cbq: The Command Line Shell for SQL++

      +
      cbq is a comprehensive command line shell for SQL++. It is a powerful, developer friendly tool that enables you to query and update data from Couchbase Server. The cbq shell enables you to perform all the operations that are supported by the Query REST API and more, such as additional scripting functionality.

      The cbq shell executable, cbq, is available in the Couchbase Server installation directory.

      The cbq shell interface accepts both shell commands as well as SQL++ commands. All the cbq shell commands start with a back-slash (\). If the command does not start with a back-slash (\), the cbq shell interprets the command as a SQL++ command.

      When starting the cbq shell you can provide a set of command line options. If no options are present then it assumes default values for expected options.

      The cbq shell commands are case insensitive. However, the command line options are case sensitive.

      For the complete list of command line options and shell commands, refer to Table 1 and Table 2.

      The cbq shell enables you to manipulate parameters based on the REST API. See Parameter Manipulation for details.

      Running the cbq Shell

      To run cbq on the local host:

      1. Log in to a Couchbase Server node that has the query service enabled.

      2. Open a command window.

      3. Change to the Couchbase tools directory.

        • Linux

        • macOS

        • Microsoft Windows

        $ cd /opt/couchbase/bin
        $ cd /Applications/Couchbase\ Server.app/Contents/Resources/couchbase-core/bin
        > cd C:\Program Files\Couchbase\Server\bin
      4. Run the following command to connect to the local query node and start the interactive query shell:

        • Linux / macOS

        • Microsoft Windows

        $ ./cbq
        > cbq

      Executing a Single Command

      You can use the --script option to execute a single SQL++ query and exit the shell:

      $ ./cbq -u Administrator -p password -e "http://localhost:8091" \
      --script="SELECT * FROM \`travel-sample\`.inventory.airline LIMIT 1;"
      Results
      Connected to : http://localhost:8091/. Type Ctrl-D or \QUIT to exit.
      
      Path to history file for the shell : ~/.cbq_history
      
      SELECT * FROM `travel-sample`.inventory.airline LIMIT 1;
      {
          ...
      }

      Support for Multi-line Queries

      The cbq shell supports multi-line queries by default, enabling you to enter a query over multiple lines. When entering a query, you can hit Enter without specifying a semi-colon (;) at the end of the line to move the cursor to the next line. The prompt > indicates that the shell is in multi-line mode. For example:

      cbq> SELECT *
         > FROM `travel-sample`.inventory.airline
         > LIMIT 1;

      When you’re done, use a semi-colon ; to indicate the end of the query, and then press Enter to execute the query.

      Handling Comments

      You can add comments in your query by preceding the comment with a # or --. The cbq shell interprets a line that starts with # or -- as a comment, logs the line into history, and returns a new prompt. No other action is taken.

      cbq> SELECT *
         > #This is the first comment
         > FROM `travel-sample`.inventory.airline
         > --This is the second comment
         > LIMIT 1;

      However, if a comment exists within a statement, it is considered as part of the SQL++ command. If the cbq shell encounters a block comment (enclosed between /* ... */) within a statement, it sends the block comment to the query service.

      cbq> SELECT * FROM `travel-sample`.inventory.airline /* Block comment */ LIMIT 1;

      File Based Operations

      The cbq shell can execute SQL++ and shell commands contained in files using file-based commands and options. See File Based Operations for more information.

      History

      The cbq shell stores the history for every session. All the commands executed in a session are stored in history. By default, history is stored in ~/.cbq_history. You can change the name of the file using the SET command to set the predefined parameter HISTFILE.

      cbq> \SET HISTFILE filename;

      By default, all the commands are stored in the specified file. You can scroll through history and retrieve the commands from history using the scrolling arrow keys. Once the query is on the command prompt, you can edit it before executing the updated query.

      Exit Status

      The cbq shell returns the exit status 0 for successful exit with no errors and 1 if an error was encountered before exiting.

      Exit On Error

      When you specify the argument --exit-on-error, the cbq shell checks the result returned after executing the query for any error and exits when the first error is encountered.

      Help

      Help displays the help information for the shell commands and for the general usage of cbq. Use the help option when bringing up the shell to display the information for all available options:

      $ ./cbq -h
      $ ./cbq --help

      Use the \HELP shell command during a session to display information for specific shell commands. If you specify one or more commands, the shell displays the usage information for the specified commands.

      cbq> \HELP command-name;

      If you do not specify a command, the cbq shell lists all the commands for which syntax help is available.

      cbq> \HELP;

      Available Command Line Options and Shell Commands

      Table 1. Command Line Options for cbq Shell
      Option Arguments Description and Examples

      -e

      --engine

      string (url)

      The connection string consists of a protocol scheme, followed by a host, and optionally a port number to connect to the query service (8093) or the Couchbase cluster (8091).

      For more details, refer to Connecting to the Cluster or Query Node.

      Shell command: \CONNECT

      Default

      http://localhost:8091

      Examples
      $ ./cbq -e couchbase://localhost
      $ ./cbq --engine http://localhost:8091
      $ ./cbq -e http://localhost:8091
      $ ./cbq -e http://[fd63:6f75:6368:1075:816:3c1d:789b:bc4]:8091
      Result
      Connected to : http://localhost:8091/. Type Ctrl-D or \QUIT to exit.
      Path to history file for the shell : /Users/myuser1/.cbq_history
      cbq>

      -ne

      --no-engine

      boolean [1]

      When specified, the cbq shell does not connect to any query service. You must explicitly connect to a query service using the \CONNECT shell command.

      Default

      false

      Example
      $ ./cbq --no-engine

      -ncfg

      --networkconfig

      string (auto, default, external)

      Specifies whether to connect to a node’s principal or alternate address.

      • auto — Select the principal address or alternate address automatically, depending on the input IP.

      • default — Use the principal address.

      • external — Use the alternate addresses.

      Default

      auto

      Example
      $ ./cbq -ncfg default -e http://localhost:8091

      -q

      --quiet

      boolean [1]

      When specified, disables the startup connection message for the cbq shell.

      Default

      false

      Example
      $ ./cbq -q -e http://localhost:8091
      Result
      cbq>

      -ad

      --advise

      boolean [1]

      Runs ADVISE on all queries in the specified file, or that are read from standard input, if a file is not provided with the -file option.

      Default

      false

      Example
      ./cbq -advise -file queries.txt
      SELECT ADVISOR(["select * from collection1 where id = 1;",
                      "select * from collection2 where name is not missing;"])
      Result
      {
          "requestID": "15ed5c93-e5f6-4193-83fa-6fdc87847552",
          "signature": {
              "$1": "object"
          },
          "results": [
          {
              "$1": {
                  "recommended_indexes": [
                      {
                          "index": "CREATE INDEX adv_id ON `collection1`(`id`)",
                          "statements": [
                              {
                                  "run_count": 1,
                                  "statement": "select * from collection1 where id = 1;"
                              }
                          ]
                      },
                      {
                          "index": "CREATE INDEX adv_name ON `collection2`(`name`)",
                          "statements": [
                              {
                                  "run_count": 1,
                                  "statement": "select * from collection2 where name is not missing;"
                              }
                          ]
                      }
                  ]
              }
          }
          ]
      }

      -a

      --analytics

      boolean [1]

      Only applicable when connecting to the Analytics Service. When specified, if you are connecting to a cluster, cbq automatically discovers and connects to an Analytics node. This option also switches on batch mode.

      Default

      false

      Example
      $ ./cbq --analytics

      -b

      --batch

      string (on, off) [2]

      This option is available only with the Analytics Service. When specified, cbq sends the queries to server only when you hit EOF or \ to indicate the end of the batch input.

      Default

      off

      Examples
      $ ./cbq --batch

      You can also set the batch mode in the interactive session using the \SET command:

      cbq> \set batch on;
      cbq> \set batch off;

      -qc

      --query_context

      string

      Sets the query context parameter. For more information, see Query Context.

      Shell command: \SET -query_context

      Default

      none

      Example
      $ ./cbq -e http://localhost:8091 -qc "travel-sample.inventory"

      -t

      --timeout

      string (duration)

      Sets the query timeout parameter. For more information, see timeout.

      Shell command: \SET -timeout

      Default

      0ms

      Example
      $ ./cbq -e http://localhost:8091 --timeout="1s"

      For further examples, see Connection Timeout Parameter.

      -u

      --user

      string

      Specifies a single user name to log in to Couchbase. When used by itself, without the -p option to specify the password, you will be prompted for the password.

      This option requires administration credentials and you cannot switch the credentials during a session.

      Couchbase recommends using the -u and -p option if your password contains special characters such as #, $, %, &, (,), or '.

      Default

      none

      Example
      $ ./cbq -e http://localhost:8091 -u=Administrator
      Result
      Enter Password:

      -p

      --password

      string

      Specifies the password for the given user name. You cannot use this option by itself. It must be used with the -u option to specify the user name.

      This option requires administration credentials and you cannot switch the credentials during a session.

      Couchbase recommends using the -u and -p option if your password contains special characters such as #, $, %, &, (,), or '.

      Default

      none

      Example
      $ ./cbq -e http://localhost:8091 -u=Administrator -p=password

      -c

      --credentials

      string

      Specify the login credentials in the form of username:password. You can specify credentials for different keyspaces by separating them with a comma.

      Shell command: \SET -creds

      REST API: -creds parameter

      Default

      none

      Example
      $ ./cbq -e http://localhost:8091 -c=travel-sample:password,Administrator:password

      -v

      --version

      boolean [1]

      When specified, provides the version of the cbq shell. To display the query engine version of Couchbase Server (this is not the same as the version of Couchbase Server itself), use one of the following SQL++ queries:

      select version();
      select min_version();

      Shell command: \VERSION

      Default

      false

      Example
      $ ./cbq --version
      Result
      GO VERSION : go1.21.6
      SHELL VERSION : 7.6.0-2176
      
      Use N1QL queries select version(); or select min_version(); to display server version.

      -h

      --help

      none

      Provides help for the command line options.

      Shell command: \HELP

      Default

      none

      Example
      $ ./cbq --help

      -s

      -script

      string

      Provides a single command mode to execute a query from the command line.

      You can also use multiple -s options on the command line. If one of the commands is incorrect, an error is displayed for that command and cbq continues to execute the remaining commands.

      Default

      none

      Examples
      $ ./cbq -u Administrator -p password -s="SELECT * FROM \`travel-sample\`.inventory.airline LIMIT 1;"
      $ ./cbq -s="\SET v 1" -s="\SET b 2" -s="\PUSH b3" -s="\SET b 5" -s="\SET" -ne
      Result
      Path to history file for the shell : ~/.cbq_history
      
       \SET v 1
       \SET b 2
       \PUSH b3
       ERROR 139 : Too few input arguments to command.
       \SET b 5
       \SET
      
       Query Parameters :
       Named Parameters :
       User Defined Session Parameters :
       Predefined Session Parameters :
       Parameter name : histfile
       Value : [".cbq_history"]
       Parameter name : batch
       Value : ["off"]
       Parameter name : quiet
       Value : [false]
       Parameter name : v
       Value : [1]
       Parameter name : b
       Value : [5]

      -f

      --file

      string (path)

      Provides an input file which contains all the commands to be run.

      Shell command: \SOURCE

      Default

      none

      Example
      $ ./cbq --file="sample.txt"

      -o

      --output

      string (path)

      Specifies an output file where the commands and their results are to be written.

      If the file doesn’t exist, it is created. If the file already exists, it is overwritten.

      Shell command: \REDIRECT

      Default

      none

      Example
      $ ./cbq -u Administrator -p password -o="results.txt" -s="SELECT * FROM \`travel-sample\`.inventory.airline LIMIT 1;"

      --pretty

      boolean [1]

      Specifies whether the output should be formatted with line breaks and indents.

      This option is set to true by default. To specify that the output should not be formatted with line breaks and indents, you must explicitly set this option to false.

      Default

      true

      Example
      $ ./cbq -u Administrator -p password --pretty=false -s="SELECT * FROM \`travel-sample\`.inventory.airline LIMIT 1;"

      --exit-on-error

      boolean [1]

      When specified, the cbq shell must exit when it encounters the first error.

      Default

      false

      Example
      $ ./cbq --exit-on-error -f="sample.txt"

      --cacert

      string (path)

      Only applicable when using an encrypted protocol scheme — either https:// or couchbases://.

      Specifies the path to the root CA certificate to verify the identity of the server.

      Default

      none

      Example
      $ ./cbq --cacert ./root/ca.pem

      --cert

      string (path)

      Only applicable when using an encrypted protocol scheme — either https:// or couchbases://.

      Specifies the path to the chain certificate.

      Default

      none

      Example
      $ ./cbq --cert ./client/client/chain.pem

      --key

      string (path)

      Only applicable when using an encrypted protocol scheme — either https:// or couchbases://.

      Specifies the path to the client key file.

      Default

      none

      Examples
      $ ./cbq --key ./client/client/client.key

      --no-ssl-verify or

      -skip-verify

      boolean [1]

      Only applicable when using an encrypted protocol scheme — either https:// or couchbases://.

      When specified, the cbq shell can skip the verification of certificates.

      Default

      false

      Examples
      $ ./cbq --no-ssl-verify -f="sample.txt"
      $ ./cbq -skip-verify https://127.0.0.1:18091
      Table 2. cbq Shell Commands
      Shell Command Arguments Description and Examples

      \CONNECT

      url

      Connects cbq shell to the specified query engine or Couchbase cluster.

      The connection string consists of a protocol scheme, followed by a host, and optionally a port number to connect to the query service (8093) or the Couchbase cluster (8091).

      For more details, refer to Connecting to the Cluster or Query Node.

      Command Line Option: -e or --engine

      Examples
      cbq> \CONNECT http://localhost:8093;
      cbq> \CONNECT http://[fd63:6f75:6368:1075:816:3c1d:789b:bc4]:8091;

      \DISCONNECT

      none

      Disconnects the cbq shell from the query service or cluster endpoint.

      Example
      cbq> \DISCONNECT;
      Result
       Couchbase query shell not connected to any endpoint.
       Use \CONNECT command to connect.

      \EXIT

      \QUIT

      none

      Exits cbq shell.

      Examples
      cbq> \EXIT;
      cbq> \QUIT;

      \SET

      parameter value

      parameter = prefix : variable name

      Sets the top most value of the stack for the given variable with the specified value.

      Variables can be of the following types:

      • Query parameters

      • Session variables

      • User-defined

      • Pre-defined and named parameters.

      When the \SET command is used without any arguments, it displays the values for all the parameters of the current session.

      Examples
      cbq> \SET -args [5, "12-14-1987"];
      cbq> \SET -args [6,7];

      \PUSH

      parameter value

      Pushes the specified value on to the given parameter stack.

      When the \PUSH command is used without any arguments, it copies the top element of every variable’s stack, and then pushes that copy to the top of the respective variable’s stack.

      While each variable stack grows by 1, the previous values are preserved.

      Examples
      cbq> \PUSH -args  [8];
      cbq> \PUSH;
      Check variable stack
      cbq> \SET;
      Result
       Query Parameters :
       Parameter name : args
       Value : [[6,7] [8] [8]]
      ...

      \UNSET

      parameter

      Deletes or resets the entire stack for the specified parameter.

      Examples
      cbq> \UNSET -args;
      cbq> \SET;
      Result
       Query Parameters :
       ...

      \POP

      parameter

      Pops the top most value from the specified parameter’s stack.

      When the \POP command is used without any arguments, it pops the top most value of every variable’s stack.

      Examples
      cbq> \POP -args;
      cbq> \SET;
      Result
       Query Parameters :
       Parameter name : args
       Value : [[6,7] [8]]

      \ALIAS

      shell-command or n1ql-statement

      Creates a command alias for the specified cbq shell command or SQL++ statement. You can then execute the alias using \\alias-name;.

      When the \ALIAS command is used without any arguments, it lists all the available aliases.

      Examples
      cbq> \ALIAS travel-limit1 SELECT * FROM `travel-sample`.inventory.airline LIMIT 1;
      cbq> \ALIAS;
      Result
      serverversion  select version()
      travel-limit1  SELECT * FROM `travel-sample`.inventory.airline LIMIT 1
      Execute alias
      cbq> \\serverversion;
      Result
      {
          "requestID": "ef63f01b-f159-437f-a4df-28d6145fa3c2",
          "signature": {
              "$1": "string"
          },
          "results": [
              {
                  "$1": "7.0.0-N1QL"
              }
          ],
          "status": "success",
          "metrics": {
              "elapsedTime": "14.54962ms",
              "executionTime": "13.164635ms",
              "resultCount": 1,
              "resultSize": 34,
              "serviceLoad": 12
          }
      }

      \UNALIAS

      alias-name

      Deletes the specified alias.

      Examples
      cbq> \UNALIAS travel-limit1;
      cbq> \ALIAS;
      Result
      serverversion  select version()

      \ECHO

      args

      where args can be parameters, aliases, or any input.

      If the input is a parameter, this command echoes (displays) the value of the parameter. The parameter must be prefixed according to its type. See Table 3 for details.

      If the input is not a parameter, the command echoes the statement as is.

      If the input is an alias, the command displays the value of an alias command.

      Examples
      cbq> \ECHO -$r;
      cbq> \ECHO \\serverversion;
      Result
      select version()

      \VERSION

      none

      Displays the version of the client shell.

      Command Line Option: -v or --version

      Example
      cbq> \VERSION;
      Result
      GO VERSION : go1.21.6
      SHELL VERSION : 7.6.0-2176
      
      Use N1QL queries select version(); or select min_version(); to display server version.

      \HELP

      command

      Displays the help information for the specified command. When used without any arguments, it lists all the commands supported by the cbq shell.

      Command Line Option: -h or --help

      Example
      cbq> \HELP ECHO;
      Result
      \ECHO args ...
      Echo the input value. args can be a name (a prefixed-parameter), an alias (command alias) or
      a value (any input statement).
      Example :
      \ECHO -$r ;
      \ECHO \\tempalias;

      \COPYRIGHT

      none

      Displays the copyright, attributions, and distribution terms.

      Example
      cbq> \COPYRIGHT;

      \SOURCE

      input-file

      Reads and executes the commands from a file. Multiple commands in the input file must be separated by ; <newline>.

      Command Line Option: -f or --file

      For example, sample.txt contains the following commands:

      SELECT * FROM `travel-sample`.inventory.airline LIMIT 1;
      \ECHO this;
      #This is a comment;
      Example
      cbq> \SOURCE sample.txt;

      \REDIRECT

      filename

      Redirects the output of all the commands to the specified file until the cbq shell receives the \REDIRECT OFF command. By default, the file is created in the directory that you were in when you started the cbq shell. You can specify a different location using relative paths.

      If the file doesn’t exist, it is created. If the file already exists, it is overwritten. You can append redirected output to an existing file using File Append Mode.

      Command Line Option: -o or --output

      Example
      cbq> \REDIRECT temp_out.txt;

      \REDIRECT OFF

      none

      Redirects the output of subsequent commands from a custom file to standard output (os.stdout).

      Example
      cbq> \REDIRECT OFF;

      Connecting to the Cluster or Query Node

      You can connect the cbq shell to Couchbase Server either through the query service or through the cluster endpoint. There are two ways to establish a connection:

      • Using an option on startup:

        -e <url to query engine or Couchbase cluster>
        --engine=<url to query engine or Couchbase cluster>
      • Using a shell command:

        cbq> \CONNECT url;

      The url may contain up to three components: the protocol scheme, the host, and a port number. The URL is optional and if it is not specified, the default URL http://localhost:8091 is used. An error is thrown if the URL is invalid.

      The cbq shell supports the http://, https://, couchbase:// and couchbases:// protocol schemes. The https:// and couchbases:// protocol schemes are encrypted. For more details, refer to Using an Encrypted Connection.

      The host may be the IP address or hostname of any node in the cluster, as cbq will automatically discover the query nodes. The cbq shell supports both IPV4 and IPV6 addresses.

      The couchbase:// and couchbases:// protocol schemes support the domain name service (DNS). When using one of these protocol schemes, the host may be a domain name which is resolved using DNS. For example, this enables you to connect to a cluster or node over the internet.

      Note that you must use the encrypted couchbases:// protocol scheme to connect to a cluster or node deployed in Couchbase Capella.

      You may optionally specify the port when using the http:// or https:// protocol schemes. When connecting to the query service, use the query port 8093, or 18093 for an encrypted connection. When connecting to the cluster, you don’t need to specify the port as the connection uses round robin to find a query service to connect to. If you want to specify a port, use the admin port 8091, or 18091 for an encrypted connection.

      You cannot specify the port when using the couchbase:// or couchbases:// protocol schemes.

      You can close the connection with an existing node or cluster without exiting the shell at any given time during the session using the \DISCONNECT; command. If the shell is not connected to any endpoint, an error with a message that the shell is not connected to any instance is thrown.

      Examples
      $ ./cbq -e http://localhost:8091 -u Administrator -p password
      Connected to : http://localhost:8091/. Type Ctrl-D or \QUIT to exit.
      
      Path to history file for the shell : ~/.cbq_history
      
      cbq> \DISCONNECT;
      Couchbase query shell not connected to any endpoint. Use \CONNECT command to connect.
      
      cbq> \CONNECT http://127.0.0.1:8091;
      Connected to : http://127.0.0.1:8091. Type Ctrl-D or \QUIT to exit.
      
      cbq> \EXIT;
      
      $ ./cbq -e http://127.0.0.1:8091 -u Administrator -p password
      Connected to : http://127.0.0.1:8091/. Type Ctrl-D or \QUIT to exit.
      
      Path to history file for the shell : ~/.cbq_history
      cbq>

      Bringing Up an Unconnected Instance

      You can bring up the shell without connecting to any query service or cluster endpoint by using the -ne or --no-engine option. After starting cbq without any service, you can connect to a specific endpoint using the CONNECT command.

      Example
      $ ./cbq -ne
      Path to history file for the shell : ~/.cbq_history
      
      cbq> \CONNECT http://Administrator:password@localhost;
      Connected to : http://Administrator:password@localhost:8091. Type Ctrl-D or \QUIT to exit.

      Exiting the cbq Shell

      You can exit the cbq shell by pressing Ctrl+D or by using one of the following commands:

      cbq> \EXIT;
      cbq> \QUIT;

      When you run the exit command, the cbq shell first saves the history, closes existing connections, saves the current session in a session file, resets all environment variables, and then closes the shell liner interface.

      Example
      $ ./cbq -u Administrator -p password
      Connected to : http://localhost:8091/. Type Ctrl-D or \QUIT to exit.
      Path to history file for the shell : ~/.cbq_history
      
      cbq> SELECT name FROM `travel-sample`.inventory.airline LIMIT 1;
      {
          "requestID": "59d1c699-11a2-47c6-b4d0-4a7de1d15a3c",
          "signature": {
              "name": "json"
          },
          "results": [
          {
              "name": "40-Mile Air"
          }
          ],
          "status": "success",
          "metrics": {
              "elapsedTime": "13.514441ms",
              "executionTime": "13.355058ms",
              "resultCount": 1,
              "resultSize": 37,
              "serviceLoad": 12
          }
      }
      
      cbq> \EXIT;
      $

      Accessing a Secure Keyspace

      If your keyspace has a password, you can pass the keyspace name and keyspace password like so:

      $ ./cbq -engine="http://<keyspacename>:<keyspacepassword>@localhost:8091/"

      For the 'travel-sample' keyspace, if you add a password to it of w1fg2Uhj89 (as by default it has none), the command to start cbq would look like this:

      $ ./cbq -engine="http://travel-sample:w1fg2Uhj89@localhost:8091/"
      These commands execute successfully only if you have loaded sample bucket 'travel-sample' either at install or from the Settings menu in the web UI.

      If you want to access all of the keyspaces in the same cbq session, you would pass in the Administrator username and password instead of the keyspace level.

      $ ./cbq -engine="http://Administrator:password@localhost:8091/"

      Providing Single User Credentials

      You can pass a single user name credential to the cbq shell on startup using the command line options:

      -u=username
      --user=username

      The shell then prompts you for a password. You can also provide a single password credential using the -p option. You cannot use this option by itself. It must be used with the -u option to specify the user name that the password is associated with.

      -p=password
      --password=password
      Example
      $ ./cbq -u=Administrator
      Enter Password:
      Connected to : http://localhost:8091/. Type Ctrl-D or \QUIT to exit.
      
      $ ./cbq -e http://localhost:8091 -u=Administrator -p=password
      Connected to : http://localhost:8091/. Type Ctrl-D or \QUIT to exit.
      cbq>

      For information on passing the user name and password credentials using other mechanisms, see Passing Credentials Using the SET Shell Command and Passing Credentials Using REST API.

      Providing Multiple Credentials for Authorization

      The cbq shell supports self-signed certificates for encrypting communication between clusters.

      Using the cbq shell, you can set the credentials for different users on startup or by using the SET shell commands to set the credentials query parameter. You can also use this to provide authentication credentials for multiple SASL buckets per session. Depending on the type of credential being set, there are multiple possible values for the credentials query parameter.

      To set the credentials for different users on startup, use one of the following options:

      -c=list-of-creds
      --credentials=list-of-creds

      The list-of-creds can take either one or multiple credentials. The credentials consist of an identity and a password separated by a colon :. To specify multiple credentials, append all the user names and passwords to the same credentials array. For example:

      -c=travel-sample:pwd1,beer-sample:pwd2

      For information on passing a single user name credential to the cbq shell, see Providing Single User Credentials.

      Passing Credentials Using the SET Shell Command

      You can provide the credential types using the SET command.

      The credentials are set for the shell session and not on a per query basis. You can use the SET, PUSH, POP and UNSET commands to reset the credentials during a session.

      To pass authentication credentials per query, set the query parameter to a new value using the SET shell command before executing the query.

      You can also switch between users and change credentials during a session. To do so, set the -creds query parameter for the session using the following command:

      cbq> \SET -creds travel-sample:b1, session:s1;

      Passing Credentials Using Query REST API

      You can use query REST API to pass credentials from clients.

      For SASL buckets, you can pass the credentials as:

      [  {
           "user":"travel-sample",
           "pass":"password"
         }  ]

      If you are using the Administrator credentials:

      [  {
              "user":"Administrator",
              "pass":"password"
         }  ]

      For multiple SASL protected buckets, you can pass an array of authentication credentials:

      [  {
              "user":"beer-sample",
              "pass":"password1"
              },
              {
              "user":"travel-sample",
              "pass":"password2"
         }  ]

      Displaying the Credentials

      You can display the credentials for the current session using the ECHO shell command. This command displays only the user names (and not the passwords).

      cbq> \ECHO -creds;
      
      Administrator:*

      You can also display a full list of variables using the SET command specified without any arguments.

      cbq> \SET;
      Query Parameters ::
      Parameter name : timeout Value  ["3ms" "4s"]
      
      Named Parameters ::
      Parameter name : r Value  [9.5 9.5]
      
      User Defined Session Parameters ::
      
      Predefined Session Parameters ::
      Parameter name : histfile Value  [".cbq_history"]

      Using an Encrypted Connection

      You can connect to the cluster or node with an encrypted protocol scheme — that is, either https:// or couchbases://. To do this, you can provide the root CA certificate, the chain certificate, and the client key file using the --cacert, --cert, and --key options. You can use the --no-ssl-verify option to skip the verification of certificates.

      When connecting to a cluster or node with an encrypted protocol scheme, the default ports are 18091 and 18093. You need not specify the port when connecting to the cluster.

      You can use the encrypted couchbases:// protocol scheme with a domain name to connect to a node or cluster deployed in Couchbase Capella. For more details, refer to Connecting to the Cluster or Query Node.

      Parameter Manipulation

      The cbq shell categorizes parameters into the following types:

      • Named Parameters

      • REST API Parameters

      • Session or Pre-defined Parameters

      • User-defined Parameters

      Parameter Configuration

      When using parameters, you can set a stack of values for each parameter. You can either push a new value onto the stack using the PUSH command, or set the current value for a parameter using the SET command. The SET command always modifies the top of a variable’s stack while the PUSH command adds to the stack. When you use PUSH with no arguments, it copies the top element of every parameter’s (except the predefined parameters) stack and pushes that copy to the top of its respective stack. As a result, each stack grows by 1, but the values are preserved. You can then use the SET command to modify the top value.

      To unset the values from a parameter’s stack, you can use the UNSET command to remove all the values from the stack and delete the corresponding parameter stack. However, if you want to delete a single value from the settings, use the POP command. When you use the POP command with no arguments, it pops the one value from the top of each parameter’s stack.

      Setting Variable Values

      Each variable has a separate stack associated with it and the prefix name argument helps distinguish between the stacks.

      The SET command always modifies the top value of a variable. You can use the SET command to set different kinds of parameters: query parameter, predefined session variables, user-defined session variables and named parameters.

      cbq> \SET <prefix><name> value;

      where name is the name of the parameter, value is the value to be set, and prefix is one of the following depending on the parameter type. The cbq shell uses the prefix to differentiate between the different types of parameters.

      Table 3. Prefixes for Parameters
      Prefix Parameter Type

      -

      Query parameter

      -$ or -@

      Named parameters

      No prefix

      Predefined (built-in) session variable

      $

      User defined session variable

      Positional parameters are set using the -args query parameter.

      You can use the cbq shell to set all the REST API settings by specifying the settings as query parameters prefixed by -. As a best practice, we recommend that you save the initial set of basic parameters and their default values using the \PUSH command (with no arguments).

      Examples
      cbq> \SET -$airport "SJC";
      cbq> \PUSH -args ["LAX", 6];
      cbq> \SET;
      Query Parameters ::
      Parameter name : args Value  [["LAX",6]]
      
      Named Parameters ::
      Parameter name : airport Value  ["SJC"]
      
      User Defined Session Parameters ::
      
      Predefined Session Parameters ::
      Parameter name : histfile Value  [".cbq_history"]
      
      cbq> \PUSH -$airport "SFO";
      cbq> \PUSH;
      cbq> \SET;
      Query Parameters ::
      Parameter name : args Value  [["LAX",6] ["LAX",6]]
      
      Named Parameters ::
      Parameter name : airport Value  ["SJC" "SFO" "SFO"]
      
      User Defined Session Parameters ::
      
      Predefined Session Parameters ::
      Parameter name : histfile Value  [".cbq_history"]
      
      cbq> \SET -args ["SFO", 8];
      cbq> \SET;
      Query Parameters ::
      Parameter name : args Value  [["LAX",6] ["SFO",8]]
      
      Named Parameters ::
      Parameter name : airport Value  ["SJC" "SFO" "SFO"]
      
      User Defined Session Parameters ::
      
      Predefined Session Parameters ::
      Parameter name : histfile Value  [".cbq_history"]
      
      cbq> \POP;
      cbq> \SET;
      Query Parameters ::
      Parameter name : args Value  [["LAX",6]]
      
      Named Parameters ::
      Parameter name : airport Value  ["SJC" "SFO"]
      
      User Defined Session Parameters ::
      
      Predefined Session Parameters ::
      Parameter name : histfile Value  [".cbq_history"]
      
      cbq> \POP -$airport;
      cbq> \SET;
      Query Parameters ::
      Parameter name : args Value  [["LAX",6]]
      
      Named Parameters ::
      Parameter name : airport Value  ["SJC"]
      
      User Defined Session Parameters ::
      
      Predefined Session Parameters ::
      Parameter name : histfile Value  [".cbq_history"]
      
      cbq> \UNSET -$airport;
      cbq> \SET;
      Query Parameters ::
      Parameter name : args Value  [["LAX",6]]
      
      Named Parameters ::
      
      User Defined Session Parameters ::
      
      Predefined Session Parameters ::
      Parameter name : histfile Value  [".cbq_history"]

      To display all the parameters defined in a session, use the SET command with no arguments.

      cbq> \SET;
      Query Parameters ::
      Parameter name : timeout Value  ["100m"]
      
      Named Parameters ::
      Parameter name : r Value  [9.5]
      
      User Defined Session Parameters ::
      
      Predefined Session Parameters ::
      Parameter name : histfile Value  [".cbq_history"]

      The following table lists the available predefined session variables.

      Table 4. Predefined Session Variables
      Variable Name Possible Values Description

      HISTFILE

      Valid file name

      Specifies the file name to store the command history. By default the file is saved in the user’s home directory.

      Default: .cbq_history

      Handling Named Parameters

      Use the \SET command to define named parameters. For each named parameter, prefix the variable name with -$ or -@. The following example creates named parameters r and date with values 9.5 and "1-1-2016" respectively.

      cbq> \SET -$r 9.5;
      cbq> \SET -@date "1-1-2016";

      Handling Positional Parameters

      Use the SET shell command with the -args query parameter to define positional parameters:

      cbq> \SET -args value;

      The value contains the different values that correspond to positions within the query. For example,

      cbq> \SET -args [ 9.5, "1-1-2016"];

      Resetting Variable Values

      You can reset the value of a variable by either popping it or deleting it altogether. To pop the top of a parameter’s stack use:

      cbq> \POP <prefix><name>;

      To pop the top of every parameter’s stack once, use the POP command without any arguments:

      cbq> \POP;

      To pop all the values of a parameter’s stack and then delete the parameter, use:

      cbq> \UNSET <prefix><name>;

      Using ECHO to Display Values of Parameters and More

      The ECHO command displays the current values of the parameters set for a session. You can use it to display any input string or command aliases that have been created using the ALIAS shell command. To display parameters, you must include their prefixes. If not, the shell considers the parameters as generic statements and displays the parameter as is.

      cbq> \ECHO input ... ;

      where input can be a parameter with prefix (<prefix><parameter-name>), an alias (\\command-alias), a SQL++ statement, or a string.

      Examples
      cbq> \ECHO hello;
      hello
      
      cbq> \ECHO \\travel-alias1;
      SELECT * FROM `travel-sample`.inventory.airline LIMIT 1;
      
      cbq> \ECHO -$r;
      9.5

      Command Alias

      Using the ALIAS shell command, you can define and store aliases for commands. This is useful when you have lengthy queries that need to be executed often. Run the following command to define an alias:

      cbq> \ALIAS command-alias command;
      Example
      cbq> \ALIAS travel-alias1 SELECT * FROM `travel-sample`.inventory.airline LIMIT 1;

      To run the command alias, use \\command-alias.

      Example
      cbq> \\travel-alias1;
      
      {
          "requestID": "b25c84d6-7b7b-440a-a286-5027e6ecbbb5",
          "signature": {
              "*": "*"
          },
          "results": [
          {
              "airline": {
                  "callsign": "MILE-AIR",
                  "country": "United States",
                  "iata": "Q5",
                  "icao": "MLA",
                  "id": 10,
                  "name": "40-Mile Air",
                  "type": "airline"
              }
          }
          ],
          "status": "success",
          ...
      }

      To list all the existing aliases, use:

      cbq> \ALIAS;
      Example
      cbq> \ALIAS;
      serverversion  select version()
      travel-alias1  SELECT * FROM `travel-sample`.inventory.airline LIMIT 1;

      You can delete a defined alias using the \UNLIAS command.

      cbq> \UNALIAS alias-name ... ;
      cbq> \UNALIAS serverversion travel-alias1;
      
      /* Check existing aliases */
      cbq> \ALIAS;
       ERROR 141 : Alias does not exist :

      This command can take multiple arguments and deletes the defined alias for every input name.

      Executing Prepared Statements

      You can use the shell command to execute prepared statements. As a pre-requisite, you must first prepare a statement using the SQL++ PREPARE statement. To execute prepared statements, follow these steps:

      1. Set the named and positional parameters that are present in the prepared statement.

      2. Prepare using the SQL++ PREPARE statement. This can be either prepared statement or a named prepared statement. If you do not specify a name for the prepared statement (PREPARE query;), a unique name is assigned. You can use this auto-assigned name when executing the prepared statement. If you specify a name (PREPARE name FROM query;), you can use this name to run the prepared statement.

      3. Execute the prepared statement using the shell command:

        EXECUTE name-of-prepared-stmt;

      Canceling a Query

      You can cancel a running query by using the Ctrl+C keys.

      Connection Timeout Parameter

      You can use the timeout parameter to limit the running time of a query. This parameter specifies the time to wait before returning an error when executing a query.

      --t=value
      --timeout=value

      Timeout can be specified in the following units: ns for nanoseconds, μs for microseconds, ms for milliseconds, s for seconds, m for minutes, and h for hours. Examples of valid values include "0.5s", "10ms", or "1m".

      You can also the SET shell command to set this parameter. An error is thrown if the timeout is invalid.

      $ ./cbq --timeout="2s"
      
      $ ./cbq -q
      cbq> \SET -TIMEOUT 1ms;

      File Based Operations

      Using the file based commands and options, the cbq shell can execute SQL++ and shell commands contained in files. There are two ways to accomplish this:

      • Using an option on startup:

        -f=input-file
        --file=input-file

        The cbq shell executes the commands present in the input file, prints them to stdout (or to a file if using redirects), and exits.

      • Using a shell command:

        cbq> \SOURCE input-file;

        Runs the commands present in the input file and prints the result to stdout.

      Consider the input file, sample.txt, containing the following commands:

      CREATE PRIMARY INDEX ON `travel-sample`.inventory.airline USING GSI;
      SELECT * from `travel-sample`.inventory.airline LIMIT 2;
      SELECT callsign from `travel-sample`.inventory.airline LIMIT 3;
      \HELP;

      To execute the commands contained in sample.txt using the -f option, run $./cbq -f=sample.txt

      Results
       Connected to : http://localhost:8091/. Type Ctrl-D or \QUIT to exit.
      
       Path to history file for the shell : ~/.cbq_history
      CREATE PRIMARY INDEX ON `travel-sample`.inventory.airline USING GSI;
      { ...
        "results": [ ],
        ...
      }
      SELECT * from `travel-sample`.inventory.airline LIMIT 2;
      { ...
        "results": [ ],
        ...
      }
      SELECT callsign from `travel-sample`.inventory.airline LIMIT 3;
      { ...
        "results": [ ],
        ...
      }
      \HELP;
      Help information for all shell commands.
      ...
      $

      To execute the commands contained in sample.txt using the shell command, run cbq> \SOURCE sample.txt;

      Results
      CREATE PRIMARY INDEX ON `travel-sample`.inventory.airline USING GSI;
      { ...
        "results": [ ],
       ...
      }
      SELECT * from `travel-sample`.inventory.airline LIMIT 2;
      { ...
        "results": [ ],
        ...
      }
      SELECT callsign from `travel-sample`.inventory.airline LIMIT 3;
      { ...
        "results": [ ],
        ...
      }
      \HELP;
      Help information for all shell commands.
      ...
      cbq>

      Redirecting Results to a File

      You can redirect all the output for a session or part of a session to a specified file by using the following option:

      -o filename
      --output=filename

      To redirect a specific set of commands during a session, you must specify the commands between \REDIRECT and \REDIRECT OFF as shown:

      cbq> \REDIRECT filename;
      command-1; command-2;, ..., command-n;
      cbq> \REDIRECT OFF;

      All the commands specified after \REDIRECT and before \REDIRECT OFF are saved into the specified output file.

      If the file doesn’t exist, it is created. If the file already exists, it is overwritten. You can append redirected output to an existing file using File Append Mode.

      Example
      cbq> \REDIRECT temp_output.txt;
      cbq> CREATE PRIMARY INDEX ON `travel-sample`.inventory.airline USING GSI;
      cbq> SELECT * FROM `travel-sample`.inventory.airline LIMIT 1;
      cbq> \HELP;
      cbq> \REDIRECT OFF;

      You can specify multiple \REDIRECT commands. When you do so, the output file changes to the specified files and switches back to stdout only when you specify \REDIRECT OFF.

      File Append Mode

      You can use file append mode to specify that cbq should append redirected output to the end of an existing file, rather than overwriting the existing file.

      To use file append mode, include a plus sign + at the start of the output path or filename.

      Example
      cbq> \REDIRECT +temp_output.txt;
      cbq> SELECT * FROM `travel-sample`.inventory.airline LIMIT 1;
      cbq> \REDIRECT OFF;

      Every time you start appending to the output file, a timestamp is added to the end of the output file, followed by any redirected commands and results.

      -- <2021-07-30T14:48:43.661+01:00> : opened in append mode
      
      SELECT * FROM `travel-sample`.inventory.airline LIMIT 1
      ...

      Note that file append mode is only available with the \REDIRECT command within a shell session. It is not available for the -o or --output command line option. When you use the -o or --output command line option, the specified output file is always overwritten.

      Server and Shell Information

      The cbq shell provides commands that convey information about the shell or cluster endpoints.

      Version

      You can find the version of the client (shell) by using either the command line option to display the current version of the shell and exit, or as a shell command to print the version of the shell during the shell session.

      Example Using the Command-line Option
      $ ./cbq -v
      GO VERSION : go1.21.6
      SHELL VERSION : 7.6.0-2176
      
      Use N1QL queries select version(); or select min_version(); to display server version.
      
      $ ./cbq --version
      GO VERSION : go1.21.6
      SHELL VERSION : 7.6.0-2176
      
      Use N1QL queries select version(); or select min_version(); to display server version.
      Example Using the Shell Command
      cbq> \VERSION;
      GO VERSION : go1.21.6
      SHELL VERSION : 7.6.0-2176
      
      Use N1QL queries select version(); or select min_version(); to display server version.

      To display the version of the query service, use the SQL++ commands SELECT version(); and SELECT min_version();.

      You can view the copyright, attributions, and distribution terms of the command line query tool using the \COPYRIGHT; command.

      cbq> \COPYRIGHT;
      Copyright (c) 2016 Couchbase, Inc. Licensed under the Apache License, Version 2.0 (the "License");
      you may not use this file except in compliance with the License. You may obtain a copy of the
      License at http://www.apache.org/licenses/LICENSE-2.0
      Unless required by applicable law or agreed to in writing, software distributed under the
      License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,
      either express or implied. See the License for the specific language governing permissions
      and limitations under the License.
      cbq>

      Shortcut Keys for cbq Shell

      The following table lists the shortcut keys used by the cbq shell.

      Table 5. Shortcut Keys for cbq Shell
      Keystroke Action

      Ctrl+A, Home

      Move cursor to beginning of line

      Ctrl+E, End

      Move cursor to end of line

      Ctrl+B, Left

      Move cursor one character left

      Ctrl+F, Right

      Move cursor one character right

      Ctrl+Left

      Move cursor to previous word

      Ctrl+Right

      Move cursor to next word

      Ctrl+D, Del

      (if line is not empty) Delete character under cursor

      Ctrl+D

      (if line is empty) End of File - usually quits application

      Ctrl+C

      Reset input (create new empty prompt)

      Ctrl+L

      Clear screen (line is unmodified)

      Ctrl+T

      Transpose previous character with current character

      Ctrl+H, BackSpace

      Delete character before cursor

      Ctrl+W

      Delete word leading up to cursor

      Ctrl+K

      Delete from cursor to end of line

      Ctrl+U

      Delete from start of line to cursor

      Ctrl+P, Up

      Previous match from history

      Ctrl+N, Down

      Next match from history

      Ctrl+R

      Reverse Search history (Ctrl+S forward, Ctrl+G cancel)

      Ctrl+Y

      Paste from Yank buffer (Alt+Y to paste next yank instead)

      Tab

      Next completion

      Shift+Tab

      (after Tab) Previous completion

      Source: https://github.com/peterh/liner


      1. Invoking a boolean option with no value sets the value to true.
      2. Invoking this option with no value sets the value to on.