`dbx-tools-core`

Dependency-free Python configuration, identity, and mise-backed executable helpers shared by dbx-tools packages.

Install from PyPI:

pip install dbx-tools-core

To install the current main branch directly from the repository instead:

pip install "dbx-tools-core @ git+https://github.com/reggie-db/dbx-tools.git@main#subdirectory=packages/py/core"

Key features

config.text() resolves scoped keys from constant data, the process environment, the nearest project .env file, the single App's config.env in validated Databricks bundle JSON, then app.yaml / app.yml env values. Root bundle variables are not a config source: they are authoring inputs interpolated into the bundle's own targets, resources, and paths, so reading one as a process setting resolves names the deployed App never sees. Reference a variable from config.env to make it one.
.env.<NODE_ENV> wins over .env, with production/prod and development/dev treated as aliases.
Bundle validation stays lazy: the Databricks CLI runs only after environment and dotenv lookup miss; app YAML runs only after bundle lookup misses. Parsed dotenv records are cached by file path, bundle output by bundle path plus Databricks profile, and app YAML by app path. Bundle value_from and App YAML valueFrom references resolve supported values from named resources. Config-file discovery and parsed results are single-attempt per source key: found paths, missing files, empty records, invalid records, and None results all cache.
Deployed Databricks Apps skip local files and bundle validation because the platform has already populated real environment variables.
DBX_TOOLS_DATABRICKS_APP_ENV=true or false forces Databricks App runtime detection; absent or unrecognized values retain automatic detection.
DBX_TOOLS_CONFIG_DOTENV, DBX_TOOLS_CONFIG_BUNDLE, and DBX_TOOLS_CONFIG_APP independently force each local source on or off, overriding the usual deployed-App skip.
Bundle reads default off when NODE_ENV=production unless DBX_TOOLS_CONFIG_BUNDLE=true explicitly enables them.
String, boolean, positive-number, positive-integer, and list helpers use the same loose configuration coercions as @dbx-tools/core.
Stable-key, FNV hash, and identifier functions preserve deterministic Node and Python identity contracts.
cache.file_lock() provides a cross-platform process and thread lock around cache refreshes and other shared filesystem operations.
cache.check_lock_check() centralizes the cached-read, lock, cached-read, load sequence so concurrent processes share one refresh.
bin.resolve() reuses executables from PATH, otherwise performs a check-lock-check mise installation and returns the path reported by mise which.
bin.execute() has the asyncio.create_subprocess_exec calling convention, adds only mise_tool=, and returns the native asyncio.subprocess.Process. If mise itself is missing on macOS or Linux, the official checksum-verifying installer is run under the same cross-process lock.

Quick start

from dbx_tools.core import config

host = config.text("HOST", {"prefix": "SMTP"})
port = config.positive_int(None, "PORT", 587, {"prefix": "SMTP"})
endpoint = config.resolve_value(
    "lakebaseEndpoint",
    {
        "data": {"LAKEBASE_ENDPOINT": flags.endpoint},
        "sources": ["env", "dotenv", "bundle", "app"],
    },
)

Mise-backed async subprocesses retain the standard library process API:

import asyncio

from dbx_tools.core import bin

process = await bin.execute(
    "uv",
    "--version",
    mise_tool="uv@0.11",
    stdout=asyncio.subprocess.PIPE,
)
stdout, _ = await process.communicate()

Use bin.ensure_tool("neo4j@5.26.12").root when a caller needs an installed tool directory rather than one executable.

The default key order for HOST with prefix SMTP is DBX_TOOLS_SMTP_HOST, SMTP_HOST, then HOST. Pass config.ENV_ONLY when a caller must read the exact process environment without local file fallbacks. Constant data is first by default. If custom sources omit config, passed data is still read and appended last, as in the environment-first example. resolve_value() tries exact, uppercase, and tokenized-uppercase names through the same scope and prefix rules. config.is_databricks_app_env() validates the App name, HTTP(S) workspace host, and TCP port unless config.DATABRICKS_APP_ENV_KEY names a recognized boolean override in the environment. config.CONFIG_DOTENV_KEY, config.CONFIG_BUNDLE_KEY, and config.CONFIG_APP_KEY name the equivalent per-source overrides. Recognized booleans win; absent or unrecognized values read files outside a Databricks App and skip them inside one. Bundle validation also stays off by default in production.

Modules

cache - platform cache paths, cross-process file locking, and check-lock-check loading;
bin — locked mise bootstrap, tool installation, executable resolution, and native asyncio subprocess creation;
config — layered environment, dotenv, Databricks bundle, and app YAML configuration;
hash.fnv_hash() — the single-string subset of TypeScript fnvHashWithOptions, including UTF-16 code-unit hashing and base-32 output;
object.to_stable_key() — strict structured identity canonicalization;
string.to_identifier() — readable identifier tokenization, with the same hyphen default as TypeScript and an explicit delimiter override for consumers such as the underscore-delimited Postgres bus channel.

The identity functions exist so Python packages do not copy the TypeScript algorithms locally and silently drift. Their shared behavior is exercised by colocated polygotTest callbacks in the owning TypeScript packages. Configuration precedence, dotenv discovery/parsing, bundle fallback, laziness, and parsed-record caching are covered by each runtime's native config tests.

Release files for dbx-tools-core 0.6.208

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for dbx-tools-core 0.6.208
File	Size	Uploaded
dbx_tools_core-0.6.208.tar.gz	14.2 kB	Sep 14, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for dbx-tools-core 0.6.208
File	Interpreter	ABI	Platform	Reset
dbx_tools_core-0.6.208-py3-none-any.whl 17.0 kB Sep 14, 2026	Python 3	none	any	Details

Total release size: 31.2 kB

Release files / dbx_tools_core-0.6.208.tar.gz

Download URL	dbx_tools_core-0.6.208.tar.gz
Size	14.2 kB
Tags	Source
SHA-256 checksum How to use checksums	`4ccec392206b2362766e4125727fcefee45d465ccc2f118353d87c17cbcb1547`
BLAKE2b-256 checksum How to use checksums	`e5f45360e76bf31ea148b38e5e01bf82e3e70924cd5bd99974f0ff169ad35989`
Upload date	Sep 14, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/reggie-db/dbx-tools
Publishing commit: github.com/reggie-db/dbx-tools/tree/3efe83356acdf4bdb8d16c077420ead8b11e9c99

Workflow

Publishing configuration: github.com/reggie-db/dbx-tools/blob/3efe83356acdf4bdb8d16c077420ead8b11e9c99/.github/workflows/release.yml
Publishing logs: github.com/reggie-db/dbx-tools/actions/runs/34902113507/attempts/1

Artifact

Subject: dbx_tools_core-0.6.208.tar.gz
SHA-256 checksum: 4ccec392206b2362766e4125727fcefee45d465ccc2f118353d87c17cbcb1547 Verifying attestations

Release files / dbx_tools_core-0.6.208-py3-none-any.whl

Download URL	dbx_tools_core-0.6.208-py3-none-any.whl
Size	17.0 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`e7109cccf1d8eefadfdcef16d0e92a0d344d39fdfb98bfd17c13912749a4db1c`
BLAKE2b-256 checksum How to use checksums	`d23d5ca14a36c58b388e198893fbba1d5453e7c93fb0f8fd5a3ff4ad0d0ce8fe`
Upload date	Sep 14, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/7.0.0 CPython/3.13.14`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/reggie-db/dbx-tools
Publishing commit: github.com/reggie-db/dbx-tools/tree/3efe83356acdf4bdb8d16c077420ead8b11e9c99

Workflow

Publishing configuration: github.com/reggie-db/dbx-tools/blob/3efe83356acdf4bdb8d16c077420ead8b11e9c99/.github/workflows/release.yml
Publishing logs: github.com/reggie-db/dbx-tools/actions/runs/34902113507/attempts/1

Artifact

Subject: dbx_tools_core-0.6.208-py3-none-any.whl
SHA-256 checksum: e7109cccf1d8eefadfdcef16d0e92a0d344d39fdfb98bfd17c13912749a4db1c Verifying attestations