votinvv-datapulse 0.0.1__tar.gz

votinvv_datapulse-0.0.1/LICENSE

@@ -0,0 +1,201 @@
+                                 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 [yyyy] [name of copyright owner]
+
+   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.

votinvv_datapulse-0.0.1/PKG-INFO

@@ -0,0 +1,77 @@
+Metadata-Version: 2.4
+Name: votinvv-datapulse
+Version: 0.0.1
+Summary: DataPulse SDK — decorators and runtime context for spec authors
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/votinvv/datapulse
+Project-URL: Repository, https://github.com/votinvv/datapulse
+Project-URL: Issues, https://github.com/votinvv/datapulse/issues
+Keywords: data-engineering,pipelines,scd2,postgresql,etl
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Database
+Classifier: Topic :: Software Development :: Libraries
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: postgres
+Requires-Dist: psycopg[binary]>=3.2; extra == "postgres"
+Provides-Extra: oracle
+Requires-Dist: oracledb>=3.0; extra == "oracle"
+Dynamic: license-file
+
+# DataPulse SDK
+
+The developer-facing library of [DataPulse](https://github.com/votinvv/datapulse) —
+declarative data pipelines over PostgreSQL with SCD2 versioning, parallel
+builds and a full audit trail.
+
+Spec authors install this package for IDE autocompletion and local runs:
+
+```
+pip install votinvv-datapulse
+```
+
+The distribution is named `votinvv-datapulse`; the import package is `datapulse`:
+
+```python
+from datapulse import spec, attr, mode, Self
+from datapulse.types import bigint, text, timestamptz
+
+
+@spec(
+    dataset="raw.customer",
+    num=1,
+    descr="Customers",
+    attrs=[
+        attr("customer_id", bigint, descr="Customer id", is_primary_key=True),
+        attr("full_name", text, descr="Full name"),
+    ],
+    modes=[mode("initial", "init", True), mode("increment", "incr", True)],
+)
+def build(param):
+    with Self() as db:
+        db.execute("insert into raw.customer_1_$ (...) select ...")
+```
+
+The SDK is pure stdlib — database drivers import lazily. To execute mappings
+locally (e.g. via `datapulse.dev.run_local`), pull the drivers you need:
+
+```
+pip install "votinvv-datapulse[postgres]"   # psycopg
+pip install "votinvv-datapulse[oracle]"     # oracledb
+```
+
+The server side — engine, HTTP API and UI — ships as the
+[`votinvv/datapulse`](https://hub.docker.com/r/votinvv/datapulse) Docker
+image; see the [project README](https://github.com/votinvv/datapulse) for the
+full picture.
+
+## License
+
+[Apache-2.0](https://github.com/votinvv/datapulse/blob/main/LICENSE)

votinvv_datapulse-0.0.1/README.md

@@ -0,0 +1,50 @@
+# DataPulse SDK
+
+The developer-facing library of [DataPulse](https://github.com/votinvv/datapulse) —
+declarative data pipelines over PostgreSQL with SCD2 versioning, parallel
+builds and a full audit trail.
+
+Spec authors install this package for IDE autocompletion and local runs:
+
+```
+pip install votinvv-datapulse
+```
+
+The distribution is named `votinvv-datapulse`; the import package is `datapulse`:
+
+```python
+from datapulse import spec, attr, mode, Self
+from datapulse.types import bigint, text, timestamptz
+
+
+@spec(
+    dataset="raw.customer",
+    num=1,
+    descr="Customers",
+    attrs=[
+        attr("customer_id", bigint, descr="Customer id", is_primary_key=True),
+        attr("full_name", text, descr="Full name"),
+    ],
+    modes=[mode("initial", "init", True), mode("increment", "incr", True)],
+)
+def build(param):
+    with Self() as db:
+        db.execute("insert into raw.customer_1_$ (...) select ...")
+```
+
+The SDK is pure stdlib — database drivers import lazily. To execute mappings
+locally (e.g. via `datapulse.dev.run_local`), pull the drivers you need:
+
+```
+pip install "votinvv-datapulse[postgres]"   # psycopg
+pip install "votinvv-datapulse[oracle]"     # oracledb
+```
+
+The server side — engine, HTTP API and UI — ships as the
+[`votinvv/datapulse`](https://hub.docker.com/r/votinvv/datapulse) Docker
+image; see the [project README](https://github.com/votinvv/datapulse) for the
+full picture.
+
+## License
+
+[Apache-2.0](https://github.com/votinvv/datapulse/blob/main/LICENSE)

votinvv_datapulse-0.0.1/datapulse/__init__.py

@@ -0,0 +1,39 @@
+"""DataPulse SDK — the developer-facing library.
+
+Installed by spec authors (`pip install data-pulse`) and imported in spec files:
+
+    from datapulse import spec, attr, mode, Self
+    from datapulse.types import bigint, text
+
+    @spec(dataset="raw.customer", num=1, attrs=[...], modes=[...])
+    def build(param):
+        with Self() as db:
+            db.execute("insert into raw.customer_1_$ (...) select ...")
+
+This package has NO runtime dependencies on Postgres, FastAPI, or the server —
+psycopg is imported lazily inside `Self` only. It carries: the declarative
+decorators, the type aliases, the `BuildParams` shape (so the IDE can
+autocomplete `param.`), the built-in `Self` context and the `secret()` helper.
+The server reuses the same decorators for validation/apply.
+
+The metadata payload a `@spec` builds lives on the function as the attribute
+`__datapulse__` — a plain dict the server reads by importing the module.
+
+User contexts (Oracle / S3 / Kafka) live in a developer-authored `context.py`
+as `@contextmanager` functions and read their secrets via `secret(...)`.
+"""
+from __future__ import annotations
+
+__version__ = "0.0.1"
+
+from .decorators import attr, mode, spec
+from .flow import CatalogSpec, flowspec
+from .params import BuildParams
+from .sources import Self
+from .secrets import secret, secret_section, SecretError
+from .types import DataType
+
+__all__ = [
+    "spec", "attr", "mode", "DataType", "flowspec", "CatalogSpec",
+    "BuildParams", "Self", "secret", "secret_section", "SecretError",
+]

votinvv_datapulse-0.0.1/datapulse/decorators.py

@@ -0,0 +1,153 @@
+"""Declarative decorators: `@spec`, `attr`, `mode`.
+
+A spec file imports `spec` and decorates its `build(param)` function. The
+decorator attaches the metadata payload as `build.__datapulse__` — a plain dict
+the server reads when the module is imported at apply time. Nothing about the
+function itself changes; it is still a normal Python callable that the engine
+will invoke as `build(param)` in production and the developer may invoke directly
+(e.g. via `datapulse.dev.run_local` from a scratch script or a test).
+
+The payload shape is the SAME dict shape the server's DDL planner consumes, so
+no translation layer is needed between SDK and server.
+
+Design choices:
+  - `attr`/`mode` are factory functions returning dicts, not classes: a spec is
+    data, and the server treats it as data. Keeping it as dicts avoids any
+    magic and keeps `__datapulse__` trivially serialisable.
+  - `order_id` defaults to declaration order (1-based), matching the XML format.
+  - `is_chunk` implies `is_primary_key` (the chunk attribute is part of the
+    logical key) — enforced here at decoration time.
+"""
+from __future__ import annotations
+
+from typing import Any, Callable
+
+from .types import MODE_TYPES
+
+# `core` holds the product's journals and helpers; they live in the same
+# database as the data, so a dataset must not collide with them.
+RESERVED_SCHEMAS = frozenset({"public", "core", "pg_catalog", "information_schema"})
+
+
+def attr(
+    code: str,
+    data_type: str,
+    *,
+    descr: str | None = None,
+    is_primary_key: bool = False,
+    is_chunk: bool = False,
+    order_id: int | None = None,
+) -> dict:
+    """Declare one attribute (column) of a spec.
+
+    `code`      — column name (PostgreSQL identifier)
+    `data_type` — one of datapulse.types.* (or a matching string)
+    `descr`     — human description (defaults to `code`)
+    `is_primary_key` — part of the logical key
+    `is_chunk`       — the chunking attribute; there must be exactly one per
+                       spec, and it must be part of the key
+    `order_id`       — physical column order (defaults to declaration order)
+    """
+    return {
+        "code": code,
+        "descr": descr or code,
+        "data_type": data_type,
+        "is_primary_key": bool(is_primary_key),
+        "is_chunk": bool(is_chunk),
+        "order_id": order_id,  # filled in by @spec if None
+    }
+
+
+def mode(code: str, type_code: str, is_clear: bool) -> dict:
+    """Declare one build mode of a spec.
+
+    `code`      — mode name (free-form; `initial`/`increment` are conventional)
+    `type_code` — one of init/incr/sect/appd/skip (datapulse.types.MODE_TYPES)
+    `is_clear`  — does a build in this mode consume EVERY source change up to its
+                  snapshot? Required, never inferred: the next incremental build
+                  starts its delta at the last clear build's snapshot, so calling
+                  a partial mode clear silently loses whatever changed in between.
+                  `initial`/`increment` are normally True; a mode that reloads a
+                  section (`sect`) or appends (`appd`) is normally False.
+    """
+    if type_code not in MODE_TYPES:
+        raise ValueError(
+            f"mode type_code must be one of {MODE_TYPES}, got {type_code!r}"
+        )
+    if not isinstance(is_clear, bool):
+        raise ValueError(
+            f"mode {code!r}: is_clear must be True or False (declare it explicitly)"
+        )
+    return {"mode_code": code, "type_code": type_code, "is_clear": is_clear}
+
+
+def spec(
+    *,
+    dataset: str,
+    num: int,
+    descr: str = "",
+    parallel: int = 1,
+    attrs: list[dict],
+    modes: list[dict] | None = None,
+    sources: list[str] | None = None,
+) -> Callable[[Callable], Callable]:
+    """Decorate `def build(param): ...` with the spec's metadata.
+
+    The metadata is attached as `build.__datapulse__` and the function is returned
+    unchanged, so the same function is what the engine calls in production and
+    what the developer may call from a scratch script or a test.
+
+    `dataset`  — qualified 'schema.name' (the data DB schema + table prefix)
+    `num`      — spec number within the dataset (1, 2, ... for parallel runs)
+    `parallel` — worker parallelism across chunks
+    `attrs`    — list of attr(...) dicts
+    `modes`    — list of mode(...) dicts
+    `sources`  — list of dataset codes this spec reads from
+    """
+    if num < 1:
+        raise ValueError(f"spec num must be >= 1, got {num}")
+    if parallel < 1:
+        raise ValueError(f"parallel must be >= 1, got {parallel}")
+    if not attrs:
+        raise ValueError("a spec must declare at least one attribute")
+    if "." not in dataset or dataset.split(".", 1)[0].lower() in RESERVED_SCHEMAS:
+        raise ValueError(
+            f"dataset must be a qualified 'schema.name' and not under "
+            f"{'/'.join(sorted(RESERVED_SCHEMAS))}: {dataset!r}"
+        )
+    if dataset.split(".", 1)[1][-1:].isdigit():
+        # The spec's full name is <dataset>_<num> (like its physical table);
+        # a digit-ending dataset name would make that suffix ambiguous.
+        raise ValueError(
+            f"the dataset name cannot end with a digit: {dataset!r} "
+            "(the _<num> suffix is reserved for the spec number)"
+        )
+
+    # Fill order_id by declaration order; enforce chunk⇒key.
+    prepared_attrs: list[dict] = []
+    for i, a in enumerate(attrs, start=1):
+        a = dict(a)  # don't mutate the caller's dict
+        if a.get("order_id") is None:
+            a["order_id"] = i
+        if a.get("is_chunk") and not a.get("is_primary_key"):
+            raise ValueError(
+                f"attribute {a['code']!r}: is_chunk implies is_primary_key (the chunk "
+                "attribute must be part of the logical key)"
+            )
+        prepared_attrs.append(a)
+
+    payload = {
+        "dataset_code": dataset,
+        "build_spec_num": int(num),
+        "descr": descr or "",
+        "parallel_cnt": int(parallel),
+        "attrs": prepared_attrs,
+        "modes": list(modes or []),
+        "sources": list(sources or []),
+    }
+
+    def decorate(build_fn: Callable[..., Any]) -> Callable[..., Any]:
+        build_fn.__datapulse__ = payload  # type: ignore[attr-defined]
+        return build_fn
+
+    return decorate

votinvv_datapulse-0.0.1/datapulse/dev.py

@@ -0,0 +1,50 @@
+"""`run_local` — run a spec's `build(param)` in the IDE, no server.
+
+The spec file itself stays clean (no executable tail); the developer calls
+run_local from a scratch script, a REPL or a test:
+
+    from datapulse.dev import run_local
+    run_local(build, mode_code="increment")
+
+`run_local` builds a local `BuildParams`, calls `build(param)`, and prints a
+one-line summary. The mapping talks to the data DB itself (via `Self`), so a
+local run needs `DATAPULSE_SECRETS_FILE` pointing at a store whose `Self`
+section reaches the developer's local data DB. There is nothing to collect and
+show anymore — the mapping wrote its slice into the interface table directly.
+"""
+from __future__ import annotations
+
+from typing import Any, Callable
+
+from .params import BuildParams
+
+
+def run_local(
+    build_fn: Callable[..., Any],
+    *,
+    mode_code: str = "initial",
+    from_build_id: int = 0,
+    to_build_id: int | None = None,
+    previous_time: Any = None,
+    keys: str | None = None,
+    parallel_cnt: int | None = None,
+) -> BuildParams:
+    """Run `build_fn` with a local `BuildParams`, print a summary, return it.
+
+    `parallel_cnt` defaults to whatever the spec declares, so a local run sees
+    the same value the engine would pass.
+    """
+    meta = getattr(build_fn, "__datapulse__", None) or {}
+    param = BuildParams(
+        mode_code=mode_code,
+        from_build_id=from_build_id,
+        to_build_id=to_build_id,
+        previous_time=previous_time,
+        keys=keys,
+        parallel_cnt=parallel_cnt if parallel_cnt is not None
+        else int(meta.get("parallel_cnt", 1)),
+    )
+    build_fn(param)
+    who = f"{meta['dataset_code']} #{meta['build_spec_num']}" if meta else build_fn.__name__
+    print(f"[run_local] {who}  mode={mode_code}  — build() returned")
+    return param

votinvv_datapulse-0.0.1/datapulse/flow.py

@@ -0,0 +1,65 @@
+"""`@flowspec` — flow templates the data engineer authors in `flowspec.py`.
+
+A flowspec is a plain function that receives the spec CATALOG and returns the
+launch plan — a list of `(spec_name, mode_code)` pairs. It lives in one
+`flowspec.py` (a settings object, applied through the same upgrade transport as
+specs and connect.py) and shows up as a launch button in the UI:
+
+    from datapulse import flowspec
+
+    @flowspec(descr="Daily increment of the whole warehouse")
+    def daily(catalog):
+        return [(s.name, "increment") for s in catalog if "increment" in s.modes]
+
+    @flowspec(descr="Micro-batch carousel", repeat=True)
+    def carousel(catalog):
+        return [(s.name, "increment") for s in catalog if "increment" in s.modes]
+
+Why code and not a declarative list: "every spec that has an increment mode" is
+a QUERY against the catalog, not a list — it stays current as specs come and
+go. The engine re-evaluates the flowspec on every launch (and on every repeat
+iteration), so a freshly applied spec joins the next run by itself.
+
+`repeat=True` relaunches the flow as user `system` each time an iteration
+completes fully — a quasi-online micro-batch pipeline. An iteration completes
+when EVERY build reached done; a failed build holds its dependent subtree until
+support marks it fixed (the flow waits — see the server docs), so a repeat
+carousel never spins over unresolved breakage.
+
+The launch screen's `keys` value is passed by the ENGINE into every build of
+the flow (`param.keys`); the flowspec function does not see or handle it.
+"""
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any, Callable
+
+
+@dataclass(frozen=True)
+class CatalogSpec:
+    """One installed spec, as a flowspec's `catalog` argument presents it."""
+
+    name: str                  # full spec name: 'raw.customer_1'
+    dataset_code: str          # 'raw.customer'
+    num: int                   # spec number within the dataset
+    modes: tuple[str, ...]     # declared mode codes, e.g. ('increment', 'initial')
+    sources: tuple[str, ...]   # dataset codes this spec reads from
+    descr: str = ""
+
+
+def flowspec(*, descr: str = "", repeat: bool = False) -> Callable[[Callable], Callable]:
+    """Declare a flow template. The function name is the flowspec's code (what
+    the UI button and the journal show); the function must accept the catalog
+    (a list of CatalogSpec) and return an iterable of (spec_name, mode_code)
+    pairs. The metadata rides on the function as `__datapulse_flowspec__` — the
+    same pattern as `@spec`."""
+
+    def decorate(fn: Callable[..., Any]) -> Callable[..., Any]:
+        fn.__datapulse_flowspec__ = {  # type: ignore[attr-defined]
+            "code": fn.__name__,
+            "descr": descr or "",
+            "repeat": bool(repeat),
+        }
+        return fn
+
+    return decorate

votinvv_datapulse-0.0.1/datapulse/params.py

@@ -0,0 +1,68 @@
+"""BuildParams — the parameter set of one build, handed by the engine to
+`build(param)`.
+
+This used to be a `BuildContext` with `read`/`write`/`connection` methods. The
+model changed: the mapping fills its interface table ITSELF — through the `Self`
+context (see `sources.py`) with plain SQL — and the engine then compares and
+merges it in (SCD2). So `build` receives only the recompute PARAMETERS, no
+data-access methods.
+
+The mental model is a PL/pgSQL load procedure: the core calls a function with a
+fixed parameter set. Here the set is packed into one `param` object (instead of
+a long signature) — adding a field does not break existing mappings, and
+`param.` autocompletes in the IDE.
+
+Fields:
+  mode_code      — the NAME of the launch mode (`initial` / `increment` /
+                   `manual` / ...); the mapping branches its logic on it. The
+                   mode's TYPE (init/incr/sect/appd/skip) is the engine's
+                   business (the interface → spec-table merge) and is not part
+                   of the parameters.
+  from_build_id  — the delta's lower bound (EXCLUSIVE): the source snapshot of
+                   this spec's last CLEAR successful build. 0 on the first build
+                   — so `build_id > 0` reads the sources whole. Never None: it
+                   is inlined into an f-string.
+  to_build_id    — the delta's upper bound and THIS build's source snapshot
+                   (INCLUSIVE). The engine freezes the sources at it: if a
+                   source build was still running when this build was created,
+                   it is `min(its build_id) - 1`, otherwise the build's own id.
+  previous_time  — when the previous clear successful build finished (a
+                   time-based watermark: `where updated_at >=
+                   param.previous_time`), or None.
+  keys           — "recompute keys": the raw input from the build launch screen.
+                   Whether to use them is the mapping's decision (the typical
+                   case is a pinpoint recompute). None when nothing was entered.
+  parallel_cnt   — the spec's parallelism. The engine slices the merge into that
+                   many chunks itself; for the mapping it is a hint — a heavy
+                   statement can be wrapped in
+                   `call core.parallel(param.parallel_cnt)`.
+
+A source's delta is read straight off its dataset view (all columns, all rows):
+
+    from raw.loan_contract
+    where build_id > param.from_build_id and build_id <= param.to_build_id
+
+A source's as-of snapshot — same `to_build_id`, via the SCD2 predicate
+(`end_build_id` is INCLUSIVE — the last build in which the version is still
+current):
+
+    where build_id <= param.to_build_id
+      and (end_build_id is null or end_build_id >= param.to_build_id)
+      and is_active
+"""
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Any
+
+
+@dataclass
+class BuildParams:
+    """The static parameter set the engine hands to `build(param)`."""
+
+    mode_code: str = "initial"
+    from_build_id: int = 0
+    to_build_id: int | None = None
+    previous_time: Any = None  # datetime/str — the engine fills it; None on the first build
+    keys: str | None = None
+    parallel_cnt: int = 1

votinvv_datapulse-0.0.1/datapulse/secrets.py

@@ -0,0 +1,63 @@
+"""`secret(context, key)` — fetches a context's secret value from the store.
+
+Secrets live OUTSIDE the code and OUTSIDE git: the context file (`context.py`)
+carries only the connection CODE, while the values (passwords, DSNs) sit in a
+separate store. Today the store is a plain JSON file, its path in the
+`DATAPULSE_SECRETS_FILE` env var:
+
+    { "RBS":  {"dsn": "...", "user": "...", "password": "..."},
+      "Self": {"host": "...", "port": 5433, "user": "...",
+               "password": "...", "dbname": "datapulse"} }
+
+The "context → set of fields" model mirrors Vault (path → fields), so the store
+backend can later be swapped for Vault without touching `secret()` call sites.
+The file is read on every access (not a hot path — only when a connection is
+opened), so a secrets edit is picked up without a restart.
+"""
+from __future__ import annotations
+
+import json
+import os
+from pathlib import Path
+
+ENV_SECRETS_FILE = "DATAPULSE_SECRETS_FILE"
+
+
+class SecretError(RuntimeError):
+    """A secret is missing or the store is unreadable/misconfigured."""
+
+
+def _store() -> dict:
+    path = os.environ.get(ENV_SECRETS_FILE)
+    if not path:
+        raise SecretError(
+            f"{ENV_SECRETS_FILE} is not set — no secrets store configured."
+        )
+    p = Path(path)
+    if not p.is_file():
+        raise SecretError(f"secrets file not found: {p}")
+    try:
+        data = json.loads(p.read_text(encoding="utf-8"))
+    except Exception as e:  # noqa: BLE001
+        raise SecretError(f"cannot read secrets file {p}: {e}")
+    if not isinstance(data, dict):
+        raise SecretError(
+            f"secrets file {p} must be a JSON object {{context: {{key: value}}}}."
+        )
+    return data
+
+
+def secret_section(context: str) -> dict:
+    """All secrets of one context as a dict. Raises SecretError if absent."""
+    section = _store().get(context)
+    if not isinstance(section, dict):
+        raise SecretError(f"no secrets for context {context!r} in the store.")
+    return dict(section)
+
+
+def secret(context: str, key: str):
+    """One secret value. Raises SecretError with a clear message if not set."""
+    section = secret_section(context)
+    if key not in section:
+        raise SecretError(f"secret {context}.{key} is not set.")
+    return section[key]

votinvv_datapulse-0.0.1/datapulse/sources.py

@@ -0,0 +1,60 @@
+"""Built-in contexts shipped by the product.
+
+`Self` — the connection to the project's DATA database. A mapping uses it to
+fill its own interface table with plain SQL:
+
+    from datapulse import Self
+
+    def build(param):
+        with Self() as db:
+            db.execute(f'''
+                insert into raw.customer_1_$ (customer_id, full_name, updated_at)
+                select customer_id, full_name, updated_at
+                from staging.customer
+                where updated_at >= '{param.previous_time}'
+            ''')
+
+The interface-table name (`schema.name_<num>_$`) is written by hand on purpose —
+the engineer copy-pastes the SELECT from DataGrip, no magic name substitution.
+The engine takes over after the interface table is filled: it compares and
+merges into the SCD2 spec table.
+
+`Self`'s credentials come from the `Self` section of the secrets store
+(see `secrets.py`). psycopg is imported LAZILY — the SDK carries no driver
+dependency; the product image ships psycopg, and a developer installs it only if
+they run a spec locally. `with Self() as db` yields a psycopg connection (it has
+`.execute()`, `.cursor()`, `.commit()`); the connection commits on a clean exit
+and rolls back on an exception.
+
+User contexts (Oracle / S3 / Kafka / ...) are NOT here — they live in the
+developer-authored `context.py` as `@contextmanager` functions. Only `Self` is
+built in, because only the product knows it always targets Postgres.
+"""
+from __future__ import annotations
+
+from contextlib import contextmanager
+
+from .secrets import secret_section
+
+
+@contextmanager
+def Self():
+    """Open a connection to the project's data DB (yields a psycopg connection)."""
+    import psycopg  # lazy: the driver is provided by the image / dev environment
+
+    p = secret_section("Self")
+    conn = psycopg.connect(
+        host=p["host"],
+        port=int(p.get("port", 5432)),
+        dbname=p["dbname"],
+        user=p["user"],
+        password=p.get("password", ""),
+    )
+    try:
+        yield conn
+        conn.commit()
+    except BaseException:
+        conn.rollback()
+        raise
+    finally:
+        conn.close()

votinvv_datapulse-0.0.1/datapulse/types.py

@@ -0,0 +1,58 @@
+"""Type aliases for spec attributes.
+
+These are plain strings kept in one place so:
+  - the IDE autocompletes `bigint`, `text`, ... (no typos in `data_type`)
+  - the server's whitelist matches exactly what the SDK advertises
+
+The server re-exports the same whitelist from its own validation module so the
+two never drift apart.
+"""
+from __future__ import annotations
+
+# PostgreSQL types a spec attribute may use. Matches the XSD `columnType`
+# whitelist the previous XML format allowed (so existing specs translate 1:1).
+bigint = "bigint"
+integer = "integer"
+smallint = "smallint"
+text = "text"
+boolean = "boolean"
+date = "date"
+timestamptz = "timestamptz"
+jsonb = "jsonb"
+uuid = "uuid"
+
+
+def varchar(n: int) -> str:
+    return f"varchar({n})"
+
+
+def numeric(precision: int, scale: int) -> str:
+    return f"numeric({precision},{scale})"
+
+
+# The full whitelist, as a regex pattern (the server validates against it too).
+ALLOWED_TYPES_PATTERN = (
+    r"^(bigint|integer|smallint|text|boolean|date|timestamptz|jsonb|uuid"
+    r"|varchar\(\d+\)|numeric\(\d+,\d+\))$"
+)
+
+# Reserved attribute names — the engine injects its own physical columns with
+# these names (SCD2), and the dataset view adds build_spec_number, so a spec
+# cannot declare them.
+RESERVED_ATTRS = frozenset({"build_id", "end_build_id", "is_active", "build_spec_number"})
+
+# Build-mode types — how the engine moves the interface table into the spec table:
+#   init — full reload: rows absent from the slice are closed (tombstoned)
+#   incr — increment: only the keys present in the slice are touched
+#   sect — section reload: like init, but scoped to the chunk values in the slice
+#   appd — append only: rows are inserted, nothing is compared or closed
+#   skip — do nothing (the mapping is not even called); counters journal as 0
+MODE_TYPES = ("init", "incr", "sect", "appd", "skip")
+
+# A convenience alias used in `attr(..., data_type=DataType.bigint)`.
+DataType = type("DataType", (), {
+    "bigint": bigint, "integer": integer, "smallint": smallint,
+    "text": text, "boolean": boolean, "date": date,
+    "timestamptz": timestamptz, "jsonb": jsonb, "uuid": uuid,
+    "varchar": staticmethod(varchar), "numeric": staticmethod(numeric),
+})

votinvv_datapulse-0.0.1/pyproject.toml

@@ -0,0 +1,48 @@
+[build-system]
+requires = ["setuptools>=77.0"]
+build-backend = "setuptools.build_meta"
+
+[project]
+# Distribution name on PyPI is votinvv-datapulse (plain "datapulse" is taken
+# by an unrelated package, and PyPI rejects near-identical names like
+# data-pulse); the import package stays `datapulse`.
+name = "votinvv-datapulse"
+dynamic = ["version"]
+description = "DataPulse SDK — decorators and runtime context for spec authors"
+readme = "README.md"
+license = "Apache-2.0"
+requires-python = ">=3.11"
+keywords = ["data-engineering", "pipelines", "scd2", "postgresql", "etl"]
+classifiers = [
+    "Development Status :: 3 - Alpha",
+    "Intended Audience :: Developers",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+    "Programming Language :: Python :: 3.14",
+    "Topic :: Database",
+    "Topic :: Software Development :: Libraries",
+]
+# No third-party runtime deps: the SDK is pure stdlib so a spec author can
+# `pip install votinvv-datapulse` and `from datapulse import spec` with nothing else.
+# Drivers import lazily (psycopg inside Self) — pull them via extras when the
+# mapping runs locally.
+
+[project.optional-dependencies]
+postgres = ["psycopg[binary]>=3.2"]
+oracle = ["oracledb>=3.0"]
+
+[project.urls]
+Homepage = "https://github.com/votinvv/datapulse"
+Repository = "https://github.com/votinvv/datapulse"
+Issues = "https://github.com/votinvv/datapulse/issues"
+
+[tool.setuptools.dynamic]
+version = { attr = "datapulse.__version__" }
+
+[tool.setuptools.packages.find]
+where = ["."]
+# Only ship the SDK package (datapulse/). The build root is sdk/, which contains
+# only the datapulse package + this pyproject.toml — nothing else to exclude.
+include = ["datapulse*"]

votinvv_datapulse-0.0.1/setup.cfg

@@ -0,0 +1,4 @@
+[egg_info]
+tag_build = 
+tag_date = 0
+

votinvv_datapulse-0.0.1/votinvv_datapulse.egg-info/PKG-INFO

@@ -0,0 +1,77 @@
+Metadata-Version: 2.4
+Name: votinvv-datapulse
+Version: 0.0.1
+Summary: DataPulse SDK — decorators and runtime context for spec authors
+License-Expression: Apache-2.0
+Project-URL: Homepage, https://github.com/votinvv/datapulse
+Project-URL: Repository, https://github.com/votinvv/datapulse
+Project-URL: Issues, https://github.com/votinvv/datapulse/issues
+Keywords: data-engineering,pipelines,scd2,postgresql,etl
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Classifier: Programming Language :: Python :: 3.14
+Classifier: Topic :: Database
+Classifier: Topic :: Software Development :: Libraries
+Requires-Python: >=3.11
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Provides-Extra: postgres
+Requires-Dist: psycopg[binary]>=3.2; extra == "postgres"
+Provides-Extra: oracle
+Requires-Dist: oracledb>=3.0; extra == "oracle"
+Dynamic: license-file
+
+# DataPulse SDK
+
+The developer-facing library of [DataPulse](https://github.com/votinvv/datapulse) —
+declarative data pipelines over PostgreSQL with SCD2 versioning, parallel
+builds and a full audit trail.
+
+Spec authors install this package for IDE autocompletion and local runs:
+
+```
+pip install votinvv-datapulse
+```
+
+The distribution is named `votinvv-datapulse`; the import package is `datapulse`:
+
+```python
+from datapulse import spec, attr, mode, Self
+from datapulse.types import bigint, text, timestamptz
+
+
+@spec(
+    dataset="raw.customer",
+    num=1,
+    descr="Customers",
+    attrs=[
+        attr("customer_id", bigint, descr="Customer id", is_primary_key=True),
+        attr("full_name", text, descr="Full name"),
+    ],
+    modes=[mode("initial", "init", True), mode("increment", "incr", True)],
+)
+def build(param):
+    with Self() as db:
+        db.execute("insert into raw.customer_1_$ (...) select ...")
+```
+
+The SDK is pure stdlib — database drivers import lazily. To execute mappings
+locally (e.g. via `datapulse.dev.run_local`), pull the drivers you need:
+
+```
+pip install "votinvv-datapulse[postgres]"   # psycopg
+pip install "votinvv-datapulse[oracle]"     # oracledb
+```
+
+The server side — engine, HTTP API and UI — ships as the
+[`votinvv/datapulse`](https://hub.docker.com/r/votinvv/datapulse) Docker
+image; see the [project README](https://github.com/votinvv/datapulse) for the
+full picture.
+
+## License
+
+[Apache-2.0](https://github.com/votinvv/datapulse/blob/main/LICENSE)

votinvv_datapulse-0.0.1/votinvv_datapulse.egg-info/SOURCES.txt

@@ -0,0 +1,16 @@
+LICENSE
+README.md
+pyproject.toml
+datapulse/__init__.py
+datapulse/decorators.py
+datapulse/dev.py
+datapulse/flow.py
+datapulse/params.py
+datapulse/secrets.py
+datapulse/sources.py
+datapulse/types.py
+votinvv_datapulse.egg-info/PKG-INFO
+votinvv_datapulse.egg-info/SOURCES.txt
+votinvv_datapulse.egg-info/dependency_links.txt
+votinvv_datapulse.egg-info/requires.txt
+votinvv_datapulse.egg-info/top_level.txt

votinvv_datapulse-0.0.1/votinvv_datapulse.egg-info/dependency_links.txt

@@ -0,0 +1 @@
+

votinvv_datapulse-0.0.1/votinvv_datapulse.egg-info/requires.txt

@@ -0,0 +1,6 @@
+
+[oracle]
+oracledb>=3.0
+
+[postgres]
+psycopg[binary]>=3.2

votinvv_datapulse-0.0.1/votinvv_datapulse.egg-info/top_level.txt

@@ -0,0 +1 @@
+datapulse