parsimony-bde 0.4.0__tar.gz

parsimony_bde-0.4.0/.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_bde-0.4.0/CHANGELOG.md

@@ -0,0 +1,21 @@
+# Changelog — parsimony-bde
+
+All notable changes to `parsimony-bde` will be documented in this file. The
+format is based on [Keep a Changelog](https://keepachangelog.com/) and
+this project adheres to [Semantic Versioning](https://semver.org/).
+
+## [Unreleased]
+
+## [0.4.0] — 2026-04-24
+
+Part of the first coordinated release of the
+[`parsimony-connectors`](https://github.com/ockham-sh/parsimony-connectors)
+monorepo under `parsimony-core==0.4`.
+
+### Changed
+
+- Connector rewritten against the kernel's `parsimony.discover` surface
+  (`iter_providers`, `load`, `load_all`) and the `@connector(env=...)`
+  decorator-level env-var declaration that replaced module-level
+  `ENV_VARS`.
+- Pin bumped to `parsimony-core>=0.4,<0.5`.

parsimony_bde-0.4.0/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_bde-0.4.0/PKG-INFO

@@ -0,0 +1,102 @@
+Metadata-Version: 2.4
+Name: parsimony-bde
+Version: 0.4.0
+Summary: Banco de España connector for the parsimony framework
+Project-URL: Homepage, https://www.bde.es
+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: bde,connectors,data,finance,parsimony
+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: pandas<3,>=2.3.0
+Requires-Dist: parsimony-core<0.5,>=0.4.0
+Requires-Dist: pydantic<3,>=2.11.1
+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: respx>=0.22.0; extra == 'dev'
+Requires-Dist: ruff>=0.15.10; extra == 'dev'
+Description-Content-Type: text/markdown
+
+# parsimony-bde
+
+Banco de España connector — Spanish macroeconomic, monetary, and financial time series via the BIEST REST API.
+
+Part of the [parsimony-connectors](https://github.com/ockham-sh/parsimony-connectors) monorepo. Distributed standalone on PyPI as `parsimony-bde`.
+
+## Connectors
+
+| Name | Kind | Description |
+|---|---|---|
+| `bde_fetch` | fetch | Fetch one or more BdE time series by series code (comma-separated). |
+| `enumerate_bde` | enumerator | Enumerate BdE series by querying well-known series codes for catalog seeding. |
+
+## Install
+
+```bash
+pip install parsimony-bde
+```
+
+Pulls in `parsimony-core>=0.4,<0.5` automatically. Verify discovery:
+
+```bash
+python -c "from parsimony import discover; print([p.name for p in discover.iter_providers()])"
+```
+
+## Configuration
+
+No configuration required — the BdE BIEST API is open and unauthenticated.
+
+## Quick start
+
+```python
+import asyncio
+from parsimony_bde import CONNECTORS
+
+async def main():
+    connectors = CONNECTORS.bind_env()
+    result = await connectors["bde_fetch"](key="D_1NBAF472")
+    print(result.data.head())
+
+asyncio.run(main())
+```
+
+For multi-plugin composition (autoloads everything installed):
+
+```python
+from parsimony import discover
+connectors = discover.load_all().bind_env()
+```
+
+## Catalog publishing
+
+This plugin publishes catalogs under the `bde` namespace. Build and push:
+
+```bash
+parsimony publish --provider bde --target "hf://<your-org>/parsimony-bde"
+```
+
+## Provider
+
+- Homepage: https://www.bde.es
+- API docs: https://www.bde.es/webbe/en/estadisticas/recursos/api-estadisticas-bde.html
+- Series browser: https://app.bde.es/bie_www/bie_wwwias/xml/Arranque.html (BIEST)
+
+## License
+
+See [LICENSE](./LICENSE).

parsimony_bde-0.4.0/README.md

@@ -0,0 +1,67 @@
+# parsimony-bde
+
+Banco de España connector — Spanish macroeconomic, monetary, and financial time series via the BIEST REST API.
+
+Part of the [parsimony-connectors](https://github.com/ockham-sh/parsimony-connectors) monorepo. Distributed standalone on PyPI as `parsimony-bde`.
+
+## Connectors
+
+| Name | Kind | Description |
+|---|---|---|
+| `bde_fetch` | fetch | Fetch one or more BdE time series by series code (comma-separated). |
+| `enumerate_bde` | enumerator | Enumerate BdE series by querying well-known series codes for catalog seeding. |
+
+## Install
+
+```bash
+pip install parsimony-bde
+```
+
+Pulls in `parsimony-core>=0.4,<0.5` automatically. Verify discovery:
+
+```bash
+python -c "from parsimony import discover; print([p.name for p in discover.iter_providers()])"
+```
+
+## Configuration
+
+No configuration required — the BdE BIEST API is open and unauthenticated.
+
+## Quick start
+
+```python
+import asyncio
+from parsimony_bde import CONNECTORS
+
+async def main():
+    connectors = CONNECTORS.bind_env()
+    result = await connectors["bde_fetch"](key="D_1NBAF472")
+    print(result.data.head())
+
+asyncio.run(main())
+```
+
+For multi-plugin composition (autoloads everything installed):
+
+```python
+from parsimony import discover
+connectors = discover.load_all().bind_env()
+```
+
+## Catalog publishing
+
+This plugin publishes catalogs under the `bde` namespace. Build and push:
+
+```bash
+parsimony publish --provider bde --target "hf://<your-org>/parsimony-bde"
+```
+
+## Provider
+
+- Homepage: https://www.bde.es
+- API docs: https://www.bde.es/webbe/en/estadisticas/recursos/api-estadisticas-bde.html
+- Series browser: https://app.bde.es/bie_www/bie_wwwias/xml/Arranque.html (BIEST)
+
+## License
+
+See [LICENSE](./LICENSE).

parsimony_bde-0.4.0/parsimony_bde/__init__.py

@@ -0,0 +1,524 @@
+"""Banco de España (BdE): fetch + catalog enumeration.
+
+API docs: https://www.bde.es/webbe/en/estadisticas/recursos/api-estadisticas-bde.html
+Series search: https://app.bde.es/bie_www/bie_wwwias/xml/Arranque.html (BIEST)
+No authentication required.
+
+The catalog enumerator pulls BdE's own published catalog CSVs — seven chapters
+(``catalogo_{be,cf,ie,pb,si,tc,ti}``) covering ~20,450 rows (≈15,500 unique
+series codes after de-duplication across overlapping chapters) spanning general
+statistics, financial accounts of the Spanish economy, international economy,
+bank lending surveys, financial indicators, exchange rates, and interest rates.
+BdE has no queryable list endpoint and no SDMX feed of its own; the CSV
+directory is the only discovery surface, and it carries the descriptive prose,
+frequency, units, date ranges, and dataset grouping needed for high-recall
+semantic search. An exhaustive probe of ``catalogo_{aa..zz}.csv`` confirms no
+other 2-letter chapter resolves to HTTP 200.
+"""
+
+from __future__ import annotations
+
+import contextlib
+import csv
+import io
+import logging
+from datetime import datetime
+from typing import Annotated, Any
+
+import httpx
+import pandas as pd
+from parsimony.connector import Connectors, connector, enumerator
+from parsimony.errors import EmptyDataError
+from parsimony.result import (
+    Column,
+    ColumnRole,
+    OutputConfig,
+    Provenance,
+    Result,
+)
+from parsimony.transport import map_http_error
+from pydantic import BaseModel, Field, field_validator
+
+logger = logging.getLogger(__name__)
+
+
+_BASE_URL = "https://app.bde.es/bierest/resources/srdatosapp"
+
+# BdE publishes its statistical catalog as seven CSV files, one per "chapter":
+#
+#   * BE — General Statistics (national accounts, prices, employment, …)
+#   * CF — Financial Accounts of the Spanish Economy (CFEE, SEC2010 sector
+#          balance sheets; ~4.7k rows, several hundred overlap with BE).
+#   * IE — International Economy (world prices, commodity indices; ~93 rows,
+#          mostly overlapping BE but retained for category-filtered search).
+#   * PB — Bank Lending Survey
+#   * SI — Financial Indicators (confidence indices, retail trade, …)
+#   * TC — Exchange Rates
+#   * TI — Interest Rates
+#
+# An exhaustive probe of ``catalogo_{aa..zz}.csv`` (676 combinations) on
+# 2026-04-24 confirms these seven are the complete published set; every other
+# 2-letter code 302s to a 404 page. Only the Spanish (``es``) variant resolves;
+# the ``en`` URL 302s to a 404. The catalog itself contains both Spanish and
+# English-translatable descriptions in its ``descripcion`` and ``titulo``
+# columns; we surface them as-is and let downstream embedders handle the
+# bilingual content.
+_CATALOG_CSV_BASE_URL = "https://www.bde.es/webbe/es/estadisticas/compartido/datos/csv"
+_CATALOG_CHAPTERS: tuple[tuple[str, str], ...] = (
+    ("be", "General Statistics"),
+    ("cf", "Financial Accounts"),
+    ("ie", "International Economy"),
+    ("pb", "Bank Lending Survey"),
+    ("si", "Financial Indicators"),
+    ("tc", "Exchange Rates"),
+    ("ti", "Interest Rates"),
+)
+
+# BdE catalog CSVs are encoded in CP1252 (Latin-1 superset). Lowercase column
+# headers map onto our schema. Index keys match the ``Nombre de la serie``
+# header (Spanish, with a trailing space the publisher kept since the 90s).
+_CSV_ENCODING = "cp1252"
+_CSV_HEADERS: tuple[str, ...] = (
+    "serie",          # Internal API code — fed to /listaSeries.
+    "seq",            # Numeric sequential id (unused).
+    "alias",          # Public alias like "TI_1_1.1".
+    "file",           # Source CSV file on bde.es.
+    "description",    # Long descriptive prose (Spanish).
+    "var_type",       # MEDIA / SUMA / FINAL aggregation kind.
+    "unit_code",      # ISO/internal unit code (EUR, %, USD/EUR, …).
+    "exponent",       # Power-of-ten scale on stored values.
+    "decimals",       # Display precision.
+    "unit_desc",      # Human-readable unit ("Millones de euros").
+    "frequency_raw",  # MENSUAL / TRIMESTRAL / DIARIA / LABORABLE / ANUAL.
+    "start_date",     # Spanish-format first observation ("MAR 1995").
+    "end_date",       # Spanish-format last observation ("DIC 2025").
+    "n_obs",          # Observation count.
+    "title",          # "/"-separated taxonomic path.
+    "source_org",     # Originating organisation (INE, BCE, BdE, …).
+    "notes",          # Methodological remarks.
+)
+
+# BdE chapter codes use Spanish frequency labels; we normalise to English to
+# match Treasury / FRED conventions and so an agent searching "monthly" hits
+# Spanish series that were originally labelled "MENSUAL".
+_FREQ_MAP_RAW = {
+    "DIARIA": "Daily",
+    "LABORABLE": "Business Daily",
+    "SEMANAL": "Weekly",
+    "QUINCENAL": "Bi-weekly",
+    "MENSUAL": "Monthly",
+    "TRIMESTRAL": "Quarterly",
+    "SEMESTRAL": "Semi-annual",
+    "ANUAL": "Annual",
+    # The /listaSeries endpoint returns single-letter codes for the same
+    # frequencies; we keep both maps in sync so bde_fetch can label rows.
+}
+
+# Frequency single-letter code (from /favoritas, /listaSeries) → English.
+_FREQ_MAP = {
+    "D": "Daily",
+    "M": "Monthly",
+    "Q": "Quarterly",
+    "A": "Annual",
+    "S": "Semi-annual",
+    "W": "Weekly",
+    "B": "Business Daily",
+}
+
+# Spanish ↔ English column mapping used by the fetch parser.
+_COLUMN_MAP = {
+    "serie": "key",
+    "descripcion": "description",
+    "descripcionCorta": "title",
+    "codFrecuencia": "freq",
+    "decimales": "decimals",
+    "simbolo": "symbol",
+    "fechaInicio": "start_date",
+    "fechaFin": "end_date",
+    "fechas": "date",
+    "valores": "value",
+}
+
+
+# ---------------------------------------------------------------------------
+# Parameter models
+# ---------------------------------------------------------------------------
+
+
+class BdeFetchParams(BaseModel):
+    """Parameters for fetching Banco de España time series."""
+
+    key: Annotated[str, "ns:bde"] = Field(
+        ...,
+        description="Comma-separated BdE series codes (e.g. D_1NBAF472)",
+    )
+    time_range: str | None = Field(
+        default=None,
+        description=("Time range: 30M, 60M, MAX, or a year (e.g. 2024). Default uses the full available range."),
+    )
+    lang: str = Field(default="en", description="Language: en or es")
+
+    @field_validator("key")
+    @classmethod
+    def _non_empty(cls, v: str) -> str:
+        v = v.strip()
+        if not v:
+            raise ValueError("At least one series code required")
+        return v
+
+    @field_validator("time_range")
+    @classmethod
+    def _valid_range(cls, v: str | None) -> str | None:
+        if v is None:
+            return v
+        v = v.strip()
+        # BdE API rejects short range codes (3M, 12M); accept 30M, 60M, MAX, or year
+        _VALID_RANGES = {"30M", "60M", "MAX"}
+        if v.upper() in _VALID_RANGES or v.isdigit():
+            return v
+        raise ValueError(f"Invalid time_range '{v}'. Use 30M, 60M, MAX, or a year (e.g. 2024).")
+
+    @field_validator("lang")
+    @classmethod
+    def _valid_lang(cls, v: str) -> str:
+        if v not in ("en", "es"):
+            raise ValueError("lang must be 'en' or 'es'")
+        return v
+
+
+class BdeEnumerateParams(BaseModel):
+    """No parameters needed — discovers series from BdE's published catalog CSVs."""
+
+    pass
+
+
+# ---------------------------------------------------------------------------
+# Output configs
+# ---------------------------------------------------------------------------
+
+BDE_ENUMERATE_OUTPUT = OutputConfig(
+    columns=[
+        Column(name="key", role=ColumnRole.KEY, namespace="bde"),
+        Column(name="title", role=ColumnRole.TITLE),
+        # ``description`` carries the upstream long-form prose. Lifted into
+        # DESCRIPTION (rather than METADATA) so the embedder sees it at index
+        # time — semantic recall on full sentences matters more than for
+        # categorical metadata.
+        Column(name="description", role=ColumnRole.DESCRIPTION),
+        # ``source`` lets agents dispatch the right fetch connector when more
+        # than one BdE source is wired. Today only ``bde_biest`` exists; the
+        # column is in place so adding (e.g.) an SDMX path later costs zero
+        # schema churn.
+        Column(name="source", role=ColumnRole.METADATA),
+        Column(name="alias", role=ColumnRole.METADATA),
+        Column(name="dataset", role=ColumnRole.METADATA),
+        Column(name="category", role=ColumnRole.METADATA),
+        Column(name="frequency", role=ColumnRole.METADATA),
+        Column(name="unit", role=ColumnRole.METADATA),
+        Column(name="decimals", role=ColumnRole.METADATA),
+        Column(name="start_date", role=ColumnRole.METADATA),
+        Column(name="end_date", role=ColumnRole.METADATA),
+        Column(name="n_obs", role=ColumnRole.METADATA),
+        Column(name="source_org", role=ColumnRole.METADATA),
+    ]
+)
+
+BDE_FETCH_OUTPUT = OutputConfig(
+    columns=[
+        Column(name="key", role=ColumnRole.KEY, param_key="key", namespace="bde"),
+        Column(name="title", role=ColumnRole.TITLE),
+        Column(name="date", dtype="datetime", role=ColumnRole.DATA),
+        Column(name="value", dtype="numeric", role=ColumnRole.DATA),
+    ]
+)
+
+
+# ---------------------------------------------------------------------------
+# Helpers
+# ---------------------------------------------------------------------------
+
+
+def _parse_bde_response(json_data: list[dict[str, Any]]) -> pd.DataFrame:
+    """Parse BdE JSON response into a long-format DataFrame.
+
+    Each element in json_data represents one series with parallel
+    fechas (dates) and valores (values) arrays.
+    """
+    all_rows: list[dict[str, Any]] = []
+
+    for series in json_data:
+        key = series.get("serie", "")
+        title = series.get("descripcionCorta", series.get("descripcion", key))
+        series.get("codFrecuencia", "")
+        dates = series.get("fechas", [])
+        values = series.get("valores", [])
+
+        if not dates or not values:
+            continue
+
+        for date_str, raw_value in zip(dates, values, strict=False):
+            try:
+                value = float(raw_value) if raw_value not in (None, "", "NaN") else None
+            except (ValueError, TypeError):
+                value = None
+
+            # Parse ISO datetime: "2024-01-31T00:00:00Z" → date
+            date_val = date_str
+            if isinstance(date_str, str) and "T" in date_str:
+                with contextlib.suppress(ValueError):
+                    date_val = datetime.strptime(date_str[:10], "%Y-%m-%d").strftime("%Y-%m-%d")
+
+            all_rows.append(
+                {
+                    "key": key,
+                    "title": title,
+                    "date": date_val,
+                    "value": value,
+                }
+            )
+
+    return pd.DataFrame(all_rows) if all_rows else pd.DataFrame(columns=["key", "title", "date", "value"])
+
+
+def _split_title_path(raw: str) -> tuple[str, str]:
+    """Split a "/"-separated BdE title-path into (dataset, leaf_title).
+
+    BdE encodes a taxonomic path in the title column. Two shapes occur:
+
+    1. **Short path** (most TI/TC/PB/SI rows) — e.g.
+       ``"Monetary policy/Eurosystem operations/Fixed rate auctions"``. The
+       leaf is the most specific bit and makes a good catalog title; the
+       prefix becomes ``dataset`` METADATA so agents can filter by family.
+    2. **Long faceted path** (most BE rows) — e.g.
+       ``"Descripción de la DSD: ... / Metodología: ... / Año Base: ... /
+       Tipo de Transformación: ..."``. Each segment is a faceted ``key: value``
+       attribute, and the "leaf" is just the last facet — meaningless on its
+       own. In this case we leave the title to the caller (who falls back to
+       ``description``) and put the whole faceted string in ``dataset``.
+
+    Heuristic for "faceted path": all segments contain ``:``. That distinguishes
+    BdE's DSD-encoded series (where every segment is ``Facet: value``) from
+    natural-language taxonomies (where slashes separate concept names).
+    """
+    if "/" not in raw:
+        return "", raw.strip()
+    parts = [p.strip() for p in raw.split("/") if p.strip()]
+    if not parts:
+        return "", raw.strip()
+    if len(parts) == 1:
+        return "", parts[0]
+    if all(":" in p for p in parts):
+        # Faceted DSD-encoded path — no semantic leaf.
+        return " › ".join(parts), ""
+    return " › ".join(parts[:-1]), parts[-1]
+
+
+def _parse_catalog_csv(text: str, *, category: str) -> list[dict[str, str]]:
+    """Parse one ``catalogo_*.csv`` payload into enumerator rows.
+
+    The CSV is comma-delimited, double-quoted, with a 17-column header BdE
+    has been stable on for years. We keep the raw ``serie`` (the API code)
+    as ``key`` and pull every descriptive column we can use for retrieval.
+    Empty/whitespace-only rows are skipped; rows missing the ``serie`` key
+    can't be fetched and are filtered out here.
+    """
+    reader = csv.reader(io.StringIO(text))
+    rows: list[dict[str, str]] = []
+
+    header_seen = False
+    for raw_row in reader:
+        if not raw_row:
+            continue
+        if not header_seen:
+            # First row is the schema; skip it. We rely on positional access.
+            header_seen = True
+            continue
+        if len(raw_row) < len(_CSV_HEADERS):
+            # Defensive: BdE has occasional malformed rows when their export
+            # job clips a description containing a literal ``"``. Skip rather
+            # than crash the enumerator.
+            logger.debug(
+                "skipping malformed BdE catalog row (got %d cols, expected %d)",
+                len(raw_row),
+                len(_CSV_HEADERS),
+            )
+            continue
+
+        record = dict(zip(_CSV_HEADERS, raw_row, strict=False))
+        serie = (record.get("serie") or "").strip()
+        if not serie:
+            continue
+
+        title_raw = (record.get("title") or "").strip()
+        dataset, leaf_title = _split_title_path(title_raw)
+        # Catalog title is the leaf if we have one, otherwise fall back to the
+        # description (always populated) so semantic_text is never empty.
+        title = leaf_title or (record.get("description") or "").strip() or serie
+
+        freq_raw = (record.get("frequency_raw") or "").strip().upper()
+        frequency = _FREQ_MAP_RAW.get(freq_raw, freq_raw.title() if freq_raw else "")
+
+        rows.append(
+            {
+                "key": serie,
+                "title": title,
+                "description": (record.get("description") or "").strip(),
+                "source": "bde_biest",
+                "alias": (record.get("alias") or "").strip(),
+                "dataset": dataset,
+                "category": category,
+                "frequency": frequency,
+                "unit": (record.get("unit_desc") or record.get("unit_code") or "").strip(),
+                "decimals": (record.get("decimals") or "").strip(),
+                "start_date": (record.get("start_date") or "").strip(),
+                "end_date": (record.get("end_date") or "").strip(),
+                "n_obs": (record.get("n_obs") or "").strip(),
+                "source_org": (record.get("source_org") or "").strip(),
+            }
+        )
+    return rows
+
+
+async def _fetch_catalog_chapter(
+    client: httpx.AsyncClient,
+    chapter: str,
+    category: str,
+) -> list[dict[str, str]]:
+    """Fetch one ``catalogo_*.csv`` and return its parsed rows.
+
+    Per-chapter failures are logged and degrade to an empty list rather than
+    aborting the whole enumeration — so a transient outage on the bank lending
+    survey CSV doesn't lose the 11k+ general statistics rows.
+    """
+    url = f"{_CATALOG_CSV_BASE_URL}/catalogo_{chapter}.csv"
+    try:
+        response = await client.get(url)
+        response.raise_for_status()
+    except httpx.HTTPError as exc:
+        logger.warning("BdE catalog chapter %r unavailable: %s", chapter, exc)
+        return []
+    # CSVs are CP1252-encoded; httpx's auto-detect can pick the wrong one if
+    # BdE forgets to send a charset header, so decode explicitly.
+    raw_bytes = response.content
+    try:
+        text = raw_bytes.decode(_CSV_ENCODING)
+    except UnicodeDecodeError:
+        # Fall back to latin-1 (strictly bigger than cp1252 in coverage); the
+        # only diff is a handful of typographic glyphs we don't index on.
+        text = raw_bytes.decode("latin-1", errors="replace")
+    return _parse_catalog_csv(text, category=category)
+
+
+# ---------------------------------------------------------------------------
+# Connectors
+# ---------------------------------------------------------------------------
+
+
+@connector(output=BDE_FETCH_OUTPUT, tags=["macro", "es"])
+async def bde_fetch(params: BdeFetchParams) -> Result:
+    """Fetch Banco de España time series by series code(s).
+
+    Uses the BdE REST API (BIEST). Returns date + value with series metadata.
+    """
+    url = f"{_BASE_URL}/listaSeries"
+
+    # BdE API: fetch each series individually and merge results.
+    # Multi-series in a single request is unreliable (412 errors).
+    keys = [k.strip() for k in params.key.split(",") if k.strip()]
+    json_data: list[dict[str, Any]] = []
+
+    async with httpx.AsyncClient(timeout=60.0) as client:
+        for key in keys:
+            req_params: dict[str, str] = {
+                "idioma": params.lang,
+                "series": key,
+            }
+            if params.time_range is not None:
+                req_params["rango"] = str(params.time_range)
+
+            response = await client.get(url, params=req_params)
+            try:
+                response.raise_for_status()
+            except httpx.HTTPStatusError as exc:
+                map_http_error(exc, provider="bde", op_name="series")
+            data = response.json()
+            if isinstance(data, list):
+                json_data.extend(data)
+
+    if not isinstance(json_data, list) or not json_data:
+        raise EmptyDataError(provider="bde", message=f"BdE returned empty or invalid response for: {params.key}")
+
+    df = _parse_bde_response(json_data)
+    if df.empty:
+        raise EmptyDataError(provider="bde", message=f"No observations parsed for: {params.key}")
+
+    return Result.from_dataframe(
+        df,
+        Provenance(
+            source="bde",
+            params={"key": params.key, "time_range": params.time_range},
+            properties={
+                "source_url": "https://www.bde.es/webbe/en/estadisticas/recursos/api-estadisticas-bde.html",
+            },
+        ),
+    )
+
+
+@enumerator(
+    output=BDE_ENUMERATE_OUTPUT,
+    tags=["macro", "es"],
+)
+async def enumerate_bde(params: BdeEnumerateParams) -> pd.DataFrame:
+    """Enumerate every BdE statistical series across the seven published catalog chapters.
+
+    BdE has no list endpoint or SDMX feed. The only authoritative discovery
+    surface is the ``catalogo_{be,cf,ie,pb,si,tc,ti}.csv`` files BdE publishes
+    alongside its statistical bulletin. Each row maps onto a series the
+    ``/listaSeries`` API can fetch by ``serie`` code, and carries the
+    descriptive prose, frequency, units, and date range needed to rank it
+    in semantic search. Per-chapter network failures degrade gracefully —
+    the enumerator returns whatever chapters succeeded rather than empty.
+    Some ``serie`` codes appear in more than one chapter (e.g. a national
+    accounts series listed under both BE and CF); we keep all occurrences so
+    agents filtering by ``category`` see the series under every taxonomy it
+    belongs to. Downstream de-duplication by ``key`` is the caller's call.
+    """
+    rows: list[dict[str, str]] = []
+    async with httpx.AsyncClient(timeout=60.0, follow_redirects=True) as client:
+        for chapter, category in _CATALOG_CHAPTERS:
+            chapter_rows = await _fetch_catalog_chapter(client, chapter, category)
+            rows.extend(chapter_rows)
+
+    columns = [
+        "key",
+        "title",
+        "description",
+        "source",
+        "alias",
+        "dataset",
+        "category",
+        "frequency",
+        "unit",
+        "decimals",
+        "start_date",
+        "end_date",
+        "n_obs",
+        "source_org",
+    ]
+    return pd.DataFrame(rows, columns=columns) if rows else pd.DataFrame(columns=columns)
+
+
+# ---------------------------------------------------------------------------
+# Exports
+# ---------------------------------------------------------------------------
+
+from parsimony_bde.search import (  # noqa: E402, F401  (after public decorators; re-exported)
+    BDE_SEARCH_OUTPUT,
+    PARSIMONY_BDE_CATALOG_URL_ENV,
+    BdeSearchParams,
+    bde_search,
+)
+
+CATALOGS: list[tuple[str, object]] = [("bde", enumerate_bde)]
+
+CONNECTORS = Connectors([bde_fetch, enumerate_bde, bde_search])

parsimony_bde-0.4.0/parsimony_bde/search.py

@@ -0,0 +1,112 @@
+"""Semantic search over the published Banco de España (BdE) catalog.
+
+Wraps the parquet+FAISS catalog at ``hf://parsimony-dev/bde`` (override with
+``PARSIMONY_BDE_CATALOG_URL`` for local testing) as an MCP tool. The agent
+calls :func:`bde_search` with a natural-language query and gets back the
+top-N matches with their codes, titles, and similarity scores.
+
+Codes returned by this tool are ``serie`` IDs that :func:`bde_fetch`
+accepts directly via its ``key`` parameter — the discover→fetch handshake.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+import os
+from typing import Annotated
+
+import pandas as pd
+from parsimony.catalog import Catalog
+from parsimony.connector import connector
+from parsimony.result import Column, ColumnRole, OutputConfig
+from pydantic import BaseModel, Field
+
+logger = logging.getLogger(__name__)
+
+#: Env var carrying the BdE catalog URL. Defaults to the canonical HF repo.
+#: Override with e.g. ``file:///path/to/catalogs/bde/repo/bde`` for local testing.
+PARSIMONY_BDE_CATALOG_URL_ENV = "PARSIMONY_BDE_CATALOG_URL"
+_DEFAULT_CATALOG_URL = "hf://parsimony-dev/bde"
+
+# Single-catalog cache. The BdE catalog is ~26 MB FAISS + 0.6 MB parquet —
+# a one-time load amortizes across every search call in the MCP session.
+_catalog: Catalog | None = None
+_catalog_lock = asyncio.Lock()
+
+
+async def _get_catalog() -> Catalog:
+    """Return the singleton BdE catalog, loading from URL on first use."""
+    global _catalog
+    if _catalog is not None:
+        return _catalog
+    async with _catalog_lock:
+        if _catalog is None:  # double-checked under the lock
+            url = os.environ.get(PARSIMONY_BDE_CATALOG_URL_ENV, _DEFAULT_CATALOG_URL)
+            logger.info("loading BdE catalog from %s", url)
+            _catalog = await Catalog.from_url(url)
+    return _catalog
+
+
+BDE_SEARCH_OUTPUT = OutputConfig(
+    columns=[
+        # ``code`` is the serie ID (e.g. ``D_1NBAF472``) that bde_fetch
+        # accepts via its ``key`` parameter — the search→fetch handshake.
+        Column(name="code", role=ColumnRole.KEY, namespace="bde"),
+        Column(name="title", role=ColumnRole.TITLE),
+        Column(name="similarity", role=ColumnRole.METADATA),
+    ]
+)
+
+
+class BdeSearchParams(BaseModel):
+    """Parameters for :func:`bde_search`."""
+
+    query: Annotated[
+        str,
+        Field(
+            min_length=1,
+            max_length=512,
+            description=(
+                "Natural-language description of the BdE series you want "
+                "(e.g. 'Spanish 10-year bond yield', 'Euribor 3-month', "
+                "'monthly HICP Spain'). Spanish or English both work."
+            ),
+        ),
+    ]
+    limit: int = Field(
+        default=10,
+        ge=1,
+        le=50,
+        description="Top-N results to return.",
+    )
+
+
+@connector(
+    output=BDE_SEARCH_OUTPUT,
+    tags=["macro", "es", "tool"],
+)
+async def bde_search(params: BdeSearchParams) -> pd.DataFrame:
+    """Semantic-search the Banco de España (BdE) catalog by natural language.
+
+    Returns the top matching ``serie`` codes from BdE's published
+    statistical catalog (~15.5k unique series across 7 chapters: general
+    statistics, financial accounts, international economy, bank lending
+    survey, financial indicators, exchange rates, interest rates).
+
+    Pass the returned ``code`` to ``bde_fetch(key=...)`` to retrieve the
+    actual time series. The catalog is bilingual — both Spanish and English
+    queries route correctly through the embedder.
+    """
+    catalog = await _get_catalog()
+    matches = await catalog.search(params.query, limit=params.limit)
+    return pd.DataFrame(
+        [
+            {
+                "code": m.code,
+                "title": m.title,
+                "similarity": round(m.similarity, 6),
+            }
+            for m in matches
+        ]
+    )

parsimony_bde-0.4.0/pyproject.toml

@@ -0,0 +1,76 @@
+[project]
+name = "parsimony-bde"
+version = "0.4.0"
+description = "Banco de España connector for the parsimony framework"
+authors = [{ name = "Ockham.sh", email = "team@ockham.sh" }]
+license = "Apache-2.0"
+readme = "README.md"
+requires-python = ">=3.11"
+keywords = ["finance", "data", "connectors", "parsimony", "bde"]
+classifiers = [
+    "Development Status :: 4 - Beta",
+    "Intended Audience :: Developers",
+    "Intended Audience :: Financial and Insurance Industry",
+    "License :: OSI Approved :: Apache Software License",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+    "Topic :: Office/Business :: Financial",
+    "Topic :: Software Development :: Libraries :: Python Modules",
+    "Typing :: Typed",
+]
+dependencies = [
+    "parsimony-core>=0.4.0,<0.5",
+    "pydantic>=2.11.1,<3",
+    "pandas>=2.3.0,<3",
+]
+
+[project.optional-dependencies]
+dev = [
+    "pytest>=9.0.3",
+    "pytest-asyncio>=1.3.0",
+    "pytest-cov>=5.0",
+    "respx>=0.22.0",
+    "ruff>=0.15.10",
+    "mypy>=1.10",
+]
+
+[project.urls]
+Homepage = "https://www.bde.es"
+Repository = "https://github.com/ockham-sh/parsimony-connectors"
+Issues = "https://github.com/ockham-sh/parsimony-connectors/issues"
+
+[project.entry-points."parsimony.providers"]
+bde = "parsimony_bde"
+
+
+[build-system]
+requires = ["hatchling"]
+build-backend = "hatchling.build"
+
+[tool.hatch.build.targets.wheel]
+packages = ["parsimony_bde"]
+
+[tool.hatch.build.targets.sdist]
+include = ["parsimony_bde", "README.md", "LICENSE", "CHANGELOG.md"]
+
+[tool.ruff]
+target-version = "py311"
+line-length = 120
+
+[tool.ruff.lint]
+select = ["E", "F", "I", "UP", "B", "SIM"]
+
+[tool.mypy]
+python_version = "3.11"
+warn_return_any = true
+warn_unused_ignores = true
+ignore_missing_imports = true
+
+[tool.pytest.ini_options]
+addopts = "--import-mode=importlib -m 'not integration'"
+asyncio_mode = "auto"
+markers = [
+    "integration: hits live APIs (may be slow, requires env vars)",
+]