Couchbase

Welcome to the Couchbase Shell cbsh documentation! If you are new, you will want to start with the quickstart section and then proceed with the introduction. If you are already familar with the shell, feel free to jump right into the command reference.

Note that while the project is maintained by Couchbase, it is not covered under the EE support contract. We are providing community support through this bug tracker.

1. Quickstart

1.1. Installation

The current latest version is 0.75. The version of cbsh is kept in line with the underlying version of nushell that is supported.

There are a couple ways you can get access to cbsh, the easiest one is to download our pre-built binaries for your platform of choice:

Once you’ve downloaded the zip file, extract it and switch into the just created directory. The following example shows it for mac, but it works very similar if you are on linux (just align the commands with the file you just downloaded):

$ unzip cbsh-x86_64-apple-darwin.zip
$ ls
cbsh LICENSE LICENSE_AGREEMENT  README.md

You can now run the cbsh binary:

❯ ./cbsh --version
The Couchbase Shell 0.75.1
Tip
 If you are running a recent macOS release (i.e. 10.15.x), you’ll likely see an error similar to "cbsh" was blocked from use because it is not from an identified developer. This is because our binaries are not yet signed. To run it nonetheless you need to either navigate to System Preferences → Security & Privacy and click Allow Anyway, or run sudo xattr -r -d com.apple.quarantine $PWD/cbsh inside your terminal. Next time you run the binary you’ll get another prompt but then it should run fine.
macOS Warning

1.2. Connecting to a Database

The first time that you run ./cbsh you will receive a prompt asking if you’d like to create a config file. If you choose yes then the shell will provide you with a series of prompts to provide information about your default database. If you choose no then it will try to connect to localhost using the Administrator username and the password password.

You can modify this through CLI arguments (see ./cbsh -h for more information).

Note: Unless you specify TLS settings then PLAIN authentication is used and your credentials are sent in plaintext.

❯ ./cbsh --username Administrator --connstr 127.0.0.1 -p
Password:
[WARN] 2023-04-14 08:32:25.180 Using PLAIN authentication for cluster dev.local, credentials will sent in plaintext - configure tls to disable this warning
[INFO] 2023-04-14 08:32:25.389 Thanks for trying CBSH!
👤 Administrator 🏠 dev.local in 🗄 default
>

Once in the shell, you can start to execute commands (see the introduction section for more information). As a quick sanity check, list the nodes in the database:

> nodes
╭───┬───────────┬────────────────┬─────────┬──────────────────────────┬───────────────────────┬───────────────────────────┬──────────────┬─────────────┬─────────╮
│ # │  cluster  │    hostname    │ status  │         services         │        version        │            os             │ memory_total │ memory_free │ capella │
├───┼───────────┼────────────────┼─────────┼──────────────────────────┼───────────────────────┼───────────────────────────┼──────────────┼─────────────┼─────────┤
│ 0 │ dev.local │ 127.0.0.1:8091 │ healthy │ search,indexing,kv,query │ 8.0.0-1246-enterprise │ x86_64-apple-darwin19.6.0 │  34359738368 │ 12026126336 │ false   │
╰───┴───────────┴────────────────┴─────────┴──────────────────────────┴───────────────────────┴───────────────────────────┴──────────────┴─────────────┴─────────╯

Or if you have the travel-sample bucket installed you can switch to it and then fetch a document:

>  doc get airline_10 --bucket travel-sample | flatten
───┬────────────┬─────────────────────┬────────────┬─────────┬─────────────┬──────┬──────┬──────────┬───────────────┬───────┬─────────
 # │     id     │         cas         │ content_id │  type   │    name     │ iata │ icao │ callsign │    country    │ error │ database
───┼────────────┼─────────────────────┼────────────┼─────────┼─────────────┼──────┼──────┼──────────┼───────────────┼───────┼─────────
 0 │ airline_10 │ 1629809626107281408 │         10 │ airline │ 40-Mile Air │ Q5   │ MLA  │ MILE-AIR │ United States │       │ default
───┴────────────┴─────────────────────┴────────────┴─────────┴─────────────┴──────┴──────┴──────────┴───────────────┴───────┴─────────

1.3. The config dotfiles

Connecting to a single database through the command line is nice when you are starting out, but later on you will likely either connect to the same database all the time or even to a multitude of them. To help with this, you can create a .cbsh dot folder in your home directory and place a config file in it that the shell will read on startup.

The downloaded zip contains an example already, but here is a small sample config to help you get started as well:

version = 1

[[database]]
identifier = "local"
connstr = "127.0.0.1"
default-bucket = "travel-sample"
username = "Administrator"
password = "password"

[[database]]
identifier = "remote"
connstr = "10.143.200.101"
default-bucket = "myapp"
username = "user"
password = "pass"

This will register two databases, one called local and one called remote. The file format is toml in case you wonder. Now when you start the shell, it will connect to local automatically and you are all set.

Please check out the reference section on additional parameters you can set as well as how to move the credentials to a separate credentials file in case you want to share your config with other people and they do not use the same credentials.

2. Introduction

Couchbase Shell is fully featured, so it does not only contain commands related to couchbase but is actually built on top of a general purpose shell called nushell. This allows you to interact with the file system or any other command available on your machine, making it a great tool for both operational and development tasks on top of Couchbase.

The following introduction only touches on the basic concepts to make you productive quickly. We recommend also checking out the great nushell documentation so you can get the most out of it.

2.1. Navigating the Shell

Commands take inputs and produce output in a structured manner, most often represented as tables. Note how both the generic ls command and the couchbase-specific buckets command both produce a table as their output:

> ls
────┬──────────────┬──────┬──────────┬────────────────
 #  │     name     │ type │   size   │    modified
────┼──────────────┼──────┼──────────┼────────────────
  0 │ CHANGELOG.md │ File │   4.8 KB │ 2 hours ago
  1 │ Cargo.lock   │ File │ 170.2 KB │ 16 minutes ago
  2 │ Cargo.toml   │ File │   1.8 KB │ 16 minutes ago
  3 │ LICENSE      │ File │  11.4 KB │ 2 days ago
  4 │ README.md    │ File │   8.6 KB │ 9 minutes ago
  5 │ docs         │ Dir  │    544 B │ 2 days ago
  6 │ examples     │ Dir  │    192 B │ 2 days ago
  7 │ jupyter      │ Dir  │    128 B │ 2 days ago
  8 │ src          │ Dir  │    256 B │ 2 days ago
  9 │ target       │ Dir  │    224 B │ 32 minutes ago
 10 │ tests        │ Dir  │    224 B │ 2 days ago
────┴──────────────┴──────┴──────────┴────────────────
> buckets
───┬──────────┬───────────────┬───────────┬──────────┬──────────────────────┬───────────┬───────────────┬────────┬───────
 # │ database │     name      │   type    │ replicas │ min_durability_level │ ram_quota │ flush_enabled │ status │ cloud
───┼──────────┼───────────────┼───────────┼──────────┼──────────────────────┼───────────┼───────────────┼────────┼───────
 0 │ default  │ beer-sample   │ couchbase │        1 │ none                 │  209.7 MB │ false         │        │ false
 1 │ default  │ default       │ couchbase │        1 │ none                 │  104.9 MB │ true          │        │ false
 2 │ default  │ targetBucket  │ couchbase │        0 │ none                 │  104.9 MB │ true          │        │ false
 3 │ default  │ travel-sample │ couchbase │        1 │ none                 │  209.7 MB │ false         │        │ false
───┴──────────┴───────────────┴───────────┴──────────┴──────────────────────┴───────────┴───────────────┴────────┴───────

You can pipe the output into other commands, for example if you only want to see buckets that have sample in their name you can utilize the where command:

> buckets | where name =~ "sample"
───┬──────────┬───────────────┬───────────┬──────────┬──────────────────────┬───────────┬───────────────┬────────┬───────
 # │ database │     name      │   type    │ replicas │ min_durability_level │ ram_quota │ flush_enabled │ status │ cloud
───┼──────────┼───────────────┼───────────┼──────────┼──────────────────────┼───────────┼───────────────┼────────┼───────
 0 │ default  │ beer-sample   │ couchbase │        1 │ none                 │  209.7 MB │ false         │        │ false
 1 │ default  │ travel-sample │ couchbase │        1 │ none                 │  209.7 MB │ false         │        │ false
───┴──────────┴───────────────┴───────────┴──────────┴──────────────────────┴───────────┴───────────────┴────────┴───────

In a similar fashion you can turn this structured table into other output formats, for example JSON:

> buckets | where name =~ "sample" | to json --pretty 2
[
  {
    "database": "default",
    "name": "beer-sample",
    "type": "couchbase",
    "replicas": 1,
    "min_durability_level": "none",
    "ram_quota": 209715200,
    "flush_enabled": false,
    "status": "",
    "cloud": false
  },
  {
    "database": "default",
    "name": "travel-sample",
    "type": "couchbase",
    "replicas": 1,
    "min_durability_level": "none",
    "ram_quota": 209715200,
    "flush_enabled": false,
    "status": "",
    "cloud": false
  }
]

Exactly this type of composition takes the unix philosophy of "do one thing well" and meshes it together with the idea of flexible structured pipelines. This allows to build powerful compositions that help you in your daily operations with Couchbase, both from a developer or operations point of view.

2.2. Getting Help

Other than using this documentation for help, each command can be called with -h or --help to get information about potential flags, arguments and subcommands. Also, some commands provide additional examples.

> buckets -h
Perform bucket management operations

Usage:
  > buckets <subcommand> {flags}

Subcommands:
  buckets config - Shows the bucket config (low level)
  buckets create - Creates a bucket
  buckets drop - Drops buckets through the HTTP API
  buckets flush - Flushes buckets through the HTTP API
  buckets get - Fetches buckets through the HTTP API
  buckets load-sample - Load a sample bucket
  buckets update - Updates a bucket

Flags:
  -h, --help: Display this help message
  --databases <string>: the databases which should be contacted

Some commands (like the one above) only act as groupings for subcommands, like from, to or doc. Since they do not serve a purpose on their own, they will render their help output automatically:

> doc
Perform document operations against a bucket or collection

Usage:
  > doc <subcommand> {flags}

Subcommands:
  doc get - Fetches a document through the data service
  doc insert - Insert a document through the data service
  doc remove - Removes a document through the data service
  doc replace - Replace a document through the data service
  doc upsert - Upsert (insert or override) a document through the data service

Flags:
  -h, --help: Display this help message

2.3. The Prompt explained

Couchbase Shell uses a custom, two line prompt to show you exactly in what environment you are working in right now. Since you can connect to different databases, switch buckets etc. it is important to know what is currently "active". Here is a sample prompt that will greet you when starting the shell:

👤 Administrator at 🏠 local in 🗄 travel-sample
>

It tells you that your user is Administrator, the current active database identifier is local and the active bucket is travel-sample.

If you have an active scope or collection set then the prompt will also update to reflect that:

👤 Administrator 🏠 dev.local in 🗄 travel-sample.myscope.mycollection
>

In the second line, your actual user prompt starts.

2.4. Loading Data into the Shell

If you want to import data into Couchbase, or just load it into the shell for further processing, there are different commands available to help you. Once the data is loaded into the shell it can be sent to one of the couchbase save commands like doc upsert and doc import. Depending on the structure of the data, and the command used, you may also need to tweak it a little bit so it can be properly stored.

2.4.1. Doc import

The doc import is the simplest way to import data through Couchbase Shell. The command expects a path to file containing data, the data can be any of the formats supported by the from command. In the following example we import a json file containing a single document. We also specify the --id-column flag because doc import will try to use an id field for the document key by default.

> cat user.json
{
  "name": "Michael",
  "age": 32,
  "height": 180
}

> doc import user.json --id-column name
╭───┬───────────┬─────────┬────────┬──────────┬───────────╮
│ # │ processed │ success │ failed │ failures │  cluster  │
├───┼───────────┼─────────┼────────┼──────────┼───────────┤
│ 0 │         1 │       1 │      0 │          │ dev.local │
╰───┴───────────┴─────────┴────────┴──────────┴───────────╯
> doc import -h
Import documents from a file through the data service

Usage:
  > doc import {flags} <filename>

Flags:
  -h, --help - Display the help message for this command
  --id-column <String> - the name of the id column if used with an input stream
  --bucket <String> - the name of the bucket
  --expiry <Number> - the expiry for the documents in seconds, or absolute
  --scope <String> - the name of the scope
  --collection <String> - the name of the collection
  --clusters <String> - the clusters which should be contacted
  --batch-size <Number> - the maximum number of items to batch send at a time

Parameters:
  filename <string>: the path to the file containing data to import

2.4.2. Manual import

The open command will look at file endings and try to decode it automatically. Imagine a file named user.json in your current directory.

> cat user.json
{
  "name": "Michael",
  "age": 32,
  "height": 180
}

> open user.json
╭────────┬─────────╮
│ name   │ Michael │
│ age    │ 32      │
│ height │ 180     │
╰────────┴─────────╯

As you can see, the open command automatically decoded the JSON document into the tabular format. If the filename would only be user, the import would look like this instead:

> open user
{
  "name": "Michael",
  "age": 32,
  "height": 180
}

If you are dealing with data that cannot be decoded automatically, you can use the various from subcommands to help with decoding. In our case we use from json:

> open user | from json
╭────────┬─────────╮
│ name   │ Michael │
│ age    │ 32      │
│ height │ 180     │
╰────────┴─────────╯
Tip
 look at the many different import formats from supports, including csv, xml, yaml and even sqlite. With this simple tool at hand you are able to load many different data formats quickly and import them into couchbase!

We cannot use this format directly with commands like doc upsert as the command expects two "columns" in the data - id and content. This means that we have to perform some translation from the above format to one that doc upsert understands. To do this we wrap the entire document into a content column and then extract the id that we want to use:

> open user.json | wrap content | insert id {|it| echo $it.content.name}
╭─────────┬───────────────────╮
│ content │ {record 3 fields} │
│ id      │ Michael           │
╰─────────┴───────────────────╯

There are many other approaches to achieving this same result. With our data in the correct format we can then upsert:

> open user.json | wrap content | insert id {|it| echo $it.content.name} | doc upsert
╭───┬───────────┬─────────┬────────┬──────────┬───────────╮
│ # │ processed │ success │ failed │ failures │  cluster  │
├───┼───────────┼─────────┼────────┼──────────┼───────────┤
│ 0 │         1 │       1 │      0 │          │ dev.local │
╰───┴───────────┴─────────┴────────┴──────────┴───────────╯

See the Importing data recipes for more information.

2.5. Exporting Data from the Shell

The export counterparts to open and from, are save and to. You can use both commands to take tabular data from the shell and store it in files of the needed target format.

Like open, save will try to discern the format from the file ending. The following example will load a JSON file and then save it as CSV:

> cat user.json
{
  "name": "Michael",
  "age": 32,
  "height": 180
}
> open user.json | save user.csv
> cat user.csv
name,age,height
Michael,32,180

This example is dealing with only one row for simplicity, but you can save as many rows as you need in one file.

As a motivating example, the following snippet runs a N1QL query and stores the result as a csv file:

> query "select airportname,city,country from `travel-sample` where type = 'airport' limit 10" | save output.csv
> cat output.csv
airportname,city,country
Calais Dunkerque,Calais,France
Peronne St Quentin,Peronne,France
Les Loges,Nangis,France
Couterne,Bagnole-de-l'orne,France
Bray,Albert,France
Le Touquet Paris Plage,Le Tourquet,France
Denain,Valenciennes,France
Glisy,Amiens,France
La Garenne,Agen,France
Cazaux,Cazaux,France

See the Exporting data recipes for more information.

3. cb-env and the Environment

Whilst multiple databases can be registered at the same time, there is only ever one database (at most) active. The same is true for buckets, scopes, and collections. When a resource is active then it used as the default to run commands against (this can be overridden on a per command basis).

You can run the cb-env command, which will tell you which resources are currently active (you are also able to tell from the prompt):

> cb-env
╭──────────────┬─────────────╮
│ username     │ charlie     │
│ display_name │ Charlie     │
│ database     │ capella     │
│ bucket       │ default     │
│ scope        │ inventory   │
│ collection   │ hotel       │
│ cluster_type │ provisioned │
╰──────────────┴─────────────╯

If you were to now run a command then we would be running it:

  • As the user "charlie"

  • Against the "capella" database

  • Against the "default" bucket

  • Against the "inventory" scope

  • Against the "hotel" collection

Npte that display_name is the name that appears in your shell prompt and is not used by commands.

You can also change the active resources with the cb-env command.

> cb-env -h
Modify the default execution environment of commands

Usage:
  > cb-env {flags}

Subcommands:
  cb-env bucket - Sets the active bucket based on its name
  cb-env capella-organization - Sets the active capella organization based on its identifier
  cb-env collection - Sets the active collection based on its name
  cb-env database - Sets the active database based on its identifier
  cb-env managed - Lists all databases currently managed by couchbase shell
  cb-env project - Sets the active project based on its name
  cb-env register - Registers a database for use with the shell
  cb-env scope - Sets the active scope based on its name
  cb-env timeouts - Sets the active timeouts for operations
  cb-env unregister - Registers a database for use with the shell

Flags:
  -h, --help - Display the help message for this command
  --capella - show default execution environment of capella
  --timeouts - show default execution environment for timeouts

For example if you change the active bucket:

> cb-env bucket beer-sample
╭────────┬─────────────╮
│ bucket │ beer-sample │
╰────────┴─────────────╯
> cb-env
╭──────────────┬─────────────╮
│ username     │ charlie     │
│ display_name │ Charlie     │
│ database     │ capella     │
│ bucket       │ beer-sample │
│ scope        │ inventory   │
│ collection   │ hotel       │
│ cluster_type │ provisioned │
╰──────────────┴─────────────╯

Both the output of cb-env and the prompt will reflect the changes.

3.1. Per command execution environments

On many commands you will notice a set of flags which allow you to override the active execution environment. Different commands support different flags, depending on the command you can expect to see any of:

  • --databases

  • --bucket

  • --scope

  • --collection

3.1.1. The --databases flag

The argument for this flag is an identifier combined with a regular expression. So imagine you have three databases setup with the following names:

> cb-env | get database
───┬────────
 0 │ prod-us-west
 1 │ prod-us-east
 2 │ prod-eu-center
 3 │ local-test
───┴────────

If you wanted to run a command against all databases in prod-us, you could use --databases prod-us.*, e.g.

> buckets --databases prod-us.*
───┬──────────────┬───────────────┬───────────┬──────────┬──────────────────────┬───────────┬───────────────┬────────┬───────
 # │   database   │     name      │   type    │ replicas │ min_durability_level │ ram_quota │ flush_enabled │ status │ capella
───┼──────────────┼───────────────┼───────────┼──────────┼──────────────────────┼───────────┼───────────────┼────────┼───────
 0 │ prod-us-east │ default       │ couchbase │        1 │ none                 │  268.4 MB │ false         │        │ false
 1 │ prod-us-west │ default       │ couchbase │        1 │ none                 │  268.4 MB │ false         │        │ false
 2 │ prod-us-west │ travel-sample │ couchbase │        1 │ none                 │  209.7 MB │ false         │        │ false
───┴──────────────┴───────────────┴───────────┴──────────┴──────────────────────┴───────────┴───────────────┴────────┴───────

In the background this gets passed to a regex engine, so you can go a little crazy with it if needed.

3.1.2. The --bucket, --scope, --collection flags

These flags are a little different to the --databases flag, they are not regular expressions and can only be used to define a single name each. Unlike --databases the name provided to these flags does not have to be already known to Couchbase Shell, they can refer to any bucket, scope, and collection that exist within your active database or defined database(s). For example:

> doc get 1 --bucket travel-sample --scope tenant_agent_00 --collection users
───┬────┬─────────────────────┬──────────────────┬───────┬──────────────
 # │ id │         cas         │     content      │ error │   database
───┼────┼─────────────────────┼──────────────────┼───────┼──────────────
 0 │ 1  │ 1638870288919035904 │ [row 11 columns] │       │ prod-us-west
───┴────┴─────────────────────┴──────────────────┴───────┴──────────────

4. Couchbase Commands

The following sections discuss the individual couchbase specific commands in greater detail. Remember, you can always mix and match them with built-in other shell commands as well as executables from your environment.

4.1. Working with databases

The cb-env managed command lists all the databases you have registered with the shell.

> cb-env managed
╭───┬────────┬───────┬────────────┬───────────────┬──────────────────────╮
│ # │ active │  tls  │ identifier │   username    │ capella_organization │
├───┼────────┼───────┼────────────┼───────────────┼──────────────────────┤
│ 0 │ true   │ false │ dev.local  │ Administrator │                      │
│ 1 │ false  │ true  │ capella    │ charlie       │                      │
╰───┴────────┴───────┴────────────┴───────────────┴──────────────────────╯

4.2. Working with buckets

The buckets command lists all the buckets from your active database:

> buckets
╭───┬───────────┬─────────────┬───────────┬──────────┬──────────────────────┬───────────┬───────────────┬────────┬───────╮
│ # │  cluster  │    name     │   type    │ replicas │ min_durability_level │ ram_quota │ flush_enabled │ status │ cloud │
├───┼───────────┼─────────────┼───────────┼──────────┼──────────────────────┼───────────┼───────────────┼────────┼───────┤
│ 0 │ dev.local │ beer-sample │ couchbase │        2 │ none                 │ 412.0 MiB │ false         │        │ false │
│ 1 │ dev.local │ default     │ couchbase │        0 │ none                 │ 512.0 MiB │ false         │        │ false │
│ 2 │ dev.local │ memd        │ memcached │        0 │ none                 │ 100.0 MiB │ false         │        │ false │
╰───┴───────────┴─────────────┴───────────┴──────────┴──────────────────────┴───────────┴───────────────┴────────┴───────╯

As an advanced command, it is also possible to get the configuration for a bucket:

> buckets config default
╭────────────────────────┬──────────────────────────────────────────────────────────────────────────────────────╮
│ authType               │ sasl                                                                                 │
│ autoCompactionSettings │ false                                                                                │
│ basicStats             │ {record 8 fields}                                                                    │
│ bucketCapabilities     │ [list 17 items]                                                                      │
│ bucketCapabilitiesVer  │                                                                                      │
│ bucketType             │ membase                                                                              │
│ collectionsManifestUid │ 21                                                                                   │
│ compressionMode        │ passive                                                                              │
│ conflictResolutionType │ seqno                                                                                │
│ controllers            │ {record 4 fields}                                                                    │
│ ddocs                  │ {record 1 field}                                                                     │
│ durabilityMinLevel     │ none                                                                                 │
│ evictionPolicy         │ valueOnly                                                                            │
│ localRandomKeyUri      │ /pools/default/buckets/default/localRandomKey                                        │
│ maxTTL                 │ 0                                                                                    │
│ name                   │ default                                                                              │
│ nodeLocator            │ vbucket                                                                              │
│ nodes                  │ [table 1 row]                                                                        │
│ numVBuckets            │ 64                                                                                   │
│ pitrEnabled            │ false                                                                                │
│ pitrGranularity        │ 600                                                                                  │
│ pitrMaxHistoryAge      │ 86400                                                                                │
│ quota                  │ {record 2 fields}                                                                    │
│ replicaIndex           │ false                                                                                │
│ replicaNumber          │ 0                                                                                    │
│ stats                  │ {record 3 fields}                                                                    │
│ storageBackend         │ couchstore                                                                           │
│ streamingUri           │ /pools/default/bucketsStreaming/default?bucket_uuid=0ef162c33e14b163630f04639b347937 │
│ threadsNumber          │ 3                                                                                    │
│ uri                    │ /pools/default/buckets/default?bucket_uuid=0ef162c33e14b163630f04639b347937          │
│ uuid                   │ 0ef162c33e14b163630f04639b347937                                                     │
│ vBucketServerMap       │ {record 4 fields}                                                                    │
╰────────────────────────┴──────────────────────────────────────────────────────────────────────────────────────╯

If you are unsure what you would use this for, you probably don’t need it.

4.3. Working with scopes and collections

The scopes and collections commands can be used for managing scopes and collection respectively.

4.3.1. Scopes

> scopes -h
Fetches scopes through the HTTP API

Usage:
  > scopes <subcommand> {flags}

Subcommands:
  scopes create - Creates scopes through the HTTP API
  scopes drop - Deletes scopes through the HTTP API

Flags:
  -h, --help: Display this help message
  --bucket <string>: the name of the bucket
  --databases <string>: the databases to query against

To list all scopes in the bucket you would use:

> scopes
╭───┬─────────────────┬───────────╮
│ # │      scope      │  cluster  │
├───┼─────────────────┼───────────┤
│ 0 │ inventory       │ dev.local │
│ 1 │ tenant_agent_00 │ dev.local │
│ 2 │ tenant_agent_01 │ dev.local │
│ 3 │ tenant_agent_02 │ dev.local │
│ 4 │ tenant_agent_03 │ dev.local │
│ 5 │ tenant_agent_04 │ dev.local │
│ 6 │ _default        │ dev.local │
╰───┴─────────────────┴───────────╯

You can also create and remove scopes:

> scopes create tenant_agent_05
> scopes
╭───┬─────────────────┬───────────╮
│ # │      scope      │  cluster  │
├───┼─────────────────┼───────────┤
│ 0 │ tenant_agent_05 │ dev.local │
│ 1 │ inventory       │ dev.local │
│ 2 │ tenant_agent_00 │ dev.local │
│ 3 │ tenant_agent_01 │ dev.local │
│ 4 │ tenant_agent_02 │ dev.local │
│ 5 │ tenant_agent_03 │ dev.local │
│ 6 │ tenant_agent_04 │ dev.local │
│ 7 │ _default        │ dev.local │
╰───┴─────────────────┴───────────╯
> scopes drop tenant_agent_05
> scopes
╭───┬─────────────────┬───────────╮
│ # │      scope      │  cluster  │
├───┼─────────────────┼───────────┤
│ 0 │ inventory       │ dev.local │
│ 1 │ tenant_agent_00 │ dev.local │
│ 2 │ tenant_agent_01 │ dev.local │
│ 3 │ tenant_agent_02 │ dev.local │
│ 4 │ tenant_agent_03 │ dev.local │
│ 5 │ tenant_agent_04 │ dev.local │
│ 6 │ _default        │ dev.local │
╰───┴─────────────────┴───────────╯

4.3.2. Collections

> collections -h
Fetches collections through the HTTP API

Usage:
  > collections <subcommand> {flags}

Subcommands:
  collections create - Creates collections through the HTTP API
  collections drop - Deletes collections through the HTTP API

Flags:
  -h, --help: Display this help message
  --bucket <string>: the name of the bucket
  --scope <string>: the name of the scope
  --databases <string>: the databases to query against

To list all collection in the bucket you would use:

> collections
╭────┬─────────────────┬────────────┬────────────┬───────────╮
│  # │      scope      │ collection │ max_expiry │  cluster  │
├────┼─────────────────┼────────────┼────────────┼───────────┤
│  0 │ inventory       │ landmark   │       0sec │ dev.local │
│  1 │ inventory       │ hotel      │       0sec │ dev.local │
│  2 │ inventory       │ airport    │       0sec │ dev.local │
│  3 │ inventory       │ airline    │       0sec │ dev.local │
│  4 │ inventory       │ route      │       0sec │ dev.local │
│  5 │ tenant_agent_00 │ bookings   │       0sec │ dev.local │
│  6 │ tenant_agent_00 │ users      │       0sec │ dev.local │
│  7 │ tenant_agent_01 │ users      │       0sec │ dev.local │
│  8 │ tenant_agent_01 │ bookings   │       0sec │ dev.local │
│  9 │ tenant_agent_02 │ users      │       0sec │ dev.local │
│ 10 │ tenant_agent_02 │ bookings   │       0sec │ dev.local │
│ 11 │ tenant_agent_03 │ users      │       0sec │ dev.local │
│ 12 │ tenant_agent_03 │ bookings   │       0sec │ dev.local │
│ 13 │ tenant_agent_04 │ users      │       0sec │ dev.local │
│ 14 │ tenant_agent_04 │ bookings   │       0sec │ dev.local │
│ 15 │ _default        │ _default   │       0sec │ dev.local │
╰────┴─────────────────┴────────────┴────────────┴───────────╯

You can also create and remove collections:

> collections create staff --scope tenant_agent_00
> collections --scope tenant_agent_00
╭───┬─────────────────┬────────────┬────────────┬───────────╮
│ # │      scope      │ collection │ max_expiry │  cluster  │
├───┼─────────────────┼────────────┼────────────┼───────────┤
│ 0 │ tenant_agent_00 │ staff      │       0sec │ dev.local │
│ 1 │ tenant_agent_00 │ bookings   │       0sec │ dev.local │
│ 2 │ tenant_agent_00 │ users      │       0sec │ dev.local │
╰───┴─────────────────┴────────────┴────────────┴───────────╯
> collections drop staff --scope tenant_agent_00
> collections --scope tenant_agent_00
╭───┬─────────────────┬────────────┬────────────┬───────────╮
│ # │      scope      │ collection │ max_expiry │  cluster  │
├───┼─────────────────┼────────────┼────────────┼───────────┤
│ 0 │ tenant_agent_00 │ bookings   │       0sec │ dev.local │
│ 1 │ tenant_agent_00 │ users      │       0sec │ dev.local │
╰───┴─────────────────┴────────────┴────────────┴───────────╯

4.4. Listing nodes

The nodes command allows you to list all the nodes of the database you are currently connected to.

> nodes
───┬──────────┬─────────────────────┬─────────┬───────────────────┬───────────────────────┬──────────────────────────┬──────────────┬─────────────
 # │ database │ hostname            │ status  │ services          │ version               │ os                       │ memory_total │ memory_free
───┼──────────┼─────────────────────┼─────────┼───────────────────┼───────────────────────┼──────────────────────────┼──────────────┼─────────────
 0 │ remote   │ 10.143.200.101:8091 │ healthy │ indexing,kv,query │ 6.5.0-4960-enterprise │ x86_64-unknown-linux-gnu │       2.1 GB │    837.7 MB
 1 │ remote   │ 10.143.200.102:8091 │ healthy │ indexing,kv,query │ 6.5.0-4960-enterprise │ x86_64-unknown-linux-gnu │       2.1 GB │      1.0 GB
───┴──────────┴─────────────────────┴─────────┴───────────────────┴───────────────────────┴──────────────────────────┴──────────────┴─────────────

4.5. Reading and Writing `doc`uments

The fastest way to interact with documents is through the key value service (as long as you know the document ID). All those commands are located as subcommands under the doc namespace.

4.5.1. Reading

You can retrieve a document with doc get:

> doc get airline_10
╭───┬────────────┬───────────────────┬─────────────────────┬───────┬───────────╮
│ # │     id     │      content      │         cas         │ error │  cluster  │
├───┼────────────┼───────────────────┼─────────────────────┼───────┼───────────┤
│ 0 │ airline_10 │ {record 7 fields} │ 1681456999724089344 │       │ dev.local │
╰───┴────────────┴───────────────────┴─────────────────────┴───────┴───────────╯

To distinguish the actual content from the metadata, the content is nested in the content field. If you want to have everything at the toplevel, you can pipe to the flatten command:

> doc get airline_10 | flatten
╭───┬────────────┬──────────┬───────────────┬──────┬──────┬────────────┬─────────────┬─────────┬─────────────────────┬───────┬───────────╮
│ # │     id     │ callsign │    country    │ iata │ icao │ content_id │    name     │  type   │         cas         │ error │  cluster  │
├───┼────────────┼──────────┼───────────────┼──────┼──────┼────────────┼─────────────┼─────────┼─────────────────────┼───────┼───────────┤
│ 0 │ airline_10 │ MILE-AIR │ United States │ Q5   │ MLA  │         10 │ 40-Mile Air │ airline │ 1681456999724089344 │       │ dev.local │
╰───┴────────────┴──────────┴───────────────┴──────┴──────┴────────────┴─────────────┴─────────┴─────────────────────┴───────┴───────────╯

If the document is not found, an empty result is returned.

To perform a bulk get operation, the incoming stream can be utilized.

> echo [airline_10 airline_10748 airline_137] | wrap id | doc get
╭───┬───────────────┬───────────────────┬─────────────────────┬───────┬───────────╮
│ # │      id       │      content      │         cas         │ error │  cluster  │
├───┼───────────────┼───────────────────┼─────────────────────┼───────┼───────────┤
│ 0 │ airline_10    │ {record 7 fields} │ 1681456999724089344 │       │ dev.local │
│ 1 │ airline_10748 │ {record 7 fields} │ 1681456996753211392 │       │ dev.local │
│ 2 │ airline_137   │ {record 7 fields} │ 1681457004278579200 │       │ dev.local │
╰───┴───────────────┴───────────────────┴─────────────────────┴───────┴───────────╯

If doc get operates on an incoming stream it will extract the document id from the id column. This behavior can be customized through the --id-column flag.

4.5.2. Mutating

Documents can be mutated with doc insert, doc upsert and doc replace.

All those three commands take similar arguments. If you only want to mutate a single document, passing in the ID and the content as arguments is the simplest way:

> doc upsert my-doc {"hello": "world"}
╭───┬───────────┬─────────┬────────┬──────────┬───────────╮
│ # │ processed │ success │ failed │ failures │  cluster  │
├───┼───────────┼─────────┼────────┼──────────┼───────────┤
│ 0 │         1 │       1 │      0 │          │ dev.local │
╰───┴───────────┴─────────┴────────┴──────────┴───────────╯

Multiple documents can be mutated through an input stream as well, defaulting to the id and content columns:

4.5.3. Removing

Documents can be removed with doc remove.

> doc remove airline_10
╭───┬───────────┬─────────┬────────┬──────────┬───────────╮
│ # │ processed │ success │ failed │ failures │  cluster  │
├───┼───────────┼─────────┼────────┼──────────┼───────────┤
│ 0 │         1 │       1 │      0 │          │ dev.local │
╰───┴───────────┴─────────┴────────┴──────────┴───────────╯

Similar to doc get, if you want to delete more than one document at the same time, provide a stream of ids with an id column:

> echo [airline_10 airline_10748 airline_137] | wrap id | doc remove
╭───┬───────────┬─────────┬────────┬───────────────┬───────────╮
│ # │ processed │ success │ failed │   failures    │  cluster  │
├───┼───────────┼─────────┼────────┼───────────────┼───────────┤
│ 0 │         3 │       2 │      1 │ Key not found │ dev.local │
╰───┴───────────┴─────────┴────────┴───────────────┴───────────╯

4.6. version

The version command lists the version of the couchbase shell.

> version
╭─────────┬────────╮
│ version │ 0.75.1 │
╰─────────┴────────╯

5. Reference

5.1. Config File Format

The ~/.cbsh/config file with examples:

# Allows us to evolve in the future without breaking old config files
version = 1

[[database]]
identifier = "default"
connstr = "127.0.0.1"
default-bucket = "travel-sample"
default-scope = "my-scope"
default-collection = "my-collection"
# The following can be part of the config or credentials
username = "Administrator"
password = "password"
# TLS defaults to on
# tls-enabled = false
# tls-cert-path = "/path/to/cert" # either accept all certs or provide a cert path
# tls-accept-all-certs = true
# tls-validate-hostnames = false

# User display name is optional and is used to display a different name to the username in the prompt itself.
# This can be useful if the username that you are provided is a long randomly generated string or similar.
# user-display-name = "Charlie"

# Timeouts broadly apply to the operations that you would expect them to.
# That is:
# * data: commands using the kv service such as `doc`
# * query: `query` commands
# * analytics: `analytics` commands
# * search: `search` commands
# * management: commands that perform management level operations, such as `users`, `bucket`, `health` etc...
data-timeout = "10s"
query-timeout = "75s"
analytics-timeout = "75s"
search-timeout = "1m 15s"
management-timeout = "75s"

5.2. Credentials File Format

The optional ~/.cbsh/credentials file with examples:

# Allows us to evolve in the future without breaking old config files
version = 1

[[database]]
identifier = "default"
username = "Administrator"
password = "password"
# TLS defaults to on, accepting all certs
# tls-enabled = true
# tls-cert-path = "/path/to/cert" # either accept all certs or provide a cert path
# tls-accept-all-certs = true
# tls-validate-hostnames = false

6. Recipes

6.1. Importing data

Couchbase Shell supports loading data from a variety of formats and sources.

The simplest way to import data is using doc import as covered in Loading data into the shell. These recipes will cover more advanced usecases.

6.1.1. A Note On Data format

The doc upsert command requires there to be only two fields/columns for an upsert. There can be more than two fields/columns in the data but only two can be used. By default, these two columns are named id and content, but these can be overridden with --id-column and --content-column.

Given the following document format we need to perform some data manipulation to get it into a format which works with doc upsert:

> cat mydoc.json
{"id":3719,"cas":1600344369374167040,"type":"airport","airportname":"Columbia Rgnl","city":"Columbia","country":"United States","faa":"COU","icao":"KCOU","tz":"America/Chicago"}

> open mydoc.json | wrap content | insert id { |it| echo $it.content.airportname }
╭─────────┬───────────────────╮
│ content │ {record 9 fields} │
│ id      │ Columbia Rgnl     │
╰─────────┴───────────────────╯

6.1.2. From file

From JSON

Single Document

> cat mydoc.json
{"id":3719,"cas":1600344369374167040,"type":"airport","airportname":"Columbia Rgnl","city":"Columbia","country":"United States","faa":"COU","icao":"KCOU","tz":"America/Chicago"}

> open mydoc.json | wrap content | insert id { |it|  echo $it.content.airportname } | doc upsert
╭───┬───────────┬─────────┬────────┬──────────┬───────────╮
│ # │ processed │ success │ failed │ failures │  cluster  │
├───┼───────────┼─────────┼────────┼──────────┼───────────┤
│ 0 │         1 │       1 │      0 │          │ dev.local │
╰───┴───────────┴─────────┴────────┴──────────┴───────────╯

Multiple Documents

> ls airports
───┬────────────────────────────┬──────┬───────┬────────────
 # │            name            │ type │ size  │  modified
───┼────────────────────────────┼──────┼───────┼────────────
 0 │ airports/airport_3719.json │ File │ 151 B │ 2 days ago
 1 │ airports/airport_3720.json │ File │ 155 B │ 2 days ago
 2 │ airports/airport_3721.json │ File │ 172 B │ 2 days ago
 3 │ airports/airport_3722.json │ File │ 161 B │ 2 days ago
 4 │ airports/airport_3723.json │ File │ 163 B │ 2 days ago
 5 │ airports/airport_3724.json │ File │ 156 B │ 2 days ago
 6 │ airports/airport_3725.json │ File │ 148 B │ 2 days ago
 7 │ airports/airport_3726.json │ File │ 164 B │ 2 days ago
 8 │ airports/airport_3727.json │ File │ 169 B │ 2 days ago
 9 │ airports/airport_3728.json │ File │ 152 B │ 2 days ago
───┴────────────────────────────┴──────┴───────┴────────────

> open airports/airport_3719.json
───┬──────┬─────────┬─────────────┬──────────┬─────────┬─────┬──────┬──────────────
 # │  id  │  type   │ airportname │   city   │ country │ faa │ icao │      tz
───┼──────┼─────────┼─────────────┼──────────┼─────────┼─────┼──────┼──────────────
 0 │ 3719 │ airport │ Columbia    │ Columbia │ United  │ COU │ KCOU │ America/Chic
   │      │         │ Rgnl        │          │ States  │     │      │ ago
───┴──────┴─────────┴─────────────┴──────────┴─────────┴─────┴──────┴──────────────

> ls airports/ | each { |it| open $it.name | wrap content | insert id { |doc| echo $doc.content.airportname} } | doc upsert
╭───┬───────────┬─────────┬────────┬──────────┬───────────╮
│ # │ processed │ success │ failed │ failures │  cluster  │
├───┼───────────┼─────────┼────────┼──────────┼───────────┤
│ 0 │        10 │      10 │      0 │          │ dev.local │
╰───┴───────────┴─────────┴────────┴──────────┴───────────╯
From CSV

Single Document

> cat mydoc.csv
id,cas,type,airportname,city,country,faa,icao,tz
3719,1600344369374167040,airport,Columbia Rgnl,Columbia,United States,COU,KCOU,America/Chicago

> open mydoc.csv | each { |it| wrap content | insert id { |doc| echo $doc.content.airportname} } | doc upsert
╭───┬───────────┬─────────┬────────┬──────────┬───────────╮
│ # │ processed │ success │ failed │ failures │  cluster  │
├───┼───────────┼─────────┼────────┼──────────┼───────────┤
│ 0 │         1 │       1 │      0 │          │ dev.local │
╰───┴───────────┴─────────┴────────┴──────────┴───────────╯

Multiple Documents

> cat airports.csv
airportname,city,country,faa,icao,id,type,tz
Calais Dunkerque,Calais,France,CQF,LFAC,1254,airport,Europe/Paris
Peronne St Quentin,Peronne,France,,LFAG,1255,airport,Europe/Paris
Les Loges,Nangis,France,,LFAI,1256,airport,Europe/Paris
Couterne,Bagnole-de-l'orne,France,,LFAO,1257,airport,Europe/Paris
Bray,Albert,France,,LFAQ,1258,airport,Europe/Paris

>  open mydoc.csv | each { |it| wrap content | insert id { |doc| echo $doc.content.airportname }  } | doc upsert
╭───┬───────────┬─────────┬────────┬──────────┬───────────╮
│ # │ processed │ success │ failed │ failures │  cluster  │
├───┼───────────┼─────────┼────────┼──────────┼───────────┤
│ 0 │        10 │      10 │      0 │          │ dev.local │
╰───┴───────────┴─────────┴────────┴──────────┴───────────╯

6.1.3. Modifying data

In some circumstances you may want to modify the data before you import it. Let’s take the example of importing from a csv file but this time the airports.csv file is missing the type column but we want to add it to our data:

> cat airports.csv
airportname,city,country,faa,icao,id,tz
Calais Dunkerque,Calais,France,CQF,LFAC,1254,Europe/Paris
Peronne St Quentin,Peronne,France,,LFAG,1255,Europe/Paris
Les Loges,Nangis,France,,LFAI,1256,Europe/Paris
Couterne,Bagnole-de-l'orne,France,,LFAO,1257,Europe/Paris
Bray,Albert,France,,LFAQ,1258,Europe/Paris

> open airports.csv | insert type airport
╭───┬────────────────────┬───────────────────┬─────────┬─────┬──────┬──────┬──────────────┬─────────╮
│ # │    airportname     │       city        │ country │ faa │ icao │  id  │      tz      │  type   │
├───┼────────────────────┼───────────────────┼─────────┼─────┼──────┼──────┼──────────────┼─────────┤
│ 0 │ Calais Dunkerque   │ Calais            │ France  │ CQF │ LFAC │ 1254 │ Europe/Paris │ airport │
│ 1 │ Peronne St Quentin │ Peronne           │ France  │     │ LFAG │ 1255 │ Europe/Paris │ airport │
│ 2 │ Les Loges          │ Nangis            │ France  │     │ LFAI │ 1256 │ Europe/Paris │ airport │
│ 3 │ Couterne           │ Bagnole-de-l'orne │ France  │     │ LFAO │ 1257 │ Europe/Paris │ airport │
│ 4 │ Bray               │ Albert            │ France  │     │ LFAQ │ 1258 │ Europe/Paris │ airport │
╰───┴────────────────────┴───────────────────┴─────────┴─────┴──────┴──────┴──────────────┴─────────╯

We can also add a column based on data from other columns, for instance adding a type column which is set to the relevant country:

> open airports.csv | each { |it| insert type $it.city }
╭───┬────────────────────┬───────────────────┬─────────┬─────┬──────┬──────┬──────────────┬───────────────────╮
│ # │    airportname     │       city        │ country │ faa │ icao │  id  │      tz      │       type        │
├───┼────────────────────┼───────────────────┼─────────┼─────┼──────┼──────┼──────────────┼───────────────────┤
│ 0 │ Calais Dunkerque   │ Calais            │ France  │ CQF │ LFAC │ 1254 │ Europe/Paris │ Calais            │
│ 1 │ Peronne St Quentin │ Peronne           │ France  │     │ LFAG │ 1255 │ Europe/Paris │ Peronne           │
│ 2 │ Les Loges          │ Nangis            │ France  │     │ LFAI │ 1256 │ Europe/Paris │ Nangis            │
│ 3 │ Couterne           │ Bagnole-de-l'orne │ France  │     │ LFAO │ 1257 │ Europe/Paris │ Bagnole-de-l'orne │
│ 4 │ Bray               │ Albert            │ France  │     │ LFAQ │ 1258 │ Europe/Paris │ Albert            │
╰───┴────────────────────┴───────────────────┴─────────┴─────┴──────┴──────┴──────────────┴───────────────────╯

6.2. Exporting data

Couchbase Shell supports exporting data to a variety of formats and sources.

6.2.1. A Note On Data format

The doc get command exposes data as three fields; id, cas, and content. The body of the document is stored within the content column. If you want to only store the document body then you can use doc get <id> | get content.

6.2.2. To file

To JSON

From KeyValue

> doc get airport_3719 --bucket travel-sample
╭───┬──────────────┬───────────────────┬─────────────────────┬───────┬───────────╮
│ # │      id      │      content      │         cas         │ error │  cluster  │
├───┼──────────────┼───────────────────┼─────────────────────┼───────┼───────────┤
│ 0 │ airport_3719 │ {record 9 fields} │ 1681456998755270656 │       │ dev.local │
╰───┴──────────────┴───────────────────┴─────────────────────┴───────┴───────────╯

> doc get airport_3719 | get content | save mydoc.json
> cat mydoc.json
{"airportname":"Columbia Rgnl","city":"Columbia","country":"United States","faa":"COU","geo":{"alt":889.0,"lat":38.818094,"lon":-92.219631},"icao":"KCOU","id":3719,"type":"airport","tz":"America/Chicago"}

From Query/Analytics

To Single Document

> query "SELECT `travel-sample`.* from `travel-sample` WHERE `type`='airport' LIMIT 5"
╭───┬────────────────────┬───────────────────┬─────────┬─────┬───────────────────┬──────┬──────┬─────────┬──────────────┬───────────╮
│ # │    airportname     │       city        │ country │ faa │        geo        │ icao │  id  │  type   │      tz      │ database  │
├───┼────────────────────┼───────────────────┼─────────┼─────┼───────────────────┼──────┼──────┼─────────┼──────────────┼───────────┤
│ 0 │ Calais Dunkerque   │ Calais            │ France  │ CQF │ {record 3 fields} │ LFAC │ 1254 │ airport │ Europe/Paris │ dev.local │
│ 1 │ Peronne St Quentin │ Peronne           │ France  │     │ {record 3 fields} │ LFAG │ 1255 │ airport │ Europe/Paris │ dev.local │
│ 2 │ Les Loges          │ Nangis            │ France  │     │ {record 3 fields} │ LFAI │ 1256 │ airport │ Europe/Paris │ dev.local │
│ 3 │ Couterne           │ Bagnole-de-l'orne │ France  │     │ {record 3 fields} │ LFAO │ 1257 │ airport │ Europe/Paris │ dev.local │
│ 4 │ Bray               │ Albert            │ France  │     │ {record 3 fields} │ LFAQ │ 1258 │ airport │ Europe/Paris │ dev.local │
╰───┴────────────────────┴───────────────────┴─────────┴─────┴───────────────────┴──────┴──────┴─────────┴──────────────┴───────────╯

> query "SELECT `travel-sample`.* from `travel-sample` WHERE `type`='airport' LIMIT 5" | save airports.json
> cat airports.json
[{"airportname":"Calais Dunkerque","city":"Calais","country":"France","faa":"CQF","geo":{"alt":12,"lat":50.962097,"lon":1.9547640000000002},"icao":"LFAC","id":1254,"type":"airport","tz":"Europe/Paris"},{"airportname":"Peronne St Quentin","city":"Peronne","country":"France","faa":null,"geo":{"alt":295,"lat":49.868547,"lon":3.0295780000000003},"icao":"LFAG","id":1255,"type":"airport","tz":"Europe/Paris"},{"airportname":"Les Loges","city":"Nangis","country":"France","faa":null,"geo":{"alt":428,"lat":48.596219,"lon":3.0067860000000004},"icao":"LFAI","id":1256,"type":"airport","tz":"Europe/Paris"},{"airportname":"Couterne","city":"Bagnole-de-l'orne","country":"France","faa":null,"geo":{"alt":718,"lat":48.545836,"lon":-0.387444},"icao":"LFAO","id":1257,"type":"airport","tz":"Europe/Paris"},{"airportname":"Bray","city":"Albert","country":"France","faa":null,"geo":{"alt":364,"lat":49.971531,"lon":2.697661},"icao":"LFAQ","id":1258,"type":"airport","tz":"Europe/Paris"}]

To Multiple Documents

> query "SELECT `travel-sample`.* from `travel-sample` WHERE `type`='airport' LIMIT 5"
╭───┬────────────────────┬───────────────────┬─────────┬─────┬───────────────────┬──────┬──────┬─────────┬──────────────┬───────────╮
│ # │    airportname     │       city        │ country │ faa │        geo        │ icao │  id  │  type   │      tz      │ database  │
├───┼────────────────────┼───────────────────┼─────────┼─────┼───────────────────┼──────┼──────┼─────────┼──────────────┼───────────┤
│ 0 │ Calais Dunkerque   │ Calais            │ France  │ CQF │ {record 3 fields} │ LFAC │ 1254 │ airport │ Europe/Paris │ dev.local │
│ 1 │ Peronne St Quentin │ Peronne           │ France  │     │ {record 3 fields} │ LFAG │ 1255 │ airport │ Europe/Paris │ dev.local │
│ 2 │ Les Loges          │ Nangis            │ France  │     │ {record 3 fields} │ LFAI │ 1256 │ airport │ Europe/Paris │ dev.local │
│ 3 │ Couterne           │ Bagnole-de-l'orne │ France  │     │ {record 3 fields} │ LFAO │ 1257 │ airport │ Europe/Paris │ dev.local │
│ 4 │ Bray               │ Albert            │ France  │     │ {record 3 fields} │ LFAQ │ 1258 │ airport │ Europe/Paris │ dev.local │
╰───┴────────────────────┴───────────────────┴─────────┴─────┴───────────────────┴──────┴──────┴─────────┴──────────────┴───────────╯

> query "SELECT `travel-sample`.* FROM `travel-sample` WHERE `type`='airport' LIMIT 5" | each { |it| echo $it | save (echo (["airport_", $it.id ,".json"] | str join)) }
> ls airport*.json
╭───┬───────────────────┬──────┬───────┬──────────╮
│ # │       name        │ type │ size  │ modified │
├───┼───────────────────┼──────┼───────┼──────────┤
│ 0 │ airport_1254.json │ file │ 277 B │ now      │
│ 1 │ airport_1255.json │ file │ 280 B │ now      │
│ 2 │ airport_1256.json │ file │ 270 B │ now      │
│ 3 │ airport_1257.json │ file │ 281 B │ now      │
│ 4 │ airport_1258.json │ file │ 265 B │ now      │
╰───┴───────────────────┴──────┴───────┴──────────╯
To CSV

From KeyValue

> doc get airport_3719 --bucket travel-sample
╭───┬──────────────┬───────────────────┬─────────────────────┬───────┬───────────╮
│ # │      id      │      content      │         cas         │ error │  cluster  │
├───┼──────────────┼───────────────────┼─────────────────────┼───────┼───────────┤
│ 0 │ airport_3719 │ {record 9 fields} │ 1681456998755270656 │       │ dev.local │
╰───┴──────────────┴───────────────────┴─────────────────────┴───────┴───────────╯

> doc get airport_3719 --bucket travel-sample | get content
╭───┬───────────────┬──────────┬───────────────┬─────┬───────────────────┬──────┬──────┬─────────┬─────────────────╮
│ # │  airportname  │   city   │    country    │ faa │        geo        │ icao │  id  │  type   │       tz        │
├───┼───────────────┼──────────┼───────────────┼─────┼───────────────────┼──────┼──────┼─────────┼─────────────────┤
│ 0 │ Columbia Rgnl │ Columbia │ United States │ COU │ {record 3 fields} │ KCOU │ 3719 │ airport │ America/Chicago │
╰───┴───────────────┴──────────┴───────────────┴─────┴───────────────────┴──────┴──────┴─────────┴─────────────────╯

The geo column in the above data contains a record, which means that the data is nested. This means that we have to flatten out the geo column due to limitations of the csv format. If we try to import the content as it is then we will see:

> doc get airport_3719 --bucket travel-sample | get content | to csv
Error: nu::shell::cant_convert (link)

  × Can't convert to CSV.
   ╭─[entry #25:1:1]
 1 │ doc get airport_3719 --bucket travel-sample | get content | to csv
   · ───┬───
   ·    ╰── can't convert table<airportname: string, city: string, country: string, faa: string, geo: record<alt: float, lat: float, lon: float>, icao: string, id: int, type: string, tz: string> to CSV
   ╰────

To flatten out the data we can simply use the flatten command:

> doc get airport_3719 --bucket travel-sample | get content | flatten
╭───┬───────────────┬──────────┬───────────────┬─────┬──────────┬─────────┬──────────┬──────┬──────┬─────────┬─────────────────╮
│ # │  airportname  │   city   │    country    │ faa │   alt    │   lat   │   lon    │ icao │  id  │  type   │       tz        │
├───┼───────────────┼──────────┼───────────────┼─────┼──────────┼─────────┼──────────┼──────┼──────┼─────────┼─────────────────┤
│ 0 │ Columbia Rgnl │ Columbia │ United States │ COU │ 889.0000 │ 38.8181 │ -92.2196 │ KCOU │ 3719 │ airport │ America/Chicago │
╰───┴───────────────┴──────────┴───────────────┴─────┴──────────┴─────────┴──────────┴──────┴──────┴─────────┴─────────────────╯

Which we can then pipe to save:

> doc get airport_3719 --bucket travel-sample | get content | flatten | save mydoc.csv
> cat mydoc.csv
airportname,city,country,faa,alt,lat,lon,icao,id,type,tz
Columbia Rgnl,Columbia,United States,COU,889,38.818094,-92.219631,KCOU,3719,airport,America/Chicago

From Query/Analytics

> query "SELECT `travel-sample`.* from `travel-sample` WHERE `type`='airport' LIMIT 5"
╭───┬────────────────────┬───────────────────┬─────────┬─────┬───────────────────┬──────┬──────┬─────────┬──────────────┬───────────╮
│ # │    airportname     │       city        │ country │ faa │        geo        │ icao │  id  │  type   │      tz      │ database  │
├───┼────────────────────┼───────────────────┼─────────┼─────┼───────────────────┼──────┼──────┼─────────┼──────────────┼───────────┤
│ 0 │ Calais Dunkerque   │ Calais            │ France  │ CQF │ {record 3 fields} │ LFAC │ 1254 │ airport │ Europe/Paris │ dev.local │
│ 1 │ Peronne St Quentin │ Peronne           │ France  │     │ {record 3 fields} │ LFAG │ 1255 │ airport │ Europe/Paris │ dev.local │
│ 2 │ Les Loges          │ Nangis            │ France  │     │ {record 3 fields} │ LFAI │ 1256 │ airport │ Europe/Paris │ dev.local │
│ 3 │ Couterne           │ Bagnole-de-l'orne │ France  │     │ {record 3 fields} │ LFAO │ 1257 │ airport │ Europe/Paris │ dev.local │
│ 4 │ Bray               │ Albert            │ France  │     │ {record 3 fields} │ LFAQ │ 1258 │ airport │ Europe/Paris │ dev.local │
╰───┴────────────────────┴───────────────────┴─────────┴─────┴───────────────────┴──────┴──────┴─────────┴──────────────┴───────────╯

Similar to the above we need to flatten out the geo data before we can save this to csv:

> query "SELECT `travel-sample`.* from `travel-sample` WHERE `type`='airport' LIMIT 5" | flatten | save mydoc.csv

6.3. Useful snippets

This section contains a collection of useful commands and sets of commands which don’t really fit into their own section of recipes.

6.3.1. Migrating scope and collection definitions

When you create a new cluster it can be useful to migrate scope and collection definitions from an old cluster. A good example here is migrating from an on-premise cluster to a Capella cluster.

To migrate scopes, except the _default scope:

scopes --databases "On-Prem-Cluster" --bucket travel-sample | select scope | where scope != "_default" | each { |it| scopes create $it.scope --databases "Capella-Cluster" }

To migrate all collections, except the _default collection:

collections --databases "On-Prem-Cluster" --bucket "travel-sample" | select scope collection | where $it.scope != "_default" | where $it.collection != "_default" | each { |it| collections create $it.collection --databases "Capella-Cluster" --bucket "travel-sample-import" --scope $it.scope

These examples can easily be extended to filter out any other scopes and collections you do not want to migrate. For example to filter more scopes you would just add more where clauses: …​ | where scope != "_default" | where scope != "inventory" | …​

6.3.2. Migrating query index definitions

When you create a new cluster it can be useful to migrate index definitions from an old cluster. A good example here is migrating from an on-premise cluster to a Capella cluster.

To migrate all of your index definitions:

query indexes --definitions --databases "On-Prem-Cluster" | get definition | each { |it| query $it --databases "Capella-Cluster" }

7. Release Notes

7.1. 0.75.1 - 2023-04-13

This release contains a number of breaking changes, which are explicitly called out below. As our versioning continues to track the underlying Nushell minor version this has required breaking changes in a patch version.

  • Updated config file to rename to ( will continue to work).

  • Breaking Updated config file to rename hostnames to connstr and changed the format to be a string.

  • Added support, and detection, for different "cluster types"; Capella and Other. This allows us to modify behaviour based on cluster type.

  • Breaking Renamed clusters health to health.

  • Breaking Renamed other clusters …​ commands to database …​

  • Replaced references to cluster with database.

  • Breaking Removed support for whoami

  • Added support for username aliases - added display_name to config.

  • Trust the system store and Capella root CA when no certificate set.

  • Updated tutorial.

  • Added support for generating a config file when one does not exist.

  • Added support for connecting to kv nodes in parallel.

  • Added support for doc import command.

  • Added prompt indicator to help distinguish if a Capella or "other" cluster type is in use.

  • Fast fail buckets commands when used with Capella.

  • Updated where config files are automatically written to.

  • Fixed issue with config.nu file on Windows.

    • Pulled all beta and alpha release versions and updated version numbering

    • Couchbase Shell versions will now map to the Nushell version being used

    • Bundle Capella root CA to allow seamlessly connecting over TLS

    • Automatically detect when query_context should be sent

    • Update when SRV lookups are performed

    • Statically link OpenSSL

    • Various logging and error enhancements

    • Remove support for Capella InVpc

    • Renamed clusters managed to cb-env managed

    • Renamed clusters register/unregister to cb-env register/unregister

    • Expose CIDR in result of clusters

    • Fetch collection id over memcached rather than http

7.2. 0.75.0 - 2023-02-09

  • Nushell pinned to 0.75

  • Pulled all beta and alpha release versions and updated version numbering

  • Couchbase Shell versions will now map to the Nushell version being used

  • Bundle Capella root CA to allow seamlessly connecting over TLS

  • Automatically detect when query_context should be sent

  • Update when SRV lookups are performed

  • Statically link OpenSSL

  • Various logging and error enhancements

  • Remove support for Capella InVpc

  • Renamed clusters managed to cb-env managed

  • Renamed clusters register/unregister to cb-env register/unregister

  • Expose CIDR in result of clusters

  • Fetch collection id over memcached rather than http