1PSQL(1) PostgreSQL 13.4 Documentation PSQL(1)
2
6 psql - PostgreSQL interactive terminal
7
9 psql [option...] [dbname [username]]
10
12 psql is a terminal-based front-end to PostgreSQL. It enables you to
13 type in queries interactively, issue them to PostgreSQL, and see the
14 query results. Alternatively, input can be from a file or from command
15 line arguments. In addition, psql provides a number of meta-commands
16 and various shell-like features to facilitate writing scripts and
17 automating a wide variety of tasks.
18
20 -a
21 --echo-all
22 Print all nonempty input lines to standard output as they are read.
23 (This does not apply to lines read interactively.) This is
24 equivalent to setting the variable ECHO to all.
25 -A
27 --no-align
28 Switches to unaligned output mode. (The default output mode is
29 aligned.) This is equivalent to \pset format unaligned.
30 -b
32 --echo-errors
33 Print failed SQL commands to standard error output. This is
34 equivalent to setting the variable ECHO to errors.
35 -c command
37 --command=command
38 Specifies that psql is to execute the given command string,
39 command. This option can be repeated and combined in any order with
40 the -f option. When either -c or -f is specified, psql does not
41 read commands from standard input; instead it terminates after
42 processing all the -c and -f options in sequence.
43 command must be either a command string that is completely parsable
45 by the server (i.e., it contains no psql-specific features), or a
46 single backslash command. Thus you cannot mix SQL and psql
47 meta-commands within a -c option. To achieve that, you could use
48 repeated -c options or pipe the string into psql, for example:
49 psql -c '\x' -c 'SELECT * FROM foo;'
51 or
53 echo '\x \\ SELECT * FROM foo;' | psql
55 (\\ is the separator meta-command.)
57 Each SQL command string passed to -c is sent to the server as a
59 single request. Because of this, the server executes it as a single
60 transaction even if the string contains multiple SQL commands,
61 unless there are explicit BEGIN/COMMIT commands included in the
62 string to divide it into multiple transactions. (See
63 Section 52.2.2.1 for more details about how the server handles
64 multi-query strings.) Also, psql only prints the result of the last
65 SQL command in the string. This is different from the behavior when
66 the same string is read from a file or fed to psql's standard
67 input, because then psql sends each SQL command separately.
68 Because of this behavior, putting more than one SQL command in a
70 single -c string often has unexpected results. It's better to use
71 repeated -c commands or feed multiple commands to psql's standard
72 input, either using echo as illustrated above, or via a shell
73 here-document, for example:
74 psql <<EOF
76 \x
77 SELECT * FROM foo;
78 EOF
79 --csv
81 Switches to CSV (Comma-Separated Values) output mode. This is
82 equivalent to \pset format csv.
83 -d dbname
85 --dbname=dbname
86 Specifies the name of the database to connect to. This is
87 equivalent to specifying dbname as the first non-option argument on
88 the command line. The dbname can be a connection string. If so,
89 connection string parameters will override any conflicting command
90 line options.
91 -e
93 --echo-queries
94 Copy all SQL commands sent to the server to standard output as
95 well. This is equivalent to setting the variable ECHO to queries.
96 -E
98 --echo-hidden
99 Echo the actual queries generated by \d and other backslash
100 commands. You can use this to study psql's internal operations.
101 This is equivalent to setting the variable ECHO_HIDDEN to on.
102 -f filename
104 --file=filename
105 Read commands from the file filename, rather than standard input.
106 This option can be repeated and combined in any order with the -c
107 option. When either -c or -f is specified, psql does not read
108 commands from standard input; instead it terminates after
109 processing all the -c and -f options in sequence. Except for that,
110 this option is largely equivalent to the meta-command \i.
111 If filename is - (hyphen), then standard input is read until an EOF
113 indication or \q meta-command. This can be used to intersperse
114 interactive input with input from files. Note however that Readline
115 is not used in this case (much as if -n had been specified).
116 Using this option is subtly different from writing psql < filename.
118 In general, both will do what you expect, but using -f enables some
119 nice features such as error messages with line numbers. There is
120 also a slight chance that using this option will reduce the
121 start-up overhead. On the other hand, the variant using the shell's
122 input redirection is (in theory) guaranteed to yield exactly the
123 same output you would have received had you entered everything by
124 hand.
125 -F separator
127 --field-separator=separator
128 Use separator as the field separator for unaligned output. This is
129 equivalent to \pset fieldsep or \f.
130 -h hostname
132 --host=hostname
133 Specifies the host name of the machine on which the server is
134 running. If the value begins with a slash, it is used as the
135 directory for the Unix-domain socket.
136 -H
138 --html
139 Switches to HTML output mode. This is equivalent to \pset format
140 html or the \H command.
141 -l
143 --list
144 List all available databases, then exit. Other non-connection
145 options are ignored. This is similar to the meta-command \list.
146 When this option is used, psql will connect to the database
148 postgres, unless a different database is named on the command line
149 (option -d or non-option argument, possibly via a service entry,
150 but not via an environment variable).
151 -L filename
153 --log-file=filename
154 Write all query output into file filename, in addition to the
155 normal output destination.
156 -n
158 --no-readline
159 Do not use Readline for line editing and do not use the command
160 history. This can be useful to turn off tab expansion when cutting
161 and pasting.
162 -o filename
164 --output=filename
165 Put all query output into file filename. This is equivalent to the
166 command \o.
167 -p port
169 --port=port
170 Specifies the TCP port or the local Unix-domain socket file
171 extension on which the server is listening for connections.
172 Defaults to the value of the PGPORT environment variable or, if not
173 set, to the port specified at compile time, usually 5432.
174 -P assignment
176 --pset=assignment
177 Specifies printing options, in the style of \pset. Note that here
178 you have to separate name and value with an equal sign instead of a
179 space. For example, to set the output format to LaTeX, you could
180 write -P format=latex.
181 -q
183 --quiet
184 Specifies that psql should do its work quietly. By default, it
185 prints welcome messages and various informational output. If this
186 option is used, none of this happens. This is useful with the -c
187 option. This is equivalent to setting the variable QUIET to on.
188 -R separator
190 --record-separator=separator
191 Use separator as the record separator for unaligned output. This is
192 equivalent to \pset recordsep.
193 -s
195 --single-step
196 Run in single-step mode. That means the user is prompted before
197 each command is sent to the server, with the option to cancel
198 execution as well. Use this to debug scripts.
199 -S
201 --single-line
202 Runs in single-line mode where a newline terminates an SQL command,
203 as a semicolon does.
204 Note
206 This mode is provided for those who insist on it, but you are
207 not necessarily encouraged to use it. In particular, if you mix
208 SQL and meta-commands on a line the order of execution might
209 not always be clear to the inexperienced user.
210 -t
212 --tuples-only
213 Turn off printing of column names and result row count footers,
214 etc. This is equivalent to \t or \pset tuples_only.
215 -T table_options
217 --table-attr=table_options
218 Specifies options to be placed within the HTML table tag. See \pset
219 tableattr for details.
220 -U username
222 --username=username
223 Connect to the database as the user username instead of the
224 default. (You must have permission to do so, of course.)
225 -v assignment
227 --set=assignment
228 --variable=assignment
229 Perform a variable assignment, like the \set meta-command. Note
230 that you must separate name and value, if any, by an equal sign on
231 the command line. To unset a variable, leave off the equal sign. To
232 set a variable with an empty value, use the equal sign but leave
233 off the value. These assignments are done during command line
234 processing, so variables that reflect connection state will get
235 overwritten later.
236 -V
238 --version
239 Print the psql version and exit.
240 -w
242 --no-password
243 Never issue a password prompt. If the server requires password
244 authentication and a password is not available from other sources
245 such as a .pgpass file, the connection attempt will fail. This
246 option can be useful in batch jobs and scripts where no user is
247 present to enter a password.
248 Note that this option will remain set for the entire session, and
250 so it affects uses of the meta-command \connect as well as the
251 initial connection attempt.
252 -W
254 --password
255 Force psql to prompt for a password before connecting to a
256 database, even if the password will not be used.
257 If the server requires password authentication and a password is
259 not available from other sources such as a .pgpass file, psql will
260 prompt for a password in any case. However, psql will waste a
261 connection attempt finding out that the server wants a password. In
262 some cases it is worth typing -W to avoid the extra connection
263 attempt.
264 Note that this option will remain set for the entire session, and
266 so it affects uses of the meta-command \connect as well as the
267 initial connection attempt.
268 -x
270 --expanded
271 Turn on the expanded table formatting mode. This is equivalent to
272 \x or \pset expanded.
273 -X,
275 --no-psqlrc
276 Do not read the start-up file (neither the system-wide psqlrc file
277 nor the user's ~/.psqlrc file).
278 -z
280 --field-separator-zero
281 Set the field separator for unaligned output to a zero byte. This
282 is equivalent to \pset fieldsep_zero.
283 -0
285 --record-separator-zero
286 Set the record separator for unaligned output to a zero byte. This
287 is useful for interfacing, for example, with xargs -0. This is
288 equivalent to \pset recordsep_zero.
289 -1
291 --single-transaction
292 This option can only be used in combination with one or more -c
293 and/or -f options. It causes psql to issue a BEGIN command before
294 the first such option and a COMMIT command after the last one,
295 thereby wrapping all the commands into a single transaction. This
296 ensures that either all the commands complete successfully, or no
297 changes are applied.
298 If the commands themselves contain BEGIN, COMMIT, or ROLLBACK, this
300 option will not have the desired effects. Also, if an individual
301 command cannot be executed inside a transaction block, specifying
302 this option will cause the whole transaction to fail.
303 -?
305 --help[=topic]
306 Show help about psql and exit. The optional topic parameter
307 (defaulting to options) selects which part of psql is explained:
308 commands describes psql's backslash commands; options describes the
309 command-line options that can be passed to psql; and variables
310 shows help about psql configuration variables.
311
313 psql returns 0 to the shell if it finished normally, 1 if a fatal error
314 of its own occurs (e.g., out of memory, file not found), 2 if the
315 connection to the server went bad and the session was not interactive,
316 and 3 if an error occurred in a script and the variable ON_ERROR_STOP
317 was set.
318
320 Connecting to a Database
321 psql is a regular PostgreSQL client application. In order to connect to
322 a database you need to know the name of your target database, the host
323 name and port number of the server, and what user name you want to
324 connect as. psql can be told about those parameters via command line
325 options, namely -d, -h, -p, and -U respectively. If an argument is
326 found that does not belong to any option it will be interpreted as the
327 database name (or the user name, if the database name is already
328 given). Not all of these options are required; there are useful
329 defaults. If you omit the host name, psql will connect via a
330 Unix-domain socket to a server on the local host, or via TCP/IP to
331 localhost on machines that don't have Unix-domain sockets. The default
332 port number is determined at compile time. Since the database server
333 uses the same default, you will not have to specify the port in most
334 cases. The default user name is your operating-system user name, as is
335 the default database name. Note that you cannot just connect to any
336 database under any user name. Your database administrator should have
337 informed you about your access rights.
338 When the defaults aren't quite right, you can save yourself some typing
340 by setting the environment variables PGDATABASE, PGHOST, PGPORT and/or
341 PGUSER to appropriate values. (For additional environment variables,
342 see Section 33.14.) It is also convenient to have a ~/.pgpass file to
343 avoid regularly having to type in passwords. See Section 33.15 for more
344 information.
345 An alternative way to specify connection parameters is in a conninfo
347 string or a URI, which is used instead of a database name. This
348 mechanism give you very wide control over the connection. For example:
349 $ psql "service=myservice sslmode=require"
351 $ psql postgresql://dbmaster:5433/mydb?sslmode=require
352 This way you can also use LDAP for connection parameter lookup as
354 described in Section 33.17. See Section 33.1.2 for more information on
355 all the available connection options.
356 If the connection could not be made for any reason (e.g., insufficient
358 privileges, server is not running on the targeted host, etc.), psql
359 will return an error and terminate.
360 If both standard input and standard output are a terminal, then psql
362 sets the client encoding to “auto”, which will detect the appropriate
363 client encoding from the locale settings (LC_CTYPE environment variable
364 on Unix systems). If this doesn't work out as expected, the client
365 encoding can be overridden using the environment variable
366 PGCLIENTENCODING.
367 Entering SQL Commands
369 In normal operation, psql provides a prompt with the name of the
370 database to which psql is currently connected, followed by the string
371 =>. For example:
372 $ psql testdb
374 psql (13.4)
375 Type "help" for help.
376 testdb=>
378 At the prompt, the user can type in SQL commands. Ordinarily, input
380 lines are sent to the server when a command-terminating semicolon is
381 reached. An end of line does not terminate a command. Thus commands can
382 be spread over several lines for clarity. If the command was sent and
383 executed without error, the results of the command are displayed on the
384 screen.
385 If untrusted users have access to a database that has not adopted a
387 secure schema usage pattern, begin your session by removing
388 publicly-writable schemas from search_path. One can add
389 options=-csearch_path= to the connection string or issue SELECT
390 pg_catalog.set_config('search_path', '', false) before other SQL
391 commands. This consideration is not specific to psql; it applies to
392 every interface for executing arbitrary SQL commands.
393 Whenever a command is executed, psql also polls for asynchronous
395 notification events generated by LISTEN(7) and NOTIFY(7).
396 While C-style block comments are passed to the server for processing
398 and removal, SQL-standard comments are removed by psql.
399 Meta-Commands
401 Anything you enter in psql that begins with an unquoted backslash is a
402 psql meta-command that is processed by psql itself. These commands make
403 psql more useful for administration or scripting. Meta-commands are
404 often called slash or backslash commands.
405 The format of a psql command is the backslash, followed immediately by
407 a command verb, then any arguments. The arguments are separated from
408 the command verb and each other by any number of whitespace characters.
409 To include whitespace in an argument you can quote it with single
411 quotes. To include a single quote in an argument, write two single
412 quotes within single-quoted text. Anything contained in single quotes
413 is furthermore subject to C-like substitutions for \n (new line), \t
414 (tab), \b (backspace), \r (carriage return), \f (form feed), \digits
415 (octal), and \xdigits (hexadecimal). A backslash preceding any other
416 character within single-quoted text quotes that single character,
417 whatever it is.
418 If an unquoted colon (:) followed by a psql variable name appears
420 within an argument, it is replaced by the variable's value, as
421 described in SQL Interpolation below. The forms :'variable_name' and
422 :"variable_name" described there work as well. The :{?variable_name}
423 syntax allows testing whether a variable is defined. It is substituted
424 by TRUE or FALSE. Escaping the colon with a backslash protects it from
425 substitution.
426 Within an argument, text that is enclosed in backquotes (`) is taken as
428 a command line that is passed to the shell. The output of the command
429 (with any trailing newline removed) replaces the backquoted text.
430 Within the text enclosed in backquotes, no special quoting or other
431 processing occurs, except that appearances of :variable_name where
432 variable_name is a psql variable name are replaced by the variable's
433 value. Also, appearances of :'variable_name' are replaced by the
434 variable's value suitably quoted to become a single shell command
435 argument. (The latter form is almost always preferable, unless you are
436 very sure of what is in the variable.) Because carriage return and line
437 feed characters cannot be safely quoted on all platforms, the
438 :'variable_name' form prints an error message and does not substitute
439 the variable value when such characters appear in the value.
440 Some commands take an SQL identifier (such as a table name) as
442 argument. These arguments follow the syntax rules of SQL: Unquoted
443 letters are forced to lowercase, while double quotes (") protect
444 letters from case conversion and allow incorporation of whitespace into
445 the identifier. Within double quotes, paired double quotes reduce to a
446 single double quote in the resulting name. For example, FOO"BAR"BAZ is
447 interpreted as fooBARbaz, and "A weird"" name" becomes A weird" name.
448 Parsing for arguments stops at the end of the line, or when another
450 unquoted backslash is found. An unquoted backslash is taken as the
451 beginning of a new meta-command. The special sequence \\ (two
452 backslashes) marks the end of arguments and continues parsing SQL
453 commands, if any. That way SQL and psql commands can be freely mixed on
454 a line. But in any case, the arguments of a meta-command cannot
455 continue beyond the end of the line.
456 Many of the meta-commands act on the current query buffer. This is
458 simply a buffer holding whatever SQL command text has been typed but
459 not yet sent to the server for execution. This will include previous
460 input lines as well as any text appearing before the meta-command on
461 the same line.
462 The following meta-commands are defined:
464 \a
466 If the current table output format is unaligned, it is switched to
467 aligned. If it is not unaligned, it is set to unaligned. This
468 command is kept for backwards compatibility. See \pset for a more
469 general solution.
470 \c or \connect [ -reuse-previous=on|off ] [ dbname [ username ] [ host
472 ] [ port ] | conninfo ]
473 Establishes a new connection to a PostgreSQL server. The connection
474 parameters to use can be specified either using a positional syntax
475 (one or more of database name, user, host, and port), or using a
476 conninfo connection string as detailed in Section 33.1.1. If no
477 arguments are given, a new connection is made using the same
478 parameters as before.
479 Specifying any of dbname, username, host or port as - is equivalent
481 to omitting that parameter.
482 The new connection can re-use connection parameters from the
484 previous connection; not only database name, user, host, and port,
485 but other settings such as sslmode. By default, parameters are
486 re-used in the positional syntax, but not when a conninfo string is
487 given. Passing a first argument of -reuse-previous=on or
488 -reuse-previous=off overrides that default. If parameters are
489 re-used, then any parameter not explicitly specified as a
490 positional parameter or in the conninfo string is taken from the
491 existing connection's parameters. An exception is that if the host
492 setting is changed from its previous value using the positional
493 syntax, any hostaddr setting present in the existing connection's
494 parameters is dropped. Also, any password used for the existing
495 connection will be re-used only if the user, host, and port
496 settings are not changed. When the command neither specifies nor
497 reuses a particular parameter, the libpq default is used.
498 If the new connection is successfully made, the previous connection
500 is closed. If the connection attempt fails (wrong user name, access
501 denied, etc.), the previous connection will be kept if psql is in
502 interactive mode. But when executing a non-interactive script,
503 processing will immediately stop with an error. This distinction
504 was chosen as a user convenience against typos on the one hand, and
505 a safety mechanism that scripts are not accidentally acting on the
506 wrong database on the other hand.
507 Examples:
509 => \c mydb myuser host.dom 6432
511 => \c service=foo
512 => \c "host=localhost port=5432 dbname=mydb connect_timeout=10 sslmode=disable"
513 => \c -reuse-previous=on sslmode=require -- changes only sslmode
514 => \c postgresql://tom@localhost/mydb?application_name=myapp
515 \C [ title ]
517 Sets the title of any tables being printed as the result of a query
518 or unset any such title. This command is equivalent to \pset title
519 title. (The name of this command derives from “caption”, as it was
520 previously only used to set the caption in an HTML table.)
521 \cd [ directory ]
523 Changes the current working directory to directory. Without
524 argument, changes to the current user's home directory.
525 Tip
527 To print your current working directory, use \! pwd.
528 \conninfo
530 Outputs information about the current database connection.
531 \copy { table [ ( column_list ) ] } from { 'filename' | program
533 'command' | stdin | pstdin } [ [ with ] ( option [, ...] ) ] [ where
534 condition ]
535 \copy { table [ ( column_list ) ] | ( query ) } to { 'filename' |
536 program 'command' | stdout | pstdout } [ [ with ] ( option [, ...] ) ]
537 Performs a frontend (client) copy. This is an operation that runs
538 an SQL COPY(7) command, but instead of the server reading or
539 writing the specified file, psql reads or writes the file and
540 routes the data between the server and the local file system. This
541 means that file accessibility and privileges are those of the local
542 user, not the server, and no SQL superuser privileges are required.
543 When program is specified, command is executed by psql and the data
545 passed from or to command is routed between the server and the
546 client. Again, the execution privileges are those of the local
547 user, not the server, and no SQL superuser privileges are required.
548 For \copy ... from stdin, data rows are read from the same source
550 that issued the command, continuing until \. is read or the stream
551 reaches EOF. This option is useful for populating tables in-line
552 within a SQL script file. For \copy ... to stdout, output is sent
553 to the same place as psql command output, and the COPY count
554 command status is not printed (since it might be confused with a
555 data row). To read/write psql's standard input or output regardless
556 of the current command source or \o option, write from pstdin or to
557 pstdout.
558 The syntax of this command is similar to that of the SQL COPY(7)
560 command. All options other than the data source/destination are as
561 specified for COPY(7). Because of this, special parsing rules apply
562 to the \copy meta-command. Unlike most other meta-commands, the
563 entire remainder of the line is always taken to be the arguments of
564 \copy, and neither variable interpolation nor backquote expansion
565 are performed in the arguments.
566 Tip
568 Another way to obtain the same result as \copy ... to is to use
569 the SQL COPY ... TO STDOUT command and terminate it with \g
570 filename or \g |program. Unlike \copy, this method allows the
571 command to span multiple lines; also, variable interpolation
572 and backquote expansion can be used.
573 Tip
575 These operations are not as efficient as the SQL COPY command
576 with a file or program data source or destination, because all
577 data must pass through the client/server connection. For large
578 amounts of data the SQL command might be preferable.
579 \copyright
581 Shows the copyright and distribution terms of PostgreSQL.
582 \crosstabview [ colV [ colH [ colD [ sortcolH ] ] ] ]
584 Executes the current query buffer (like \g) and shows the results
585 in a crosstab grid. The query must return at least three columns.
586 The output column identified by colV becomes a vertical header and
587 the output column identified by colH becomes a horizontal header.
588 colD identifies the output column to display within the grid.
589 sortcolH identifies an optional sort column for the horizontal
590 header.
591 Each column specification can be a column number (starting at 1) or
593 a column name. The usual SQL case folding and quoting rules apply
594 to column names. If omitted, colV is taken as column 1 and colH as
595 column 2. colH must differ from colV. If colD is not specified,
596 then there must be exactly three columns in the query result, and
597 the column that is neither colV nor colH is taken to be colD.
598 The vertical header, displayed as the leftmost column, contains the
600 values found in column colV, in the same order as in the query
601 results, but with duplicates removed.
602 The horizontal header, displayed as the first row, contains the
604 values found in column colH, with duplicates removed. By default,
605 these appear in the same order as in the query results. But if the
606 optional sortcolH argument is given, it identifies a column whose
607 values must be integer numbers, and the values from colH will
608 appear in the horizontal header sorted according to the
609 corresponding sortcolH values.
610 Inside the crosstab grid, for each distinct value x of colH and
612 each distinct value y of colV, the cell located at the intersection
613 (x,y) contains the value of the colD column in the query result row
614 for which the value of colH is x and the value of colV is y. If
615 there is no such row, the cell is empty. If there are multiple such
616 rows, an error is reported.
617 \d[S+] [ pattern ]
619 For each relation (table, view, materialized view, index, sequence,
620 or foreign table) or composite type matching the pattern, show all
621 columns, their types, the tablespace (if not the default) and any
622 special attributes such as NOT NULL or defaults. Associated
623 indexes, constraints, rules, and triggers are also shown. For
624 foreign tables, the associated foreign server is shown as well.
625 (“Matching the pattern” is defined in Patterns below.)
626 For some types of relation, \d shows additional information for
628 each column: column values for sequences, indexed expressions for
629 indexes, and foreign data wrapper options for foreign tables.
630 The command form \d+ is identical, except that more information is
632 displayed: any comments associated with the columns of the table
633 are shown, as is the presence of OIDs in the table, the view
634 definition if the relation is a view, a non-default replica
635 identity setting.
636 By default, only user-created objects are shown; supply a pattern
638 or the S modifier to include system objects.
639 Note
641 If \d is used without a pattern argument, it is equivalent to
642 \dtvmsE which will show a list of all visible tables, views,
643 materialized views, sequences and foreign tables. This is
644 purely a convenience measure.
645 \da[S] [ pattern ]
647 Lists aggregate functions, together with their return type and the
648 data types they operate on. If pattern is specified, only
649 aggregates whose names match the pattern are shown. By default,
650 only user-created objects are shown; supply a pattern or the S
651 modifier to include system objects.
652 \dA[+] [ pattern ]
654 Lists access methods. If pattern is specified, only access methods
655 whose names match the pattern are shown. If + is appended to the
656 command name, each access method is listed with its associated
657 handler function and description.
658 \dAc[+] [access-method-pattern [input-type-pattern]]
660 Lists operator classes (see Section 37.16.1). If
661 access-method-pattern is specified, only operator classes
662 associated with access methods whose names match that pattern are
663 listed. If input-type-pattern is specified, only operator classes
664 associated with input types whose names match that pattern are
665 listed. If + is appended to the command name, each operator class
666 is listed with its associated operator family and owner.
667 \dAf[+] [access-method-pattern [input-type-pattern]]
669 Lists operator families (see Section 37.16.5). If
670 access-method-pattern is specified, only operator families
671 associated with access methods whose names match that pattern are
672 listed. If input-type-pattern is specified, only operator families
673 associated with input types whose names match that pattern are
674 listed. If + is appended to the command name, each operator family
675 is listed with its owner.
676 \dAo[+] [access-method-pattern [operator-family-pattern]]
678 Lists operators associated with operator families (see
679 Section 37.16.2). If access-method-pattern is specified, only
680 members of operator families associated with access methods whose
681 names match that pattern are listed. If operator-family-pattern is
682 specified, only members of operator families whose names match that
683 pattern are listed. If + is appended to the command name, each
684 operator is listed with its sort operator family (if it is an
685 ordering operator).
686 \dAp[+] [access-method-pattern [operator-family-pattern]]
688 Lists support functions associated with operator families (see
689 Section 37.16.3). If access-method-pattern is specified, only
690 functions of operator families associated with access methods whose
691 names match that pattern are listed. If operator-family-pattern is
692 specified, only functions of operator families whose names match
693 that pattern are listed. If + is appended to the command name,
694 functions are displayed verbosely, with their actual parameter
695 lists.
696 \db[+] [ pattern ]
698 Lists tablespaces. If pattern is specified, only tablespaces whose
699 names match the pattern are shown. If + is appended to the command
700 name, each tablespace is listed with its associated options,
701 on-disk size, permissions and description.
702 \dc[S+] [ pattern ]
704 Lists conversions between character-set encodings. If pattern is
705 specified, only conversions whose names match the pattern are
706 listed. By default, only user-created objects are shown; supply a
707 pattern or the S modifier to include system objects. If + is
708 appended to the command name, each object is listed with its
709 associated description.
710 \dC[+] [ pattern ]
712 Lists type casts. If pattern is specified, only casts whose source
713 or target types match the pattern are listed. If + is appended to
714 the command name, each object is listed with its associated
715 description.
716 \dd[S] [ pattern ]
718 Shows the descriptions of objects of type constraint, operator
719 class, operator family, rule, and trigger. All other comments may
720 be viewed by the respective backslash commands for those object
721 types.
722 \dd displays descriptions for objects matching the pattern, or of
724 visible objects of the appropriate type if no argument is given.
725 But in either case, only objects that have a description are
726 listed. By default, only user-created objects are shown; supply a
727 pattern or the S modifier to include system objects.
728 Descriptions for objects can be created with the COMMENT(7) SQL
730 command.
731 \dD[S+] [ pattern ]
733 Lists domains. If pattern is specified, only domains whose names
734 match the pattern are shown. By default, only user-created objects
735 are shown; supply a pattern or the S modifier to include system
736 objects. If + is appended to the command name, each object is
737 listed with its associated permissions and description.
738 \ddp [ pattern ]
740 Lists default access privilege settings. An entry is shown for each
741 role (and schema, if applicable) for which the default privilege
742 settings have been changed from the built-in defaults. If pattern
743 is specified, only entries whose role name or schema name matches
744 the pattern are listed.
745 The ALTER DEFAULT PRIVILEGES (ALTER_DEFAULT_PRIVILEGES(7)) command
747 is used to set default access privileges. The meaning of the
748 privilege display is explained in Section 5.7.
749 \dE[S+] [ pattern ]
751 \di[S+] [ pattern ]
752 \dm[S+] [ pattern ]
753 \ds[S+] [ pattern ]
754 \dt[S+] [ pattern ]
755 \dv[S+] [ pattern ]
756 In this group of commands, the letters E, i, m, s, t, and v stand
757 for foreign table, index, materialized view, sequence, table, and
758 view, respectively. You can specify any or all of these letters, in
759 any order, to obtain a listing of objects of these types. For
760 example, \dti lists tables and indexes. If + is appended to the
761 command name, each object is listed with its persistence status
762 (permanent, temporary, or unlogged), physical size on disk, and
763 associated description if any. If pattern is specified, only
764 objects whose names match the pattern are listed. By default, only
765 user-created objects are shown; supply a pattern or the S modifier
766 to include system objects.
767 \des[+] [ pattern ]
769 Lists foreign servers (mnemonic: “external servers”). If pattern is
770 specified, only those servers whose name matches the pattern are
771 listed. If the form \des+ is used, a full description of each
772 server is shown, including the server's access privileges, type,
773 version, options, and description.
774 \det[+] [ pattern ]
776 Lists foreign tables (mnemonic: “external tables”). If pattern is
777 specified, only entries whose table name or schema name matches the
778 pattern are listed. If the form \det+ is used, generic options and
779 the foreign table description are also displayed.
780 \deu[+] [ pattern ]
782 Lists user mappings (mnemonic: “external users”). If pattern is
783 specified, only those mappings whose user names match the pattern
784 are listed. If the form \deu+ is used, additional information about
785 each mapping is shown.
786 Caution
788 \deu+ might also display the user name and password of the
789 remote user, so care should be taken not to disclose them.
790 \dew[+] [ pattern ]
792 Lists foreign-data wrappers (mnemonic: “external wrappers”). If
793 pattern is specified, only those foreign-data wrappers whose name
794 matches the pattern are listed. If the form \dew+ is used, the
795 access privileges, options, and description of the foreign-data
796 wrapper are also shown.
797 \df[anptwS+] [ pattern ]
799 Lists functions, together with their result data types, argument
800 data types, and function types, which are classified as “agg”
801 (aggregate), “normal”, “procedure”, “trigger”, or “window”. To
802 display only functions of specific type(s), add the corresponding
803 letters a, n, p, t, or w to the command. If pattern is specified,
804 only functions whose names match the pattern are shown. By default,
805 only user-created objects are shown; supply a pattern or the S
806 modifier to include system objects. If the form \df+ is used,
807 additional information about each function is shown, including
808 volatility, parallel safety, owner, security classification, access
809 privileges, language, source code and description.
810 Tip
812 To look up functions taking arguments or returning values of a
813 specific data type, use your pager's search capability to
814 scroll through the \df output.
815 \dF[+] [ pattern ]
817 Lists text search configurations. If pattern is specified, only
818 configurations whose names match the pattern are shown. If the form
819 \dF+ is used, a full description of each configuration is shown,
820 including the underlying text search parser and the dictionary list
821 for each parser token type.
822 \dFd[+] [ pattern ]
824 Lists text search dictionaries. If pattern is specified, only
825 dictionaries whose names match the pattern are shown. If the form
826 \dFd+ is used, additional information is shown about each selected
827 dictionary, including the underlying text search template and the
828 option values.
829 \dFp[+] [ pattern ]
831 Lists text search parsers. If pattern is specified, only parsers
832 whose names match the pattern are shown. If the form \dFp+ is used,
833 a full description of each parser is shown, including the
834 underlying functions and the list of recognized token types.
835 \dFt[+] [ pattern ]
837 Lists text search templates. If pattern is specified, only
838 templates whose names match the pattern are shown. If the form
839 \dFt+ is used, additional information is shown about each template,
840 including the underlying function names.
841 \dg[S+] [ pattern ]
843 Lists database roles. (Since the concepts of “users” and “groups”
844 have been unified into “roles”, this command is now equivalent to
845 \du.) By default, only user-created roles are shown; supply the S
846 modifier to include system roles. If pattern is specified, only
847 those roles whose names match the pattern are listed. If the form
848 \dg+ is used, additional information is shown about each role;
849 currently this adds the comment for each role.
850 \dl
852 This is an alias for \lo_list, which shows a list of large objects.
853 \dL[S+] [ pattern ]
855 Lists procedural languages. If pattern is specified, only languages
856 whose names match the pattern are listed. By default, only
857 user-created languages are shown; supply the S modifier to include
858 system objects. If + is appended to the command name, each language
859 is listed with its call handler, validator, access privileges, and
860 whether it is a system object.
861 \dn[S+] [ pattern ]
863 Lists schemas (namespaces). If pattern is specified, only schemas
864 whose names match the pattern are listed. By default, only
865 user-created objects are shown; supply a pattern or the S modifier
866 to include system objects. If + is appended to the command name,
867 each object is listed with its associated permissions and
868 description, if any.
869 \do[S+] [ pattern ]
871 Lists operators with their operand and result types. If pattern is
872 specified, only operators whose names match the pattern are listed.
873 By default, only user-created objects are shown; supply a pattern
874 or the S modifier to include system objects. If + is appended to
875 the command name, additional information about each operator is
876 shown, currently just the name of the underlying function.
877 \dO[S+] [ pattern ]
879 Lists collations. If pattern is specified, only collations whose
880 names match the pattern are listed. By default, only user-created
881 objects are shown; supply a pattern or the S modifier to include
882 system objects. If + is appended to the command name, each
883 collation is listed with its associated description, if any. Note
884 that only collations usable with the current database's encoding
885 are shown, so the results may vary in different databases of the
886 same installation.
887 \dp [ pattern ]
889 Lists tables, views and sequences with their associated access
890 privileges. If pattern is specified, only tables, views and
891 sequences whose names match the pattern are listed.
892 The GRANT(7) and REVOKE(7) commands are used to set access
894 privileges. The meaning of the privilege display is explained in
895 Section 5.7.
896 \dP[itn+] [ pattern ]
898 Lists partitioned relations. If pattern is specified, only entries
899 whose name matches the pattern are listed. The modifiers t (tables)
900 and i (indexes) can be appended to the command, filtering the kind
901 of relations to list. By default, partitioned tables and indexes
902 are listed.
903 If the modifier n (“nested”) is used, or a pattern is specified,
905 then non-root partitioned relations are included, and a column is
906 shown displaying the parent of each partitioned relation.
907 If + is appended to the command name, the sum of the sizes of each
909 relation's partitions is also displayed, along with the relation's
910 description. If n is combined with +, two sizes are shown: one
911 including the total size of directly-attached leaf partitions, and
912 another showing the total size of all partitions, including
913 indirectly attached sub-partitions.
914 \drds [ role-pattern [ database-pattern ] ]
916 Lists defined configuration settings. These settings can be
917 role-specific, database-specific, or both. role-pattern and
918 database-pattern are used to select specific roles and databases to
919 list, respectively. If omitted, or if * is specified, all settings
920 are listed, including those not role-specific or database-specific,
921 respectively.
922 The ALTER ROLE (ALTER_ROLE(7)) and ALTER DATABASE
924 (ALTER_DATABASE(7)) commands are used to define per-role and
925 per-database configuration settings.
926 \dRp[+] [ pattern ]
928 Lists replication publications. If pattern is specified, only those
929 publications whose names match the pattern are listed. If + is
930 appended to the command name, the tables associated with each
931 publication are shown as well.
932 \dRs[+] [ pattern ]
934 Lists replication subscriptions. If pattern is specified, only
935 those subscriptions whose names match the pattern are listed. If +
936 is appended to the command name, additional properties of the
937 subscriptions are shown.
938 \dT[S+] [ pattern ]
940 Lists data types. If pattern is specified, only types whose names
941 match the pattern are listed. If + is appended to the command name,
942 each type is listed with its internal name and size, its allowed
943 values if it is an enum type, and its associated permissions. By
944 default, only user-created objects are shown; supply a pattern or
945 the S modifier to include system objects.
946 \du[S+] [ pattern ]
948 Lists database roles. (Since the concepts of “users” and “groups”
949 have been unified into “roles”, this command is now equivalent to
950 \dg.) By default, only user-created roles are shown; supply the S
951 modifier to include system roles. If pattern is specified, only
952 those roles whose names match the pattern are listed. If the form
953 \du+ is used, additional information is shown about each role;
954 currently this adds the comment for each role.
955 \dx[+] [ pattern ]
957 Lists installed extensions. If pattern is specified, only those
958 extensions whose names match the pattern are listed. If the form
959 \dx+ is used, all the objects belonging to each matching extension
960 are listed.
961 \dy[+] [ pattern ]
963 Lists event triggers. If pattern is specified, only those event
964 triggers whose names match the pattern are listed. If + is appended
965 to the command name, each object is listed with its associated
966 description.
967 \e or \edit [ filename ] [ line_number ]
969 If filename is specified, the file is edited; after the editor
970 exits, the file's content is copied into the current query buffer.
971 If no filename is given, the current query buffer is copied to a
972 temporary file which is then edited in the same fashion. Or, if the
973 current query buffer is empty, the most recently executed query is
974 copied to a temporary file and edited in the same fashion.
975 The new contents of the query buffer are then re-parsed according
977 to the normal rules of psql, treating the whole buffer as a single
978 line. Any complete queries are immediately executed; that is, if
979 the query buffer contains or ends with a semicolon, everything up
980 to that point is executed and removed from the query buffer.
981 Whatever remains in the query buffer is redisplayed. Type semicolon
982 or \g to send it, or \r to cancel it by clearing the query buffer.
983 Treating the buffer as a single line primarily affects
985 meta-commands: whatever is in the buffer after a meta-command will
986 be taken as argument(s) to the meta-command, even if it spans
987 multiple lines. (Thus you cannot make meta-command-using scripts
988 this way. Use \i for that.)
989 If a line number is specified, psql will position the cursor on the
991 specified line of the file or query buffer. Note that if a single
992 all-digits argument is given, psql assumes it is a line number, not
993 a file name.
994 Tip
996 See Environment, below, for how to configure and customize your
997 editor.
998 \echo text [ ... ]
1000 Prints the evaluated arguments to standard output, separated by
1001 spaces and followed by a newline. This can be useful to intersperse
1002 information in the output of scripts. For example:
1003 => \echo `date`
1005 Tue Oct 26 21:40:57 CEST 1999
1006 If the first argument is an unquoted -n the trailing newline is not
1008 written (nor is the first argument).
1009 Tip
1011 If you use the \o command to redirect your query output you
1012 might wish to use \qecho instead of this command. See also
1013 \warn.
1014 \ef [ function_description [ line_number ] ]
1016 This command fetches and edits the definition of the named function
1017 or procedure, in the form of a CREATE OR REPLACE FUNCTION or CREATE
1018 OR REPLACE PROCEDURE command. Editing is done in the same way as
1019 for \edit. After the editor exits, the updated command is executed
1020 immediately if you added a semicolon to it. Otherwise it is
1021 redisplayed; type semicolon or \g to send it, or \r to cancel.
1022 The target function can be specified by name alone, or by name and
1024 arguments, for example foo(integer, text). The argument types must
1025 be given if there is more than one function of the same name.
1026 If no function is specified, a blank CREATE FUNCTION template is
1028 presented for editing.
1029 If a line number is specified, psql will position the cursor on the
1031 specified line of the function body. (Note that the function body
1032 typically does not begin on the first line of the file.)
1033 Unlike most other meta-commands, the entire remainder of the line
1035 is always taken to be the argument(s) of \ef, and neither variable
1036 interpolation nor backquote expansion are performed in the
1037 arguments.
1038 Tip
1040 See Environment, below, for how to configure and customize your
1041 editor.
1042 \encoding [ encoding ]
1044 Sets the client character set encoding. Without an argument, this
1045 command shows the current encoding.
1046 \errverbose
1048 Repeats the most recent server error message at maximum verbosity,
1049 as though VERBOSITY were set to verbose and SHOW_CONTEXT were set
1050 to always.
1051 \ev [ view_name [ line_number ] ]
1053 This command fetches and edits the definition of the named view, in
1054 the form of a CREATE OR REPLACE VIEW command. Editing is done in
1055 the same way as for \edit. After the editor exits, the updated
1056 command is executed immediately if you added a semicolon to it.
1057 Otherwise it is redisplayed; type semicolon or \g to send it, or \r
1058 to cancel.
1059 If no view is specified, a blank CREATE VIEW template is presented
1061 for editing.
1062 If a line number is specified, psql will position the cursor on the
1064 specified line of the view definition.
1065 Unlike most other meta-commands, the entire remainder of the line
1067 is always taken to be the argument(s) of \ev, and neither variable
1068 interpolation nor backquote expansion are performed in the
1069 arguments.
1070 \f [ string ]
1072 Sets the field separator for unaligned query output. The default is
1073 the vertical bar (|). It is equivalent to \pset fieldsep.
1074 \g [ (option=value [...]) ] [ filename ]
1076 \g [ (option=value [...]) ] [ |command ]
1077 Sends the current query buffer to the server for execution.
1078 If parentheses appear after \g, they surround a space-separated
1080 list of option=value formatting-option clauses, which are
1081 interpreted in the same way as \pset option value commands, but
1082 take effect only for the duration of this query. In this list,
1083 spaces are not allowed around = signs, but are required between
1084 option clauses. If =value is omitted, the named option is changed
1085 in the same way as for \pset option with no explicit value.
1086 If a filename or |command argument is given, the query's output is
1088 written to the named file or piped to the given shell command,
1089 instead of displaying it as usual. The file or command is written
1090 to only if the query successfully returns zero or more tuples, not
1091 if the query fails or is a non-data-returning SQL command.
1092 If the current query buffer is empty, the most recently sent query
1094 is re-executed instead. Except for that behavior, \g without any
1095 arguments is essentially equivalent to a semicolon. With arguments,
1096 \g provides a “one-shot” alternative to the \o command, and
1097 additionally allows one-shot adjustments of the output formatting
1098 options normally set by \pset.
1099 When the last argument begins with |, the entire remainder of the
1101 line is taken to be the command to execute, and neither variable
1102 interpolation nor backquote expansion are performed in it. The rest
1103 of the line is simply passed literally to the shell.
1104 \gdesc
1106 Shows the description (that is, the column names and data types) of
1107 the result of the current query buffer. The query is not actually
1108 executed; however, if it contains some type of syntax error, that
1109 error will be reported in the normal way.
1110 If the current query buffer is empty, the most recently sent query
1112 is described instead.
1113 \gexec
1115 Sends the current query buffer to the server, then treats each
1116 column of each row of the query's output (if any) as a SQL
1117 statement to be executed. For example, to create an index on each
1118 column of my_table:
1119 => SELECT format('create index on my_table(%I)', attname)
1121 -> FROM pg_attribute
1122 -> WHERE attrelid = 'my_table'::regclass AND attnum > 0
1123 -> ORDER BY attnum
1124 -> \gexec
1125 CREATE INDEX
1126 CREATE INDEX
1127 CREATE INDEX
1128 CREATE INDEX
1129 The generated queries are executed in the order in which the rows
1131 are returned, and left-to-right within each row if there is more
1132 than one column. NULL fields are ignored. The generated queries are
1133 sent literally to the server for processing, so they cannot be psql
1134 meta-commands nor contain psql variable references. If any
1135 individual query fails, execution of the remaining queries
1136 continues unless ON_ERROR_STOP is set. Execution of each query is
1137 subject to ECHO processing. (Setting ECHO to all or queries is
1138 often advisable when using \gexec.) Query logging, single-step
1139 mode, timing, and other query execution features apply to each
1140 generated query as well.
1141 If the current query buffer is empty, the most recently sent query
1143 is re-executed instead.
1144 \gset [ prefix ]
1146 Sends the current query buffer to the server and stores the query's
1147 output into psql variables (see Variables below). The query to be
1148 executed must return exactly one row. Each column of the row is
1149 stored into a separate variable, named the same as the column. For
1150 example:
1151 => SELECT 'hello' AS var1, 10 AS var2
1153 -> \gset
1154 => \echo :var1 :var2
1155 hello 10
1156 If you specify a prefix, that string is prepended to the query's
1158 column names to create the variable names to use:
1159 => SELECT 'hello' AS var1, 10 AS var2
1161 -> \gset result_
1162 => \echo :result_var1 :result_var2
1163 hello 10
1164 If a column result is NULL, the corresponding variable is unset
1166 rather than being set.
1167 If the query fails or does not return one row, no variables are
1169 changed.
1170 If the current query buffer is empty, the most recently sent query
1172 is re-executed instead.
1173 \gx [ (option=value [...]) ] [ filename ]
1175 \gx [ (option=value [...]) ] [ |command ]
1176 \gx is equivalent to \g, except that it forces expanded output mode
1177 for this query, as if expanded=on were included in the list of
1178 \pset options. See also \x.
1179 \h or \help [ command ]
1181 Gives syntax help on the specified SQL command. If command is not
1182 specified, then psql will list all the commands for which syntax
1183 help is available. If command is an asterisk (*), then syntax help
1184 on all SQL commands is shown.
1185 Unlike most other meta-commands, the entire remainder of the line
1187 is always taken to be the argument(s) of \help, and neither
1188 variable interpolation nor backquote expansion are performed in the
1189 arguments.
1190 Note
1192 To simplify typing, commands that consists of several words do
1193 not have to be quoted. Thus it is fine to type \help alter
1194 table.
1195 \H or \html
1197 Turns on HTML query output format. If the HTML format is already
1198 on, it is switched back to the default aligned text format. This
1199 command is for compatibility and convenience, but see \pset about
1200 setting other output options.
1201 \i or \include filename
1203 Reads input from the file filename and executes it as though it had
1204 been typed on the keyboard.
1205 If filename is - (hyphen), then standard input is read until an EOF
1207 indication or \q meta-command. This can be used to intersperse
1208 interactive input with input from files. Note that Readline
1209 behavior will be used only if it is active at the outermost level.
1210 Note
1212 If you want to see the lines on the screen as they are read you
1213 must set the variable ECHO to all.
1214 \if expression
1216 \elif expression
1217 \else
1218 \endif
1219 This group of commands implements nestable conditional blocks. A
1220 conditional block must begin with an \if and end with an \endif. In
1221 between there may be any number of \elif clauses, which may
1222 optionally be followed by a single \else clause. Ordinary queries
1223 and other types of backslash commands may (and usually do) appear
1224 between the commands forming a conditional block.
1225 The \if and \elif commands read their argument(s) and evaluate them
1227 as a boolean expression. If the expression yields true then
1228 processing continues normally; otherwise, lines are skipped until a
1229 matching \elif, \else, or \endif is reached. Once an \if or \elif
1230 test has succeeded, the arguments of later \elif commands in the
1231 same block are not evaluated but are treated as false. Lines
1232 following an \else are processed only if no earlier matching \if or
1233 \elif succeeded.
1234 The expression argument of an \if or \elif command is subject to
1236 variable interpolation and backquote expansion, just like any other
1237 backslash command argument. After that it is evaluated like the
1238 value of an on/off option variable. So a valid value is any
1239 unambiguous case-insensitive match for one of: true, false, 1, 0,
1240 on, off, yes, no. For example, t, T, and tR will all be considered
1241 to be true.
1242 Expressions that do not properly evaluate to true or false will
1244 generate a warning and be treated as false.
1245 Lines being skipped are parsed normally to identify queries and
1247 backslash commands, but queries are not sent to the server, and
1248 backslash commands other than conditionals (\if, \elif, \else,
1249 \endif) are ignored. Conditional commands are checked only for
1250 valid nesting. Variable references in skipped lines are not
1251 expanded, and backquote expansion is not performed either.
1252 All the backslash commands of a given conditional block must appear
1254 in the same source file. If EOF is reached on the main input file
1255 or an \include-ed file before all local \if-blocks have been
1256 closed, then psql will raise an error.
1257 Here is an example:
1259 -- check for the existence of two separate records in the database and store
1261 -- the results in separate psql variables
1262 SELECT
1263 EXISTS(SELECT 1 FROM customer WHERE customer_id = 123) as is_customer,
1264 EXISTS(SELECT 1 FROM employee WHERE employee_id = 456) as is_employee
1265 \gset
1266 \if :is_customer
1267 SELECT * FROM customer WHERE customer_id = 123;
1268 \elif :is_employee
1269 \echo 'is not a customer but is an employee'
1270 SELECT * FROM employee WHERE employee_id = 456;
1271 \else
1272 \if yes
1273 \echo 'not a customer or employee'
1274 \else
1275 \echo 'this will never print'
1276 \endif
1277 \endif
1278 \ir or \include_relative filename
1280 The \ir command is similar to \i, but resolves relative file names
1281 differently. When executing in interactive mode, the two commands
1282 behave identically. However, when invoked from a script, \ir
1283 interprets file names relative to the directory in which the script
1284 is located, rather than the current working directory.
1285 \l[+] or \list[+] [ pattern ]
1287 List the databases in the server and show their names, owners,
1288 character set encodings, and access privileges. If pattern is
1289 specified, only databases whose names match the pattern are listed.
1290 If + is appended to the command name, database sizes, default
1291 tablespaces, and descriptions are also displayed. (Size information
1292 is only available for databases that the current user can connect
1293 to.)
1294 \lo_export loid filename
1296 Reads the large object with OID loid from the database and writes
1297 it to filename. Note that this is subtly different from the server
1298 function lo_export, which acts with the permissions of the user
1299 that the database server runs as and on the server's file system.
1300 Tip
1302 Use \lo_list to find out the large object's OID.
1303 \lo_import filename [ comment ]
1305 Stores the file into a PostgreSQL large object. Optionally, it
1306 associates the given comment with the object. Example:
1307 foo=> \lo_import '/home/peter/pictures/photo.xcf' 'a picture of me'
1309 lo_import 152801
1310 The response indicates that the large object received object ID
1312 152801, which can be used to access the newly-created large object
1313 in the future. For the sake of readability, it is recommended to
1314 always associate a human-readable comment with every object. Both
1315 OIDs and comments can be viewed with the \lo_list command.
1316 Note that this command is subtly different from the server-side
1318 lo_import because it acts as the local user on the local file
1319 system, rather than the server's user and file system.
1320 \lo_list
1322 Shows a list of all PostgreSQL large objects currently stored in
1323 the database, along with any comments provided for them.
1324 \lo_unlink loid
1326 Deletes the large object with OID loid from the database.
1327 Tip
1329 Use \lo_list to find out the large object's OID.
1330 \o or \out [ filename ]
1332 \o or \out [ |command ]
1333 Arranges to save future query results to the file filename or pipe
1334 future results to the shell command command. If no argument is
1335 specified, the query output is reset to the standard output.
1336 If the argument begins with |, then the entire remainder of the
1338 line is taken to be the command to execute, and neither variable
1339 interpolation nor backquote expansion are performed in it. The rest
1340 of the line is simply passed literally to the shell.
1341 “Query results” includes all tables, command responses, and notices
1343 obtained from the database server, as well as output of various
1344 backslash commands that query the database (such as \d); but not
1345 error messages.
1346 Tip
1348 To intersperse text output in between query results, use
1349 \qecho.
1350 \p or \print
1352 Print the current query buffer to the standard output. If the
1353 current query buffer is empty, the most recently executed query is
1354 printed instead.
1355 \password [ username ]
1357 Changes the password of the specified user (by default, the current
1358 user). This command prompts for the new password, encrypts it, and
1359 sends it to the server as an ALTER ROLE command. This makes sure
1360 that the new password does not appear in cleartext in the command
1361 history, the server log, or elsewhere.
1362 \prompt [ text ] name
1364 Prompts the user to supply text, which is assigned to the variable
1365 name. An optional prompt string, text, can be specified. (For
1366 multiword prompts, surround the text with single quotes.)
1367 By default, \prompt uses the terminal for input and output.
1369 However, if the -f command line switch was used, \prompt uses
1370 standard input and standard output.
1371 \pset [ option [ value ] ]
1373 This command sets options affecting the output of query result
1374 tables. option indicates which option is to be set. The semantics
1375 of value vary depending on the selected option. For some options,
1376 omitting value causes the option to be toggled or unset, as
1377 described under the particular option. If no such behavior is
1378 mentioned, then omitting value just results in the current setting
1379 being displayed.
1380 \pset without any arguments displays the current status of all
1382 printing options.
1383 Adjustable printing options are:
1385 border
1387 The value must be a number. In general, the higher the number
1388 the more borders and lines the tables will have, but details
1389 depend on the particular format. In HTML format, this will
1390 translate directly into the border=... attribute. In most
1391 other formats only values 0 (no border), 1 (internal dividing
1392 lines), and 2 (table frame) make sense, and values above 2 will
1393 be treated the same as border = 2. The latex and
1394 latex-longtable formats additionally allow a value of 3 to add
1395 dividing lines between data rows.
1396 columns
1398 Sets the target width for the wrapped format, and also the
1399 width limit for determining whether output is wide enough to
1400 require the pager or switch to the vertical display in expanded
1401 auto mode. Zero (the default) causes the target width to be
1402 controlled by the environment variable COLUMNS, or the detected
1403 screen width if COLUMNS is not set. In addition, if columns is
1404 zero then the wrapped format only affects screen output. If
1405 columns is nonzero then file and pipe output is wrapped to that
1406 width as well.
1407 csv_fieldsep
1409 Specifies the field separator to be used in CSV output format.
1410 If the separator character appears in a field's value, that
1411 field is output within double quotes, following standard CSV
1412 rules. The default is a comma.
1413 expanded (or x)
1415 If value is specified it must be either on or off, which will
1416 enable or disable expanded mode, or auto. If value is omitted
1417 the command toggles between the on and off settings. When
1418 expanded mode is enabled, query results are displayed in two
1419 columns, with the column name on the left and the data on the
1420 right. This mode is useful if the data wouldn't fit on the
1421 screen in the normal “horizontal” mode. In the auto setting,
1422 the expanded mode is used whenever the query output has more
1423 than one column and is wider than the screen; otherwise, the
1424 regular mode is used. The auto setting is only effective in the
1425 aligned and wrapped formats. In other formats, it always
1426 behaves as if the expanded mode is off.
1427 fieldsep
1429 Specifies the field separator to be used in unaligned output
1430 format. That way one can create, for example, tab-separated
1431 output, which other programs might prefer. To set a tab as
1432 field separator, type \pset fieldsep '\t'. The default field
1433 separator is '|' (a vertical bar).
1434 fieldsep_zero
1436 Sets the field separator to use in unaligned output format to a
1437 zero byte.
1438 footer
1440 If value is specified it must be either on or off which will
1441 enable or disable display of the table footer (the (n rows)
1442 count). If value is omitted the command toggles footer display
1443 on or off.
1444 format
1446 Sets the output format to one of aligned, asciidoc, csv, html,
1447 latex, latex-longtable, troff-ms, unaligned, or wrapped. Unique
1448 abbreviations are allowed.
1449 aligned format is the standard, human-readable, nicely
1451 formatted text output; this is the default.
1452 unaligned format writes all columns of a row on one line,
1454 separated by the currently active field separator. This is
1455 useful for creating output that might be intended to be read in
1456 by other programs, for example, tab-separated or
1457 comma-separated format. However, the field separator character
1458 is not treated specially if it appears in a column's value; so
1459 CSV format may be better suited for such purposes.
1460 csv format
1462 writes column values separated by commas, applying the quoting
1464 rules described in RFC 4180. This output is compatible with the
1465 CSV format of the server's COPY command. A header line with
1466 column names is generated unless the tuples_only parameter is
1467 on. Titles and footers are not printed. Each row is terminated
1468 by the system-dependent end-of-line character, which is
1469 typically a single newline (\n) for Unix-like systems or a
1470 carriage return and newline sequence (\r\n) for Microsoft
1471 Windows. Field separator characters other than comma can be
1472 selected with \pset csv_fieldsep.
1473 wrapped format is like aligned but wraps wide data values
1475 across lines to make the output fit in the target column width.
1476 The target width is determined as described under the columns
1477 option. Note that psql will not attempt to wrap column header
1478 titles; therefore, wrapped format behaves the same as aligned
1479 if the total width needed for column headers exceeds the
1480 target.
1481 The asciidoc, html, latex, latex-longtable, and troff-ms
1483 formats put out tables that are intended to be included in
1484 documents using the respective mark-up language. They are not
1485 complete documents! This might not be necessary in HTML, but in
1486 LaTeX you must have a complete document wrapper. The latex
1487 format uses LaTeX's tabular environment. The latex-longtable
1488 format requires the LaTeX longtable and booktabs packages.
1489 linestyle
1491 Sets the border line drawing style to one of ascii, old-ascii,
1492 or unicode. Unique abbreviations are allowed. (That would mean
1493 one letter is enough.) The default setting is ascii. This
1494 option only affects the aligned and wrapped output formats.
1495 ascii style uses plain ASCII characters. Newlines in data are
1497 shown using a + symbol in the right-hand margin. When the
1498 wrapped format wraps data from one line to the next without a
1499 newline character, a dot (.) is shown in the right-hand margin
1500 of the first line, and again in the left-hand margin of the
1501 following line.
1502 old-ascii style uses plain ASCII characters, using the
1504 formatting style used in PostgreSQL 8.4 and earlier. Newlines
1505 in data are shown using a : symbol in place of the left-hand
1506 column separator. When the data is wrapped from one line to the
1507 next without a newline character, a ; symbol is used in place
1508 of the left-hand column separator.
1509 unicode style uses Unicode box-drawing characters. Newlines in
1511 data are shown using a carriage return symbol in the right-hand
1512 margin. When the data is wrapped from one line to the next
1513 without a newline character, an ellipsis symbol is shown in the
1514 right-hand margin of the first line, and again in the left-hand
1515 margin of the following line.
1516 When the border setting is greater than zero, the linestyle
1518 option also determines the characters with which the border
1519 lines are drawn. Plain ASCII characters work everywhere, but
1520 Unicode characters look nicer on displays that recognize them.
1521 null
1523 Sets the string to be printed in place of a null value. The
1524 default is to print nothing, which can easily be mistaken for
1525 an empty string. For example, one might prefer \pset null
1526 '(null)'.
1527 numericlocale
1529 If value is specified it must be either on or off which will
1530 enable or disable display of a locale-specific character to
1531 separate groups of digits to the left of the decimal marker. If
1532 value is omitted the command toggles between regular and
1533 locale-specific numeric output.
1534 pager
1536 Controls use of a pager program for query and psql help output.
1537 If the environment variable PSQL_PAGER or PAGER is set, the
1538 output is piped to the specified program. Otherwise a
1539 platform-dependent default program (such as more) is used.
1540 When the pager option is off, the pager program is not used.
1542 When the pager option is on, the pager is used when
1543 appropriate, i.e., when the output is to a terminal and will
1544 not fit on the screen. The pager option can also be set to
1545 always, which causes the pager to be used for all terminal
1546 output regardless of whether it fits on the screen. \pset
1547 pager without a value toggles pager use on and off.
1548 pager_min_lines
1550 If pager_min_lines is set to a number greater than the page
1551 height, the pager program will not be called unless there are
1552 at least this many lines of output to show. The default setting
1553 is 0.
1554 recordsep
1556 Specifies the record (line) separator to use in unaligned
1557 output format. The default is a newline character.
1558 recordsep_zero
1560 Sets the record separator to use in unaligned output format to
1561 a zero byte.
1562 tableattr (or T)
1564 In HTML format, this specifies attributes to be placed inside
1565 the table tag. This could for example be cellpadding or
1566 bgcolor. Note that you probably don't want to specify border
1567 here, as that is already taken care of by \pset border. If no
1568 value is given, the table attributes are unset.
1569 In latex-longtable format, this controls the proportional width
1571 of each column containing a left-aligned data type. It is
1572 specified as a whitespace-separated list of values, e.g., '0.2
1573 0.2 0.6'. Unspecified output columns use the last specified
1574 value.
1575 title (or C)
1577 Sets the table title for any subsequently printed tables. This
1578 can be used to give your output descriptive tags. If no value
1579 is given, the title is unset.
1580 tuples_only (or t)
1582 If value is specified it must be either on or off which will
1583 enable or disable tuples-only mode. If value is omitted the
1584 command toggles between regular and tuples-only output. Regular
1585 output includes extra information such as column headers,
1586 titles, and various footers. In tuples-only mode, only actual
1587 table data is shown.
1588 unicode_border_linestyle
1590 Sets the border drawing style for the unicode line style to one
1591 of single or double.
1592 unicode_column_linestyle
1594 Sets the column drawing style for the unicode line style to one
1595 of single or double.
1596 unicode_header_linestyle
1598 Sets the header drawing style for the unicode line style to one
1599 of single or double.
1600 Illustrations of how these different formats look can be seen in
1602 Examples, below.
1603 Tip
1605 There are various shortcut commands for \pset. See \a, \C, \f,
1606 \H, \t, \T, and \x.
1607 \q or \quit
1609 Quits the psql program. In a script file, only execution of that
1610 script is terminated.
1611 \qecho text [ ... ]
1613 This command is identical to \echo except that the output will be
1614 written to the query output channel, as set by \o.
1615 \r or \reset
1617 Resets (clears) the query buffer.
1618 \s [ filename ]
1620 Print psql's command line history to filename. If filename is
1621 omitted, the history is written to the standard output (using the
1622 pager if appropriate). This command is not available if psql was
1623 built without Readline support.
1624 \set [ name [ value [ ... ] ] ]
1626 Sets the psql variable name to value, or if more than one value is
1627 given, to the concatenation of all of them. If only one argument is
1628 given, the variable is set to an empty-string value. To unset a
1629 variable, use the \unset command.
1630 \set without any arguments displays the names and values of all
1632 currently-set psql variables.
1633 Valid variable names can contain letters, digits, and underscores.
1635 See Variables below for details. Variable names are case-sensitive.
1636 Certain variables are special, in that they control psql's behavior
1638 or are automatically set to reflect connection state. These
1639 variables are documented in Variables, below.
1640 Note
1642 This command is unrelated to the SQL command SET(7).
1643 \setenv name [ value ]
1645 Sets the environment variable name to value, or if the value is not
1646 supplied, unsets the environment variable. Example:
1647 testdb=> \setenv PAGER less
1649 testdb=> \setenv LESS -imx4F
1650 \sf[+] function_description
1652 This command fetches and shows the definition of the named function
1653 or procedure, in the form of a CREATE OR REPLACE FUNCTION or CREATE
1654 OR REPLACE PROCEDURE command. The definition is printed to the
1655 current query output channel, as set by \o.
1656 The target function can be specified by name alone, or by name and
1658 arguments, for example foo(integer, text). The argument types must
1659 be given if there is more than one function of the same name.
1660 If + is appended to the command name, then the output lines are
1662 numbered, with the first line of the function body being line 1.
1663 Unlike most other meta-commands, the entire remainder of the line
1665 is always taken to be the argument(s) of \sf, and neither variable
1666 interpolation nor backquote expansion are performed in the
1667 arguments.
1668 \sv[+] view_name
1670 This command fetches and shows the definition of the named view, in
1671 the form of a CREATE OR REPLACE VIEW command. The definition is
1672 printed to the current query output channel, as set by \o.
1673 If + is appended to the command name, then the output lines are
1675 numbered from 1.
1676 Unlike most other meta-commands, the entire remainder of the line
1678 is always taken to be the argument(s) of \sv, and neither variable
1679 interpolation nor backquote expansion are performed in the
1680 arguments.
1681 \t
1683 Toggles the display of output column name headings and row count
1684 footer. This command is equivalent to \pset tuples_only and is
1685 provided for convenience.
1686 \T table_options
1688 Specifies attributes to be placed within the table tag in HTML
1689 output format. This command is equivalent to \pset tableattr
1690 table_options.
1691 \timing [ on | off ]
1693 With a parameter, turns displaying of how long each SQL statement
1694 takes on or off. Without a parameter, toggles the display between
1695 on and off. The display is in milliseconds; intervals longer than 1
1696 second are also shown in minutes:seconds format, with hours and
1697 days fields added if needed.
1698 \unset name
1700 Unsets (deletes) the psql variable name.
1701 Most variables that control psql's behavior cannot be unset;
1703 instead, an \unset command is interpreted as setting them to their
1704 default values. See Variables below.
1705 \w or \write filename
1707 \w or \write |command
1708 Writes the current query buffer to the file filename or pipes it to
1709 the shell command command. If the current query buffer is empty,
1710 the most recently executed query is written instead.
1711 If the argument begins with |, then the entire remainder of the
1713 line is taken to be the command to execute, and neither variable
1714 interpolation nor backquote expansion are performed in it. The rest
1715 of the line is simply passed literally to the shell.
1716 \warn text [ ... ]
1718 This command is identical to \echo except that the output will be
1719 written to psql's standard error channel, rather than standard
1720 output.
1721 \watch [ seconds ]
1723 Repeatedly execute the current query buffer (as \g does) until
1724 interrupted or the query fails. Wait the specified number of
1725 seconds (default 2) between executions. Each query result is
1726 displayed with a header that includes the \pset title string (if
1727 any), the time as of query start, and the delay interval.
1728 If the current query buffer is empty, the most recently sent query
1730 is re-executed instead.
1731 \x [ on | off | auto ]
1733 Sets or toggles expanded table formatting mode. As such it is
1734 equivalent to \pset expanded.
1735 \z [ pattern ]
1737 Lists tables, views and sequences with their associated access
1738 privileges. If a pattern is specified, only tables, views and
1739 sequences whose names match the pattern are listed.
1740 This is an alias for \dp (“display privileges”).
1742 \! [ command ]
1744 With no argument, escapes to a sub-shell; psql resumes when the
1745 sub-shell exits. With an argument, executes the shell command
1746 command.
1747 Unlike most other meta-commands, the entire remainder of the line
1749 is always taken to be the argument(s) of \!, and neither variable
1750 interpolation nor backquote expansion are performed in the
1751 arguments. The rest of the line is simply passed literally to the
1752 shell.
1753 \? [ topic ]
1755 Shows help information. The optional topic parameter (defaulting to
1756 commands) selects which part of psql is explained: commands
1757 describes psql's backslash commands; options describes the
1758 command-line options that can be passed to psql; and variables
1759 shows help about psql configuration variables.
1760 \;
1762 Backslash-semicolon is not a meta-command in the same way as the
1763 preceding commands; rather, it simply causes a semicolon to be
1764 added to the query buffer without any further processing.
1765 Normally, psql will dispatch a SQL command to the server as soon as
1767 it reaches the command-ending semicolon, even if more input remains
1768 on the current line. Thus for example entering
1769 select 1; select 2; select 3;
1771 will result in the three SQL commands being individually sent to
1773 the server, with each one's results being displayed before
1774 continuing to the next command. However, a semicolon entered as \;
1775 will not trigger command processing, so that the command before it
1776 and the one after are effectively combined and sent to the server
1777 in one request. So for example
1778 select 1\; select 2\; select 3;
1780 results in sending the three SQL commands to the server in a single
1782 request, when the non-backslashed semicolon is reached. The server
1783 executes such a request as a single transaction, unless there are
1784 explicit BEGIN/COMMIT commands included in the string to divide it
1785 into multiple transactions. (See Section 52.2.2.1 for more details
1786 about how the server handles multi-query strings.) psql prints
1787 only the last query result it receives for each request; in this
1788 example, although all three SELECTs are indeed executed, psql only
1789 prints the 3.
1790 Patterns
1792 The various \d commands accept a pattern parameter to specify the
1793 object name(s) to be displayed. In the simplest case, a pattern is
1794 just the exact name of the object. The characters within a pattern
1795 are normally folded to lower case, just as in SQL names; for
1796 example, \dt FOO will display the table named foo. As in SQL names,
1797 placing double quotes around a pattern stops folding to lower case.
1798 Should you need to include an actual double quote character in a
1799 pattern, write it as a pair of double quotes within a double-quote
1800 sequence; again this is in accord with the rules for SQL quoted
1801 identifiers. For example, \dt "FOO""BAR" will display the table
1802 named FOO"BAR (not foo"bar). Unlike the normal rules for SQL names,
1803 you can put double quotes around just part of a pattern, for
1804 instance \dt FOO"FOO"BAR will display the table named fooFOObar.
1805 Whenever the pattern parameter is omitted completely, the \d
1807 commands display all objects that are visible in the current schema
1808 search path — this is equivalent to using * as the pattern. (An
1809 object is said to be visible if its containing schema is in the
1810 search path and no object of the same kind and name appears earlier
1811 in the search path. This is equivalent to the statement that the
1812 object can be referenced by name without explicit schema
1813 qualification.) To see all objects in the database regardless of
1814 visibility, use *.* as the pattern.
1815 Within a pattern, * matches any sequence of characters (including
1817 no characters) and ? matches any single character. (This notation
1818 is comparable to Unix shell file name patterns.) For example, \dt
1819 int* displays tables whose names begin with int. But within double
1820 quotes, * and ? lose these special meanings and are just matched
1821 literally.
1822 A pattern that contains a dot (.) is interpreted as a schema name
1824 pattern followed by an object name pattern. For example, \dt
1825 foo*.*bar* displays all tables whose table name includes bar that
1826 are in schemas whose schema name starts with foo. When no dot
1827 appears, then the pattern matches only objects that are visible in
1828 the current schema search path. Again, a dot within double quotes
1829 loses its special meaning and is matched literally.
1830 Advanced users can use regular-expression notations such as
1832 character classes, for example [0-9] to match any digit. All
1833 regular expression special characters work as specified in
1834 Section 9.7.3, except for . which is taken as a separator as
1835 mentioned above, * which is translated to the regular-expression
1836 notation .*, ? which is translated to ., and $ which is matched
1837 literally. You can emulate these pattern characters at need by
1838 writing ? for ., (R+|) for R*, or (R|) for R?. $ is not needed as
1839 a regular-expression character since the pattern must match the
1840 whole name, unlike the usual interpretation of regular expressions
1841 (in other words, $ is automatically appended to your pattern).
1842 Write * at the beginning and/or end if you don't wish the pattern
1843 to be anchored. Note that within double quotes, all regular
1844 expression special characters lose their special meanings and are
1845 matched literally. Also, the regular expression special characters
1846 are matched literally in operator name patterns (i.e., the argument
1847 of \do).
1848 Advanced Features
1850 Variables
1851 psql provides variable substitution features similar to common Unix
1852 command shells. Variables are simply name/value pairs, where the
1853 value can be any string of any length. The name must consist of
1854 letters (including non-Latin letters), digits, and underscores.
1855 To set a variable, use the psql meta-command \set. For example,
1857 testdb=> \set foo bar
1859 sets the variable foo to the value bar. To retrieve the content of
1861 the variable, precede the name with a colon, for example:
1862 testdb=> \echo :foo
1864 bar
1865 This works in both regular SQL commands and meta-commands; there is
1867 more detail in SQL Interpolation, below.
1868 If you call \set without a second argument, the variable is set to
1870 an empty-string value. To unset (i.e., delete) a variable, use the
1871 command \unset. To show the values of all variables, call \set
1872 without any argument.
1873 Note
1875 The arguments of \set are subject to the same substitution
1876 rules as with other commands. Thus you can construct
1877 interesting references such as \set :foo 'something' and get
1878 “soft links” or “variable variables” of Perl or PHP fame,
1879 respectively. Unfortunately (or fortunately?), there is no way
1880 to do anything useful with these constructs. On the other hand,
1881 \set bar :foo is a perfectly valid way to copy a variable.
1882 A number of these variables are treated specially by psql. They
1884 represent certain option settings that can be changed at run time
1885 by altering the value of the variable, or in some cases represent
1886 changeable state of psql. By convention, all specially treated
1887 variables' names consist of all upper-case ASCII letters (and
1888 possibly digits and underscores). To ensure maximum compatibility
1889 in the future, avoid using such variable names for your own
1890 purposes.
1891 Variables that control psql's behavior generally cannot be unset or
1893 set to invalid values. An \unset command is allowed but is
1894 interpreted as setting the variable to its default value. A \set
1895 command without a second argument is interpreted as setting the
1896 variable to on, for control variables that accept that value, and
1897 is rejected for others. Also, control variables that accept the
1898 values on and off will also accept other common spellings of
1899 Boolean values, such as true and false.
1900 The specially treated variables are:
1902 AUTOCOMMIT
1904 When on (the default), each SQL command is automatically
1905 committed upon successful completion. To postpone commit in
1906 this mode, you must enter a BEGIN or START TRANSACTION SQL
1907 command. When off or unset, SQL commands are not committed
1908 until you explicitly issue COMMIT or END. The autocommit-off
1909 mode works by issuing an implicit BEGIN for you, just before
1910 any command that is not already in a transaction block and is
1911 not itself a BEGIN or other transaction-control command, nor a
1912 command that cannot be executed inside a transaction block
1913 (such as VACUUM).
1914 Note
1916 In autocommit-off mode, you must explicitly abandon any
1917 failed transaction by entering ABORT or ROLLBACK. Also keep
1918 in mind that if you exit the session without committing,
1919 your work will be lost.
1920 Note
1922 The autocommit-on mode is PostgreSQL's traditional
1923 behavior, but autocommit-off is closer to the SQL spec. If
1924 you prefer autocommit-off, you might wish to set it in the
1925 system-wide psqlrc file or your ~/.psqlrc file.
1926 COMP_KEYWORD_CASE
1928 Determines which letter case to use when completing an SQL key
1929 word. If set to lower or upper, the completed word will be in
1930 lower or upper case, respectively. If set to preserve-lower or
1931 preserve-upper (the default), the completed word will be in the
1932 case of the word already entered, but words being completed
1933 without anything entered will be in lower or upper case,
1934 respectively.
1935 DBNAME
1937 The name of the database you are currently connected to. This
1938 is set every time you connect to a database (including program
1939 start-up), but can be changed or unset.
1940 ECHO
1942 If set to all, all nonempty input lines are printed to standard
1943 output as they are read. (This does not apply to lines read
1944 interactively.) To select this behavior on program start-up,
1945 use the switch -a. If set to queries, psql prints each query to
1946 standard output as it is sent to the server. The switch to
1947 select this behavior is -e. If set to errors, then only failed
1948 queries are displayed on standard error output. The switch for
1949 this behavior is -b. If set to none (the default), then no
1950 queries are displayed.
1951 ECHO_HIDDEN
1953 When this variable is set to on and a backslash command queries
1954 the database, the query is first shown. This feature helps you
1955 to study PostgreSQL internals and provide similar functionality
1956 in your own programs. (To select this behavior on program
1957 start-up, use the switch -E.) If you set this variable to the
1958 value noexec, the queries are just shown but are not actually
1959 sent to the server and executed. The default value is off.
1960 ENCODING
1962 The current client character set encoding. This is set every
1963 time you connect to a database (including program start-up),
1964 and when you change the encoding with \encoding, but it can be
1965 changed or unset.
1966 ERROR
1968 true if the last SQL query failed, false if it succeeded. See
1969 also SQLSTATE.
1970 FETCH_COUNT
1972 If this variable is set to an integer value greater than zero,
1973 the results of SELECT queries are fetched and displayed in
1974 groups of that many rows, rather than the default behavior of
1975 collecting the entire result set before display. Therefore only
1976 a limited amount of memory is used, regardless of the size of
1977 the result set. Settings of 100 to 1000 are commonly used when
1978 enabling this feature. Keep in mind that when using this
1979 feature, a query might fail after having already displayed some
1980 rows.
1981 Tip
1983 Although you can use any output format with this feature,
1984 the default aligned format tends to look bad because each
1985 group of FETCH_COUNT rows will be formatted separately,
1986 leading to varying column widths across the row groups. The
1987 other output formats work better.
1988 HIDE_TABLEAM
1990 If this variable is set to true, a table's access method
1991 details are not displayed. This is mainly useful for regression
1992 tests.
1993 HISTCONTROL
1995 If this variable is set to ignorespace, lines which begin with
1996 a space are not entered into the history list. If set to a
1997 value of ignoredups, lines matching the previous history line
1998 are not entered. A value of ignoreboth combines the two
1999 options. If set to none (the default), all lines read in
2000 interactive mode are saved on the history list.
2001 Note
2003 This feature was shamelessly plagiarized from Bash.
2004 HISTFILE
2006 The file name that will be used to store the history list. If
2007 unset, the file name is taken from the PSQL_HISTORY environment
2008 variable. If that is not set either, the default is
2009 ~/.psql_history, or %APPDATA%\postgresql\psql_history on
2010 Windows. For example, putting:
2011 \set HISTFILE ~/.psql_history- :DBNAME
2013 in ~/.psqlrc will cause psql to maintain a separate history for
2015 each database.
2016 Note
2018 This feature was shamelessly plagiarized from Bash.
2019 HISTSIZE
2021 The maximum number of commands to store in the command history
2022 (default 500). If set to a negative value, no limit is applied.
2023 Note
2025 This feature was shamelessly plagiarized from Bash.
2026 HOST
2028 The database server host you are currently connected to. This
2029 is set every time you connect to a database (including program
2030 start-up), but can be changed or unset.
2031 IGNOREEOF
2033 If set to 1 or less, sending an EOF character (usually
2034 Control+D) to an interactive session of psql will terminate the
2035 application. If set to a larger numeric value, that many
2036 consecutive EOF characters must be typed to make an interactive
2037 session terminate. If the variable is set to a non-numeric
2038 value, it is interpreted as 10. The default is 0.
2039 Note
2041 This feature was shamelessly plagiarized from Bash.
2042 LASTOID
2044 The value of the last affected OID, as returned from an INSERT
2045 or \lo_import command. This variable is only guaranteed to be
2046 valid until after the result of the next SQL command has been
2047 displayed. PostgreSQL servers since version 12 do not support
2048 OID system columns anymore, thus LASTOID will always be 0
2049 following INSERT when targeting such servers.
2050 LAST_ERROR_MESSAGE
2052 LAST_ERROR_SQLSTATE
2053 The primary error message and associated SQLSTATE code for the
2054 most recent failed query in the current psql session, or an
2055 empty string and 00000 if no error has occurred in the current
2056 session.
2057 ON_ERROR_ROLLBACK
2059 When set to on, if a statement in a transaction block generates
2060 an error, the error is ignored and the transaction continues.
2061 When set to interactive, such errors are only ignored in
2062 interactive sessions, and not when reading script files. When
2063 set to off (the default), a statement in a transaction block
2064 that generates an error aborts the entire transaction. The
2065 error rollback mode works by issuing an implicit SAVEPOINT for
2066 you, just before each command that is in a transaction block,
2067 and then rolling back to the savepoint if the command fails.
2068 ON_ERROR_STOP
2070 By default, command processing continues after an error. When
2071 this variable is set to on, processing will instead stop
2072 immediately. In interactive mode, psql will return to the
2073 command prompt; otherwise, psql will exit, returning error code
2074 3 to distinguish this case from fatal error conditions, which
2075 are reported using error code 1. In either case, any currently
2076 running scripts (the top-level script, if any, and any other
2077 scripts which it may have in invoked) will be terminated
2078 immediately. If the top-level command string contained multiple
2079 SQL commands, processing will stop with the current command.
2080 PORT
2082 The database server port to which you are currently connected.
2083 This is set every time you connect to a database (including
2084 program start-up), but can be changed or unset.
2085 PROMPT1
2087 PROMPT2
2088 PROMPT3
2089 These specify what the prompts psql issues should look like.
2090 See Prompting below.
2091 QUIET
2093 Setting this variable to on is equivalent to the command line
2094 option -q. It is probably not too useful in interactive mode.
2095 ROW_COUNT
2097 The number of rows returned or affected by the last SQL query,
2098 or 0 if the query failed or did not report a row count.
2099 SERVER_VERSION_NAME
2101 SERVER_VERSION_NUM
2102 The server's version number as a string, for example 9.6.2,
2103 10.1 or 11beta1, and in numeric form, for example 90602 or
2104 100001. These are set every time you connect to a database
2105 (including program start-up), but can be changed or unset.
2106 SHOW_CONTEXT
2108 This variable can be set to the values never, errors, or always
2109 to control whether CONTEXT fields are displayed in messages
2110 from the server. The default is errors (meaning that context
2111 will be shown in error messages, but not in notice or warning
2112 messages). This setting has no effect when VERBOSITY is set to
2113 terse or sqlstate. (See also \errverbose, for use when you want
2114 a verbose version of the error you just got.)
2115 SINGLELINE
2117 Setting this variable to on is equivalent to the command line
2118 option -S.
2119 SINGLESTEP
2121 Setting this variable to on is equivalent to the command line
2122 option -s.
2123 SQLSTATE
2125 The error code (see Appendix A) associated with the last SQL
2126 query's failure, or 00000 if it succeeded.
2127 USER
2129 The database user you are currently connected as. This is set
2130 every time you connect to a database (including program
2131 start-up), but can be changed or unset.
2132 VERBOSITY
2134 This variable can be set to the values default, verbose, terse,
2135 or sqlstate to control the verbosity of error reports. (See
2136 also \errverbose, for use when you want a verbose version of
2137 the error you just got.)
2138 VERSION
2140 VERSION_NAME
2141 VERSION_NUM
2142 These variables are set at program start-up to reflect psql's
2143 version, respectively as a verbose string, a short string
2144 (e.g., 9.6.2, 10.1, or 11beta1), and a number (e.g., 90602 or
2145 100001). They can be changed or unset.
2146 SQL Interpolation
2148 A key feature of psql variables is that you can substitute
2149 (“interpolate”) them into regular SQL statements, as well as the
2150 arguments of meta-commands. Furthermore, psql provides facilities
2151 for ensuring that variable values used as SQL literals and
2152 identifiers are properly quoted. The syntax for interpolating a
2153 value without any quoting is to prepend the variable name with a
2154 colon (:). For example,
2155 testdb=> \set foo 'my_table'
2157 testdb=> SELECT * FROM :foo;
2158 would query the table my_table. Note that this may be unsafe: the
2160 value of the variable is copied literally, so it can contain
2161 unbalanced quotes, or even backslash commands. You must make sure
2162 that it makes sense where you put it.
2163 When a value is to be used as an SQL literal or identifier, it is
2165 safest to arrange for it to be quoted. To quote the value of a
2166 variable as an SQL literal, write a colon followed by the variable
2167 name in single quotes. To quote the value as an SQL identifier,
2168 write a colon followed by the variable name in double quotes. These
2169 constructs deal correctly with quotes and other special characters
2170 embedded within the variable value. The previous example would be
2171 more safely written this way:
2172 testdb=> \set foo 'my_table'
2174 testdb=> SELECT * FROM :"foo";
2175 Variable interpolation will not be performed within quoted SQL
2177 literals and identifiers. Therefore, a construction such as ':foo'
2178 doesn't work to produce a quoted literal from a variable's value
2179 (and it would be unsafe if it did work, since it wouldn't correctly
2180 handle quotes embedded in the value).
2181 One example use of this mechanism is to copy the contents of a file
2183 into a table column. First load the file into a variable and then
2184 interpolate the variable's value as a quoted string:
2185 testdb=> \set content `cat my_file.txt`
2187 testdb=> INSERT INTO my_table VALUES (:'content');
2188 (Note that this still won't work if my_file.txt contains NUL bytes.
2190 psql does not support embedded NUL bytes in variable values.)
2191 Since colons can legally appear in SQL commands, an apparent
2193 attempt at interpolation (that is, :name, :'name', or :"name") is
2194 not replaced unless the named variable is currently set. In any
2195 case, you can escape a colon with a backslash to protect it from
2196 substitution.
2197 The :{?name} special syntax returns TRUE or FALSE depending on
2199 whether the variable exists or not, and is thus always substituted,
2200 unless the colon is backslash-escaped.
2201 The colon syntax for variables is standard SQL for embedded query
2203 languages, such as ECPG. The colon syntaxes for array slices and
2204 type casts are PostgreSQL extensions, which can sometimes conflict
2205 with the standard usage. The colon-quote syntax for escaping a
2206 variable's value as an SQL literal or identifier is a psql
2207 extension.
2208 Prompting
2210 The prompts psql issues can be customized to your preference. The
2211 three variables PROMPT1, PROMPT2, and PROMPT3 contain strings and
2212 special escape sequences that describe the appearance of the
2213 prompt. Prompt 1 is the normal prompt that is issued when psql
2214 requests a new command. Prompt 2 is issued when more input is
2215 expected during command entry, for example because the command was
2216 not terminated with a semicolon or a quote was not closed. Prompt 3
2217 is issued when you are running an SQL COPY FROM STDIN command and
2218 you need to type in a row value on the terminal.
2219 The value of the selected prompt variable is printed literally,
2221 except where a percent sign (%) is encountered. Depending on the
2222 next character, certain other text is substituted instead. Defined
2223 substitutions are:
2224 %M
2226 The full host name (with domain name) of the database server,
2227 or [local] if the connection is over a Unix domain socket, or
2228 [local:/dir/name], if the Unix domain socket is not at the
2229 compiled in default location.
2230 %m
2232 The host name of the database server, truncated at the first
2233 dot, or [local] if the connection is over a Unix domain socket.
2234 %>
2236 The port number at which the database server is listening.
2237 %n
2239 The database session user name. (The expansion of this value
2240 might change during a database session as the result of the
2241 command SET SESSION AUTHORIZATION.)
2242 %/
2244 The name of the current database.
2245 %~
2247 Like %/, but the output is ~ (tilde) if the database is your
2248 default database.
2249 %#
2251 If the session user is a database superuser, then a #,
2252 otherwise a >. (The expansion of this value might change during
2253 a database session as the result of the command SET SESSION
2254 AUTHORIZATION.)
2255 %p
2257 The process ID of the backend currently connected to.
2258 %R
2260 In prompt 1 normally =, but @ if the session is in an inactive
2261 branch of a conditional block, or ^ if in single-line mode, or
2262 ! if the session is disconnected from the database (which can
2263 happen if \connect fails). In prompt 2 %R is replaced by a
2264 character that depends on why psql expects more input: - if the
2265 command simply wasn't terminated yet, but * if there is an
2266 unfinished /* ... */ comment, a single quote if there is an
2267 unfinished quoted string, a double quote if there is an
2268 unfinished quoted identifier, a dollar sign if there is an
2269 unfinished dollar-quoted string, or ( if there is an unmatched
2270 left parenthesis. In prompt 3 %R doesn't produce anything.
2271 %x
2273 Transaction status: an empty string when not in a transaction
2274 block, or * when in a transaction block, or ! when in a failed
2275 transaction block, or ? when the transaction state is
2276 indeterminate (for example, because there is no connection).
2277 %l
2279 The line number inside the current statement, starting from 1.
2280 %digits
2282 The character with the indicated octal code is substituted.
2283 %:name:
2285 The value of the psql variable name. See Variables, above, for
2286 details.
2287 %`command`
2289 The output of command, similar to ordinary “back-tick”
2290 substitution.
2291 %[ ... %]
2293 Prompts can contain terminal control characters which, for
2294 example, change the color, background, or style of the prompt
2295 text, or change the title of the terminal window. In order for
2296 the line editing features of Readline to work properly, these
2297 non-printing control characters must be designated as invisible
2298 by surrounding them with %[ and %]. Multiple pairs of these can
2299 occur within the prompt. For example:
2300 testdb=> \set PROMPT1 '%[%033[1;33;40m%]%n@%/%R%[%033[0m%]%# '
2302 results in a boldfaced (1;) yellow-on-black (33;40) prompt on
2304 VT100-compatible, color-capable terminals.
2305 %w
2307 Whitespace of the same width as the most recent output of
2308 PROMPT1. This can be used as a PROMPT2 setting, so that
2309 multi-line statements are aligned with the first line, but
2310 there is no visible secondary prompt.
2311 To insert a percent sign into your prompt, write %%. The default
2312 prompts are '%/%R%x%# ' for prompts 1 and 2, and '>> ' for prompt
2313 3.
2314 Note
2316 This feature was shamelessly plagiarized from tcsh.
2317 Command-Line Editing
2319 psql supports the Readline library for convenient line editing and
2320 retrieval. The command history is automatically saved when psql
2321 exits and is reloaded when psql starts up. Tab-completion is also
2322 supported, although the completion logic makes no claim to be an
2323 SQL parser. The queries generated by tab-completion can also
2324 interfere with other SQL commands, e.g., SET TRANSACTION ISOLATION
2325 LEVEL. If for some reason you do not like the tab completion, you
2326 can turn it off by putting this in a file named .inputrc in your
2327 home directory:
2328 $if psql
2330 set disable-completion on
2331 $endif
2332 (This is not a psql but a Readline feature. Read its documentation
2334 for further details.)
2335
2337 COLUMNS
2338 If \pset columns is zero, controls the width for the wrapped format
2339 and width for determining if wide output requires the pager or
2340 should be switched to the vertical format in expanded auto mode.
2341 PGDATABASE
2343 PGHOST
2344 PGPORT
2345 PGUSER
2346 Default connection parameters (see Section 33.14).
2347 PG_COLOR
2349 Specifies whether to use color in diagnostic messages. Possible
2350 values are always, auto and never.
2351 PSQL_EDITOR
2353 EDITOR
2354 VISUAL
2355 Editor used by the \e, \ef, and \ev commands. These variables are
2356 examined in the order listed; the first that is set is used. If
2357 none of them is set, the default is to use vi on Unix systems or
2358 notepad.exe on Windows systems.
2359 PSQL_EDITOR_LINENUMBER_ARG
2361 When \e, \ef, or \ev is used with a line number argument, this
2362 variable specifies the command-line argument used to pass the
2363 starting line number to the user's editor. For editors such as
2364 Emacs or vi, this is a plus sign. Include a trailing space in the
2365 value of the variable if there needs to be space between the option
2366 name and the line number. Examples:
2367 PSQL_EDITOR_LINENUMBER_ARG='+'
2369 PSQL_EDITOR_LINENUMBER_ARG='--line '
2370 The default is + on Unix systems (corresponding to the default
2372 editor vi, and useful for many other common editors); but there is
2373 no default on Windows systems.
2374 PSQL_HISTORY
2376 Alternative location for the command history file. Tilde (~)
2377 expansion is performed.
2378 PSQL_PAGER
2380 PAGER
2381 If a query's results do not fit on the screen, they are piped
2382 through this command. Typical values are more or less. Use of the
2383 pager can be disabled by setting PSQL_PAGER or PAGER to an empty
2384 string, or by adjusting the pager-related options of the \pset
2385 command. These variables are examined in the order listed; the
2386 first that is set is used. If none of them is set, the default is
2387 to use more on most platforms, but less on Cygwin.
2388 PSQLRC
2390 Alternative location of the user's .psqlrc file. Tilde (~)
2391 expansion is performed.
2392 SHELL
2394 Command executed by the \! command.
2395 TMPDIR
2397 Directory for storing temporary files. The default is /tmp.
2398 This utility, like most other PostgreSQL utilities, also uses the
2400 environment variables supported by libpq (see Section 33.14).
2401
2403 psqlrc and ~/.psqlrc
2404 Unless it is passed an -X option, psql attempts to read and execute
2405 commands from the system-wide startup file (psqlrc) and then the
2406 user's personal startup file (~/.psqlrc), after connecting to the
2407 database but before accepting normal commands. These files can be
2408 used to set up the client and/or the server to taste, typically
2409 with \set and SET commands.
2410 The system-wide startup file is named psqlrc and is sought in the
2412 installation's “system configuration” directory, which is most
2413 reliably identified by running pg_config --sysconfdir. By default
2414 this directory will be ../etc/ relative to the directory containing
2415 the PostgreSQL executables. The name of this directory can be set
2416 explicitly via the PGSYSCONFDIR environment variable.
2417 The user's personal startup file is named .psqlrc and is sought in
2419 the invoking user's home directory. On Windows, which lacks such a
2420 concept, the personal startup file is named
2421 %APPDATA%\postgresql\psqlrc.conf. The location of the user's
2422 startup file can be set explicitly via the PSQLRC environment
2423 variable.
2424 Both the system-wide startup file and the user's personal startup
2426 file can be made psql-version-specific by appending a dash and the
2427 PostgreSQL major or minor release number to the file name, for
2428 example ~/.psqlrc-9.2 or ~/.psqlrc-9.2.5. The most specific
2429 version-matching file will be read in preference to a
2430 non-version-specific file.
2431 .psql_history
2433 The command-line history is stored in the file ~/.psql_history, or
2434 %APPDATA%\postgresql\psql_history on Windows.
2435 The location of the history file can be set explicitly via the
2437 HISTFILE psql variable or the PSQL_HISTORY environment variable.
2438
2440 • psql works best with servers of the same or an older major version.
2441 Backslash commands are particularly likely to fail if the server is
2442 of a newer version than psql itself. However, backslash commands of
2443 the \d family should work with servers of versions back to 7.4,
2444 though not necessarily with servers newer than psql itself. The
2445 general functionality of running SQL commands and displaying query
2446 results should also work with servers of a newer major version, but
2447 this cannot be guaranteed in all cases.
2448 If you want to use psql to connect to several servers of different
2450 major versions, it is recommended that you use the newest version
2451 of psql. Alternatively, you can keep around a copy of psql from
2452 each major version and be sure to use the version that matches the
2453 respective server. But in practice, this additional complication
2454 should not be necessary.
2455 • Before PostgreSQL 9.6, the -c option implied -X (--no-psqlrc); this
2457 is no longer the case.
2458 • Before PostgreSQL 8.4, psql allowed the first argument of a
2460 single-letter backslash command to start directly after the
2461 command, without intervening whitespace. Now, some whitespace is
2462 required.
2463
2465 psql is built as a “console application”. Since the Windows console
2466 windows use a different encoding than the rest of the system, you must
2467 take special care when using 8-bit characters within psql. If psql
2468 detects a problematic console code page, it will warn you at startup.
2469 To change the console code page, two things are necessary:
2470 • Set the code page by entering cmd.exe /c chcp 1252. (1252 is a code
2472 page that is appropriate for German; replace it with your value.)
2473 If you are using Cygwin, you can put this command in /etc/profile.
2474 • Set the console font to Lucida Console, because the raster font
2476 does not work with the ANSI code page.
2477
2479 The first example shows how to spread a command over several lines of
2480 input. Notice the changing prompt:
2481 testdb=> CREATE TABLE my_table (
2483 testdb(> first integer not null default 0,
2484 testdb(> second text)
2485 testdb-> ;
2486 CREATE TABLE
2487 Now look at the table definition again:
2489 testdb=> \d my_table
2491 Table "public.my_table"
2492 Column | Type | Collation | Nullable | Default
2493 --------+---------+-----------+----------+---------
2494 first | integer | | not null | 0
2495 second | text | | |
2496 Now we change the prompt to something more interesting:
2498 testdb=> \set PROMPT1 '%n@%m %~%R%# '
2500 peter@localhost testdb=>
2501 Let's assume you have filled the table with data and want to take a
2503 look at it:
2504 peter@localhost testdb=> SELECT * FROM my_table;
2506 first | second
2507 -------+--------
2508 1 | one
2509 2 | two
2510 3 | three
2511 4 | four
2512 (4 rows)
2513 You can display tables in different ways by using the \pset command:
2515 peter@localhost testdb=> \pset border 2
2517 Border style is 2.
2518 peter@localhost testdb=> SELECT * FROM my_table;
2519 +-------+--------+
2520 | first | second |
2521 +-------+--------+
2522 | 1 | one |
2523 | 2 | two |
2524 | 3 | three |
2525 | 4 | four |
2526 +-------+--------+
2527 (4 rows)
2528 peter@localhost testdb=> \pset border 0
2530 Border style is 0.
2531 peter@localhost testdb=> SELECT * FROM my_table;
2532 first second
2533 ----- ------
2534 1 one
2535 2 two
2536 3 three
2537 4 four
2538 (4 rows)
2539 peter@localhost testdb=> \pset border 1
2541 Border style is 1.
2542 peter@localhost testdb=> \pset format csv
2543 Output format is csv.
2544 peter@localhost testdb=> \pset tuples_only
2545 Tuples only is on.
2546 peter@localhost testdb=> SELECT second, first FROM my_table;
2547 one,1
2548 two,2
2549 three,3
2550 four,4
2551 peter@localhost testdb=> \pset format unaligned
2552 Output format is unaligned.
2553 peter@localhost testdb=> \pset fieldsep '\t'
2554 Field separator is " ".
2555 peter@localhost testdb=> SELECT second, first FROM my_table;
2556 one 1
2557 two 2
2558 three 3
2559 four 4
2560 Alternatively, use the short commands:
2562 peter@localhost testdb=> \a \t \x
2564 Output format is aligned.
2565 Tuples only is off.
2566 Expanded display is on.
2567 peter@localhost testdb=> SELECT * FROM my_table;
2568 -[ RECORD 1 ]-
2569 first | 1
2570 second | one
2571 -[ RECORD 2 ]-
2572 first | 2
2573 second | two
2574 -[ RECORD 3 ]-
2575 first | 3
2576 second | three
2577 -[ RECORD 4 ]-
2578 first | 4
2579 second | four
2580 Also, these output format options can be set for just one query by
2582 using \g:
2583 peter@localhost testdb=> SELECT * FROM my_table
2585 peter@localhost testdb-> \g (format=aligned tuples_only=off expanded=on)
2586 -[ RECORD 1 ]-
2587 first | 1
2588 second | one
2589 -[ RECORD 2 ]-
2590 first | 2
2591 second | two
2592 -[ RECORD 3 ]-
2593 first | 3
2594 second | three
2595 -[ RECORD 4 ]-
2596 first | 4
2597 second | four
2598 When suitable, query results can be shown in a crosstab representation
2600 with the \crosstabview command:
2601 testdb=> SELECT first, second, first > 2 AS gt2 FROM my_table;
2603 first | second | gt2
2604 -------+--------+-----
2605 1 | one | f
2606 2 | two | f
2607 3 | three | t
2608 4 | four | t
2609 (4 rows)
2610 testdb=> \crosstabview first second
2612 first | one | two | three | four
2613 -------+-----+-----+-------+------
2614 1 | f | | |
2615 2 | | f | |
2616 3 | | | t |
2617 4 | | | | t
2618 (4 rows)
2619 This second example shows a multiplication table with rows sorted in
2621 reverse numerical order and columns with an independent, ascending
2622 numerical order.
2623 testdb=> SELECT t1.first as "A", t2.first+100 AS "B", t1.first*(t2.first+100) as "AxB",
2625 testdb(> row_number() over(order by t2.first) AS ord
2626 testdb(> FROM my_table t1 CROSS JOIN my_table t2 ORDER BY 1 DESC
2627 testdb(> \crosstabview "A" "B" "AxB" ord
2628 A | 101 | 102 | 103 | 104
2629 ---+-----+-----+-----+-----
2630 4 | 404 | 408 | 412 | 416
2631 3 | 303 | 306 | 309 | 312
2632 2 | 202 | 204 | 206 | 208
2633 1 | 101 | 102 | 103 | 104
2634 (4 rows)
2635PostgreSQL 13.4 2021 PSQL(1)