Metadata-Version: 2.5
Name: vgi-polars
Version: 0.4.1
Summary: Polars connector for VGI (Vector Gateway Interface) — scan VGI catalog tables and call VGI functions from Polars
Project-URL: Homepage, https://github.com/Query-farm/vgi-polars
Project-URL: Repository, https://github.com/Query-farm/vgi-polars
Project-URL: Issues, https://github.com/Query-farm/vgi-polars/issues
Project-URL: VGI Protocol, https://github.com/Query-farm/vgi-python
Author-email: Rusty Conover <rusty@query.farm>
Maintainer-email: Query Farm LLC <hello@query.farm>
License:                                  Apache License
                                   Version 2.0, January 2004
                                http://www.apache.org/licenses/
        
           TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
        
           1. Definitions.
        
              "License" shall mean the terms and conditions for use, reproduction,
              and distribution as defined by Sections 1 through 9 of this document.
        
              "Licensor" shall mean the copyright owner or entity authorized by
              the copyright owner that is granting the License.
        
              "Legal Entity" shall mean the union of the acting entity and all
              other entities that control, are controlled by, or are under common
              control with that entity. For the purposes of this definition,
              "control" means (i) the power, direct or indirect, to cause the
              direction or management of such entity, whether by contract or
              otherwise, or (ii) ownership of fifty percent (50%) or more of the
              outstanding shares, or (iii) beneficial ownership of such entity.
        
              "You" (or "Your") shall mean an individual or Legal Entity
              exercising permissions granted by this License.
        
              "Source" form shall mean the preferred form for making modifications,
              including but not limited to software source code, documentation
              source, and configuration files.
        
              "Object" form shall mean any form resulting from mechanical
              transformation or translation of a Source form, including but
              not limited to compiled object code, generated documentation,
              and conversions to other media types.
        
              "Work" shall mean the work of authorship, whether in Source or
              Object form, made available under the License, as indicated by a
              copyright notice that is included in or attached to the work
              (an example is provided in the Appendix below).
        
              "Derivative Works" shall mean any work, whether in Source or Object
              form, that is based on (or derived from) the Work and for which the
              editorial revisions, annotations, elaborations, or other modifications
              represent, as a whole, an original work of authorship. For the
              purposes of this License, Derivative Works shall not include works
              that remain separable from, or merely link (or bind by name) to the
              interfaces of, the Work and Derivative Works thereof.
        
              "Contribution" shall mean any work of authorship, including
              the original version of the Work and any modifications or additions
              to that Work or Derivative Works thereof, that is intentionally
              submitted to Licensor for inclusion in the Work by the copyright owner
              or by an individual or Legal Entity authorized to submit on behalf of
              the copyright owner. For the purposes of this definition, "submitted"
              means any form of electronic, verbal, or written communication sent
              to the Licensor or its representatives, including but not limited to
              communication on electronic mailing lists, source code control systems,
              and issue tracking systems that are managed by, or on behalf of, the
              Licensor for the purpose of discussing and improving the Work, but
              excluding communication that is conspicuously marked or otherwise
              designated in writing by the copyright owner as "Not a Contribution."
        
              "Contributor" shall mean Licensor and any individual or Legal Entity
              on behalf of whom a Contribution has been received by Licensor and
              subsequently incorporated within the Work.
        
           2. Grant of Copyright License. Subject to the terms and conditions of
              this License, each Contributor hereby grants to You a perpetual,
              worldwide, non-exclusive, no-charge, royalty-free, irrevocable
              copyright license to reproduce, prepare Derivative Works of,
              publicly display, publicly perform, sublicense, and distribute the
              Work and such Derivative Works in Source or Object form.
        
           3. Grant of Patent License. Subject to the terms and conditions of
              this License, each Contributor hereby grants to You a perpetual,
              worldwide, non-exclusive, no-charge, royalty-free, irrevocable
              (except as stated in this section) patent license to make, have made,
              use, offer to sell, sell, import, and otherwise transfer the Work,
              where such license applies only to those patent claims licensable
              by such Contributor that are necessarily infringed by their
              Contribution(s) alone or by combination of their Contribution(s)
              with the Work to which such Contribution(s) was submitted. If You
              institute patent litigation against any entity (including a
              cross-claim or counterclaim in a lawsuit) alleging that the Work
              or a Contribution incorporated within the Work constitutes direct
              or contributory patent infringement, then any patent licenses
              granted to You under this License for that Work shall terminate
              as of the date such litigation is filed.
        
           4. Redistribution. You may reproduce and distribute copies of the
              Work or Derivative Works thereof in any medium, with or without
              modifications, and in Source or Object form, provided that You
              meet the following conditions:
        
              (a) You must give any other recipients of the Work or
                  Derivative Works a copy of this License; and
        
              (b) You must cause any modified files to carry prominent notices
                  stating that You changed the files; and
        
              (c) You must retain, in the Source form of any Derivative Works
                  that You distribute, all copyright, patent, trademark, and
                  attribution notices from the Source form of the Work,
                  excluding those notices that do not pertain to any part of
                  the Derivative Works; and
        
              (d) If the Work includes a "NOTICE" text file as part of its
                  distribution, then any Derivative Works that You distribute must
                  include a readable copy of the attribution notices contained
                  within such NOTICE file, excluding those notices that do not
                  pertain to any part of the Derivative Works, in at least one
                  of the following places: within a NOTICE text file distributed
                  as part of the Derivative Works; within the Source form or
                  documentation, if provided along with the Derivative Works; or,
                  within a display generated by the Derivative Works, if and
                  wherever such third-party notices normally appear. The contents
                  of the NOTICE file are for informational purposes only and
                  do not modify the License. You may add Your own attribution
                  notices within Derivative Works that You distribute, alongside
                  or as an addendum to the NOTICE text from the Work, provided
                  that such additional attribution notices cannot be construed
                  as modifying the License.
        
              You may add Your own copyright statement to Your modifications and
              may provide additional or different license terms and conditions
              for use, reproduction, or distribution of Your modifications, or
              for any such Derivative Works as a whole, provided Your use,
              reproduction, and distribution of the Work otherwise complies with
              the conditions stated in this License.
        
           5. Submission of Contributions. Unless You explicitly state otherwise,
              any Contribution intentionally submitted for inclusion in the Work
              by You to the Licensor shall be under the terms and conditions of
              this License, without any additional terms or conditions.
              Notwithstanding the above, nothing herein shall supersede or modify
              the terms of any separate license agreement you may have executed
              with Licensor regarding such Contributions.
        
           6. Trademarks. This License does not grant permission to use the trade
              names, trademarks, service marks, or product names of the Licensor,
              except as required for reasonable and customary use in describing the
              origin of the Work and reproducing the content of the NOTICE file.
        
           7. Disclaimer of Warranty. Unless required by applicable law or
              agreed to in writing, Licensor provides the Work (and each
              Contributor provides its Contributions) on an "AS IS" BASIS,
              WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
              implied, including, without limitation, any warranties or conditions
              of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
              PARTICULAR PURPOSE. You are solely responsible for determining the
              appropriateness of using or redistributing the Work and assume any
              risks associated with Your exercise of permissions under this License.
        
           8. Limitation of Liability. In no event and under no legal theory,
              whether in tort (including negligence), contract, or otherwise,
              unless required by applicable law (such as deliberate and grossly
              negligent acts) or agreed to in writing, shall any Contributor be
              liable to You for damages, including any direct, indirect, special,
              incidental, or consequential damages of any character arising as a
              result of this License or out of the use or inability to use the
              Work (including but not limited to damages for loss of goodwill,
              work stoppage, computer failure or malfunction, or any and all
              other commercial damages or losses), even if such Contributor
              has been advised of the possibility of such damages.
        
           9. Accepting Warranty or Additional Liability. While redistributing
              the Work or Derivative Works thereof, You may choose to offer,
              and charge a fee for, acceptance of support, warranty, indemnity,
              or other liability obligations and/or rights consistent with this
              License. However, in accepting such obligations, You may act only
              on Your own behalf and on Your sole responsibility, not on behalf
              of any other Contributor, and only if You agree to indemnify,
              defend, and hold each Contributor harmless for any liability
              incurred by, or claims asserted against, such Contributor by reason
              of your accepting any such warranty or additional liability.
        
           END OF TERMS AND CONDITIONS
        
           APPENDIX: How to apply the Apache License to your work.
        
              To apply the Apache License to your work, attach the following
              boilerplate notice, with the fields enclosed by brackets "[]"
              replaced with your own identifying information. (Don't include
              the brackets!)  The text should be enclosed in the appropriate
              comment syntax for the file format. We also recommend that a
              file or class name and description of purpose be included on the
              same "printed page" as the copyright notice for easier
              identification within third-party archives.
        
           Copyright 2026 Query Farm LLC
        
           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.
License-File: LICENSE
Keywords: analytics,apache-arrow,arrow,data-engineering,polars,pyarrow,vgi
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Developers
Classifier: License :: OSI Approved :: Apache Software License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3.14
Classifier: Topic :: Database
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Classifier: Typing :: Typed
Requires-Python: >=3.13
Requires-Dist: polars>=1.41.1
Requires-Dist: pyarrow>=18
Requires-Dist: vgi-python>=0.29.0
Provides-Extra: http
Requires-Dist: vgi-python[http]; extra == 'http'
Description-Content-Type: text/markdown

<p align="center">
  <img src="https://raw.githubusercontent.com/Query-farm/vgi-polars/main/docs/vgi-logo.png?v=1" alt="VGI logo" width="260">
  &nbsp;&nbsp;+&nbsp;&nbsp;
  <img src="https://raw.githubusercontent.com/Query-farm/vgi-polars/main/docs/polars-logo.svg?v=1" alt="Polars logo" width="180">
</p>

# vgi-polars

[![PyPI](https://img.shields.io/pypi/v/vgi-polars.svg)](https://pypi.org/project/vgi-polars/)
[![Python](https://img.shields.io/pypi/pyversions/vgi-polars.svg)](https://pypi.org/project/vgi-polars/)
[![License](https://img.shields.io/badge/license-Apache--2.0-blue.svg)](LICENSE)
[![CI](https://github.com/Query-farm/vgi-polars/actions/workflows/ci.yml/badge.svg)](https://github.com/Query-farm/vgi-polars/actions/workflows/ci.yml)

A [Polars](https://pola.rs) client for [VGI](https://github.com/Query-farm/vgi-python)
(Vector Gateway Interface). Lets a `polars.LazyFrame`/`polars.DataFrame` scan a VGI
catalog's tables and call its scalar, table-in-out, and aggregate functions — the same
role the [VGI DuckDB extension](https://github.com/Query-farm/vgi) plays for DuckDB,
but for Polars, with no DuckDB dependency at all.

This is not a new VGI protocol implementation. It's a thin adapter over
[vgi-python](https://github.com/Query-farm/vgi-python)'s existing pure-Python,
Arrow-native reference client (`vgi.client.Client`) — the same wire-protocol code the
DuckDB extension speaks — combined with Polars'
[`polars.io.plugins.register_io_source`](https://docs.pola.rs/api/python/stable/reference/api/polars.io.plugins.register_io_source.html)
extension point, which was purpose-built for exactly this "external source with
pushdown" shape.

## Installation

```bash
pip install vgi-polars
```

HTTP-transport support (talking to a VGI worker over `http://`/`https://`, and the
Orchard remote-secret-provider path) needs an extra:

```bash
pip install "vgi-polars[http]"
```

Subprocess and TCP transports need no extra.

## Quick start

```python
import polars as pl
import vgi_polars as vp

with vp.attach("path/to/my-vgi-worker", name="my_catalog") as cat:
    print(cat.schemas())
    print(cat.tables("main"))

    t = cat.table("main", "events")
    print(t.schema)                # polars.Schema, no scan performed
    print(t.scan().filter(pl.col("value") > 90).collect())

    my_fn = cat.scalar_function("main", "my_function")
    df = pl.DataFrame({"a": [1, 2, 3], "b": [10, 20, 30]})
    print(df.with_columns(my_fn(pl.col("a"), pl.col("b")).alias("result")))
```

`attach()` auto-detects transport from the location string's scheme — a bare command
is a subprocess worker, `http://`/`https://` is HTTP, `tcp://host:port` is raw
Arrow-IPC framing over TCP:

```python
cat = vp.attach("http://localhost:8080", name="my_catalog")
```

Any VGI worker — written in Python, Rust, Go, Java, or TypeScript — speaks to
vgi-polars unchanged; VGI is a cross-language protocol, not a Python-specific one.
See [vgi-python](https://github.com/Query-farm/vgi-python) for reference worker
implementations and the protocol documentation.

## Live example: earthquakes

No local worker needed — this attaches over HTTPS to a live, public VGI worker
serving the [USGS Earthquake Hazards Program](https://earthquake.usgs.gov/)'s
rolling 30-day feed as an ordinary table (a weather example follows below).
More live example workers at [query.farm/vgi](https://query.farm/vgi/).

```python
import polars as pl
import vgi_polars as vp

cat = vp.attach("https://vgi-earthquakes.rusty-bb6.workers.dev", name="earthquakes")
recent = cat.table("main", "recent")

print(
    recent.scan()
    .filter(pl.col("mag") >= 5)
    .sort("mag", descending=True)
    .head(8)
    .select("time", pl.col("mag").round(1), "place")
    .collect()
)
```

```text
shape: (8, 3)
┌─────────────────────────────┬─────┬─────────────────────────────────┐
│ time                        ┆ mag ┆ place                           │
│ ---                         ┆ --- ┆ ---                             │
│ datetime[μs, UTC]           ┆ f64 ┆ str                             │
╞═════════════════════════════╪═════╪═════════════════════════════════╡
│ 2026-08-14 21:58:21.564 UTC ┆ 7.7 ┆ 68 km NNW of Ende, Indonesia    │
│ 2026-08-10 12:34:28.125 UTC ┆ 7.4 ┆ 5 km S of San José del Palmar,… │
│ …                           ┆ …   ┆ …                               │
└─────────────────────────────┴─────┴─────────────────────────────────┘
```

The `.filter()`/`.sort()`/`.head()`/`.select()` chain runs entirely against the
`LazyFrame` `.scan()` returns — nothing is fetched until `.collect()`. If the worker
declares filter/projection pushdown support, `mag >= 5` and the column selection are
sent to it to reduce what crosses the wire; either way, the *complete* original
predicate and projection are always re-applied locally after scanning too (see
[Pushdown is an optimization, never a correctness delegation](#pushdown-is-an-optimization-never-a-correctness-delegation)
below), so the result is identical whether or not pushdown happened to work.

## Live example: weather

This worker exposes no plain catalog tables at all — every function
(`geocoding`, `forecast_hourly`, ...) is a *blended row-transform* function,
callable either as a bare literal call or joined against an existing
`LazyFrame`/`DataFrame` (DuckDB's `FROM t, LATERAL f(t.x)`, for Polars). Both
shapes chain together below: a literal `geocoding` call resolves a place name
to coordinates, then `forecast_hourly` is called *against that result*,
correlating each output row back to the place that produced it.

Try it live: [Colab notebook](https://colab.research.google.com/drive/1eGmuXmpaLQ4MfStPTWlDXabfX5aAZf8a?usp=sharing).

```python
import polars as pl
import vgi_polars as vp

cat = vp.attach("https://vgi-open-meteo.rusty-bb6.workers.dev", name="open_meteo")
geocoding = cat.row_transform_function("main", "geocoding")
forecast_hourly = cat.row_transform_function("main", "forecast_hourly")

# Literal call: fn(None, ...) -- no LazyFrame, just arguments.
place = geocoding(None, "Glen Allen, VA", count=1, country_code="US")

# Correlated call: fn(lf, pl.col(...), ...) -- joined against `place`'s
# result, one forecast per input row (here, just the one place).
forecast = forecast_hourly(
    place,
    pl.col("latitude"),
    pl.col("longitude"),
    forecast_days=1,
    temperature_unit="fahrenheit",
).collect()

print(forecast.select("name", "time", pl.col("temperature_2m").round(1).alias("temp_f"), "weather_code").head(6))
```

```text
shape: (6, 4)
┌────────────┬─────────────────────────┬────────┬──────────────┐
│ name       ┆ time                    ┆ temp_f ┆ weather_code │
│ ---        ┆ ---                     ┆ ---    ┆ ---          │
│ str        ┆ datetime[μs, UTC]       ┆ f64    ┆ i32          │
╞════════════╪═════════════════════════╪════════╪══════════════╡
│ Glen Allen ┆ 2026-08-27 00:00:00 UTC ┆ 76.4   ┆ 0            │
│ Glen Allen ┆ 2026-08-27 01:00:00 UTC ┆ 73.8   ┆ 1            │
│ Glen Allen ┆ 2026-08-27 02:00:00 UTC ┆ 73.1   ┆ 0            │
│ Glen Allen ┆ 2026-08-27 03:00:00 UTC ┆ 72.7   ┆ 0            │
│ Glen Allen ┆ 2026-08-27 04:00:00 UTC ┆ 71.4   ┆ 2            │
│ Glen Allen ┆ 2026-08-27 05:00:00 UTC ┆ 71.0   ┆ 0            │
└────────────┴─────────────────────────┴────────┴──────────────┘
```

`place` is itself a `LazyFrame` — `forecast_hourly` never sees raw coordinates,
it sees the result of another VGI call, and every one of `place`'s own columns
(`name`, `country`, `population`, ...) rides along onto each forecast row for
free. Swap in `forecast_daily`, `historical_hourly`, `marine_hourly`, or any
of the worker's other functions the same way — one registration serves the
literal, column, and correlated-join call shapes uniformly.

## Pushdown is an optimization, never a correctness delegation

This is the single design principle vgi-polars won't compromise on, so it's worth
stating plainly: **a worker's pushdown support is never trusted for correctness, only
for performance.**

Polars' `register_io_source` extension point — the mechanism `.scan()` is built on —
does **not** re-verify a predicate or projection an `io_source` claims to have
applied. An `io_source` that silently ignores the `predicate` it's handed still gets
every row back, unfiltered, in the final `.collect()` result; Polars has no fallback
check. This was confirmed empirically, not assumed: a `register_io_source` callback
that received a filter and did nothing with it produced unfiltered results with no
warning or error anywhere in the pipeline.

VGI workers, meanwhile, are written and tested against the DuckDB extension, which
*always* re-verifies a pushed-down predicate against DuckDB's own query engine — so a
worker can declare `filter_pushdown`/`projection_pushdown` support and still apply
either only approximately (e.g. a worker that pushes an equality filter but silently
ignores a range filter it doesn't know how to translate) and never notice, because
DuckDB was always going to catch the difference downstream. Polars won't.

Two systems that each individually assume "the other side will catch what I miss"
add up to neither side catching anything. So vgi-polars breaks that: every scan
**always** applies the complete, original `with_columns` selection, `predicate`, and
row-limit truncation locally, after fetching, regardless of what was pushed down or
what the worker claims to have handled. A partial or entirely-failed pushdown
translation is therefore only ever a performance loss — sending more rows/columns
than strictly necessary — never a correctness one.

## Status

**Implemented:**

- Catalog attach/detach with versioning introspection
- Schema and table discovery, incl. per-column statistics
- Table scan (eager + lazy) with best-effort projection/filter pushdown, incl. `is_in`
- `required_filters` cost-safety enforcement
- Sequential split-scan redemption
- Transparent multi-branch-table scanning (`pl.concat` under the hood)
- A minimal in-memory/TTL result cache
- Time-travel scans (`AT` clauses)
- Scalar function calls with scoped secrets and per-chunk input dedup
- Streaming and buffered table-in-out functions
- Blended (row-transform) table functions (`cat.row_transform_function(schema, name)`) —
  a worker function called with caller-supplied literal or column arguments and no
  separate table input, both the correlated-join shape (DuckDB's `SELECT * FROM t,
  LATERAL geocode(t.place)` equivalent: `geocode(lf, pl.col("place"))`) and the bare
  literal-call shape (`geocode(None, 'some place')`)
- Aggregate functions
- Native scan-function delegation (`read_parquet` -> `pl.scan_parquet`, `read_csv` ->
  `pl.scan_csv`, `iceberg_scan` -> `pl.scan_iceberg`) — a worker that ships no data of
  its own and instead tells the caller to run a native reader itself (VGI's
  `ScanFunctionResult` mechanism; see e.g.
  [vgi-overture-maps](https://github.com/Query-farm/vgi-overture-maps-typescript),
  a pure-metadata Overture Maps catalog). Real Polars-native pushdown (row-group
  pruning, cloud range reads), not anything vgi-polars hand-rolls
- Subprocess, HTTP, and TCP transports

**Not implemented:**

- Writes
- Companion-catalog federation
- Per-table time-travel discovery
- The `container://`/`github://` transport schemes (a substantially larger effort — a
  from-scratch Python transport layer, not an extension of the existing scheme table).
  `launch:`/`unix://` (a launcher-managed shared worker) is implemented in the
  underlying `vgi-python` client (`Client.from_launch`, v0.29.4+) but not yet wired
  into `attach()`'s scheme detection here.

## Development

```bash
git clone https://github.com/Query-farm/vgi-polars.git
git clone https://github.com/Query-farm/vgi-python.git   # sibling checkout — see below
cd vgi-polars
uv sync
uv run pytest -v
```

`vgi-python` is pulled from that local sibling checkout (`[tool.uv.sources]` in
`pyproject.toml`, path `../vgi-python`) rather than the published PyPI release,
because this repo tracks vgi-python's client-side surface as both projects develop
together — the same pattern `vgi-spark`'s `settings.gradle.kts` uses to
composite-build a sibling `vgi-java` checkout. Integration tests need a
`vgi-fixture-worker` binary from that checkout; `VGI_PYTHON` (default
`~/Development/vgi-python`) picks which one, and `VGI_TEST_WORKER` overrides the
binary path directly if you need something other than the default venv location.

`uv run mypy src/`, `uv run ruff check src/ tests/`, and `uv run ruff format --check
src/ tests/` mirror what CI runs; `tests/test_docstrings.py` runs `pydoclint` as part
of the normal `pytest` run rather than as a separate step.

## License

Apache License, Version 2.0 — see [LICENSE](LICENSE).
