#> 3 y alongside of the p BRAZIL 2 1 nationsSF
#> 4 eas hang ironic, sil CANADA 3 1 nationsSF
#> 5 y above the carefull EGYPT 4 4 nationsSF
A preview of the built-in support for spatial ops
Via: https://github.com/k255/drill-gis
A common use case is to select data within boundary of given polygon:
drill_query(dc, "
select columns[2] as city, columns[4] as lon, columns[3] as lat
from cp.`sample-data/CA-cities.csv`
where
ST_Within(
ST_Point(columns[4], columns[3]),
ST_GeomFromText(
'POLYGON((-121.95 37.28, -121.94 37.35, -121.84 37.35, -121.84 37.28, -121.95 37.28))'
)
)
")
#> Parsed with column specification:
#> cols(
#> city = col_character(),
#> lon = col_double(),
#> lat = col_double()
#> )
#> # A tibble: 7 x 3
#> city lon lat
#> * <chr> <dbl> <dbl>
#> 1 Burbank -121.9316 37.32328
#> 2 San Jose -121.8950 37.33939
#> 3 Lick -121.8458 37.28716
#> 4 Willow Glen -121.8897 37.30855
#> 5 Buena Vista -121.9166 37.32133
#> 6 Parkmoor -121.9308 37.32105
#> 7 Fruitdale -121.9327 37.31086
JDBC
library(RJDBC)
#> Loading required package: rJava
# Use this if connecting to a cluster with zookeeper
# con <- drill_jdbc("drill-node:2181", "drillbits1")
# Use the following if running drill-embedded
con <- drill_jdbc("localhost:31010", use_zk=FALSE)
#> Using [jdbc:drill:drillbit=bigd:31010]...
drill_query(con, "SELECT * FROM cp.`employee.json`")
#> # A tibble: 1,155 x 16
#> employee_id full_name first_name last_name position_id position_title store_id department_id
#> * <dbl> <chr> <chr> <chr> <dbl> <chr> <dbl> <dbl>
#> 1 1 Sheri Nowmer Sheri Nowmer 1 President 0 1
#> 2 2 Derrick Whelply Derrick Whelply 2 VP Country Manager 0 1
#> 3 4 Michael Spence Michael Spence 2 VP Country Manager 0 1
#> 4 5 Maya Gutierrez Maya Gutierrez 2 VP Country Manager 0 1
#> 5 6 Roberta Damstra Roberta Damstra 3 VP Information Systems 0 2
#> 6 7 Rebecca Kanagaki Rebecca Kanagaki 4 VP Human Resources 0 3
#> 7 8 Kim Brunner Kim Brunner 11 Store Manager 9 11
#> 8 9 Brenda Blumberg Brenda Blumberg 11 Store Manager 21 11
#> 9 10 Darren Stanz Darren Stanz 5 VP Finance 0 5
#> 10 11 Jonathan Murraiin Jonathan Murraiin 11 Store Manager 1 11
#> # ... with 1,145 more rows, and 8 more variables: birth_date <chr>, hire_date <chr>, salary <dbl>, supervisor_id <dbl>,
#> # education_level <chr>, marital_status <chr>, gender <chr>, management_role <chr>
# but it can work via JDBC function calls, too
dbGetQuery(con, "SELECT * FROM cp.`employee.json`") %>%
tibble::as_tibble()
#> # A tibble: 1,155 x 16
#> employee_id full_name first_name last_name position_id position_title store_id department_id
#> * <dbl> <chr> <chr> <chr> <dbl> <chr> <dbl> <dbl>
#> 1 1 Sheri Nowmer Sheri Nowmer 1 President 0 1
#> 2 2 Derrick Whelply Derrick Whelply 2 VP Country Manager 0 1
#> 3 4 Michael Spence Michael Spence 2 VP Country Manager 0 1
#> 4 5 Maya Gutierrez Maya Gutierrez 2 VP Country Manager 0 1
#> 5 6 Roberta Damstra Roberta Damstra 3 VP Information Systems 0 2
#> 6 7 Rebecca Kanagaki Rebecca Kanagaki 4 VP Human Resources 0 3
#> 7 8 Kim Brunner Kim Brunner 11 Store Manager 9 11
#> 8 9 Brenda Blumberg Brenda Blumberg 11 Store Manager 21 11
#> 9 10 Darren Stanz Darren Stanz 5 VP Finance 0 5
#> 10 11 Jonathan Murraiin Jonathan Murraiin 11 Store Manager 1 11
#> # ... with 1,145 more rows, and 8 more variables: birth_date <chr>, hire_date <chr>, salary <dbl>, supervisor_id <dbl>,
#> # education_level <chr>, marital_status <chr>, gender <chr>, management_role <chr>
Test Results
library(sergeant)
library(testthat)
#>
#> Attaching package: 'testthat'
#> The following object is masked from 'package:dplyr':
#>
#> matches
date()
#> [1] "Mon Jul 17 12:35:05 2017"
devtools::test()
#> Loading sergeant
#> Testing sergeant
#> dplyr: ...
#> rest: ................
#>
#> DONE ===================================================================================================================
Code of Conduct
Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.