1PSQL(1) PostgreSQL 12.6 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
82 Switches to CSV (Comma-Separated Values) output mode. This is
83 equivalent to \pset format csv.
84 -d dbname
86 --dbname=dbname
87 Specifies the name of the database to connect to. This is
88 equivalent to specifying dbname as the first non-option argument on
89 the command line. The dbname can be a connection string. If so,
90 connection string parameters will override any conflicting command
91 line options.
92 -e
94 --echo-queries
95 Copy all SQL commands sent to the server to standard output as
96 well. This is equivalent to setting the variable ECHO to queries.
97 -E
99 --echo-hidden
100 Echo the actual queries generated by \d and other backslash
101 commands. You can use this to study psql's internal operations.
102 This is equivalent to setting the variable ECHO_HIDDEN to on.
103 -f filename
105 --file=filename
106 Read commands from the file filename, rather than standard input.
107 This option can be repeated and combined in any order with the -c
108 option. When either -c or -f is specified, psql does not read
109 commands from standard input; instead it terminates after
110 processing all the -c and -f options in sequence. Except for that,
111 this option is largely equivalent to the meta-command \i.
112 If filename is - (hyphen), then standard input is read until an EOF
114 indication or \q meta-command. This can be used to intersperse
115 interactive input with input from files. Note however that Readline
116 is not used in this case (much as if -n had been specified).
117 Using this option is subtly different from writing psql < filename.
119 In general, both will do what you expect, but using -f enables some
120 nice features such as error messages with line numbers. There is
121 also a slight chance that using this option will reduce the
122 start-up overhead. On the other hand, the variant using the shell's
123 input redirection is (in theory) guaranteed to yield exactly the
124 same output you would have received had you entered everything by
125 hand.
126 -F separator
128 --field-separator=separator
129 Use separator as the field separator for unaligned output. This is
130 equivalent to \pset fieldsep or \f.
131 -h hostname
133 --host=hostname
134 Specifies the host name of the machine on which the server is
135 running. If the value begins with a slash, it is used as the
136 directory for the Unix-domain socket.
137 -H
139 --html
140 Switches to HTML output mode. This is equivalent to \pset format
141 html or the \H command.
142 -l
144 --list
145 List all available databases, then exit. Other non-connection
146 options are ignored. This is similar to the meta-command \list.
147 When this option is used, psql will connect to the database
149 postgres, unless a different database is named on the command line
150 (option -d or non-option argument, possibly via a service entry,
151 but not via an environment variable).
152 -L filename
154 --log-file=filename
155 Write all query output into file filename, in addition to the
156 normal output destination.
157 -n
159 --no-readline
160 Do not use Readline for line editing and do not use the command
161 history. This can be useful to turn off tab expansion when cutting
162 and pasting.
163 -o filename
165 --output=filename
166 Put all query output into file filename. This is equivalent to the
167 command \o.
168 -p port
170 --port=port
171 Specifies the TCP port or the local Unix-domain socket file
172 extension on which the server is listening for connections.
173 Defaults to the value of the PGPORT environment variable or, if not
174 set, to the port specified at compile time, usually 5432.
175 -P assignment
177 --pset=assignment
178 Specifies printing options, in the style of \pset. Note that here
179 you have to separate name and value with an equal sign instead of a
180 space. For example, to set the output format to LaTeX, you could
181 write -P format=latex.
182 -q
184 --quiet
185 Specifies that psql should do its work quietly. By default, it
186 prints welcome messages and various informational output. If this
187 option is used, none of this happens. This is useful with the -c
188 option. This is equivalent to setting the variable QUIET to on.
189 -R separator
191 --record-separator=separator
192 Use separator as the record separator for unaligned output. This is
193 equivalent to \pset recordsep.
194 -s
196 --single-step
197 Run in single-step mode. That means the user is prompted before
198 each command is sent to the server, with the option to cancel
199 execution as well. Use this to debug scripts.
200 -S
202 --single-line
203 Runs in single-line mode where a newline terminates an SQL command,
204 as a semicolon does.
205 Note
207 This mode is provided for those who insist on it, but you are
208 not necessarily encouraged to use it. In particular, if you mix
209 SQL and meta-commands on a line the order of execution might
210 not always be clear to the inexperienced user.
211 -t
213 --tuples-only
214 Turn off printing of column names and result row count footers,
215 etc. This is equivalent to \t or \pset tuples_only.
216 -T table_options
218 --table-attr=table_options
219 Specifies options to be placed within the HTML table tag. See \pset
220 tableattr for details.
221 -U username
223 --username=username
224 Connect to the database as the user username instead of the
225 default. (You must have permission to do so, of course.)
226 -v assignment
228 --set=assignment
229 --variable=assignment
230 Perform a variable assignment, like the \set meta-command. Note
231 that you must separate name and value, if any, by an equal sign on
232 the command line. To unset a variable, leave off the equal sign. To
233 set a variable with an empty value, use the equal sign but leave
234 off the value. These assignments are done during command line
235 processing, so variables that reflect connection state will get
236 overwritten later.
237 -V
239 --version
240 Print the psql version and exit.
241 -w
243 --no-password
244 Never issue a password prompt. If the server requires password
245 authentication and a password is not available from other sources
246 such as a .pgpass file, the connection attempt will fail. This
247 option can be useful in batch jobs and scripts where no user is
248 present to enter a password.
249 Note that this option will remain set for the entire session, and
251 so it affects uses of the meta-command \connect as well as the
252 initial connection attempt.
253 -W
255 --password
256 Force psql to prompt for a password before connecting to a
257 database, even if the password will not be used.
258 If the server requires password authentication and a password is
260 not available from other sources such as a .pgpass file, psql will
261 prompt for a password in any case. However, psql will waste a
262 connection attempt finding out that the server wants a password. In
263 some cases it is worth typing -W to avoid the extra connection
264 attempt.
265 Note that this option will remain set for the entire session, and
267 so it affects uses of the meta-command \connect as well as the
268 initial connection attempt.
269 -x
271 --expanded
272 Turn on the expanded table formatting mode. This is equivalent to
273 \x or \pset expanded.
274 -X,
276 --no-psqlrc
277 Do not read the start-up file (neither the system-wide psqlrc file
278 nor the user's ~/.psqlrc file).
279 -z
281 --field-separator-zero
282 Set the field separator for unaligned output to a zero byte. This
283 is equivalent to \pset fieldsep_zero.
284 -0
286 --record-separator-zero
287 Set the record separator for unaligned output to a zero byte. This
288 is useful for interfacing, for example, with xargs -0. This is
289 equivalent to \pset recordsep_zero.
290 -1
292 --single-transaction
293 This option can only be used in combination with one or more -c
294 and/or -f options. It causes psql to issue a BEGIN command before
295 the first such option and a COMMIT command after the last one,
296 thereby wrapping all the commands into a single transaction. This
297 ensures that either all the commands complete successfully, or no
298 changes are applied.
299 If the commands themselves contain BEGIN, COMMIT, or ROLLBACK, this
301 option will not have the desired effects. Also, if an individual
302 command cannot be executed inside a transaction block, specifying
303 this option will cause the whole transaction to fail.
304 -?
306 --help[=topic]
307 Show help about psql and exit. The optional topic parameter
308 (defaulting to options) selects which part of psql is explained:
309 commands describes psql's backslash commands; options describes the
310 command-line options that can be passed to psql; and variables
311 shows help about psql configuration variables.
312
314 psql returns 0 to the shell if it finished normally, 1 if a fatal error
315 of its own occurs (e.g., out of memory, file not found), 2 if the
316 connection to the server went bad and the session was not interactive,
317 and 3 if an error occurred in a script and the variable ON_ERROR_STOP
318 was set.
319
321 Connecting to a Database
322 psql is a regular PostgreSQL client application. In order to connect to
323 a database you need to know the name of your target database, the host
324 name and port number of the server, and what user name you want to
325 connect as. psql can be told about those parameters via command line
326 options, namely -d, -h, -p, and -U respectively. If an argument is
327 found that does not belong to any option it will be interpreted as the
328 database name (or the user name, if the database name is already
329 given). Not all of these options are required; there are useful
330 defaults. If you omit the host name, psql will connect via a
331 Unix-domain socket to a server on the local host, or via TCP/IP to
332 localhost on machines that don't have Unix-domain sockets. The default
333 port number is determined at compile time. Since the database server
334 uses the same default, you will not have to specify the port in most
335 cases. The default user name is your operating-system user name, as is
336 the default database name. Note that you cannot just connect to any
337 database under any user name. Your database administrator should have
338 informed you about your access rights.
339 When the defaults aren't quite right, you can save yourself some typing
341 by setting the environment variables PGDATABASE, PGHOST, PGPORT and/or
342 PGUSER to appropriate values. (For additional environment variables,
343 see Section 33.14.) It is also convenient to have a ~/.pgpass file to
344 avoid regularly having to type in passwords. See Section 33.15 for more
345 information.
346 An alternative way to specify connection parameters is in a conninfo
348 string or a URI, which is used instead of a database name. This
349 mechanism give you very wide control over the connection. For example:
350 $ psql "service=myservice sslmode=require"
352 $ psql postgresql://dbmaster:5433/mydb?sslmode=require
353 This way you can also use LDAP for connection parameter lookup as
355 described in Section 33.17. See Section 33.1.2 for more information on
356 all the available connection options.
357 If the connection could not be made for any reason (e.g., insufficient
359 privileges, server is not running on the targeted host, etc.), psql
360 will return an error and terminate.
361 If both standard input and standard output are a terminal, then psql
363 sets the client encoding to “auto”, which will detect the appropriate
364 client encoding from the locale settings (LC_CTYPE environment variable
365 on Unix systems). If this doesn't work out as expected, the client
366 encoding can be overridden using the environment variable
367 PGCLIENTENCODING.
368 Entering SQL Commands
370 In normal operation, psql provides a prompt with the name of the
371 database to which psql is currently connected, followed by the string
372 =>. For example:
373 $ psql testdb
375 psql (12.6)
376 Type "help" for help.
377 testdb=>
379 At the prompt, the user can type in SQL commands. Ordinarily, input
381 lines are sent to the server when a command-terminating semicolon is
382 reached. An end of line does not terminate a command. Thus commands can
383 be spread over several lines for clarity. If the command was sent and
384 executed without error, the results of the command are displayed on the
385 screen.
386 If untrusted users have access to a database that has not adopted a
388 secure schema usage pattern, begin your session by removing
389 publicly-writable schemas from search_path. One can add
390 options=-csearch_path= to the connection string or issue SELECT
391 pg_catalog.set_config('search_path', '', false) before other SQL
392 commands. This consideration is not specific to psql; it applies to
393 every interface for executing arbitrary SQL commands.
394 Whenever a command is executed, psql also polls for asynchronous
396 notification events generated by LISTEN(7) and NOTIFY(7).
397 While C-style block comments are passed to the server for processing
399 and removal, SQL-standard comments are removed by psql.
400 Meta-Commands
402 Anything you enter in psql that begins with an unquoted backslash is a
403 psql meta-command that is processed by psql itself. These commands make
404 psql more useful for administration or scripting. Meta-commands are
405 often called slash or backslash commands.
406 The format of a psql command is the backslash, followed immediately by
408 a command verb, then any arguments. The arguments are separated from
409 the command verb and each other by any number of whitespace characters.
410 To include whitespace in an argument you can quote it with single
412 quotes. To include a single quote in an argument, write two single
413 quotes within single-quoted text. Anything contained in single quotes
414 is furthermore subject to C-like substitutions for \n (new line), \t
415 (tab), \b (backspace), \r (carriage return), \f (form feed), \digits
416 (octal), and \xdigits (hexadecimal). A backslash preceding any other
417 character within single-quoted text quotes that single character,
418 whatever it is.
419 If an unquoted colon (:) followed by a psql variable name appears
421 within an argument, it is replaced by the variable's value, as
422 described in SQL Interpolation. The forms :'variable_name' and
423 :"variable_name" described there work as well. The :{?variable_name}
424 syntax allows testing whether a variable is defined. It is substituted
425 by TRUE or FALSE. Escaping the colon with a backslash protects it from
426 substitution.
427 Within an argument, text that is enclosed in backquotes (`) is taken as
429 a command line that is passed to the shell. The output of the command
430 (with any trailing newline removed) replaces the backquoted text.
431 Within the text enclosed in backquotes, no special quoting or other
432 processing occurs, except that appearances of :variable_name where
433 variable_name is a psql variable name are replaced by the variable's
434 value. Also, appearances of :'variable_name' are replaced by the
435 variable's value suitably quoted to become a single shell command
436 argument. (The latter form is almost always preferable, unless you are
437 very sure of what is in the variable.) Because carriage return and line
438 feed characters cannot be safely quoted on all platforms, the
439 :'variable_name' form prints an error message and does not substitute
440 the variable value when such characters appear in the value.
441 Some commands take an SQL identifier (such as a table name) as
443 argument. These arguments follow the syntax rules of SQL: Unquoted
444 letters are forced to lowercase, while double quotes (") protect
445 letters from case conversion and allow incorporation of whitespace into
446 the identifier. Within double quotes, paired double quotes reduce to a
447 single double quote in the resulting name. For example, FOO"BAR"BAZ is
448 interpreted as fooBARbaz, and "A weird"" name" becomes A weird" name.
449 Parsing for arguments stops at the end of the line, or when another
451 unquoted backslash is found. An unquoted backslash is taken as the
452 beginning of a new meta-command. The special sequence \\ (two
453 backslashes) marks the end of arguments and continues parsing SQL
454 commands, if any. That way SQL and psql commands can be freely mixed on
455 a line. But in any case, the arguments of a meta-command cannot
456 continue beyond the end of the line.
457 Many of the meta-commands act on the current query buffer. This is
459 simply a buffer holding whatever SQL command text has been typed but
460 not yet sent to the server for execution. This will include previous
461 input lines as well as any text appearing before the meta-command on
462 the same line.
463 The following meta-commands are defined:
465 \a
467 If the current table output format is unaligned, it is switched to
468 aligned. If it is not unaligned, it is set to unaligned. This
469 command is kept for backwards compatibility. See \pset for a more
470 general solution.
471 \c or \connect [ -reuse-previous=on|off ] [ dbname [ username ] [ host
473 ] [ port ] | conninfo ]
474 Establishes a new connection to a PostgreSQL server. The connection
475 parameters to use can be specified either using a positional syntax
476 (one or more of database name, user, host, and port), or using a
477 conninfo connection string as detailed in Section 33.1.1. If no
478 arguments are given, a new connection is made using the same
479 parameters as before.
480 Specifying any of dbname, username, host or port as - is equivalent
482 to omitting that parameter.
483 The new connection can re-use connection parameters from the
485 previous connection; not only database name, user, host, and port,
486 but other settings such as sslmode. By default, parameters are
487 re-used in the positional syntax, but not when a conninfo string is
488 given. Passing a first argument of -reuse-previous=on or
489 -reuse-previous=off overrides that default. If parameters are
490 re-used, then any parameter not explicitly specified as a
491 positional parameter or in the conninfo string is taken from the
492 existing connection's parameters. An exception is that if the host
493 setting is changed from its previous value using the positional
494 syntax, any hostaddr setting present in the existing connection's
495 parameters is dropped. Also, any password used for the existing
496 connection will be re-used only if the user, host, and port
497 settings are not changed. When the command neither specifies nor
498 reuses a particular parameter, the libpq default is used.
499 If the new connection is successfully made, the previous connection
501 is closed. If the connection attempt fails (wrong user name, access
502 denied, etc.), the previous connection will be kept if psql is in
503 interactive mode. But when executing a non-interactive script,
504 processing will immediately stop with an error. This distinction
505 was chosen as a user convenience against typos on the one hand, and
506 a safety mechanism that scripts are not accidentally acting on the
507 wrong database on the other hand.
508 Examples:
510 => \c mydb myuser host.dom 6432
512 => \c service=foo
513 => \c "host=localhost port=5432 dbname=mydb connect_timeout=10 sslmode=disable"
514 => \c -reuse-previous=on sslmode=require -- changes only sslmode
515 => \c postgresql://tom@localhost/mydb?application_name=myapp
516 \C [ title ]
518 Sets the title of any tables being printed as the result of a query
519 or unset any such title. This command is equivalent to \pset title
520 title. (The name of this command derives from “caption”, as it was
521 previously only used to set the caption in an HTML table.)
522 \cd [ directory ]
524 Changes the current working directory to directory. Without
525 argument, changes to the current user's home directory.
526 Tip
528 To print your current working directory, use \! pwd.
529 \conninfo
531 Outputs information about the current database connection.
532 \copy { table [ ( column_list ) ] } from { 'filename' | program
534 'command' | stdin | pstdin } [ [ with ] ( option [, ...] ) ] [ where
535 condition ]
536 \copy { table [ ( column_list ) ] | ( query ) } to { 'filename' |
537 program 'command' | stdout | pstdout } [ [ with ] ( option [, ...] ) ]
538 Performs a frontend (client) copy. This is an operation that runs
539 an SQL COPY(7) command, but instead of the server reading or
540 writing the specified file, psql reads or writes the file and
541 routes the data between the server and the local file system. This
542 means that file accessibility and privileges are those of the local
543 user, not the server, and no SQL superuser privileges are required.
544 When program is specified, command is executed by psql and the data
546 passed from or to command is routed between the server and the
547 client. Again, the execution privileges are those of the local
548 user, not the server, and no SQL superuser privileges are required.
549 For \copy ... from stdin, data rows are read from the same source
551 that issued the command, continuing until \. is read or the stream
552 reaches EOF. This option is useful for populating tables in-line
553 within a SQL script file. For \copy ... to stdout, output is sent
554 to the same place as psql command output, and the COPY count
555 command status is not printed (since it might be confused with a
556 data row). To read/write psql's standard input or output regardless
557 of the current command source or \o option, write from pstdin or to
558 pstdout.
559 The syntax of this command is similar to that of the SQL COPY(7)
561 command. All options other than the data source/destination are as
562 specified for COPY(7). Because of this, special parsing rules apply
563 to the \copy meta-command. Unlike most other meta-commands, the
564 entire remainder of the line is always taken to be the arguments of
565 \copy, and neither variable interpolation nor backquote expansion
566 are performed in the arguments.
567 Tip
569 Another way to obtain the same result as \copy ... to is to use
570 the SQL COPY ... TO STDOUT command and terminate it with \g
571 filename or \g |program. Unlike \copy, this method allows the
572 command to span multiple lines; also, variable interpolation
573 and backquote expansion can be used.
574 Tip
576 These operations are not as efficient as the SQL COPY command
577 with a file or program data source or destination, because all
578 data must pass through the client/server connection. For large
579 amounts of data the SQL command might be preferable.
580 \copyright
582 Shows the copyright and distribution terms of PostgreSQL.
583 \crosstabview [ colV [ colH [ colD [ sortcolH ] ] ] ]
585 Executes the current query buffer (like \g) and shows the results
586 in a crosstab grid. The query must return at least three columns.
587 The output column identified by colV becomes a vertical header and
588 the output column identified by colH becomes a horizontal header.
589 colD identifies the output column to display within the grid.
590 sortcolH identifies an optional sort column for the horizontal
591 header.
592 Each column specification can be a column number (starting at 1) or
594 a column name. The usual SQL case folding and quoting rules apply
595 to column names. If omitted, colV is taken as column 1 and colH as
596 column 2. colH must differ from colV. If colD is not specified,
597 then there must be exactly three columns in the query result, and
598 the column that is neither colV nor colH is taken to be colD.
599 The vertical header, displayed as the leftmost column, contains the
601 values found in column colV, in the same order as in the query
602 results, but with duplicates removed.
603 The horizontal header, displayed as the first row, contains the
605 values found in column colH, with duplicates removed. By default,
606 these appear in the same order as in the query results. But if the
607 optional sortcolH argument is given, it identifies a column whose
608 values must be integer numbers, and the values from colH will
609 appear in the horizontal header sorted according to the
610 corresponding sortcolH values.
611 Inside the crosstab grid, for each distinct value x of colH and
613 each distinct value y of colV, the cell located at the intersection
614 (x,y) contains the value of the colD column in the query result row
615 for which the value of colH is x and the value of colV is y. If
616 there is no such row, the cell is empty. If there are multiple such
617 rows, an error is reported.
618 \d[S+] [ pattern ]
620 For each relation (table, view, materialized view, index, sequence,
621 or foreign table) or composite type matching the pattern, show all
622 columns, their types, the tablespace (if not the default) and any
623 special attributes such as NOT NULL or defaults. Associated
624 indexes, constraints, rules, and triggers are also shown. For
625 foreign tables, the associated foreign server is shown as well.
626 (“Matching the pattern” is defined in Patterns below.)
627 For some types of relation, \d shows additional information for
629 each column: column values for sequences, indexed expressions for
630 indexes, and foreign data wrapper options for foreign tables.
631 The command form \d+ is identical, except that more information is
633 displayed: any comments associated with the columns of the table
634 are shown, as is the presence of OIDs in the table, the view
635 definition if the relation is a view, a non-default replica
636 identity setting.
637 By default, only user-created objects are shown; supply a pattern
639 or the S modifier to include system objects.
640 Note
642 If \d is used without a pattern argument, it is equivalent to
643 \dtvmsE which will show a list of all visible tables, views,
644 materialized views, sequences and foreign tables. This is
645 purely a convenience measure.
646 \da[S] [ pattern ]
648 Lists aggregate functions, together with their return type and the
649 data types they operate on. If pattern is specified, only
650 aggregates whose names match the pattern are shown. By default,
651 only user-created objects are shown; supply a pattern or the S
652 modifier to include system objects.
653 \dA[+] [ pattern ]
655 Lists access methods. If pattern is specified, only access methods
656 whose names match the pattern are shown. If + is appended to the
657 command name, each access method is listed with its associated
658 handler function and description.
659 \db[+] [ pattern ]
661 Lists tablespaces. If pattern is specified, only tablespaces whose
662 names match the pattern are shown. If + is appended to the command
663 name, each tablespace is listed with its associated options,
664 on-disk size, permissions and description.
665 \dc[S+] [ pattern ]
667 Lists conversions between character-set encodings. If pattern is
668 specified, only conversions whose names match the pattern are
669 listed. By default, only user-created objects are shown; supply a
670 pattern or the S modifier to include system objects. If + is
671 appended to the command name, each object is listed with its
672 associated description.
673 \dC[+] [ pattern ]
675 Lists type casts. If pattern is specified, only casts whose source
676 or target types match the pattern are listed. If + is appended to
677 the command name, each object is listed with its associated
678 description.
679 \dd[S] [ pattern ]
681 Shows the descriptions of objects of type constraint, operator
682 class, operator family, rule, and trigger. All other comments may
683 be viewed by the respective backslash commands for those object
684 types.
685 \dd displays descriptions for objects matching the pattern, or of
687 visible objects of the appropriate type if no argument is given.
688 But in either case, only objects that have a description are
689 listed. By default, only user-created objects are shown; supply a
690 pattern or the S modifier to include system objects.
691 Descriptions for objects can be created with the COMMENT(7) SQL
693 command.
694 \dD[S+] [ pattern ]
696 Lists domains. If pattern is specified, only domains whose names
697 match the pattern are shown. By default, only user-created objects
698 are shown; supply a pattern or the S modifier to include system
699 objects. If + is appended to the command name, each object is
700 listed with its associated permissions and description.
701 \ddp [ pattern ]
703 Lists default access privilege settings. An entry is shown for each
704 role (and schema, if applicable) for which the default privilege
705 settings have been changed from the built-in defaults. If pattern
706 is specified, only entries whose role name or schema name matches
707 the pattern are listed.
708 The ALTER DEFAULT PRIVILEGES (ALTER_DEFAULT_PRIVILEGES(7)) command
710 is used to set default access privileges. The meaning of the
711 privilege display is explained in Section 5.7.
712 \dE[S+] [ pattern ]
714 \di[S+] [ pattern ]
715 \dm[S+] [ pattern ]
716 \ds[S+] [ pattern ]
717 \dt[S+] [ pattern ]
718 \dv[S+] [ pattern ]
719 In this group of commands, the letters E, i, m, s, t, and v stand
720 for foreign table, index, materialized view, sequence, table, and
721 view, respectively. You can specify any or all of these letters, in
722 any order, to obtain a listing of objects of these types. For
723 example, \dit lists indexes and tables. If + is appended to the
724 command name, each object is listed with its physical size on disk
725 and its associated description, if any. If pattern is specified,
726 only objects whose names match the pattern are listed. By default,
727 only user-created objects are shown; supply a pattern or the S
728 modifier to include system objects.
729 \des[+] [ pattern ]
731 Lists foreign servers (mnemonic: “external servers”). If pattern is
732 specified, only those servers whose name matches the pattern are
733 listed. If the form \des+ is used, a full description of each
734 server is shown, including the server's access privileges, type,
735 version, options, and description.
736 \det[+] [ pattern ]
738 Lists foreign tables (mnemonic: “external tables”). If pattern is
739 specified, only entries whose table name or schema name matches the
740 pattern are listed. If the form \det+ is used, generic options and
741 the foreign table description are also displayed.
742 \deu[+] [ pattern ]
744 Lists user mappings (mnemonic: “external users”). If pattern is
745 specified, only those mappings whose user names match the pattern
746 are listed. If the form \deu+ is used, additional information about
747 each mapping is shown.
748 Caution
750 \deu+ might also display the user name and password of the
751 remote user, so care should be taken not to disclose them.
752 \dew[+] [ pattern ]
754 Lists foreign-data wrappers (mnemonic: “external wrappers”). If
755 pattern is specified, only those foreign-data wrappers whose name
756 matches the pattern are listed. If the form \dew+ is used, the
757 access privileges, options, and description of the foreign-data
758 wrapper are also shown.
759 \df[anptwS+] [ pattern ]
761 Lists functions, together with their result data types, argument
762 data types, and function types, which are classified as “agg”
763 (aggregate), “normal”, “procedure”, “trigger”, or “window”. To
764 display only functions of specific type(s), add the corresponding
765 letters a, n, p, t, or w to the command. If pattern is specified,
766 only functions whose names match the pattern are shown. By default,
767 only user-created objects are shown; supply a pattern or the S
768 modifier to include system objects. If the form \df+ is used,
769 additional information about each function is shown, including
770 volatility, parallel safety, owner, security classification, access
771 privileges, language, source code and description.
772 Tip
774 To look up functions taking arguments or returning values of a
775 specific data type, use your pager's search capability to
776 scroll through the \df output.
777 \dF[+] [ pattern ]
779 Lists text search configurations. If pattern is specified, only
780 configurations whose names match the pattern are shown. If the form
781 \dF+ is used, a full description of each configuration is shown,
782 including the underlying text search parser and the dictionary list
783 for each parser token type.
784 \dFd[+] [ pattern ]
786 Lists text search dictionaries. If pattern is specified, only
787 dictionaries whose names match the pattern are shown. If the form
788 \dFd+ is used, additional information is shown about each selected
789 dictionary, including the underlying text search template and the
790 option values.
791 \dFp[+] [ pattern ]
793 Lists text search parsers. If pattern is specified, only parsers
794 whose names match the pattern are shown. If the form \dFp+ is used,
795 a full description of each parser is shown, including the
796 underlying functions and the list of recognized token types.
797 \dFt[+] [ pattern ]
799 Lists text search templates. If pattern is specified, only
800 templates whose names match the pattern are shown. If the form
801 \dFt+ is used, additional information is shown about each template,
802 including the underlying function names.
803 \dg[S+] [ pattern ]
805 Lists database roles. (Since the concepts of “users” and “groups”
806 have been unified into “roles”, this command is now equivalent to
807 \du.) By default, only user-created roles are shown; supply the S
808 modifier to include system roles. If pattern is specified, only
809 those roles whose names match the pattern are listed. If the form
810 \dg+ is used, additional information is shown about each role;
811 currently this adds the comment for each role.
812 \dl
814 This is an alias for \lo_list, which shows a list of large objects.
815 \dL[S+] [ pattern ]
817 Lists procedural languages. If pattern is specified, only languages
818 whose names match the pattern are listed. By default, only
819 user-created languages are shown; supply the S modifier to include
820 system objects. If + is appended to the command name, each language
821 is listed with its call handler, validator, access privileges, and
822 whether it is a system object.
823 \dn[S+] [ pattern ]
825 Lists schemas (namespaces). If pattern is specified, only schemas
826 whose names match the pattern are listed. By default, only
827 user-created objects are shown; supply a pattern or the S modifier
828 to include system objects. If + is appended to the command name,
829 each object is listed with its associated permissions and
830 description, if any.
831 \do[S+] [ pattern ]
833 Lists operators with their operand and result types. If pattern is
834 specified, only operators whose names match the pattern are listed.
835 By default, only user-created objects are shown; supply a pattern
836 or the S modifier to include system objects. If + is appended to
837 the command name, additional information about each operator is
838 shown, currently just the name of the underlying function.
839 \dO[S+] [ pattern ]
841 Lists collations. If pattern is specified, only collations whose
842 names match the pattern are listed. By default, only user-created
843 objects are shown; supply a pattern or the S modifier to include
844 system objects. If + is appended to the command name, each
845 collation is listed with its associated description, if any. Note
846 that only collations usable with the current database's encoding
847 are shown, so the results may vary in different databases of the
848 same installation.
849 \dp [ pattern ]
851 Lists tables, views and sequences with their associated access
852 privileges. If pattern is specified, only tables, views and
853 sequences whose names match the pattern are listed.
854 The GRANT(7) and REVOKE(7) commands are used to set access
856 privileges. The meaning of the privilege display is explained in
857 Section 5.7.
858 \dP[itn+] [ pattern ]
860 Lists partitioned relations. If pattern is specified, only entries
861 whose name matches the pattern are listed. The modifiers t (tables)
862 and i (indexes) can be appended to the command, filtering the kind
863 of relations to list. By default, partitioned tables and indexes
864 are listed.
865 If the modifier n (“nested”) is used, or a pattern is specified,
867 then non-root partitioned relations are included, and a column is
868 shown displaying the parent of each partitioned relation.
869 If + is appended to the command name, the sum of the sizes of each
871 relation's partitions is also displayed, along with the relation's
872 description. If n is combined with +, two sizes are shown: one
873 including the total size of directly-attached leaf partitions, and
874 another showing the total size of all partitions, including
875 indirectly attached sub-partitions.
876 \drds [ role-pattern [ database-pattern ] ]
878 Lists defined configuration settings. These settings can be
879 role-specific, database-specific, or both. role-pattern and
880 database-pattern are used to select specific roles and databases to
881 list, respectively. If omitted, or if * is specified, all settings
882 are listed, including those not role-specific or database-specific,
883 respectively.
884 The ALTER ROLE (ALTER_ROLE(7)) and ALTER DATABASE
886 (ALTER_DATABASE(7)) commands are used to define per-role and
887 per-database configuration settings.
888 \dRp[+] [ pattern ]
890 Lists replication publications. If pattern is specified, only those
891 publications whose names match the pattern are listed. If + is
892 appended to the command name, the tables associated with each
893 publication are shown as well.
894 \dRs[+] [ pattern ]
896 Lists replication subscriptions. If pattern is specified, only
897 those subscriptions whose names match the pattern are listed. If +
898 is appended to the command name, additional properties of the
899 subscriptions are shown.
900 \dT[S+] [ pattern ]
902 Lists data types. If pattern is specified, only types whose names
903 match the pattern are listed. If + is appended to the command name,
904 each type is listed with its internal name and size, its allowed
905 values if it is an enum type, and its associated permissions. By
906 default, only user-created objects are shown; supply a pattern or
907 the S modifier to include system objects.
908 \du[S+] [ pattern ]
910 Lists database roles. (Since the concepts of “users” and “groups”
911 have been unified into “roles”, this command is now equivalent to
912 \dg.) By default, only user-created roles are shown; supply the S
913 modifier to include system roles. If pattern is specified, only
914 those roles whose names match the pattern are listed. If the form
915 \du+ is used, additional information is shown about each role;
916 currently this adds the comment for each role.
917 \dx[+] [ pattern ]
919 Lists installed extensions. If pattern is specified, only those
920 extensions whose names match the pattern are listed. If the form
921 \dx+ is used, all the objects belonging to each matching extension
922 are listed.
923 \dy[+] [ pattern ]
925 Lists event triggers. If pattern is specified, only those event
926 triggers whose names match the pattern are listed. If + is appended
927 to the command name, each object is listed with its associated
928 description.
929 \e or \edit [ filename ] [ line_number ]
931 If filename is specified, the file is edited; after the editor
932 exits, the file's content is copied into the current query buffer.
933 If no filename is given, the current query buffer is copied to a
934 temporary file which is then edited in the same fashion. Or, if the
935 current query buffer is empty, the most recently executed query is
936 copied to a temporary file and edited in the same fashion.
937 The new contents of the query buffer are then re-parsed according
939 to the normal rules of psql, treating the whole buffer as a single
940 line. Any complete queries are immediately executed; that is, if
941 the query buffer contains or ends with a semicolon, everything up
942 to that point is executed. Whatever remains will wait in the query
943 buffer; type semicolon or \g to send it, or \r to cancel it by
944 clearing the query buffer. Treating the buffer as a single line
945 primarily affects meta-commands: whatever is in the buffer after a
946 meta-command will be taken as argument(s) to the meta-command, even
947 if it spans multiple lines. (Thus you cannot make
948 meta-command-using scripts this way. Use \i for that.)
949 If a line number is specified, psql will position the cursor on the
951 specified line of the file or query buffer. Note that if a single
952 all-digits argument is given, psql assumes it is a line number, not
953 a file name.
954 Tip
956 See under ENVIRONMENT for how to configure and customize your
957 editor.
958 \echo text [ ... ]
960 Prints the arguments to the standard output, separated by one space
961 and followed by a newline. This can be useful to intersperse
962 information in the output of scripts. For example:
963 => \echo `date`
965 Tue Oct 26 21:40:57 CEST 1999
966 If the first argument is an unquoted -n the trailing newline is not
968 written.
969 Tip
971 If you use the \o command to redirect your query output you
972 might wish to use \qecho instead of this command.
973 \ef [ function_description [ line_number ] ]
975 This command fetches and edits the definition of the named function
976 or procedure, in the form of a CREATE OR REPLACE FUNCTION or CREATE
977 OR REPLACE PROCEDURE command. Editing is done in the same way as
978 for \edit. After the editor exits, the updated command waits in the
979 query buffer; type semicolon or \g to send it, or \r to cancel.
980 The target function can be specified by name alone, or by name and
982 arguments, for example foo(integer, text). The argument types must
983 be given if there is more than one function of the same name.
984 If no function is specified, a blank CREATE FUNCTION template is
986 presented for editing.
987 If a line number is specified, psql will position the cursor on the
989 specified line of the function body. (Note that the function body
990 typically does not begin on the first line of the file.)
991 Unlike most other meta-commands, the entire remainder of the line
993 is always taken to be the argument(s) of \ef, and neither variable
994 interpolation nor backquote expansion are performed in the
995 arguments.
996 Tip
998 See under ENVIRONMENT for how to configure and customize your
999 editor.
1000 \encoding [ encoding ]
1002 Sets the client character set encoding. Without an argument, this
1003 command shows the current encoding.
1004 \errverbose
1006 Repeats the most recent server error message at maximum verbosity,
1007 as though VERBOSITY were set to verbose and SHOW_CONTEXT were set
1008 to always.
1009 \ev [ view_name [ line_number ] ]
1011 This command fetches and edits the definition of the named view, in
1012 the form of a CREATE OR REPLACE VIEW command. Editing is done in
1013 the same way as for \edit. After the editor exits, the updated
1014 command waits in the query buffer; type semicolon or \g to send it,
1015 or \r to cancel.
1016 If no view is specified, a blank CREATE VIEW template is presented
1018 for editing.
1019 If a line number is specified, psql will position the cursor on the
1021 specified line of the view definition.
1022 Unlike most other meta-commands, the entire remainder of the line
1024 is always taken to be the argument(s) of \ev, and neither variable
1025 interpolation nor backquote expansion are performed in the
1026 arguments.
1027 \f [ string ]
1029 Sets the field separator for unaligned query output. The default is
1030 the vertical bar (|). It is equivalent to \pset fieldsep.
1031 \g [ filename ]
1033 \g [ |command ]
1034 Sends the current query buffer to the server for execution. If an
1035 argument is given, the query's output is written to the named file
1036 or piped to the given shell command, instead of displaying it as
1037 usual. The file or command is written to only if the query
1038 successfully returns zero or more tuples, not if the query fails or
1039 is a non-data-returning SQL command.
1040 If the current query buffer is empty, the most recently sent query
1042 is re-executed instead. Except for that behavior, \g without an
1043 argument is essentially equivalent to a semicolon. A \g with
1044 argument is a “one-shot” alternative to the \o command.
1045 If the argument begins with |, then the entire remainder of the
1047 line is taken to be the command to execute, and neither variable
1048 interpolation nor backquote expansion are performed in it. The rest
1049 of the line is simply passed literally to the shell.
1050 \gdesc
1052 Shows the description (that is, the column names and data types) of
1053 the result of the current query buffer. The query is not actually
1054 executed; however, if it contains some type of syntax error, that
1055 error will be reported in the normal way.
1056 If the current query buffer is empty, the most recently sent query
1058 is described instead.
1059 \gexec
1061 Sends the current query buffer to the server, then treats each
1062 column of each row of the query's output (if any) as a SQL
1063 statement to be executed. For example, to create an index on each
1064 column of my_table:
1065 => SELECT format('create index on my_table(%I)', attname)
1067 -> FROM pg_attribute
1068 -> WHERE attrelid = 'my_table'::regclass AND attnum > 0
1069 -> ORDER BY attnum
1070 -> \gexec
1071 CREATE INDEX
1072 CREATE INDEX
1073 CREATE INDEX
1074 CREATE INDEX
1075 The generated queries are executed in the order in which the rows
1077 are returned, and left-to-right within each row if there is more
1078 than one column. NULL fields are ignored. The generated queries are
1079 sent literally to the server for processing, so they cannot be psql
1080 meta-commands nor contain psql variable references. If any
1081 individual query fails, execution of the remaining queries
1082 continues unless ON_ERROR_STOP is set. Execution of each query is
1083 subject to ECHO processing. (Setting ECHO to all or queries is
1084 often advisable when using \gexec.) Query logging, single-step
1085 mode, timing, and other query execution features apply to each
1086 generated query as well.
1087 If the current query buffer is empty, the most recently sent query
1089 is re-executed instead.
1090 \gset [ prefix ]
1092 Sends the current query buffer to the server and stores the query's
1093 output into psql variables (see Variables). The query to be
1094 executed must return exactly one row. Each column of the row is
1095 stored into a separate variable, named the same as the column. For
1096 example:
1097 => SELECT 'hello' AS var1, 10 AS var2
1099 -> \gset
1100 => \echo :var1 :var2
1101 hello 10
1102 If you specify a prefix, that string is prepended to the query's
1104 column names to create the variable names to use:
1105 => SELECT 'hello' AS var1, 10 AS var2
1107 -> \gset result_
1108 => \echo :result_var1 :result_var2
1109 hello 10
1110 If a column result is NULL, the corresponding variable is unset
1112 rather than being set.
1113 If the query fails or does not return one row, no variables are
1115 changed.
1116 If the current query buffer is empty, the most recently sent query
1118 is re-executed instead.
1119 \gx [ filename ]
1121 \gx [ |command ]
1122 \gx is equivalent to \g, but forces expanded output mode for this
1123 query. See \x.
1124 \h or \help [ command ]
1126 Gives syntax help on the specified SQL command. If command is not
1127 specified, then psql will list all the commands for which syntax
1128 help is available. If command is an asterisk (*), then syntax help
1129 on all SQL commands is shown.
1130 Unlike most other meta-commands, the entire remainder of the line
1132 is always taken to be the argument(s) of \help, and neither
1133 variable interpolation nor backquote expansion are performed in the
1134 arguments.
1135 Note
1137 To simplify typing, commands that consists of several words do
1138 not have to be quoted. Thus it is fine to type \help alter
1139 table.
1140 \H or \html
1142 Turns on HTML query output format. If the HTML format is already
1143 on, it is switched back to the default aligned text format. This
1144 command is for compatibility and convenience, but see \pset about
1145 setting other output options.
1146 \i or \include filename
1148 Reads input from the file filename and executes it as though it had
1149 been typed on the keyboard.
1150 If filename is - (hyphen), then standard input is read until an EOF
1152 indication or \q meta-command. This can be used to intersperse
1153 interactive input with input from files. Note that Readline
1154 behavior will be used only if it is active at the outermost level.
1155 Note
1157 If you want to see the lines on the screen as they are read you
1158 must set the variable ECHO to all.
1159 \if expression
1161 \elif expression
1162 \else
1163 \endif
1164 This group of commands implements nestable conditional blocks. A
1165 conditional block must begin with an \if and end with an \endif. In
1166 between there may be any number of \elif clauses, which may
1167 optionally be followed by a single \else clause. Ordinary queries
1168 and other types of backslash commands may (and usually do) appear
1169 between the commands forming a conditional block.
1170 The \if and \elif commands read their argument(s) and evaluate them
1172 as a boolean expression. If the expression yields true then
1173 processing continues normally; otherwise, lines are skipped until a
1174 matching \elif, \else, or \endif is reached. Once an \if or \elif
1175 test has succeeded, the arguments of later \elif commands in the
1176 same block are not evaluated but are treated as false. Lines
1177 following an \else are processed only if no earlier matching \if or
1178 \elif succeeded.
1179 The expression argument of an \if or \elif command is subject to
1181 variable interpolation and backquote expansion, just like any other
1182 backslash command argument. After that it is evaluated like the
1183 value of an on/off option variable. So a valid value is any
1184 unambiguous case-insensitive match for one of: true, false, 1, 0,
1185 on, off, yes, no. For example, t, T, and tR will all be considered
1186 to be true.
1187 Expressions that do not properly evaluate to true or false will
1189 generate a warning and be treated as false.
1190 Lines being skipped are parsed normally to identify queries and
1192 backslash commands, but queries are not sent to the server, and
1193 backslash commands other than conditionals (\if, \elif, \else,
1194 \endif) are ignored. Conditional commands are checked only for
1195 valid nesting. Variable references in skipped lines are not
1196 expanded, and backquote expansion is not performed either.
1197 All the backslash commands of a given conditional block must appear
1199 in the same source file. If EOF is reached on the main input file
1200 or an \include-ed file before all local \if-blocks have been
1201 closed, then psql will raise an error.
1202 Here is an example:
1204 -- check for the existence of two separate records in the database and store
1206 -- the results in separate psql variables
1207 SELECT
1208 EXISTS(SELECT 1 FROM customer WHERE customer_id = 123) as is_customer,
1209 EXISTS(SELECT 1 FROM employee WHERE employee_id = 456) as is_employee
1210 \gset
1211 \if :is_customer
1212 SELECT * FROM customer WHERE customer_id = 123;
1213 \elif :is_employee
1214 \echo 'is not a customer but is an employee'
1215 SELECT * FROM employee WHERE employee_id = 456;
1216 \else
1217 \if yes
1218 \echo 'not a customer or employee'
1219 \else
1220 \echo 'this will never print'
1221 \endif
1222 \endif
1223 \ir or \include_relative filename
1225 The \ir command is similar to \i, but resolves relative file names
1226 differently. When executing in interactive mode, the two commands
1227 behave identically. However, when invoked from a script, \ir
1228 interprets file names relative to the directory in which the script
1229 is located, rather than the current working directory.
1230 \l[+] or \list[+] [ pattern ]
1232 List the databases in the server and show their names, owners,
1233 character set encodings, and access privileges. If pattern is
1234 specified, only databases whose names match the pattern are listed.
1235 If + is appended to the command name, database sizes, default
1236 tablespaces, and descriptions are also displayed. (Size information
1237 is only available for databases that the current user can connect
1238 to.)
1239 \lo_export loid filename
1241 Reads the large object with OID loid from the database and writes
1242 it to filename. Note that this is subtly different from the server
1243 function lo_export, which acts with the permissions of the user
1244 that the database server runs as and on the server's file system.
1245 Tip
1247 Use \lo_list to find out the large object's OID.
1248 \lo_import filename [ comment ]
1250 Stores the file into a PostgreSQL large object. Optionally, it
1251 associates the given comment with the object. Example:
1252 foo=> \lo_import '/home/peter/pictures/photo.xcf' 'a picture of me'
1254 lo_import 152801
1255 The response indicates that the large object received object ID
1257 152801, which can be used to access the newly-created large object
1258 in the future. For the sake of readability, it is recommended to
1259 always associate a human-readable comment with every object. Both
1260 OIDs and comments can be viewed with the \lo_list command.
1261 Note that this command is subtly different from the server-side
1263 lo_import because it acts as the local user on the local file
1264 system, rather than the server's user and file system.
1265 \lo_list
1267 Shows a list of all PostgreSQL large objects currently stored in
1268 the database, along with any comments provided for them.
1269 \lo_unlink loid
1271 Deletes the large object with OID loid from the database.
1272 Tip
1274 Use \lo_list to find out the large object's OID.
1275 \o or \out [ filename ]
1277 \o or \out [ |command ]
1278 Arranges to save future query results to the file filename or pipe
1279 future results to the shell command command. If no argument is
1280 specified, the query output is reset to the standard output.
1281 If the argument begins with |, then the entire remainder of the
1283 line is taken to be the command to execute, and neither variable
1284 interpolation nor backquote expansion are performed in it. The rest
1285 of the line is simply passed literally to the shell.
1286 “Query results” includes all tables, command responses, and notices
1288 obtained from the database server, as well as output of various
1289 backslash commands that query the database (such as \d); but not
1290 error messages.
1291 Tip
1293 To intersperse text output in between query results, use
1294 \qecho.
1295 \p or \print
1297 Print the current query buffer to the standard output. If the
1298 current query buffer is empty, the most recently executed query is
1299 printed instead.
1300 \password [ username ]
1302 Changes the password of the specified user (by default, the current
1303 user). This command prompts for the new password, encrypts it, and
1304 sends it to the server as an ALTER ROLE command. This makes sure
1305 that the new password does not appear in cleartext in the command
1306 history, the server log, or elsewhere.
1307 \prompt [ text ] name
1309 Prompts the user to supply text, which is assigned to the variable
1310 name. An optional prompt string, text, can be specified. (For
1311 multiword prompts, surround the text with single quotes.)
1312 By default, \prompt uses the terminal for input and output.
1314 However, if the -f command line switch was used, \prompt uses
1315 standard input and standard output.
1316 \pset [ option [ value ] ]
1318 This command sets options affecting the output of query result
1319 tables. option indicates which option is to be set. The semantics
1320 of value vary depending on the selected option. For some options,
1321 omitting value causes the option to be toggled or unset, as
1322 described under the particular option. If no such behavior is
1323 mentioned, then omitting value just results in the current setting
1324 being displayed.
1325 \pset without any arguments displays the current status of all
1327 printing options.
1328 Adjustable printing options are:
1330 border
1332 The value must be a number. In general, the higher the number
1333 the more borders and lines the tables will have, but details
1334 depend on the particular format. In HTML format, this will
1335 translate directly into the border=... attribute. In most
1336 other formats only values 0 (no border), 1 (internal dividing
1337 lines), and 2 (table frame) make sense, and values above 2 will
1338 be treated the same as border = 2. The latex and
1339 latex-longtable formats additionally allow a value of 3 to add
1340 dividing lines between data rows.
1341 columns
1343 Sets the target width for the wrapped format, and also the
1344 width limit for determining whether output is wide enough to
1345 require the pager or switch to the vertical display in expanded
1346 auto mode. Zero (the default) causes the target width to be
1347 controlled by the environment variable COLUMNS, or the detected
1348 screen width if COLUMNS is not set. In addition, if columns is
1349 zero then the wrapped format only affects screen output. If
1350 columns is nonzero then file and pipe output is wrapped to that
1351 width as well.
1352 csv_fieldsep
1354 Specifies the field separator to be used in CSV output format.
1355 If the separator character appears in a field's value, that
1356 field is output within double quotes, following standard CSV
1357 rules. The default is a comma.
1358 expanded (or x)
1360 If value is specified it must be either on or off, which will
1361 enable or disable expanded mode, or auto. If value is omitted
1362 the command toggles between the on and off settings. When
1363 expanded mode is enabled, query results are displayed in two
1364 columns, with the column name on the left and the data on the
1365 right. This mode is useful if the data wouldn't fit on the
1366 screen in the normal “horizontal” mode. In the auto setting,
1367 the expanded mode is used whenever the query output has more
1368 than one column and is wider than the screen; otherwise, the
1369 regular mode is used. The auto setting is only effective in the
1370 aligned and wrapped formats. In other formats, it always
1371 behaves as if the expanded mode is off.
1372 fieldsep
1374 Specifies the field separator to be used in unaligned output
1375 format. That way one can create, for example, tab-separated
1376 output, which other programs might prefer. To set a tab as
1377 field separator, type \pset fieldsep '\t'. The default field
1378 separator is '|' (a vertical bar).
1379 fieldsep_zero
1381 Sets the field separator to use in unaligned output format to a
1382 zero byte.
1383 footer
1385 If value is specified it must be either on or off which will
1386 enable or disable display of the table footer (the (n rows)
1387 count). If value is omitted the command toggles footer display
1388 on or off.
1389 format
1391 Sets the output format to one of aligned, asciidoc, csv, html,
1392 latex, latex-longtable, troff-ms, unaligned, or wrapped. Unique
1393 abbreviations are allowed.
1394 aligned format is the standard, human-readable, nicely
1396 formatted text output; this is the default.
1397 unaligned format writes all columns of a row on one line,
1399 separated by the currently active field separator. This is
1400 useful for creating output that might be intended to be read in
1401 by other programs, for example, tab-separated or
1402 comma-separated format. However, the field separator character
1403 is not treated specially if it appears in a column's value; so
1404 CSV format may be better suited for such purposes.
1405 csv format
1407 writes column values separated by commas, applying the quoting
1409 rules described in RFC 4180. This output is compatible with the
1410 CSV format of the server's COPY command. A header line with
1411 column names is generated unless the tuples_only parameter is
1412 on. Titles and footers are not printed. Each row is terminated
1413 by the system-dependent end-of-line character, which is
1414 typically a single newline (\n) for Unix-like systems or a
1415 carriage return and newline sequence (\r\n) for Microsoft
1416 Windows. Field separator characters other than comma can be
1417 selected with \pset csv_fieldsep.
1418 wrapped format is like aligned but wraps wide data values
1420 across lines to make the output fit in the target column width.
1421 The target width is determined as described under the columns
1422 option. Note that psql will not attempt to wrap column header
1423 titles; therefore, wrapped format behaves the same as aligned
1424 if the total width needed for column headers exceeds the
1425 target.
1426 The asciidoc, html, latex, latex-longtable, and troff-ms
1428 formats put out tables that are intended to be included in
1429 documents using the respective mark-up language. They are not
1430 complete documents! This might not be necessary in HTML, but in
1431 LaTeX you must have a complete document wrapper. The latex
1432 format uses LaTeX's tabular environment. The latex-longtable
1433 format requires the LaTeX longtable and booktabs packages.
1434 linestyle
1436 Sets the border line drawing style to one of ascii, old-ascii,
1437 or unicode. Unique abbreviations are allowed. (That would mean
1438 one letter is enough.) The default setting is ascii. This
1439 option only affects the aligned and wrapped output formats.
1440 ascii style uses plain ASCII characters. Newlines in data are
1442 shown using a + symbol in the right-hand margin. When the
1443 wrapped format wraps data from one line to the next without a
1444 newline character, a dot (.) is shown in the right-hand margin
1445 of the first line, and again in the left-hand margin of the
1446 following line.
1447 old-ascii style uses plain ASCII characters, using the
1449 formatting style used in PostgreSQL 8.4 and earlier. Newlines
1450 in data are shown using a : symbol in place of the left-hand
1451 column separator. When the data is wrapped from one line to the
1452 next without a newline character, a ; symbol is used in place
1453 of the left-hand column separator.
1454 unicode style uses Unicode box-drawing characters. Newlines in
1456 data are shown using a carriage return symbol in the right-hand
1457 margin. When the data is wrapped from one line to the next
1458 without a newline character, an ellipsis symbol is shown in the
1459 right-hand margin of the first line, and again in the left-hand
1460 margin of the following line.
1461 When the border setting is greater than zero, the linestyle
1463 option also determines the characters with which the border
1464 lines are drawn. Plain ASCII characters work everywhere, but
1465 Unicode characters look nicer on displays that recognize them.
1466 null
1468 Sets the string to be printed in place of a null value. The
1469 default is to print nothing, which can easily be mistaken for
1470 an empty string. For example, one might prefer \pset null
1471 '(null)'.
1472 numericlocale
1474 If value is specified it must be either on or off which will
1475 enable or disable display of a locale-specific character to
1476 separate groups of digits to the left of the decimal marker. If
1477 value is omitted the command toggles between regular and
1478 locale-specific numeric output.
1479 pager
1481 Controls use of a pager program for query and psql help output.
1482 If the environment variable PSQL_PAGER or PAGER is set, the
1483 output is piped to the specified program. Otherwise a
1484 platform-dependent default program (such as more) is used.
1485 When the pager option is off, the pager program is not used.
1487 When the pager option is on, the pager is used when
1488 appropriate, i.e., when the output is to a terminal and will
1489 not fit on the screen. The pager option can also be set to
1490 always, which causes the pager to be used for all terminal
1491 output regardless of whether it fits on the screen. \pset
1492 pager without a value toggles pager use on and off.
1493 pager_min_lines
1495 If pager_min_lines is set to a number greater than the page
1496 height, the pager program will not be called unless there are
1497 at least this many lines of output to show. The default setting
1498 is 0.
1499 recordsep
1501 Specifies the record (line) separator to use in unaligned
1502 output format. The default is a newline character.
1503 recordsep_zero
1505 Sets the record separator to use in unaligned output format to
1506 a zero byte.
1507 tableattr (or T)
1509 In HTML format, this specifies attributes to be placed inside
1510 the table tag. This could for example be cellpadding or
1511 bgcolor. Note that you probably don't want to specify border
1512 here, as that is already taken care of by \pset border. If no
1513 value is given, the table attributes are unset.
1514 In latex-longtable format, this controls the proportional width
1516 of each column containing a left-aligned data type. It is
1517 specified as a whitespace-separated list of values, e.g., '0.2
1518 0.2 0.6'. Unspecified output columns use the last specified
1519 value.
1520 title (or C)
1522 Sets the table title for any subsequently printed tables. This
1523 can be used to give your output descriptive tags. If no value
1524 is given, the title is unset.
1525 tuples_only (or t)
1527 If value is specified it must be either on or off which will
1528 enable or disable tuples-only mode. If value is omitted the
1529 command toggles between regular and tuples-only output. Regular
1530 output includes extra information such as column headers,
1531 titles, and various footers. In tuples-only mode, only actual
1532 table data is shown.
1533 unicode_border_linestyle
1535 Sets the border drawing style for the unicode line style to one
1536 of single or double.
1537 unicode_column_linestyle
1539 Sets the column drawing style for the unicode line style to one
1540 of single or double.
1541 unicode_header_linestyle
1543 Sets the header drawing style for the unicode line style to one
1544 of single or double.
1545 Illustrations of how these different formats look can be seen in
1547 the EXAMPLES section.
1548 Tip
1550 There are various shortcut commands for \pset. See \a, \C, \f,
1551 \H, \t, \T, and \x.
1552 \q or \quit
1554 Quits the psql program. In a script file, only execution of that
1555 script is terminated.
1556 \qecho text [ ... ]
1558 This command is identical to \echo except that the output will be
1559 written to the query output channel, as set by \o.
1560 \r or \reset
1562 Resets (clears) the query buffer.
1563 \s [ filename ]
1565 Print psql's command line history to filename. If filename is
1566 omitted, the history is written to the standard output (using the
1567 pager if appropriate). This command is not available if psql was
1568 built without Readline support.
1569 \set [ name [ value [ ... ] ] ]
1571 Sets the psql variable name to value, or if more than one value is
1572 given, to the concatenation of all of them. If only one argument is
1573 given, the variable is set to an empty-string value. To unset a
1574 variable, use the \unset command.
1575 \set without any arguments displays the names and values of all
1577 currently-set psql variables.
1578 Valid variable names can contain letters, digits, and underscores.
1580 See the section Variables below for details. Variable names are
1581 case-sensitive.
1582 Certain variables are special, in that they control psql's behavior
1584 or are automatically set to reflect connection state. These
1585 variables are documented in Variables, below.
1586 Note
1588 This command is unrelated to the SQL command SET(7).
1589 \setenv name [ value ]
1591 Sets the environment variable name to value, or if the value is not
1592 supplied, unsets the environment variable. Example:
1593 testdb=> \setenv PAGER less
1595 testdb=> \setenv LESS -imx4F
1596 \sf[+] function_description
1598 This command fetches and shows the definition of the named function
1599 or procedure, in the form of a CREATE OR REPLACE FUNCTION or CREATE
1600 OR REPLACE PROCEDURE command. The definition is printed to the
1601 current query output channel, as set by \o.
1602 The target function can be specified by name alone, or by name and
1604 arguments, for example foo(integer, text). The argument types must
1605 be given if there is more than one function of the same name.
1606 If + is appended to the command name, then the output lines are
1608 numbered, with the first line of the function body being line 1.
1609 Unlike most other meta-commands, the entire remainder of the line
1611 is always taken to be the argument(s) of \sf, and neither variable
1612 interpolation nor backquote expansion are performed in the
1613 arguments.
1614 \sv[+] view_name
1616 This command fetches and shows the definition of the named view, in
1617 the form of a CREATE OR REPLACE VIEW command. The definition is
1618 printed to the current query output channel, as set by \o.
1619 If + is appended to the command name, then the output lines are
1621 numbered from 1.
1622 Unlike most other meta-commands, the entire remainder of the line
1624 is always taken to be the argument(s) of \sv, and neither variable
1625 interpolation nor backquote expansion are performed in the
1626 arguments.
1627 \t
1629 Toggles the display of output column name headings and row count
1630 footer. This command is equivalent to \pset tuples_only and is
1631 provided for convenience.
1632 \T table_options
1634 Specifies attributes to be placed within the table tag in HTML
1635 output format. This command is equivalent to \pset tableattr
1636 table_options.
1637 \timing [ on | off ]
1639 With a parameter, turns displaying of how long each SQL statement
1640 takes on or off. Without a parameter, toggles the display between
1641 on and off. The display is in milliseconds; intervals longer than 1
1642 second are also shown in minutes:seconds format, with hours and
1643 days fields added if needed.
1644 \unset name
1646 Unsets (deletes) the psql variable name.
1647 Most variables that control psql's behavior cannot be unset;
1649 instead, an \unset command is interpreted as setting them to their
1650 default values. See Variables, below.
1651 \w or \write filename
1653 \w or \write |command
1654 Writes the current query buffer to the file filename or pipes it to
1655 the shell command command. If the current query buffer is empty,
1656 the most recently executed query is written instead.
1657 If the argument begins with |, then the entire remainder of the
1659 line is taken to be the command to execute, and neither variable
1660 interpolation nor backquote expansion are performed in it. The rest
1661 of the line is simply passed literally to the shell.
1662 \watch [ seconds ]
1664 Repeatedly execute the current query buffer (as \g does) until
1665 interrupted or the query fails. Wait the specified number of
1666 seconds (default 2) between executions. Each query result is
1667 displayed with a header that includes the \pset title string (if
1668 any), the time as of query start, and the delay interval.
1669 If the current query buffer is empty, the most recently sent query
1671 is re-executed instead.
1672 \x [ on | off | auto ]
1674 Sets or toggles expanded table formatting mode. As such it is
1675 equivalent to \pset expanded.
1676 \z [ pattern ]
1678 Lists tables, views and sequences with their associated access
1679 privileges. If a pattern is specified, only tables, views and
1680 sequences whose names match the pattern are listed.
1681 This is an alias for \dp (“display privileges”).
1683 \! [ command ]
1685 With no argument, escapes to a sub-shell; psql resumes when the
1686 sub-shell exits. With an argument, executes the shell command
1687 command.
1688 Unlike most other meta-commands, the entire remainder of the line
1690 is always taken to be the argument(s) of \!, and neither variable
1691 interpolation nor backquote expansion are performed in the
1692 arguments. The rest of the line is simply passed literally to the
1693 shell.
1694 \? [ topic ]
1696 Shows help information. The optional topic parameter (defaulting to
1697 commands) selects which part of psql is explained: commands
1698 describes psql's backslash commands; options describes the
1699 command-line options that can be passed to psql; and variables
1700 shows help about psql configuration variables.
1701 \;
1703 Backslash-semicolon is not a meta-command in the same way as the
1704 preceding commands; rather, it simply causes a semicolon to be
1705 added to the query buffer without any further processing.
1706 Normally, psql will dispatch a SQL command to the server as soon as
1708 it reaches the command-ending semicolon, even if more input remains
1709 on the current line. Thus for example entering
1710 select 1; select 2; select 3;
1712 will result in the three SQL commands being individually sent to
1714 the server, with each one's results being displayed before
1715 continuing to the next command. However, a semicolon entered as \;
1716 will not trigger command processing, so that the command before it
1717 and the one after are effectively combined and sent to the server
1718 in one request. So for example
1719 select 1\; select 2\; select 3;
1721 results in sending the three SQL commands to the server in a single
1723 request, when the non-backslashed semicolon is reached. The server
1724 executes such a request as a single transaction, unless there are
1725 explicit BEGIN/COMMIT commands included in the string to divide it
1726 into multiple transactions. (See Section 52.2.2.1 for more details
1727 about how the server handles multi-query strings.) psql prints
1728 only the last query result it receives for each request; in this
1729 example, although all three SELECTs are indeed executed, psql only
1730 prints the 3.
1731 Patterns
1733 The various \d commands accept a pattern parameter to specify the
1734 object name(s) to be displayed. In the simplest case, a pattern is
1735 just the exact name of the object. The characters within a pattern
1736 are normally folded to lower case, just as in SQL names; for
1737 example, \dt FOO will display the table named foo. As in SQL names,
1738 placing double quotes around a pattern stops folding to lower case.
1739 Should you need to include an actual double quote character in a
1740 pattern, write it as a pair of double quotes within a double-quote
1741 sequence; again this is in accord with the rules for SQL quoted
1742 identifiers. For example, \dt "FOO""BAR" will display the table
1743 named FOO"BAR (not foo"bar). Unlike the normal rules for SQL names,
1744 you can put double quotes around just part of a pattern, for
1745 instance \dt FOO"FOO"BAR will display the table named fooFOObar.
1746 Whenever the pattern parameter is omitted completely, the \d
1748 commands display all objects that are visible in the current schema
1749 search path — this is equivalent to using * as the pattern. (An
1750 object is said to be visible if its containing schema is in the
1751 search path and no object of the same kind and name appears earlier
1752 in the search path. This is equivalent to the statement that the
1753 object can be referenced by name without explicit schema
1754 qualification.) To see all objects in the database regardless of
1755 visibility, use *.* as the pattern.
1756 Within a pattern, * matches any sequence of characters (including
1758 no characters) and ? matches any single character. (This notation
1759 is comparable to Unix shell file name patterns.) For example, \dt
1760 int* displays tables whose names begin with int. But within double
1761 quotes, * and ? lose these special meanings and are just matched
1762 literally.
1763 A pattern that contains a dot (.) is interpreted as a schema name
1765 pattern followed by an object name pattern. For example, \dt
1766 foo*.*bar* displays all tables whose table name includes bar that
1767 are in schemas whose schema name starts with foo. When no dot
1768 appears, then the pattern matches only objects that are visible in
1769 the current schema search path. Again, a dot within double quotes
1770 loses its special meaning and is matched literally.
1771 Advanced users can use regular-expression notations such as
1773 character classes, for example [0-9] to match any digit. All
1774 regular expression special characters work as specified in
1775 Section 9.7.3, except for . which is taken as a separator as
1776 mentioned above, * which is translated to the regular-expression
1777 notation .*, ? which is translated to ., and $ which is matched
1778 literally. You can emulate these pattern characters at need by
1779 writing ? for ., (R+|) for R*, or (R|) for R?. $ is not needed as
1780 a regular-expression character since the pattern must match the
1781 whole name, unlike the usual interpretation of regular expressions
1782 (in other words, $ is automatically appended to your pattern).
1783 Write * at the beginning and/or end if you don't wish the pattern
1784 to be anchored. Note that within double quotes, all regular
1785 expression special characters lose their special meanings and are
1786 matched literally. Also, the regular expression special characters
1787 are matched literally in operator name patterns (i.e., the argument
1788 of \do).
1789 Advanced Features
1791 Variables
1792 psql provides variable substitution features similar to common Unix
1793 command shells. Variables are simply name/value pairs, where the
1794 value can be any string of any length. The name must consist of
1795 letters (including non-Latin letters), digits, and underscores.
1796 To set a variable, use the psql meta-command \set. For example,
1798 testdb=> \set foo bar
1800 sets the variable foo to the value bar. To retrieve the content of
1802 the variable, precede the name with a colon, for example:
1803 testdb=> \echo :foo
1805 bar
1806 This works in both regular SQL commands and meta-commands; there is
1808 more detail in SQL Interpolation, below.
1809 If you call \set without a second argument, the variable is set to
1811 an empty-string value. To unset (i.e., delete) a variable, use the
1812 command \unset. To show the values of all variables, call \set
1813 without any argument.
1814 Note
1816 The arguments of \set are subject to the same substitution
1817 rules as with other commands. Thus you can construct
1818 interesting references such as \set :foo 'something' and get
1819 “soft links” or “variable variables” of Perl or PHP fame,
1820 respectively. Unfortunately (or fortunately?), there is no way
1821 to do anything useful with these constructs. On the other hand,
1822 \set bar :foo is a perfectly valid way to copy a variable.
1823 A number of these variables are treated specially by psql. They
1825 represent certain option settings that can be changed at run time
1826 by altering the value of the variable, or in some cases represent
1827 changeable state of psql. By convention, all specially treated
1828 variables' names consist of all upper-case ASCII letters (and
1829 possibly digits and underscores). To ensure maximum compatibility
1830 in the future, avoid using such variable names for your own
1831 purposes.
1832 Variables that control psql's behavior generally cannot be unset or
1834 set to invalid values. An \unset command is allowed but is
1835 interpreted as setting the variable to its default value. A \set
1836 command without a second argument is interpreted as setting the
1837 variable to on, for control variables that accept that value, and
1838 is rejected for others. Also, control variables that accept the
1839 values on and off will also accept other common spellings of
1840 Boolean values, such as true and false.
1841 The specially treated variables are:
1843 AUTOCOMMIT
1845 When on (the default), each SQL command is automatically
1846 committed upon successful completion. To postpone commit in
1847 this mode, you must enter a BEGIN or START TRANSACTION SQL
1848 command. When off or unset, SQL commands are not committed
1849 until you explicitly issue COMMIT or END. The autocommit-off
1850 mode works by issuing an implicit BEGIN for you, just before
1851 any command that is not already in a transaction block and is
1852 not itself a BEGIN or other transaction-control command, nor a
1853 command that cannot be executed inside a transaction block
1854 (such as VACUUM).
1855 Note
1857 In autocommit-off mode, you must explicitly abandon any
1858 failed transaction by entering ABORT or ROLLBACK. Also keep
1859 in mind that if you exit the session without committing,
1860 your work will be lost.
1861 Note
1863 The autocommit-on mode is PostgreSQL's traditional
1864 behavior, but autocommit-off is closer to the SQL spec. If
1865 you prefer autocommit-off, you might wish to set it in the
1866 system-wide psqlrc file or your ~/.psqlrc file.
1867 COMP_KEYWORD_CASE
1869 Determines which letter case to use when completing an SQL key
1870 word. If set to lower or upper, the completed word will be in
1871 lower or upper case, respectively. If set to preserve-lower or
1872 preserve-upper (the default), the completed word will be in the
1873 case of the word already entered, but words being completed
1874 without anything entered will be in lower or upper case,
1875 respectively.
1876 DBNAME
1878 The name of the database you are currently connected to. This
1879 is set every time you connect to a database (including program
1880 start-up), but can be changed or unset.
1881 ECHO
1883 If set to all, all nonempty input lines are printed to standard
1884 output as they are read. (This does not apply to lines read
1885 interactively.) To select this behavior on program start-up,
1886 use the switch -a. If set to queries, psql prints each query to
1887 standard output as it is sent to the server. The switch to
1888 select this behavior is -e. If set to errors, then only failed
1889 queries are displayed on standard error output. The switch for
1890 this behavior is -b. If set to none (the default), then no
1891 queries are displayed.
1892 ECHO_HIDDEN
1894 When this variable is set to on and a backslash command queries
1895 the database, the query is first shown. This feature helps you
1896 to study PostgreSQL internals and provide similar functionality
1897 in your own programs. (To select this behavior on program
1898 start-up, use the switch -E.) If you set this variable to the
1899 value noexec, the queries are just shown but are not actually
1900 sent to the server and executed. The default value is off.
1901 ENCODING
1903 The current client character set encoding. This is set every
1904 time you connect to a database (including program start-up),
1905 and when you change the encoding with \encoding, but it can be
1906 changed or unset.
1907 ERROR
1909 true if the last SQL query failed, false if it succeeded. See
1910 also SQLSTATE.
1911 FETCH_COUNT
1913 If this variable is set to an integer value greater than zero,
1914 the results of SELECT queries are fetched and displayed in
1915 groups of that many rows, rather than the default behavior of
1916 collecting the entire result set before display. Therefore only
1917 a limited amount of memory is used, regardless of the size of
1918 the result set. Settings of 100 to 1000 are commonly used when
1919 enabling this feature. Keep in mind that when using this
1920 feature, a query might fail after having already displayed some
1921 rows.
1922 Tip
1924 Although you can use any output format with this feature,
1925 the default aligned format tends to look bad because each
1926 group of FETCH_COUNT rows will be formatted separately,
1927 leading to varying column widths across the row groups. The
1928 other output formats work better.
1929 HIDE_TABLEAM
1931 If this variable is set to true, a table's access method
1932 details are not displayed. This is mainly useful for regression
1933 tests.
1934 HISTCONTROL
1936 If this variable is set to ignorespace, lines which begin with
1937 a space are not entered into the history list. If set to a
1938 value of ignoredups, lines matching the previous history line
1939 are not entered. A value of ignoreboth combines the two
1940 options. If set to none (the default), all lines read in
1941 interactive mode are saved on the history list.
1942 Note
1944 This feature was shamelessly plagiarized from Bash.
1945 HISTFILE
1947 The file name that will be used to store the history list. If
1948 unset, the file name is taken from the PSQL_HISTORY environment
1949 variable. If that is not set either, the default is
1950 ~/.psql_history, or %APPDATA%\postgresql\psql_history on
1951 Windows. For example, putting:
1952 \set HISTFILE ~/.psql_history- :DBNAME
1954 in ~/.psqlrc will cause psql to maintain a separate history for
1956 each database.
1957 Note
1959 This feature was shamelessly plagiarized from Bash.
1960 HISTSIZE
1962 The maximum number of commands to store in the command history
1963 (default 500). If set to a negative value, no limit is applied.
1964 Note
1966 This feature was shamelessly plagiarized from Bash.
1967 HOST
1969 The database server host you are currently connected to. This
1970 is set every time you connect to a database (including program
1971 start-up), but can be changed or unset.
1972 IGNOREEOF
1974 If set to 1 or less, sending an EOF character (usually
1975 Control+D) to an interactive session of psql will terminate the
1976 application. If set to a larger numeric value, that many
1977 consecutive EOF characters must be typed to make an interactive
1978 session terminate. If the variable is set to a non-numeric
1979 value, it is interpreted as 10. The default is 0.
1980 Note
1982 This feature was shamelessly plagiarized from Bash.
1983 LASTOID
1985 The value of the last affected OID, as returned from an INSERT
1986 or \lo_import command. This variable is only guaranteed to be
1987 valid until after the result of the next SQL command has been
1988 displayed. PostgreSQL servers since version 12 do not support
1989 OID system columns anymore, thus LASTOID will always be 0
1990 following INSERT when targeting such servers.
1991 LAST_ERROR_MESSAGE
1993 LAST_ERROR_SQLSTATE
1994 The primary error message and associated SQLSTATE code for the
1995 most recent failed query in the current psql session, or an
1996 empty string and 00000 if no error has occurred in the current
1997 session.
1998 ON_ERROR_ROLLBACK
2000 When set to on, if a statement in a transaction block generates
2001 an error, the error is ignored and the transaction continues.
2002 When set to interactive, such errors are only ignored in
2003 interactive sessions, and not when reading script files. When
2004 set to off (the default), a statement in a transaction block
2005 that generates an error aborts the entire transaction. The
2006 error rollback mode works by issuing an implicit SAVEPOINT for
2007 you, just before each command that is in a transaction block,
2008 and then rolling back to the savepoint if the command fails.
2009 ON_ERROR_STOP
2011 By default, command processing continues after an error. When
2012 this variable is set to on, processing will instead stop
2013 immediately. In interactive mode, psql will return to the
2014 command prompt; otherwise, psql will exit, returning error code
2015 3 to distinguish this case from fatal error conditions, which
2016 are reported using error code 1. In either case, any currently
2017 running scripts (the top-level script, if any, and any other
2018 scripts which it may have in invoked) will be terminated
2019 immediately. If the top-level command string contained multiple
2020 SQL commands, processing will stop with the current command.
2021 PORT
2023 The database server port to which you are currently connected.
2024 This is set every time you connect to a database (including
2025 program start-up), but can be changed or unset.
2026 PROMPT1
2028 PROMPT2
2029 PROMPT3
2030 These specify what the prompts psql issues should look like.
2031 See Prompting below.
2032 QUIET
2034 Setting this variable to on is equivalent to the command line
2035 option -q. It is probably not too useful in interactive mode.
2036 ROW_COUNT
2038 The number of rows returned or affected by the last SQL query,
2039 or 0 if the query failed or did not report a row count.
2040 SERVER_VERSION_NAME
2042 SERVER_VERSION_NUM
2043 The server's version number as a string, for example 9.6.2,
2044 10.1 or 11beta1, and in numeric form, for example 90602 or
2045 100001. These are set every time you connect to a database
2046 (including program start-up), but can be changed or unset.
2047 SHOW_CONTEXT
2049 This variable can be set to the values never, errors, or always
2050 to control whether CONTEXT fields are displayed in messages
2051 from the server. The default is errors (meaning that context
2052 will be shown in error messages, but not in notice or warning
2053 messages). This setting has no effect when VERBOSITY is set to
2054 terse or sqlstate. (See also \errverbose, for use when you want
2055 a verbose version of the error you just got.)
2056 SINGLELINE
2058 Setting this variable to on is equivalent to the command line
2059 option -S.
2060 SINGLESTEP
2062 Setting this variable to on is equivalent to the command line
2063 option -s.
2064 SQLSTATE
2066 The error code (see Appendix A) associated with the last SQL
2067 query's failure, or 00000 if it succeeded.
2068 USER
2070 The database user you are currently connected as. This is set
2071 every time you connect to a database (including program
2072 start-up), but can be changed or unset.
2073 VERBOSITY
2075 This variable can be set to the values default, verbose, terse,
2076 or sqlstate to control the verbosity of error reports. (See
2077 also \errverbose, for use when you want a verbose version of
2078 the error you just got.)
2079 VERSION
2081 VERSION_NAME
2082 VERSION_NUM
2083 These variables are set at program start-up to reflect psql's
2084 version, respectively as a verbose string, a short string
2085 (e.g., 9.6.2, 10.1, or 11beta1), and a number (e.g., 90602 or
2086 100001). They can be changed or unset.
2087 SQL Interpolation
2089 A key feature of psql variables is that you can substitute
2090 (“interpolate”) them into regular SQL statements, as well as the
2091 arguments of meta-commands. Furthermore, psql provides facilities
2092 for ensuring that variable values used as SQL literals and
2093 identifiers are properly quoted. The syntax for interpolating a
2094 value without any quoting is to prepend the variable name with a
2095 colon (:). For example,
2096 testdb=> \set foo 'my_table'
2098 testdb=> SELECT * FROM :foo;
2099 would query the table my_table. Note that this may be unsafe: the
2101 value of the variable is copied literally, so it can contain
2102 unbalanced quotes, or even backslash commands. You must make sure
2103 that it makes sense where you put it.
2104 When a value is to be used as an SQL literal or identifier, it is
2106 safest to arrange for it to be quoted. To quote the value of a
2107 variable as an SQL literal, write a colon followed by the variable
2108 name in single quotes. To quote the value as an SQL identifier,
2109 write a colon followed by the variable name in double quotes. These
2110 constructs deal correctly with quotes and other special characters
2111 embedded within the variable value. The previous example would be
2112 more safely written this way:
2113 testdb=> \set foo 'my_table'
2115 testdb=> SELECT * FROM :"foo";
2116 Variable interpolation will not be performed within quoted SQL
2118 literals and identifiers. Therefore, a construction such as ':foo'
2119 doesn't work to produce a quoted literal from a variable's value
2120 (and it would be unsafe if it did work, since it wouldn't correctly
2121 handle quotes embedded in the value).
2122 One example use of this mechanism is to copy the contents of a file
2124 into a table column. First load the file into a variable and then
2125 interpolate the variable's value as a quoted string:
2126 testdb=> \set content `cat my_file.txt`
2128 testdb=> INSERT INTO my_table VALUES (:'content');
2129 (Note that this still won't work if my_file.txt contains NUL bytes.
2131 psql does not support embedded NUL bytes in variable values.)
2132 Since colons can legally appear in SQL commands, an apparent
2134 attempt at interpolation (that is, :name, :'name', or :"name") is
2135 not replaced unless the named variable is currently set. In any
2136 case, you can escape a colon with a backslash to protect it from
2137 substitution.
2138 The :{?name} special syntax returns TRUE or FALSE depending on
2140 whether the variable exists or not, and is thus always substituted,
2141 unless the colon is backslash-escaped.
2142 The colon syntax for variables is standard SQL for embedded query
2144 languages, such as ECPG. The colon syntaxes for array slices and
2145 type casts are PostgreSQL extensions, which can sometimes conflict
2146 with the standard usage. The colon-quote syntax for escaping a
2147 variable's value as an SQL literal or identifier is a psql
2148 extension.
2149 Prompting
2151 The prompts psql issues can be customized to your preference. The
2152 three variables PROMPT1, PROMPT2, and PROMPT3 contain strings and
2153 special escape sequences that describe the appearance of the
2154 prompt. Prompt 1 is the normal prompt that is issued when psql
2155 requests a new command. Prompt 2 is issued when more input is
2156 expected during command entry, for example because the command was
2157 not terminated with a semicolon or a quote was not closed. Prompt 3
2158 is issued when you are running an SQL COPY FROM STDIN command and
2159 you need to type in a row value on the terminal.
2160 The value of the selected prompt variable is printed literally,
2162 except where a percent sign (%) is encountered. Depending on the
2163 next character, certain other text is substituted instead. Defined
2164 substitutions are:
2165 %M
2167 The full host name (with domain name) of the database server,
2168 or [local] if the connection is over a Unix domain socket, or
2169 [local:/dir/name], if the Unix domain socket is not at the
2170 compiled in default location.
2171 %m
2173 The host name of the database server, truncated at the first
2174 dot, or [local] if the connection is over a Unix domain socket.
2175 %>
2177 The port number at which the database server is listening.
2178 %n
2180 The database session user name. (The expansion of this value
2181 might change during a database session as the result of the
2182 command SET SESSION AUTHORIZATION.)
2183 %/
2185 The name of the current database.
2186 %~
2188 Like %/, but the output is ~ (tilde) if the database is your
2189 default database.
2190 %#
2192 If the session user is a database superuser, then a #,
2193 otherwise a >. (The expansion of this value might change during
2194 a database session as the result of the command SET SESSION
2195 AUTHORIZATION.)
2196 %p
2198 The process ID of the backend currently connected to.
2199 %R
2201 In prompt 1 normally =, but @ if the session is in an inactive
2202 branch of a conditional block, or ^ if in single-line mode, or
2203 ! if the session is disconnected from the database (which can
2204 happen if \connect fails). In prompt 2 %R is replaced by a
2205 character that depends on why psql expects more input: - if the
2206 command simply wasn't terminated yet, but * if there is an
2207 unfinished /* ... */ comment, a single quote if there is an
2208 unfinished quoted string, a double quote if there is an
2209 unfinished quoted identifier, a dollar sign if there is an
2210 unfinished dollar-quoted string, or ( if there is an unmatched
2211 left parenthesis. In prompt 3 %R doesn't produce anything.
2212 %x
2214 Transaction status: an empty string when not in a transaction
2215 block, or * when in a transaction block, or ! when in a failed
2216 transaction block, or ? when the transaction state is
2217 indeterminate (for example, because there is no connection).
2218 %l
2220 The line number inside the current statement, starting from 1.
2221 %digits
2223 The character with the indicated octal code is substituted.
2224 %:name:
2226 The value of the psql variable name. See the section Variables
2227 for details.
2228 %`command`
2230 The output of command, similar to ordinary “back-tick”
2231 substitution.
2232 %[ ... %]
2234 Prompts can contain terminal control characters which, for
2235 example, change the color, background, or style of the prompt
2236 text, or change the title of the terminal window. In order for
2237 the line editing features of Readline to work properly, these
2238 non-printing control characters must be designated as invisible
2239 by surrounding them with %[ and %]. Multiple pairs of these can
2240 occur within the prompt. For example:
2241 testdb=> \set PROMPT1 '%[%033[1;33;40m%]%n@%/%R%[%033[0m%]%# '
2243 results in a boldfaced (1;) yellow-on-black (33;40) prompt on
2245 VT100-compatible, color-capable terminals.
2246 To insert a percent sign into your prompt, write %%. The default
2247 prompts are '%/%R%# ' for prompts 1 and 2, and '>> ' for prompt 3.
2248 Note
2250 This feature was shamelessly plagiarized from tcsh.
2251 Command-Line Editing
2253 psql supports the Readline library for convenient line editing and
2254 retrieval. The command history is automatically saved when psql
2255 exits and is reloaded when psql starts up. Tab-completion is also
2256 supported, although the completion logic makes no claim to be an
2257 SQL parser. The queries generated by tab-completion can also
2258 interfere with other SQL commands, e.g., SET TRANSACTION ISOLATION
2259 LEVEL. If for some reason you do not like the tab completion, you
2260 can turn it off by putting this in a file named .inputrc in your
2261 home directory:
2262 $if psql
2264 set disable-completion on
2265 $endif
2266 (This is not a psql but a Readline feature. Read its documentation
2268 for further details.)
2269
2271 COLUMNS
2272 If \pset columns is zero, controls the width for the wrapped format
2273 and width for determining if wide output requires the pager or
2274 should be switched to the vertical format in expanded auto mode.
2275 PGDATABASE
2277 PGHOST
2278 PGPORT
2279 PGUSER
2280 Default connection parameters (see Section 33.14).
2281 PG_COLOR
2283 Specifies whether to use color in diagnostic messages. Possible
2284 values are always, auto and never.
2285 PSQL_EDITOR
2287 EDITOR
2288 VISUAL
2289 Editor used by the \e, \ef, and \ev commands. These variables are
2290 examined in the order listed; the first that is set is used. If
2291 none of them is set, the default is to use vi on Unix systems or
2292 notepad.exe on Windows systems.
2293 PSQL_EDITOR_LINENUMBER_ARG
2295 When \e, \ef, or \ev is used with a line number argument, this
2296 variable specifies the command-line argument used to pass the
2297 starting line number to the user's editor. For editors such as
2298 Emacs or vi, this is a plus sign. Include a trailing space in the
2299 value of the variable if there needs to be space between the option
2300 name and the line number. Examples:
2301 PSQL_EDITOR_LINENUMBER_ARG='+'
2303 PSQL_EDITOR_LINENUMBER_ARG='--line '
2304 The default is + on Unix systems (corresponding to the default
2306 editor vi, and useful for many other common editors); but there is
2307 no default on Windows systems.
2308 PSQL_HISTORY
2310 Alternative location for the command history file. Tilde (~)
2311 expansion is performed.
2312 PSQL_PAGER
2314 PAGER
2315 If a query's results do not fit on the screen, they are piped
2316 through this command. Typical values are more or less. Use of the
2317 pager can be disabled by setting PSQL_PAGER or PAGER to an empty
2318 string, or by adjusting the pager-related options of the \pset
2319 command. These variables are examined in the order listed; the
2320 first that is set is used. If none of them is set, the default is
2321 to use more on most platforms, but less on Cygwin.
2322 PSQLRC
2324 Alternative location of the user's .psqlrc file. Tilde (~)
2325 expansion is performed.
2326 SHELL
2328 Command executed by the \! command.
2329 TMPDIR
2331 Directory for storing temporary files. The default is /tmp.
2332 This utility, like most other PostgreSQL utilities, also uses the
2334 environment variables supported by libpq (see Section 33.14).
2335
2337 psqlrc and ~/.psqlrc
2338 Unless it is passed an -X option, psql attempts to read and execute
2339 commands from the system-wide startup file (psqlrc) and then the
2340 user's personal startup file (~/.psqlrc), after connecting to the
2341 database but before accepting normal commands. These files can be
2342 used to set up the client and/or the server to taste, typically
2343 with \set and SET commands.
2344 The system-wide startup file is named psqlrc and is sought in the
2346 installation's “system configuration” directory, which is most
2347 reliably identified by running pg_config --sysconfdir. By default
2348 this directory will be ../etc/ relative to the directory containing
2349 the PostgreSQL executables. The name of this directory can be set
2350 explicitly via the PGSYSCONFDIR environment variable.
2351 The user's personal startup file is named .psqlrc and is sought in
2353 the invoking user's home directory. On Windows, which lacks such a
2354 concept, the personal startup file is named
2355 %APPDATA%\postgresql\psqlrc.conf. The location of the user's
2356 startup file can be set explicitly via the PSQLRC environment
2357 variable.
2358 Both the system-wide startup file and the user's personal startup
2360 file can be made psql-version-specific by appending a dash and the
2361 PostgreSQL major or minor release number to the file name, for
2362 example ~/.psqlrc-9.2 or ~/.psqlrc-9.2.5. The most specific
2363 version-matching file will be read in preference to a
2364 non-version-specific file.
2365 .psql_history
2367 The command-line history is stored in the file ~/.psql_history, or
2368 %APPDATA%\postgresql\psql_history on Windows.
2369 The location of the history file can be set explicitly via the
2371 HISTFILE psql variable or the PSQL_HISTORY environment variable.
2372
2374 · psql works best with servers of the same or an older major version.
2375 Backslash commands are particularly likely to fail if the server is
2376 of a newer version than psql itself. However, backslash commands of
2377 the \d family should work with servers of versions back to 7.4,
2378 though not necessarily with servers newer than psql itself. The
2379 general functionality of running SQL commands and displaying query
2380 results should also work with servers of a newer major version, but
2381 this cannot be guaranteed in all cases.
2382 If you want to use psql to connect to several servers of different
2384 major versions, it is recommended that you use the newest version
2385 of psql. Alternatively, you can keep around a copy of psql from
2386 each major version and be sure to use the version that matches the
2387 respective server. But in practice, this additional complication
2388 should not be necessary.
2389 · Before PostgreSQL 9.6, the -c option implied -X (--no-psqlrc); this
2391 is no longer the case.
2392 · Before PostgreSQL 8.4, psql allowed the first argument of a
2394 single-letter backslash command to start directly after the
2395 command, without intervening whitespace. Now, some whitespace is
2396 required.
2397
2399 psql is built as a “console application”. Since the Windows console
2400 windows use a different encoding than the rest of the system, you must
2401 take special care when using 8-bit characters within psql. If psql
2402 detects a problematic console code page, it will warn you at startup.
2403 To change the console code page, two things are necessary:
2404 · Set the code page by entering cmd.exe /c chcp 1252. (1252 is a code
2406 page that is appropriate for German; replace it with your value.)
2407 If you are using Cygwin, you can put this command in /etc/profile.
2408 · Set the console font to Lucida Console, because the raster font
2410 does not work with the ANSI code page.
2411
2413 The first example shows how to spread a command over several lines of
2414 input. Notice the changing prompt:
2415 testdb=> CREATE TABLE my_table (
2417 testdb(> first integer not null default 0,
2418 testdb(> second text)
2419 testdb-> ;
2420 CREATE TABLE
2421 Now look at the table definition again:
2423 testdb=> \d my_table
2425 Table "public.my_table"
2426 Column | Type | Collation | Nullable | Default
2427 --------+---------+-----------+----------+---------
2428 first | integer | | not null | 0
2429 second | text | | |
2430 Now we change the prompt to something more interesting:
2432 testdb=> \set PROMPT1 '%n@%m %~%R%# '
2434 peter@localhost testdb=>
2435 Let's assume you have filled the table with data and want to take a
2437 look at it:
2438 peter@localhost testdb=> SELECT * FROM my_table;
2440 first | second
2441 -------+--------
2442 1 | one
2443 2 | two
2444 3 | three
2445 4 | four
2446 (4 rows)
2447 You can display tables in different ways by using the \pset command:
2450 peter@localhost testdb=> \pset border 2
2452 Border style is 2.
2453 peter@localhost testdb=> SELECT * FROM my_table;
2454 +-------+--------+
2455 | first | second |
2456 +-------+--------+
2457 | 1 | one |
2458 | 2 | two |
2459 | 3 | three |
2460 | 4 | four |
2461 +-------+--------+
2462 (4 rows)
2463 peter@localhost testdb=> \pset border 0
2465 Border style is 0.
2466 peter@localhost testdb=> SELECT * FROM my_table;
2467 first second
2468 ----- ------
2469 1 one
2470 2 two
2471 3 three
2472 4 four
2473 (4 rows)
2474 peter@localhost testdb=> \pset border 1
2476 Border style is 1.
2477 peter@localhost testdb=> \pset format csv
2478 Output format is csv.
2479 peter@localhost testdb=> \pset tuples_only
2480 Tuples only is on.
2481 peter@localhost testdb=> SELECT second, first FROM my_table;
2482 one,1
2483 two,2
2484 three,3
2485 four,4
2486 peter@localhost testdb=> \pset format unaligned
2487 Output format is unaligned.
2488 peter@localhost testdb=> \pset fieldsep '\t'
2489 Field separator is " ".
2490 peter@localhost testdb=> SELECT second, first FROM my_table;
2491 one 1
2492 two 2
2493 three 3
2494 four 4
2495 Alternatively, use the short commands:
2497 peter@localhost testdb=> \a \t \x
2499 Output format is aligned.
2500 Tuples only is off.
2501 Expanded display is on.
2502 peter@localhost testdb=> SELECT * FROM my_table;
2503 -[ RECORD 1 ]-
2504 first | 1
2505 second | one
2506 -[ RECORD 2 ]-
2507 first | 2
2508 second | two
2509 -[ RECORD 3 ]-
2510 first | 3
2511 second | three
2512 -[ RECORD 4 ]-
2513 first | 4
2514 second | four
2515 When suitable, query results can be shown in a crosstab representation
2517 with the \crosstabview command:
2518 testdb=> SELECT first, second, first > 2 AS gt2 FROM my_table;
2520 first | second | gt2
2521 -------+--------+-----
2522 1 | one | f
2523 2 | two | f
2524 3 | three | t
2525 4 | four | t
2526 (4 rows)
2527 testdb=> \crosstabview first second
2529 first | one | two | three | four
2530 -------+-----+-----+-------+------
2531 1 | f | | |
2532 2 | | f | |
2533 3 | | | t |
2534 4 | | | | t
2535 (4 rows)
2536 This second example shows a multiplication table with rows sorted in
2538 reverse numerical order and columns with an independent, ascending
2539 numerical order.
2540 testdb=> SELECT t1.first as "A", t2.first+100 AS "B", t1.first*(t2.first+100) as "AxB",
2542 testdb(> row_number() over(order by t2.first) AS ord
2543 testdb(> FROM my_table t1 CROSS JOIN my_table t2 ORDER BY 1 DESC
2544 testdb(> \crosstabview "A" "B" "AxB" ord
2545 A | 101 | 102 | 103 | 104
2546 ---+-----+-----+-----+-----
2547 4 | 404 | 408 | 412 | 416
2548 3 | 303 | 306 | 309 | 312
2549 2 | 202 | 204 | 206 | 208
2550 1 | 101 | 102 | 103 | 104
2551 (4 rows)
2552PostgreSQL 12.6 2021 PSQL(1)