statcatlib 0.12.0__tar.gz → 0.12.1__tar.gz

{statcatlib-0.12.0 → statcatlib-0.12.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: statcatlib
-Version: 0.12.0
+Version: 0.12.1
 Summary: Queryable open data catalog engine for DCAT-AP, StatDCAT-AP, CKAN, and SDMX
 Keywords: sdmx,dcat,open-data,catalog,duckdb,statistical-data,ckan
 Author: gabrielgellner
@@ -18,7 +18,7 @@ Classifier: Topic :: Internet :: WWW/HTTP :: Indexing/Search
 Requires-Dist: attrs>=26.1.0
 Requires-Dist: duckdb>=2.0.0
 Requires-Dist: httpx>=0.28
-Requires-Dist: sdmxlib>=0.81.0,<0.82
+Requires-Dist: sdmxlib>=0.82.0,<0.83
 Requires-Dist: lancedb>=0.39 ; extra == 'search'
 Requires-Dist: sentence-transformers>=3.0 ; extra == 'search'
 Requires-Dist: typer>=0.15 ; extra == 'search'

{statcatlib-0.12.0 → statcatlib-0.12.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "statcatlib"
-version = "0.12.0"
+version = "0.12.1"
 description = "Queryable open data catalog engine for DCAT-AP, StatDCAT-AP, CKAN, and SDMX"
 license = "Apache-2.0"
 license-files = ["LICENSE"]
@@ -29,7 +29,7 @@ dependencies = [
     "attrs>=26.1.0",
     "duckdb>=2.0.0",
     "httpx>=0.28",
-    "sdmxlib>=0.81.0,<0.82",
+    "sdmxlib>=0.82.0,<0.83",
 ]
 
 [[project.authors]]

{statcatlib-0.12.0 → statcatlib-0.12.1}/pyproject.toml.orig

@@ -1,6 +1,6 @@
 [project]
 name = "statcatlib"
-version = "0.12.0"
+version = "0.12.1"
 description = "Queryable open data catalog engine for DCAT-AP, StatDCAT-AP, CKAN, and SDMX"
 license = "Apache-2.0"
 license-files = ["LICENSE"]
@@ -27,9 +27,9 @@ dependencies = [
     # by DuckDB 1.x, so by statcatlib 0.8 or earlier.
     "duckdb>=2.0.0",
     "httpx>=0.28",
-    # <0.82: every suite library is pre-1.0, so a minor bump may break
+    # <0.83: every suite library is pre-1.0, so a minor bump may break
     # (GH-67).
-    "sdmxlib>=0.81.0,<0.82",
+    "sdmxlib>=0.82.0,<0.83",
 ]
 
 [project.optional-dependencies]

{statcatlib-0.12.0 → statcatlib-0.12.1}/src/statcatlib/EGM.md

@@ -275,9 +275,11 @@ read.
   lands on: a rollback whose `if_live` is no longer live, a release that is
   no longer a fast-forward, and a seal record computed before the embargo was
   turned on are each refused or recomputed.
-- `Product.view` hashes the live edition's `manifest.json` and raises
-  `IncompatibleCatalogError` when it differs from the `manifest_sha256` the
-  Inventory records. Every view a `Catalog` returns shares one
+- `Product.view` passes the Inventory's `manifest_sha256` to
+  `sdmxlib.LocalRegistry.edition`, which hashes the bytes of
+  `manifest.json` it parses. A missing, malformed or different manifest
+  raises sdmxlib's `ManifestError`, which `view` re-raises as
+  `IncompatibleCatalogError` naming `statcat verify-catalog`, chained. Every view a `Catalog` returns shares one
   `sdmxlib.EditionEngine`, and `Catalog.close` closes them all.
 - A seal check only reads: `seal` runs every check against `wc.view()`
   before writing anything, and any finding raises `SealCheckError`.
@@ -300,7 +302,7 @@ read.
   at INFO, and writes and removes a `.statcat-link-probe.*` file and its hard
   link.
 - `Product.view` attaches the edition to the catalog's engine
-  (`LocalRegistry.edition`). `Product.working_copy` writes as
+  (`LocalRegistry.edition(path, engine=, manifest_sha256=)`). `Product.working_copy` writes as
   `store/EGM.md` describes.
 - `synthetic_catalog` writes a whole catalog: `catalog.json`, sdmxlib's
   synthetic edition, the catalog rows, the search index when lancedb is

{statcatlib-0.12.0 → statcatlib-0.12.1}/src/statcatlib/catalog.py

@@ -39,7 +39,7 @@ from sdmxlib.inventory import CATALOG_NAME, LiveEdition, LiveInventory, decode_r
 from statcatlib.bundle import description_from_root
 from statcatlib.store.inventory import Change, InventoryContentionError, InventoryRuleError, InventoryWriter
 from statcatlib.store.store import VIEW_ATTRIBUTES, EditionView, IncompatibleCatalogError, _Store
-from statcatlib.store.working_copy import Edition, SealCheckError, WorkingCopy, manifest_sha256
+from statcatlib.store.working_copy import Edition, SealCheckError, WorkingCopy
 
 if TYPE_CHECKING:
     from collections.abc import Callable, Iterable, Mapping
@@ -618,10 +618,11 @@ class Product:
     def view(self, *, embedding_model: EmbeddingModel | None = None, query_prefix: str = "") -> EditionView:
         """A read-only view of the product's live edition, as of the catalog's last `Catalog.refresh`.
 
-        It checks the edition's ``manifest.json`` against the SHA-256 the
-        Inventory records, then attaches the edition to the catalog's engine
-        (``LocalRegistry.edition``), which checks the manifest's store schema.
-        The view stays on this edition after a refresh or a rollback.
+        It attaches the edition to the catalog's engine with
+        ``LocalRegistry.edition``, which reads ``manifest.json`` once, checks
+        its SHA-256 against the one the Inventory records, and checks its
+        store schema. The view stays on this edition after a refresh or a
+        rollback.
 
         Args:
             embedding_model: The encoder the index was built with, for vector
@@ -633,27 +634,26 @@ class Product:
             The view.
 
         Raises:
-            IncompatibleCatalogError: The edition's ``manifest.json`` is not
-                the one the Inventory recorded. Its ``statcat`` schema at
+            IncompatibleCatalogError: The edition's ``manifest.json`` is
+                missing, unreadable, or not the one the Inventory recorded
+                (sdmxlib's `ManifestError`, chained). Its ``statcat`` schema at
                 another version, or a search index this build does not read,
                 raise it too, and a product with no live edition raises
                 `NoLiveEditionError`.
         """
         live = self._head()
         edition_dir = self._dir / live.edition_id
+        cat = self._catalog
         try:
-            found = manifest_sha256(edition_dir)
-        except FileNotFoundError:
-            found = "missing"
-        if found != live.manifest_sha256:
+            registry = sdmxlib.LocalRegistry.edition(
+                edition_dir, engine=cat._engine_for_views(), manifest_sha256=live.manifest_sha256
+            )
+        except ManifestError as exc:
             msg = (
-                f"{self._product_id}'s live edition {live.edition_id} has a manifest.json ({found}) that is not "
-                f"the one the Inventory recorded ({live.manifest_sha256}); run `statcat verify-catalog` on "
-                f"{self._catalog._root}"
+                f"{self._product_id}'s live edition {live.edition_id} does not hold the manifest.json the Inventory "
+                f"recorded ({exc}); run `statcat verify-catalog` on {cat._root}"
             )
-            raise IncompatibleCatalogError(msg)
-        cat = self._catalog
-        registry = sdmxlib.LocalRegistry.edition(edition_dir, engine=cat._engine_for_views())
+            raise IncompatibleCatalogError(msg) from exc
         opened = False
         try:
             store = _Store._over(

{statcatlib-0.12.0 → statcatlib-0.12.1}/src/statcatlib/store/EGM.md

@@ -238,10 +238,11 @@ constants, `IncompatibleCatalogError`, `IncompatibleIndexError`, `Row` and
   `ManifestError` from `sdmxlib.edition.seal` comes before the journal entry;
   the working copy stays closed, and `working_copy(resume=True)` reopens it.
 - `InventoryWriter` is the only code that writes `catalog.json`,
-  `inventory/` and `CURRENT`. It creates each file in five steps (staged
-  file, fsync, `os.link` to the name, unlink the staged file, fsync the
-  directory), so a name never holds half a file and of two writers exactly
-  one creates a root; a leaf that exists already counts as written, since
+  `inventory/` and `CURRENT`. It creates each file with sdmxlib's
+  `sdmxlib.storage.create_exclusive` (staged file, fsync, `os.link` to the
+  name, unlink the staged file, fsync the directory), so a name never holds
+  half a file and of two writers exactly one creates a root: the loser's
+  `FileExistsError` becomes `InventoryConflictError`; a leaf that exists already counts as written, since
   its name is the hash of its bytes. It writes leaves, then the root, then
   `CURRENT`, and replaces `CURRENT` only with a higher `seq` in the same
   `restore_epoch`. `commit` re-reads the head and calls *change* again after
@@ -276,9 +277,10 @@ constants, `IncompatibleCatalogError`, `IncompatibleIndexError`, `Row` and
   unsealed edition directory.
 - `settle_journal` can commit an Inventory version and deletes journal
   entries.
-- `InventoryWriter` writes `catalog.json`, leaves under `inventory/leaves/`,
-  roots under `inventory/`, and `CURRENT`, each through a staged
-  `.<name>.<pid>.<uuid>.tmp` file, and fsyncs files and directories.
+- `InventoryWriter` writes `catalog.json`, leaves under `inventory/leaves/`
+  and roots under `inventory/` through `sdmxlib.storage.create_exclusive`,
+  and `CURRENT` through its own staged `.<name>.<pid>.<uuid>.tmp` file and
+  rename; each file and directory is fsynced.
 - `discard_orphans()` drops LanceDB tables the manifest does not name.
 - Emits `warnings.warn` for a quantised (stale) vector index and for changed search annotations, and logs a WARNING for an edition the journal rule cannot record.
 

{statcatlib-0.12.0 → statcatlib-0.12.1}/src/statcatlib/store/inventory.py

@@ -5,9 +5,10 @@ transition rules (`sdmxlib.inventory`). statcatlib writes them, and this module
 is the only code that writes ``catalog.json``, ``inventory/`` and ``CURRENT``.
 It adds three things to sdmxlib's helpers:
 
-- **exclusive create**: a file is staged, fsynced, hard-linked to its name and
-  the directory fsynced, so a name never holds half a file, and two writers
-  never both create one root;
+- **exclusive create**: each document is written by
+  `sdmxlib.storage.create_exclusive`, which stages the file, fsyncs it,
+  hard-links it to its name and fsyncs the directory, so a name never holds
+  half a file, and two writers never both create one root;
 - **the ``CURRENT`` advance**: ``CURRENT`` is replaced only by a higher
   ``seq``, so a slow writer never moves it back;
 - **the retry loop**: `InventoryWriter.commit` re-reads the head and re-applies
@@ -53,12 +54,13 @@ from sdmxlib.inventory import (
     sha256_hex,
     validate_root,
 )
+from sdmxlib.storage import create_exclusive
 
 if TYPE_CHECKING:
     from collections.abc import Callable, Mapping, Sequence
 
-    from sdmxlib.inventory._rules import Rule
-    from sdmxlib.storage._manifest import JsonValue
+    from sdmxlib.inventory import Rule
+    from sdmxlib.storage import JsonValue
 
 __all__ = [
     "MAX_RACES",
@@ -237,32 +239,12 @@ class InventoryWriter:
 
     # -- writing ---------------------------------------------------------
 
-    def _create_exclusive(self, path: Path, data: bytes) -> None:
-        """Create *path* holding *data*, complete before it is visible.
-
-        Five steps: write a staged file named ``.<name>.<pid>.<uuid>.tmp``,
-        fsync it, ``os.link`` it to *path*, unlink the staged file, and fsync
-        the directory. The link raises `FileExistsError` when *path* exists,
-        so of two writers exactly one creates it.
-
-        Args:
-            path: The file to create.
-            data: Its bytes.
-        """
-        staged = _stage(path, data)
-        try:
-            os.link(staged, path)
-        finally:
-            with contextlib.suppress(FileNotFoundError):
-                staged.unlink()
-        _fsync_dir(path.parent)
-
     def _write_leaf(self, path: Path, data: bytes) -> None:
         """Create a leaf. One already there is the same leaf: its name is the hash of its bytes."""
         if path.exists():
             return
         with contextlib.suppress(FileExistsError):
-            self._create_exclusive(path, data)
+            create_exclusive(path, data)
 
     def _advance_current(self, root: Root, data: bytes) -> None:
         """Point ``CURRENT`` at *root*, unless it already names a version at least as new.
@@ -296,7 +278,7 @@ class InventoryWriter:
         document = CatalogDocument(
             format_version=CATALOG_FORMAT, catalog_id=catalog_id or str(uuid.uuid7()), restore_epoch=0
         )
-        self._create_exclusive(self._root / CATALOG_NAME, encode(document))
+        create_exclusive(self._root / CATALOG_NAME, encode(document))
         return document
 
     def try_commit(
@@ -350,7 +332,7 @@ class InventoryWriter:
         validate_root(root)
         data = encode(root)
         try:
-            self._create_exclusive(self._root / root_path(root.seq), data)
+            create_exclusive(self._root / root_path(root.seq), data)
         except FileExistsError as exc:
             msg = f"another writer created Inventory version {root.seq} first"
             raise InventoryConflictError(msg) from exc