背景

psql 可谓是最强大的数据库命令行客户端:

支持丰富的LIST对象信息的功能(包括数据库、权限、类型、聚合、域、转换、外部表、外部server、大对象、全文检索、逻辑订阅、插件、事件触发器….等数据库所有对象),

支持编程

支持query buffer管理

支持IO操作

其他,支持格式、变量、操作系统命令等。

支持行列变换。

支持大对象操作。

简化的psql周期图

https://postgresweekly.com/issues/308

pic

psql 详情

  1. psql \?
  2. postgres=# \?

General

  1. \copyright show PostgreSQL usage and distribution terms
  2. \crosstabview [COLUMNS] execute query and display results in crosstab
  3. \errverbose show most recent error message at maximum verbosity
  4. \g [FILE] or ; execute query (and send results to file or |pipe)
  5. \gdesc describe result of query, without executing it
  6. \gexec execute query, then execute each value in its result
  7. \gset [PREFIX] execute query and store results in psql variables
  8. \gx [FILE] as \g, but forces expanded output mode
  9. \q quit psql
  10. \watch [SEC] execute query every SEC seconds

Help

  1. \? [commands] show help on backslash commands
  2. \? options show help on psql command-line options
  3. \? variables show help on special variables
  4. \h [NAME] help on syntax of SQL commands, * for all commands

Query Buffer

  1. \e [FILE] [LINE] edit the query buffer (or file) with external editor
  2. \ef [FUNCNAME [LINE]] edit function definition with external editor
  3. \ev [VIEWNAME [LINE]] edit view definition with external editor
  4. \p show the contents of the query buffer
  5. \r reset (clear) the query buffer
  6. \s [FILE] display history or save it to file
  7. \w FILE write query buffer to file

Input/Output

  1. \copy ... perform SQL COPY with data stream to the client host
  2. \echo [STRING] write string to standard output
  3. \i FILE execute commands from file
  4. \ir FILE as \i, but relative to location of current script
  5. \o [FILE] send all query results to file or |pipe
  6. \qecho [STRING] write string to query output stream (see \o)

Conditional

  1. \if EXPR begin conditional block
  2. \elif EXPR alternative within current conditional block
  3. \else final alternative within current conditional block
  4. \endif end conditional block

Informational

  1. (options: S = show system objects, + = additional detail)
  2. \d[S+] list tables, views, and sequences
  3. \d[S+] NAME describe table, view, sequence, or index
  4. \da[S] [PATTERN] list aggregates
  5. \dA[+] [PATTERN] list access methods
  6. \db[+] [PATTERN] list tablespaces
  7. \dc[S+] [PATTERN] list conversions
  8. \dC[+] [PATTERN] list casts
  9. \dd[S] [PATTERN] show object descriptions not displayed elsewhere
  10. \dD[S+] [PATTERN] list domains
  11. \ddp [PATTERN] list default privileges
  12. \dE[S+] [PATTERN] list foreign tables
  13. \det[+] [PATTERN] list foreign tables
  14. \des[+] [PATTERN] list foreign servers
  15. \deu[+] [PATTERN] list user mappings
  16. \dew[+] [PATTERN] list foreign-data wrappers
  17. \df[anptw][S+] [PATRN] list [only agg/normal/procedures/trigger/window] functions
  18. \dF[+] [PATTERN] list text search configurations
  19. \dFd[+] [PATTERN] list text search dictionaries
  20. \dFp[+] [PATTERN] list text search parsers
  21. \dFt[+] [PATTERN] list text search templates
  22. \dg[S+] [PATTERN] list roles
  23. \di[S+] [PATTERN] list indexes
  24. \dl list large objects, same as \lo_list
  25. \dL[S+] [PATTERN] list procedural languages
  26. \dm[S+] [PATTERN] list materialized views
  27. \dn[S+] [PATTERN] list schemas
  28. \do[S] [PATTERN] list operators
  29. \dO[S+] [PATTERN] list collations
  30. \dp [PATTERN] list table, view, and sequence access privileges
  31. \dP[tin+] [PATTERN] list [only table/index] partitioned relations
  32. \drds [PATRN1 [PATRN2]] list per-database role settings
  33. \dRp[+] [PATTERN] list replication publications
  34. \dRs[+] [PATTERN] list replication subscriptions
  35. \ds[S+] [PATTERN] list sequences
  36. \dt[S+] [PATTERN] list tables
  37. \dT[S+] [PATTERN] list data types
  38. \du[S+] [PATTERN] list roles
  39. \dv[S+] [PATTERN] list views
  40. \dx[+] [PATTERN] list extensions
  41. \dy [PATTERN] list event triggers
  42. \l[+] [PATTERN] list databases
  43. \sf[+] FUNCNAME show a function's definition
  44. \sv[+] VIEWNAME show a view's definition
  45. \z [PATTERN] same as \dp

Formatting

  1. \a toggle between unaligned and aligned output mode
  2. \C [STRING] set table title, or unset if none
  3. \f [STRING] show or set field separator for unaligned query output
  4. \H toggle HTML output mode (currently off)
  5. \pset [NAME [VALUE]] set table output option
  6. (border|columns|csv_fieldsep|expanded|fieldsep|
  7. fieldsep_zero|footer|format|linestyle|null|
  8. numericlocale|pager|pager_min_lines|recordsep|
  9. recordsep_zero|tableattr|title|tuples_only|
  10. unicode_border_linestyle|unicode_column_linestyle|
  11. unicode_header_linestyle)
  12. \t [on|off] show only rows (currently off)
  13. \T [STRING] set HTML <table> tag attributes, or unset if none
  14. \x [on|off|auto] toggle expanded output (currently off)

Connection

  1. \c[onnect] {[DBNAME|- USER|- HOST|- PORT|-] | conninfo}
  2. connect to new database (currently "postgres")
  3. \conninfo display information about current connection
  4. \encoding [ENCODING] show or set client encoding
  5. \password [USERNAME] securely change the password for a user

Operating System

  1. \cd [DIR] change the current working directory
  2. \setenv NAME [VALUE] set or unset environment variable
  3. \timing [on|off] toggle timing of commands (currently off)
  4. \! [COMMAND] execute command in shell or start interactive shell

Variables

  1. \prompt [TEXT] NAME prompt user to set internal variable
  2. \set [NAME [VALUE]] set internal variable, or list all if no parameters
  3. \unset NAME unset (delete) internal variable

Large Objects

  1. \lo_export LOBOID FILE
  2. \lo_import FILE [COMMENT]
  3. \lo_list
  4. \lo_unlink LOBOID large object operations

例子

1、列出psql 帮助

  1. \? [commands] show help on backslash commands
  2. \? options show help on psql command-line options
  3. \? variables show help on special variables

2、列出SQL COMMAND语法帮助

  1. \h [NAME] help on syntax of SQL commands, * for all commands
  1. postgres=# \h listen
  2. Command: LISTEN
  3. Description: listen for a notification
  4. Syntax:
  5. LISTEN channel
  6. URL: https://www.postgresql.org/docs/12/sql-listen.html

3、对查询结果进行行列变换

  1. \crosstabview [ colV [ colH [ colD [ sortcolH ] ] ] ]
  2. Executes the current query buffer (like \g) and shows the results in a crosstab grid.
  1. create table abc (uid int, class text, score float);
  2. postgres=# insert into abc select random()*100, (array['语文','数学','英语','物理','化学'])[ceil(random()*5)::int], round((random()*100)::numeric,0) from generate_series(1,500);
  3. INSERT 0 500
  4. postgres=# select * from abc limit 10;
  5. uid | class | score
  6. -----+-------+-------
  7. 87 | 数学 | 29
  8. 54 | 语文 | 95
  9. 1 | 语文 | 57
  10. 43 | 化学 | 4
  11. 51 | 数学 | 37
  12. 88 | 数学 | 14
  13. 77 | 物理 | 54
  14. 44 | 数学 | 5
  15. 36 | 语文 | 96
  16. 5 | 语文 | 73
  17. (10 rows)
  18. postgres=# select class, level, count(*) from (select *,width_bucket(score,0,101,5) as level from abc) t group by 1,2 order by 1,2;
  19. class | level | count
  20. -------+-------+-------
  21. 化学 | 1 | 25
  22. 化学 | 2 | 30
  23. 化学 | 3 | 22
  24. 化学 | 4 | 19
  25. 化学 | 5 | 19
  26. 数学 | 1 | 15
  27. 数学 | 2 | 20
  28. 数学 | 3 | 12
  29. 数学 | 4 | 20
  30. 数学 | 5 | 17
  31. 物理 | 1 | 17
  32. 物理 | 2 | 16
  33. 物理 | 3 | 22
  34. 物理 | 4 | 26
  35. 物理 | 5 | 16
  36. 英语 | 1 | 15
  37. 英语 | 2 | 16
  38. 英语 | 3 | 21
  39. 英语 | 4 | 22
  40. 英语 | 5 | 15
  41. 语文 | 1 | 24
  42. 语文 | 2 | 24
  43. 语文 | 3 | 23
  44. 语文 | 4 | 18
  45. 语文 | 5 | 26
  46. (25 rows)

指定X轴、Y周进行变换,执行前一条SQL(当前query buffer的SQL)。

  1. postgres=# \crosstabview level class
  2. level | 化学 | 数学 | 物理 | 英语 | 语文
  3. -------+------+------+------+------+------
  4. 1 | 25 | 15 | 17 | 15 | 24
  5. 2 | 30 | 20 | 16 | 16 | 24
  6. 3 | 22 | 12 | 22 | 21 | 23
  7. 4 | 19 | 20 | 26 | 22 | 18
  8. 5 | 19 | 17 | 16 | 15 | 26
  9. (5 rows)

4、打印当前query buffer

  1. \p
  2. \w FILE write query buffer to file
  1. postgres=# select 1;
  2. ?column?
  3. ----------
  4. 1
  5. (1 row)
  6. postgres=# \p
  7. select 1;

将QUERY BUFFER写入文件

  1. postgres=# \w '/tmp/abc'

执行操作系统命令

  1. \!
  1. postgres=# \! cat '/tmp/abc'
  2. select 1;

5、打印所有query buffer(实际上是.psql_history )

  1. \s [FILE] display history or save it to file
  1. postgres=# \s '/tmp/abc'
  2. Wrote history to file "/tmp/abc".
  1. postgres=# \! head -n 10 '/tmp/abc'
  2. create table h (id int, info text0;
  3. ;
  4. );
  5. create table h (id int, info text);
  6. insert into h select random()*1000 , 'test' from generate_series(1,1000000);
  7. select * from pg_stat_all_indexes where index;
  8. create index idx_h_1 on h(id);
  9. select * from pg_stat_all_indexes where indexrelname='idx_h_1';
  10. explain (analyze,verbose,timing,costs,buffers) select * from h where id=2;
  11. select * from pg_stat_all_indexes where indexrelname='idx_h_1';

6、执行SQL

  1. \g [FILE] or ; execute query (and send results to file or |pipe)
  2. \gdesc describe result of query, without executing it
  3. \gexec execute query, then execute each value in its result
  4. \gset [PREFIX] execute query and store results in psql variables
  5. \gx [FILE] as \g, but forces expanded output mode

直接执行

  1. select 1
  2. \g
  3. select 1
  4. ;

将执行结果存储到PSQL所在的系统文件中

  1. select 1
  2. \g '/tmp/abc'

列出query buffer中的SQL的返回结果的结构

  1. postgres=# \p
  2. select 1;
  3. postgres=# \gdesc
  4. Column | Type
  5. ----------+---------
  6. ?column? | integer
  7. (1 row)
  8. postgres=# select 'abc',1,2;
  9. ?column? | ?column? | ?column?
  10. ----------+----------+----------
  11. abc | 1 | 2
  12. (1 row)
  13. postgres=# \p
  14. select 'abc',1,2;
  15. postgres=# \gdesc
  16. Column | Type
  17. ----------+---------
  18. ?column? | text
  19. ?column? | integer
  20. ?column? | integer
  21. (3 rows)

执行QUERY,并执行QUERY的结果。通常用于批量操作。

  1. postgres=# select 'create table t_'||i||'( id int, info text, crt_time timestamp)' from generate_series(1,10) i
  2. postgres-# \gexec
  3. CREATE TABLE
  4. CREATE TABLE
  5. CREATE TABLE
  6. CREATE TABLE
  7. CREATE TABLE
  8. CREATE TABLE
  9. CREATE TABLE
  10. CREATE TABLE
  11. CREATE TABLE
  12. CREATE TABLE
  13. postgres=# \dt t_*
  14. List of relations
  15. Schema | Name | Type | Owner
  16. --------+------+-------+----------
  17. public | t_1 | table | postgres
  18. public | t_10 | table | postgres
  19. public | t_2 | table | postgres
  20. public | t_3 | table | postgres
  21. public | t_4 | table | postgres
  22. public | t_5 | table | postgres
  23. public | t_6 | table | postgres
  24. public | t_7 | table | postgres
  25. public | t_8 | table | postgres
  26. public | t_9 | table | postgres
  27. (10 rows)

将结果设置到COLUMN ALIAS的变量中

  1. postgres=# select 'abcd' as col1, 1 as col2
  2. postgres-# \gset
  3. postgres=# \echo :col1
  4. abcd
  5. postgres=# \echo :col2
  6. 1

列出已设置的PSQL变量

  1. postgres=# \set
  2. AUTOCOMMIT = 'on'
  3. COMP_KEYWORD_CASE = 'preserve-upper'
  4. DBNAME = 'postgres'
  5. ECHO = 'none'
  6. ECHO_HIDDEN = 'off'
  7. ENCODING = 'UTF8'
  8. ERROR = 'false'
  9. FETCH_COUNT = '0'
  10. HIDE_TABLEAM = 'off'
  11. HISTCONTROL = 'none'
  12. HISTSIZE = '500'
  13. HOST = '/data01/pg12/pg_root12000'
  14. IGNOREEOF = '0'
  15. LASTOID = '0'
  16. LAST_ERROR_MESSAGE = ''
  17. LAST_ERROR_SQLSTATE = '00000'
  18. ON_ERROR_ROLLBACK = 'off'
  19. ON_ERROR_STOP = 'off'
  20. PORT = '12000'
  21. PROMPT1 = '%/%R%# '
  22. PROMPT2 = '%/%R%# '
  23. PROMPT3 = '>> '
  24. QUIET = 'off'
  25. ROW_COUNT = '1'
  26. SERVER_VERSION_NAME = '12beta1'
  27. SERVER_VERSION_NUM = '120000'
  28. SHOW_CONTEXT = 'errors'
  29. SINGLELINE = 'off'
  30. SINGLESTEP = 'off'
  31. SQLSTATE = '00000'
  32. USER = 'postgres'
  33. VERBOSITY = 'default'
  34. VERSION = 'PostgreSQL 12beta1 on x86_64-pc-linux-gnu, compiled by gcc (GCC) 4.8.5 20150623 (Red Hat 4.8.5-36), 64-bit'
  35. VERSION_NAME = '12beta1'
  36. VERSION_NUM = '120000'
  37. col1 = 'abcd'
  38. col2 = '1'

扩展方式返回结果

  1. \x
  2. query;
  3. \gx
  1. postgres=# \p
  2. select 'abcd' as col1, 1 as col2
  3. postgres=# \g
  4. col1 | col2
  5. ------+------
  6. abcd | 1
  7. (1 row)
  8. postgres=# \gx
  9. -[ RECORD 1 ]
  10. col1 | abcd
  11. col2 | 1

7、每隔N秒执行一次query buffer中的SQL

  1. postgres=# \p
  2. select 'abcd' as col1, 1 as col2
  3. postgres=# \watch 1
  4. Sat 08 Jun 2019 10:33:55 AM CST (every 1s)
  5. col1 | col2
  6. ------+------
  7. abcd | 1
  8. (1 row)
  9. Sat 08 Jun 2019 10:33:56 AM CST (every 1s)
  10. col1 | col2
  11. ------+------
  12. abcd | 1
  13. (1 row)

8、条件判定逻辑

gset执行并设置变量到column alias

  1. -- check for the existence of two separate records in the database and store
  2. -- the results in separate psql variables
  3. SELECT
  4. EXISTS(SELECT 1 FROM customer WHERE customer_id = 123) as is_customer,
  5. EXISTS(SELECT 1 FROM employee WHERE employee_id = 456) as is_employee
  6. \gset
  7. \if :is_customer
  8. SELECT * FROM customer WHERE customer_id = 123;
  9. \elif :is_employee
  10. \echo 'is not a customer but is an employee'
  11. SELECT * FROM employee WHERE employee_id = 456;
  12. \else
  13. \if yes
  14. \echo 'not a customer or employee'
  15. \else
  16. \echo 'this will never print'
  17. \endif
  18. \endif

9、大对象操作

  1. postgres=# \lo_import '/tmp/abcd'
  2. could not open file "/tmp/abcd": No such file or directory
  3. postgres=# \lo_import '/tmp/abc'
  4. lo_import 16516
  5. postgres=# \lo_list
  6. Large objects
  7. ID | Owner | Description
  8. -------+----------+-------------
  9. 16516 | postgres |
  10. (1 row)
  11. postgres=# \lo_export 16516 '/tmp/copy_abc'
  12. lo_export
  13. postgres=# \! diff '/tmp/copy_abc' '/tmp/abc'
  14. postgres=# \! head -n 2 '/tmp/copy_abc'
  15. create table h (id int, info text0;
  16. ;
  17. postgres=# \! head -n 2 '/tmp/abc'
  18. create table h (id int, info text0;
  19. ;
  20. postgres=# select * from pg_largeobject_metadata;
  21. oid | lomowner | lomacl
  22. -------+----------+--------
  23. 16516 | 10 |
  24. (1 row)

清理大对象

  1. postgres=# \lo_unlink 16516
  2. lo_unlink 16516
  3. postgres=# \lo_list
  4. Large objects
  5. ID | Owner | Description
  6. ----+-------+-------------
  7. (0 rows)
  8. postgres=# select * from pg_largeobject_metadata ;
  9. oid | lomowner | lomacl
  10. -----+----------+--------
  11. (0 rows)
  12. postgres=# select * from pg_largeobject;
  13. loid | pageno | data
  14. ------+--------+------
  15. (0 rows)

10、COPY PSQL的客户端服务器上的文件到数据库,或COPY数据库数据(表或者SQL)的结果到执行PSQL的客户端。

《PostgreSQL 服务端COPY和客户端COPY - 暨PG有哪些服务端操作接口》

《PostgreSQL copy (quote,DELIMITER,…) single byte char 的输入》

11、连接其他数据库

  1. postgres=# create role abc login;
  2. CREATE ROLE
  3. postgres=# \c 'hostaddr=127.0.0.1 dbname=postgres user=abc port=12000'
  4. You are now connected to database "postgres" as user "abc".
  5. 连接当前实例的其他DBUSER,可以简写
  6. \c dbname username
  7. 例如
  8. postgres=> \c postgres postgres
  9. You are now connected to database "postgres" as user "postgres".

12、列出当前连接信息

  1. postgres=# \conninfo
  2. You are connected to database "postgres" as user "postgres" via socket in "/data01/pg12/pg_root12000" at port "12000".

13、列出对象信息、权限

S包括系统对象

  1. Informational
  2. (options: S = show system objects, + = additional detail)
  3. \d[S+] list tables, views, and sequences
  4. \d[S+] NAME describe table, view, sequence, or index
  5. \da[S] [PATTERN] list aggregates
  6. \dA[+] [PATTERN] list access methods
  7. \db[+] [PATTERN] list tablespaces
  8. \dc[S+] [PATTERN] list conversions
  9. \dC[+] [PATTERN] list casts
  10. \dd[S] [PATTERN] show object descriptions not displayed elsewhere
  11. \dD[S+] [PATTERN] list domains
  12. \ddp [PATTERN] list default privileges
  13. \dE[S+] [PATTERN] list foreign tables
  14. \det[+] [PATTERN] list foreign tables
  15. \des[+] [PATTERN] list foreign servers
  16. \deu[+] [PATTERN] list user mappings
  17. \dew[+] [PATTERN] list foreign-data wrappers
  18. \df[anptw][S+] [PATRN] list [only agg/normal/procedures/trigger/window] functions
  19. \dF[+] [PATTERN] list text search configurations
  20. \dFd[+] [PATTERN] list text search dictionaries
  21. \dFp[+] [PATTERN] list text search parsers
  22. \dFt[+] [PATTERN] list text search templates
  23. \dg[S+] [PATTERN] list roles
  24. \di[S+] [PATTERN] list indexes
  25. \dl list large objects, same as \lo_list
  26. \dL[S+] [PATTERN] list procedural languages
  27. \dm[S+] [PATTERN] list materialized views
  28. \dn[S+] [PATTERN] list schemas
  29. \do[S] [PATTERN] list operators
  30. \dO[S+] [PATTERN] list collations
  31. \dp [PATTERN] list table, view, and sequence access privileges
  32. \dP[tin+] [PATTERN] list [only table/index] partitioned relations
  33. \drds [PATRN1 [PATRN2]] list per-database role settings
  34. \dRp[+] [PATTERN] list replication publications
  35. \dRs[+] [PATTERN] list replication subscriptions
  36. \ds[S+] [PATTERN] list sequences
  37. \dt[S+] [PATTERN] list tables
  38. \dT[S+] [PATTERN] list data types
  39. \du[S+] [PATTERN] list roles
  40. \dv[S+] [PATTERN] list views
  41. \dx[+] [PATTERN] list extensions
  42. \dy [PATTERN] list event triggers
  43. \l[+] [PATTERN] list databases
  44. \sf[+] FUNCNAME show a function's definition
  45. \sv[+] VIEWNAME show a view's definition
  46. \z [PATTERN] same as \dp

14、执行psql服务器上的文件

  1. \i FILE execute commands from file
  2. \ir FILE as \i, but relative to location of current script
  1. postgres=# \a
  2. Output format is unaligned.
  3. postgres=# \t on
  4. postgres=# select format('create table abc_%s (id int);', i) from generate_series(1,10) i
  5. postgres-# \g '/tmp/abcd'
  6. postgres=# \! cat '/tmp/abcd'
  7. create table abc_1 (id int);
  8. create table abc_2 (id int);
  9. create table abc_3 (id int);
  10. create table abc_4 (id int);
  11. create table abc_5 (id int);
  12. create table abc_6 (id int);
  13. create table abc_7 (id int);
  14. create table abc_8 (id int);
  15. create table abc_9 (id int);
  16. create table abc_10 (id int);
  1. postgres=# \i '/tmp/abcd'
  2. CREATE TABLE
  3. CREATE TABLE
  4. CREATE TABLE
  5. CREATE TABLE
  6. CREATE TABLE
  7. CREATE TABLE
  8. CREATE TABLE
  9. CREATE TABLE
  10. CREATE TABLE
  11. CREATE TABLE
  12. postgres=# \dt
  13. public|abc_1|table|postgres
  14. public|abc_10|table|postgres
  15. public|abc_2|table|postgres
  16. public|abc_3|table|postgres
  17. public|abc_4|table|postgres
  18. public|abc_5|table|postgres
  19. public|abc_6|table|postgres
  20. public|abc_7|table|postgres
  21. public|abc_8|table|postgres
  22. public|abc_9|table|postgres

小结

psql是非常强大的PostgreSQL客户端,可以极大提高数据库管理人员的管理便捷性。

参考

man psql