parsimony-sdmx 0.2.0__tar.gz

parsimony_sdmx-0.2.0/.gitignore

@@ -0,0 +1,40 @@
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+
+.Python
+build/
+dist/
+*.egg-info/
+*.egg
+
+.venv/
+.env
+.env.*
+!.env.example
+
+.pytest_cache/
+.mypy_cache/
+.ruff_cache/
+.coverage
+htmlcov/
+coverage.xml
+
+uv.lock
+
+.vscode/
+.council/
+PLAN-*.md
+.idea/
+*.swp
+.DS_Store
+
+outputs/
+
+# Recorded HTTP cassettes must never be committed — respx mocks are hand-authored
+# from upstream API documentation. A pre-commit / CI regex scan is the belt; this
+# ignore is the braces. Override per-file via `!` if you need a hand-authored
+# fixture checked in.
+packages/*/tests/fixtures/**
+!packages/*/tests/fixtures/README.md

parsimony_sdmx-0.2.0/CHANGELOG.md

@@ -0,0 +1,5 @@
+# Changelog — parsimony-sdmx
+
+## Unreleased
+
+- Initial scaffold.

parsimony_sdmx-0.2.0/LICENSE

@@ -0,0 +1,190 @@
+                                 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 the 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 the 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 any 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
+
+   Copyright 2026 Ockham.sh
+
+   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.

parsimony_sdmx-0.2.0/PKG-INFO

@@ -0,0 +1,145 @@
+Metadata-Version: 2.4
+Name: parsimony-sdmx
+Version: 0.2.0
+Summary: SDMX protocol connector (ECB, Eurostat, IMF, OECD, BIS, World Bank, ILO) for the parsimony framework
+Project-URL: Homepage, https://parsimony.dev
+Project-URL: Repository, https://github.com/ockham-sh/parsimony-connectors
+Project-URL: Issues, https://github.com/ockham-sh/parsimony-connectors/issues
+Author-email: "Ockham.sh" <team@ockham.sh>
+License-Expression: Apache-2.0
+License-File: LICENSE
+Keywords: connectors,data,finance,parsimony,sdmx
+Classifier: Development Status :: 4 - Beta
+Classifier: Intended Audience :: Developers
+Classifier: Intended Audience :: Financial and Insurance Industry
+Classifier: License :: OSI Approved :: Apache Software License
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Topic :: Office/Business :: Financial
+Classifier: Topic :: Software Development :: Libraries :: Python Modules
+Classifier: Typing :: Typed
+Requires-Python: >=3.11
+Requires-Dist: beautifulsoup4>=4.12
+Requires-Dist: lxml>=5.0
+Requires-Dist: pandas<3,>=2.3.0
+Requires-Dist: parsimony-core<0.3,>=0.1.0a0
+Requires-Dist: psutil>=5.9
+Requires-Dist: pyarrow>=16.0
+Requires-Dist: pydantic<3,>=2.11.1
+Requires-Dist: requests>=2.32
+Requires-Dist: sdmx1>=2.16
+Provides-Extra: dev
+Requires-Dist: mypy>=1.10; extra == 'dev'
+Requires-Dist: pytest-asyncio>=1.3.0; extra == 'dev'
+Requires-Dist: pytest-cov>=5.0; extra == 'dev'
+Requires-Dist: pytest>=9.0.3; extra == 'dev'
+Requires-Dist: pyyaml>=6.0; extra == 'dev'
+Requires-Dist: respx>=0.22.0; extra == 'dev'
+Requires-Dist: ruff>=0.15.10; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# parsimony-sdmx
+
+Flat SDMX catalog builder. Harvests dataflows from statistical agencies (ECB, Eurostat, IMF, World Bank) and writes two parquet files per agency — a dataset-level catalog and one series-level catalog per dataset — ready to feed a FAISS or SQL index.
+
+## Layout
+
+```
+outputs/{AGENCY}/datasets.parquet        # columns: dataset_id, agency_id, title
+outputs/{AGENCY}/series/{DATASET}.parquet # columns: id, dataset_id, title
+```
+
+Titles are composed as `"CODE1: label1 - CODE2: label2 - …"` over the DSD's non-`TIME_PERIOD` dimensions in DSD order. ECB series additionally append `TITLE` / `TITLE_COMPL` fetched from the per-series XML endpoint. HTML-embedded descriptions (common on Eurostat) are stripped before they reach parquet.
+
+## Install
+
+Requires Python ≥ 3.12 and [uv](https://docs.astral.sh/uv/).
+
+```bash
+make install              # uv sync --all-extras
+```
+
+## Usage
+
+```bash
+# Print every dataset the agency exposes
+parsimony-sdmx -a ESTAT --list-datasets
+
+# Write only outputs/{AGENCY}/datasets.parquet (no series fetched)
+parsimony-sdmx -a ECB --catalog
+
+# Fetch one dataset
+parsimony-sdmx -a ECB -d YC
+
+# Fetch every dataset the agency exposes
+parsimony-sdmx -a ESTAT --all
+
+# Preview what an --all run would write, without fetching
+parsimony-sdmx -a ESTAT --all --dry-run
+
+# Rebuild datasets whose parquet already exists (resume contract: file present → skip)
+parsimony-sdmx -a ECB -d YC --force
+```
+
+Or via Make shortcuts:
+
+```bash
+make catalog AGENCY=ECB            # datasets.parquet only
+make fetch AGENCY=ECB DATASET=YC   # single dataset
+make fetch-all AGENCY=ESTAT        # every dataset for the agency
+make list AGENCY=IMF_DATA          # enumerate datasets
+```
+
+Supported agencies: `ECB`, `ESTAT`, `IMF_DATA`, `WB_WDI`. Exit codes: `0` every dataset ok/empty, `1` at least one failed.
+
+## Architecture
+
+Four packages under `parsimony_sdmx/`:
+
+- **`core/`** — pure, I/O-free: record dataclasses, title composition, codelist resolution, outcome types, domain exceptions.
+- **`io/`** — boundary effects: atomic parquet writers, hardened lxml iterparse, HTTPS-only bounded HTTP session, path safety helpers, exception classification.
+- **`providers/`** — per-agency adapters behind a narrow `CatalogProvider` protocol; ECB/ESTAT/IMF share a common sdmx1 flow helper, WB diverges with a path × decade sweep because its SDMX endpoint doesn't expose `series_keys`.
+- **`cli/`** — argparse front-end, orchestrator that forks one subprocess per dataset (`mp.spawn`) for memory isolation, a psutil-backed memory monitor that kills the largest child above a threshold and writes OOM markers, atomic `.tmp/` cleanup, and an operator-readable end-of-run summary.
+
+### Why subprocess-per-dataset
+
+`sdmx1` caches structure messages at module level with no public invalidation hook. Running every dataset in its own subprocess is the only reliable way to start each fetch with a clean cache. Large catalogs (8 k+ Eurostat dataflows) have hit real OOMs in production; the memory monitor catches runaway workers before the kernel OOM killer does, preserving a classifiable failure instead of an opaque `exit 137`.
+
+### Resume
+
+Filesystem-backed: a dataset is skipped if `outputs/{AGENCY}/series/{DATASET}.parquet` already exists. Writes land in `.tmp/` first and are `os.replace`-d atomically, so the canonical path exists iff the previous run completed. `--force` overrides (and logs a count of overwrites).
+
+## Development
+
+```bash
+make check         # ruff + mypy strict + fast tests
+make test          # fast tests only (skip subprocess-path tests)
+make test-slow     # subprocess tests (fork real mp.Process children)
+make test-all      # everything
+make format        # ruff format + --fix
+make clean         # wipe caches + build artifacts
+```
+
+Hardening enforced by default:
+
+- `mypy --strict` across source and tests
+- `ruff` with `E, F, W, I, B, UP, S` (security rules on)
+- Hardened `lxml.iterparse` (no entity resolution, no DTD load, no network)
+- HTTPS-only `bounded_get` with a configurable byte cap
+- Path traversal guards on every on-disk write
+
+## Project layout
+
+```
+parsimony_sdmx/
+├── core/      # pure domain logic
+├── io/        # boundary-layer effects
+├── providers/ # per-agency adapters
+└── cli/       # argparse → orchestrator → worker → parquet
+
+tests/        # flat: test_<module>.py per source module
+```
+
+Test suite: 312 tests. 4 of them (`@pytest.mark.slow`) fork real subprocesses to exercise the orchestrator's timeout / OOM classification / clean-exit paths; the other 308 run in < 2 s.

parsimony_sdmx-0.2.0/README.md

@@ -0,0 +1,103 @@
+# parsimony-sdmx
+
+Flat SDMX catalog builder. Harvests dataflows from statistical agencies (ECB, Eurostat, IMF, World Bank) and writes two parquet files per agency — a dataset-level catalog and one series-level catalog per dataset — ready to feed a FAISS or SQL index.
+
+## Layout
+
+```
+outputs/{AGENCY}/datasets.parquet        # columns: dataset_id, agency_id, title
+outputs/{AGENCY}/series/{DATASET}.parquet # columns: id, dataset_id, title
+```
+
+Titles are composed as `"CODE1: label1 - CODE2: label2 - …"` over the DSD's non-`TIME_PERIOD` dimensions in DSD order. ECB series additionally append `TITLE` / `TITLE_COMPL` fetched from the per-series XML endpoint. HTML-embedded descriptions (common on Eurostat) are stripped before they reach parquet.
+
+## Install
+
+Requires Python ≥ 3.12 and [uv](https://docs.astral.sh/uv/).
+
+```bash
+make install              # uv sync --all-extras
+```
+
+## Usage
+
+```bash
+# Print every dataset the agency exposes
+parsimony-sdmx -a ESTAT --list-datasets
+
+# Write only outputs/{AGENCY}/datasets.parquet (no series fetched)
+parsimony-sdmx -a ECB --catalog
+
+# Fetch one dataset
+parsimony-sdmx -a ECB -d YC
+
+# Fetch every dataset the agency exposes
+parsimony-sdmx -a ESTAT --all
+
+# Preview what an --all run would write, without fetching
+parsimony-sdmx -a ESTAT --all --dry-run
+
+# Rebuild datasets whose parquet already exists (resume contract: file present → skip)
+parsimony-sdmx -a ECB -d YC --force
+```
+
+Or via Make shortcuts:
+
+```bash
+make catalog AGENCY=ECB            # datasets.parquet only
+make fetch AGENCY=ECB DATASET=YC   # single dataset
+make fetch-all AGENCY=ESTAT        # every dataset for the agency
+make list AGENCY=IMF_DATA          # enumerate datasets
+```
+
+Supported agencies: `ECB`, `ESTAT`, `IMF_DATA`, `WB_WDI`. Exit codes: `0` every dataset ok/empty, `1` at least one failed.
+
+## Architecture
+
+Four packages under `parsimony_sdmx/`:
+
+- **`core/`** — pure, I/O-free: record dataclasses, title composition, codelist resolution, outcome types, domain exceptions.
+- **`io/`** — boundary effects: atomic parquet writers, hardened lxml iterparse, HTTPS-only bounded HTTP session, path safety helpers, exception classification.
+- **`providers/`** — per-agency adapters behind a narrow `CatalogProvider` protocol; ECB/ESTAT/IMF share a common sdmx1 flow helper, WB diverges with a path × decade sweep because its SDMX endpoint doesn't expose `series_keys`.
+- **`cli/`** — argparse front-end, orchestrator that forks one subprocess per dataset (`mp.spawn`) for memory isolation, a psutil-backed memory monitor that kills the largest child above a threshold and writes OOM markers, atomic `.tmp/` cleanup, and an operator-readable end-of-run summary.
+
+### Why subprocess-per-dataset
+
+`sdmx1` caches structure messages at module level with no public invalidation hook. Running every dataset in its own subprocess is the only reliable way to start each fetch with a clean cache. Large catalogs (8 k+ Eurostat dataflows) have hit real OOMs in production; the memory monitor catches runaway workers before the kernel OOM killer does, preserving a classifiable failure instead of an opaque `exit 137`.
+
+### Resume
+
+Filesystem-backed: a dataset is skipped if `outputs/{AGENCY}/series/{DATASET}.parquet` already exists. Writes land in `.tmp/` first and are `os.replace`-d atomically, so the canonical path exists iff the previous run completed. `--force` overrides (and logs a count of overwrites).
+
+## Development
+
+```bash
+make check         # ruff + mypy strict + fast tests
+make test          # fast tests only (skip subprocess-path tests)
+make test-slow     # subprocess tests (fork real mp.Process children)
+make test-all      # everything
+make format        # ruff format + --fix
+make clean         # wipe caches + build artifacts
+```
+
+Hardening enforced by default:
+
+- `mypy --strict` across source and tests
+- `ruff` with `E, F, W, I, B, UP, S` (security rules on)
+- Hardened `lxml.iterparse` (no entity resolution, no DTD load, no network)
+- HTTPS-only `bounded_get` with a configurable byte cap
+- Path traversal guards on every on-disk write
+
+## Project layout
+
+```
+parsimony_sdmx/
+├── core/      # pure domain logic
+├── io/        # boundary-layer effects
+├── providers/ # per-agency adapters
+└── cli/       # argparse → orchestrator → worker → parquet
+
+tests/        # flat: test_<module>.py per source module
+```
+
+Test suite: 312 tests. 4 of them (`@pytest.mark.slow`) fork real subprocesses to exercise the orchestrator's timeout / OOM classification / clean-exit paths; the other 308 run in < 2 s.

parsimony_sdmx-0.2.0/parsimony_sdmx/__init__.py

@@ -0,0 +1,43 @@
+"""``parsimony-sdmx`` — SDMX connector plugin for the ``parsimony`` kernel.
+
+Exports:
+
+- :data:`CONNECTORS` — the plugin surface discovered via the
+  ``parsimony.providers`` entry point group. Three items: two enumerators
+  (dataset-level + per-dataset series) and one live fetch connector. Both
+  enumerators carry a ``catalog=`` declaration so ``parsimony bundles``
+  drives publish; SDMX no longer ships its own bundle CLI.
+- :data:`ENV_VARS` — empty. SDMX endpoints are public.
+- :data:`PROVIDER_METADATA` — bundle topology and supported agencies.
+
+Discovery is driven by the kernel via entry points declared in
+``pyproject.toml``. No manual registration required — users ``pip install
+parsimony-sdmx`` and the plugin appears in ``parsimony list-plugins``.
+"""
+
+from __future__ import annotations
+
+from typing import Any
+
+from parsimony_sdmx.connectors import CONNECTORS, ENV_VARS
+from parsimony_sdmx.connectors._agencies import ALL_AGENCIES
+
+__version__ = "0.2.0"
+
+
+PROVIDER_METADATA: dict[str, Any] = {
+    "agencies": [a.value for a in ALL_AGENCIES],
+    "namespace_templates": [
+        "sdmx_datasets",
+        "sdmx_series_{agency}_{dataset_id}",
+    ],
+    "plugin_version": __version__,
+}
+
+
+__all__ = [
+    "CONNECTORS",
+    "ENV_VARS",
+    "PROVIDER_METADATA",
+    "__version__",
+]

parsimony_sdmx-0.2.0/parsimony_sdmx/_catalog_planning.py

@@ -0,0 +1,95 @@
+"""Plan generators for ``parsimony.bundles`` discovery.
+
+The new bundle pipeline (``parsimony.bundles``) drives every plugin's
+publish flow through ``CatalogDynamicSpec.plan`` — an async generator
+that yields one :class:`~parsimony.bundles.CatalogPlan` per bundle the
+plugin wants built. SDMX has thousands of per-dataset series bundles, so
+the plan generator walks the on-disk flat-catalog parquet files and
+emits one plan item per ``(agency, dataset_id)`` pair.
+
+The on-disk root is :data:`DEFAULT_OUTPUTS_ROOT` (sibling to the package),
+overridable via the ``PARSIMONY_SDMX_OUTPUTS_ROOT`` env var. Missing
+agencies are silently skipped — callers running ``parsimony bundles
+plan`` against a workspace where only one agency has been built locally
+should see only that agency's bundles.
+
+This module is import-cheap: it imports ``pyarrow`` lazily inside
+:func:`plan_sdmx_series` so importing the plugin's surface (which the
+``parsimony list-plugins`` discovery does eagerly) doesn't pay arrow's
+cost when no one is publishing yet.
+"""
+
+from __future__ import annotations
+
+import os
+from collections.abc import AsyncIterator
+from pathlib import Path
+
+from parsimony.bundles import CatalogPlan
+
+from parsimony_sdmx.connectors._agencies import (
+    ALL_AGENCIES,
+    AgencyId,
+    to_namespace_token,
+)
+
+
+def _outputs_root() -> Path:
+    """Resolve the flat-catalog outputs root from env var or default.
+
+    The default is imported lazily — :mod:`parsimony_sdmx.connectors.enumerate_datasets`
+    pulls in pyarrow+pandas at module load, which we don't want when the
+    plan generator is only being inspected (e.g. by the discovery walk).
+    """
+    env = os.environ.get("PARSIMONY_SDMX_OUTPUTS_ROOT")
+    if env:
+        return Path(env)
+    from parsimony_sdmx.connectors.enumerate_datasets import DEFAULT_OUTPUTS_ROOT
+
+    return DEFAULT_OUTPUTS_ROOT
+
+
+def _series_namespace(agency: AgencyId, dataset_id: str) -> str:
+    """Compose the per-dataset series namespace from agency + dataset id.
+
+    The template literal is inlined here (rather than imported from
+    :mod:`parsimony_sdmx.connectors.enumerate_series`) to avoid a
+    circular import — ``enumerate_series`` declares ``catalog=`` with
+    a callable that lives in this module.
+    """
+    return f"sdmx_series_{to_namespace_token(agency)}_{dataset_id.lower()}"
+
+
+async def plan_sdmx_series() -> AsyncIterator[CatalogPlan]:
+    """Yield one plan per ``(agency, dataset_id)`` pair found on disk.
+
+    Reads each agency's ``outputs/{AGENCY}/datasets.parquet`` and emits a
+    :class:`CatalogPlan` for every row. Empty / absent agency files are
+    skipped silently — local workspaces don't always have every agency.
+
+    Plan params shape::
+
+        {"agency": "ECB", "dataset_id": "YC"}
+
+    These map 1:1 to :class:`~parsimony_sdmx.connectors.enumerate_series.EnumerateSeriesParams`
+    so the ``parsimony bundles`` runner adapter constructs the model
+    directly via ``EnumerateSeriesParams(**plan.params)``.
+    """
+    import pyarrow.parquet as pq
+
+    root = _outputs_root()
+    for agency in ALL_AGENCIES:
+        path = root / agency.value / "datasets.parquet"
+        if not path.exists():
+            continue
+        table = pq.read_table(path, columns=["dataset_id"])
+        for dataset_id in table.column("dataset_id").to_pylist():
+            yield CatalogPlan(
+                namespace=_series_namespace(agency, dataset_id),
+                params={"agency": agency.value, "dataset_id": dataset_id},
+            )
+
+
+__all__ = [
+    "plan_sdmx_series",
+]