registream

Core infrastructure for the RegiStream ecosystem — configuration, updates, telemetry, and shared utilities for register data packages.

Quick Start

Python

pip install registream

import pandas as pd
import registream.autolabel

df = pd.read_stata("my_register_data.dta")
df.autolabel(domain="scb", lang="eng")

df.variable_labels()       # {"age": "Age in years", "sex": "Biological sex", ...}
df.value_labels()          # {"sex": {1: "Male", 2: "Female"}, ...}
df.lab.head()              # Display with labels substituted (no mutation)
df.lookup("kon")           # Look up variable metadata

Stata

net install registream, from("https://registream.org/install/stata/latest") replace
net install autolabel,  from("https://registream.org/install/stata/latest") replace

use my_register_data, clear
autolabel variables, domain(scb) lang(eng)
autolabel values, domain(scb) lang(eng)

R

# install.packages("remotes")
remotes::install_github("registream/registream", subdir = "r")
remotes::install_github("registream/autolabel",  subdir = "r")

library(autolabel)
df <- haven::read_dta("my_register_data.dta")
labelled <- autolabel(df, domain = "scb", lang = "eng")
rs_lookup("kon", domain = "scb", lang = "eng")
rs_lab(labelled)              # display with labels substituted

To install just the core:

pip install registream-core

net install registream, from("https://registream.org/install/stata/latest") replace

Commands

Python

from registream.autolabel import update_datasets

update_datasets("scb", "eng")               # Download metadata cache
df.autolabel(domain="scb", lang="eng")       # Apply labels
df.lookup("kon")                             # Look up variable metadata
df.lab.head()                                # Labeled display view

from registream.citation import cite
from registream.info import info
print(cite())                                # Citation for publications
print(info())                                # Configuration display

Command-line equivalents:

python -m registream version                # installed core version
python -m registream info                   # config + cache snapshot
python -m registream cite                   # APA + BibTeX citation
python -m registream.autolabel {version,info,cite}

Stata

registream info                          Display configuration and settings
registream config [, options]            View or edit configuration
registream update                        Check for package updates
registream version                       Show installed version
registream cite                          Display citation for publications
registream stats                         View local usage statistics

First-Run Setup

On first use, registream prompts you to choose a setup mode:

Mode	Internet	Update checks	Telemetry	Best for
Offline	No	No	No	Air-gapped / MONA environments
Standard	Yes	Yes	No	Most users (recommended)
Full	Yes	Yes	Yes	Help improve RegiStream

Modules

RegiStream is a modular ecosystem. The core package provides shared infrastructure; modules provide domain-specific functionality.

Module	Description	Stata	Python	R
autolabel	Variable and value labels from register metadata	net install autolabel, from("https://registream.org/install/stata/latest") replace	pip install registream-autolabel	remotes::install_github("registream/autolabel", subdir="r")
datamirror	Checkpoint-constrained synthetic data	net install datamirror, from("https://registream.org/install/stata/latest") replace	coming	coming

For Python, pip install registream installs core + autolabel in one step (meta-package; matches the Stata install-everything UX of net install registream followed by net install autolabel). Install individual packages with pip install registream-core or pip install registream-autolabel.

Modules are independent Stata packages from v3.0.0 onward; install only what you need.

Modules depend on the core package.

Privacy

• Local logging (on by default): Commands are logged to ~/.registream/usage_{stata,python,r}.csv. Never leaves your computer.
• Telemetry (off by default): Anonymous usage data sent to help improve RegiStream. Opt-in only.

For Developers

See DEVELOPMENT.md for architecture, build system, and contribution guide.

Citation

from registream.citation import cite
print(cite())

registream cite

Authors

Jeffrey Clark and Jie Wen

License

BSD 3-Clause. See LICENSE.