Title: Client for 'LaminDB'
Version: 1.2.0
Description: Interact with 'LaminDB'. 'LaminDB' is an open-source data framework for biology. This package allows you to query and download data from 'LaminDB' instances.
License: Apache License (≥ 2)
URL: https://laminr.lamin.ai, https://github.com/laminlabs/laminr
BugReports: https://github.com/laminlabs/laminr/issues
Depends: R (≥ 4.1.0)
Imports: callr, cli, lifecycle, pkgload, purrr, R.utils, R6, reticulate (≥ 1.41.0), rlang, sessioninfo, utils, withr
Suggests: anndata, arrow, jsonlite, knitr, magick, readr, rmarkdown, rstudioapi, rsvg, testthat (≥ 3.0.0), yaml
VignetteBuilder: knitr
Config/Needs/website: rmarkdown, Seurat
Config/testthat/edition: 3
Encoding: UTF-8
RoxygenNote: 7.3.3
NeedsCompilation: no
Packaged: 2025-10-09 09:37:11 UTC; sunnysun
Author: Robrecht Cannoodt ORCID iD [aut, cre], Luke Zappia ORCID iD [aut], Data Intuitive [aut], Lamin Labs [aut, cph]
Maintainer: Robrecht Cannoodt <robrecht@lamin.ai>
Repository: CRAN
Date/Publication: 2025-10-09 09:50:10 UTC

laminr: Client for 'LaminDB'

Description

Interact with 'LaminDB'. 'LaminDB' is an open-source data framework for biology. This package allows you to query and download data from 'LaminDB' instances.

Author(s)

Maintainer: Robrecht Cannoodt robrecht@lamin.ai (ORCID)

Authors:

See Also

Useful links:

Get current LaminDB instance

Description

Get the currently connected LaminDB instance

Usage

get_current_lamin_instance()

Details

This is done via a get_current_lamin_settings() to avoid importing Python lamindb

Value

The slug of the current LaminDB instance, or NULL invisibly if no instance is found

Get current LaminDB settings

Description

Get the current LaminDB settings as an R list

Usage

get_current_lamin_settings(minimal = FALSE)

Arguments

minimal

If TRUE, quickly extract a minimal list of important settings instead of converting the complete settings object

Details

This is done using callr::r() to avoid importing Python lamindb in the global environment

Value

A list of the current LaminDB settings

Get current LaminDB user

Description

Get the currently logged in LaminDB user

Usage

get_current_lamin_user()

Details

This is done via get_current_lamin_settings() to avoid importing Python lamindb

Value

The handle of the current LaminDB user, or NULL invisibly if no user is found

Import Python modules

Description

This function can be used to import LaminDB Python modules with additional checks and nicer error messages.

Usage

import_module(module, ...)

Arguments

module

The name of the Python module to import

...

Arguments passed on to require_module

options

A vector of defined optional dependencies for the module that is being required

version

A string specifying the version of the module to require

source

A source for the module requirement, for example ⁠git+https://github.com/owner/module.git⁠

python_version

A string defining the Python version to require. Passed to reticulate::py_require()

silent

Whether to suppress the message showing what has been required

Details

Python dependencies are set using require_module() before importing the module and used to create an ephemeral environment unless another environment is found (see vignette("versions", package = "reticulate")).

Requirements for the lamindb module can be controlled using environment variables differently, see https://docs.lamin.ai/setup-laminr for details.

Value

An object representing a Python package

See Also

Examples

## Not run: 
# Import lamindb to start interacting with an instance
ln <- import_module("lamindb")

# Import lamindb with optional dependencies
ln <- import_module("lamindb", options = c("bionty", "wetlab"))

# Import other LaminDB modules
bt <- import_module("bionty")
wl <- import_module("wetlab")
cc <- import_module("clinicore")

# Import any Python module
np <- import_module("numpy")

## End(Not run)

Install LaminDB

Description

[Deprecated]

This function is deprecated and is replaced by a system which automatically installs packages as needed. See import_module(), require_module() and reticulate::py_require() for details.

Create a Python environment containing lamindb or install lamindb into an existing environment.

Usage

install_lamindb(
  ...,
  envname = "r-lamindb",
  extra_packages = NULL,
  new_env = identical(envname, "r-lamindb"),
  use = TRUE
)

Arguments

...

Additional arguments passed to reticulate::py_install()

envname

String giving the name of the environment to install packages into

extra_packages

A vector giving the names of additional Python packages to install

new_env

Whether to remove any existing virtualenv with the same name before creating a new one with the requested packages

use

Whether to attempt use the new environment

Value

NULL, invisibly

Examples

## Not run: 
# Using import_module() will automatically install packages
ln <- import_module("lamindb")

# Create a Python environment with lamindb
# This approach is deprecated
install_lamindb()

# Add additional packages to the environment
install_lamindb(extra_packages = c("bionty", "wetlab"))

# Install into a different environment
install_lamindb(envvname = "your-env")

## End(Not run)

Lamin CLI functions

Description

[Deprecated]

Lamin CLI calls are available from R by importing the lamin_cli Python module using lc <- import_module("lamin_cli"). The previous CLI functions are now deprecated, see examples for how to switch to the new syntax.

Usage

# Import the module instead of using deprecated functions
# lc <- import_module("lamin_cli")

# Deprecated functions

lamin_connect(instance)

lamin_disconnect()

lamin_login(user = NULL, api_key = NULL)

lamin_logout()

lamin_init(storage, name = NULL, db = NULL, modules = NULL)

lamin_init_temp(
  name = "laminr-temp",
  db = NULL,
  modules = NULL,
  add_timestamp = TRUE,
  envir = parent.frame()
)

lamin_delete(instance, force = FALSE)

lamin_save(filepath, key = NULL, description = NULL, registry = NULL)

lamin_settings()

Arguments

instance

Either a slug giving the instance to connect to (⁠<owner>/<name>⁠) or an instance URL (⁠https://lamin.ai/owner/name⁠). For lamin_delete(), the slug for the instance to delete.

user

Handle for the user to login as

api_key

API key for a user

storage

A local directory, AWS S3 bucket or Google Cloud Storage bucket

name

A name for the instance

db

A Postgres database connection URL, use NULL for SQLite

modules

A vector of modules to include (e.g. "bionty")

add_timestamp

Whether to append a timestamp to name to make it unique

envir

An environment passed to withr::defer()

force

Whether to force deletion without asking for confirmation

filepath

Path to the file or folder to save

key

The key for the saved item

description

The description for the saved item

registry

The registry for the saved item

Details

lamin_connect()

Running this will set the LaminDB auto-connect option to True so you auto-connect to instance when importing Python lamindb.

lamin_login()

Depending on the input, one of these commands will be run (in this order):

  1. If user is set then ⁠lamin login <user>⁠

  2. Else if api_key is set then set the LAMIN_API_KEY environment variable temporarily with withr::with_envvar() and run ⁠lamin login⁠

  3. Else if there is a stored user handle run ⁠lamin login <handle>⁠

  4. Else if the LAMIN_API_KEY environment variable is set run ⁠lamin login⁠

Otherwise, exit with an error

lamin_init_temp()

For lamin_init_temp(), a time stamp is appended to name (if add_timestamp = TRUE) and then a new instance is initialised with lamin_init() using a temporary directory. A lamin_delete() call is registered as an exit handler with withr::defer() to clean up the instance when envir finishes.

The lamin_init_temp() function is mostly for internal use and in most cases users will want lamin_init().

Examples

## Not run: 
# Import Lamin modules
lc <- import_module("lamin_cli")
ln <- import_module("lamindb")

# Examples of replacing CLI functions with the lamin_cli module

## End(Not run)
## Not run: 
# Connect to a LaminDB instance
lamin_connect(instance)
# ->
lc$connect(instance)

## End(Not run)
## Not run: 
# Disconnect from a LaminDB instance
lamin_disconnect()
# ->
lc$disconnect()

## End(Not run)
## Not run: 
# Log in as a LaminDB user
lamin_login(...)
# ->
lc$login(...)

## End(Not run)
## Not run: 
# Log out of LaminDB
lamin_logout()
# ->
lc$logout()

## End(Not run)
## Not run: 
# Create a new LaminDB instance
lamin_init(...)
# ->
lc$init(...)

## End(Not run)
## Not run: 
# Create a temporary LaminDB instance
lamin_init_temp(...)
# ->
create_temporary_instance()

## End(Not run)
## Not run: 
# Delete a LaminDB entity
lamin_delete(...)
# ->
lc$delete(...)

## End(Not run)
## Not run: 
# Save to a LaminDB instance
lamin_save(...)
# ->
lc$save(...)

## End(Not run)
## Not run: 
# Access Lamin settings
lamin_settings()
# ->
ln$setup$settings
# OR
ln$settings
# Alternatively
get_current_lamin_settings()

## End(Not run)

laminr status

Description

Overview of the current status of the laminr package and its dependencies. Can be useful for debugging.

Usage

laminr_status()

Details

Provides information that can be useful for debugging. To run the function when an error occurs, set ⁠options(error = function() { print(laminr::laminr_status() })⁠. Note that this should be used with some caution as it will print the status whenever any error occurs.

Value

A laminr_status object

Examples

laminr_status()

Require a Python module

Description

This function can be used to require that Python modules are available for laminr with additional checks and nicer error messages.

Usage

require_module(
  module,
  options = NULL,
  version = NULL,
  source = NULL,
  python_version = NULL,
  silent = FALSE
)

Arguments

module

The name of the Python module to require

options

A vector of defined optional dependencies for the module that is being required

version

A string specifying the version of the module to require

source

A source for the module requirement, for example ⁠git+https://github.com/owner/module.git⁠

python_version

A string defining the Python version to require. Passed to reticulate::py_require()

silent

Whether to suppress the message showing what has been required

Details

Python dependencies are set using reticulate::py_require(). If a connection to Python is already initialized and the requested module is already in the list of requirements then a further call to reticulate::py_require() will not be made to avoid errors/warnings. This means that required versions etc. need to be set before Python is initialized.

Arguments

Value

The result of reticulate::py_require

See Also

reticulate::py_require()

Examples

## Not run: 
# Require lamindb
require_module("lamindb")

# Require a specific version of lamindb
require_module("lamindb", version = ">=1.2")

# Require require lamindb with options
require_module("lamindb", options = c("bionty", "wetlab"))

# Require the development version of lamindb from GitHub
require_module("lamindb", source = "git+https://github.com/laminlabs/lamindb.git")

# Require lamindb with a specific Python version
require_module("lamindb", python_version = "3.12")

## End(Not run)

Use a temporary LaminDB instance

Description

Create and connect to a temporary LaminDB instance to use for the current session. This function is primarily intended for developers to use during testing and documentation but can also be useful for users to debug issues or create reproducible examples.

Usage

use_temporary_instance(
  name = "laminr-temp",
  modules = NULL,
  add_timestamp = TRUE,
  envir = parent.frame()
)

Arguments

name

A name for the temporary instance

modules

A vector of modules to include (e.g. "bionty")

add_timestamp

Whether to append a time stamp to name to make it unique

envir

An environment passed to withr::defer()

Details

This function creates and connects to a temporary LaminDB instance. A temporary storage folder is created and used to initialize a new instance. An exit handler is registered with withr::defer() that deletes the instance and storage, then reconnects to the previous instance when envir finishes.

Switching to a temporary instance is not possible when another instance is already connected.