tablassert 7.4.8__tar.gz → 7.4.10__tar.gz

{tablassert-7.4.8 → tablassert-7.4.10}/CHANGELOG.md

@@ -2,6 +2,16 @@
 
 All notable changes to this project are documented in this file.
 
+## 7.4.10 - 2026-05-29
+
+### Changes
+- Enforced explicit `biolink:` namespace prefix on predicates and qualifiers emitted by `compile_subgraph()` in `lib.py`. Both `self.statement.predicate` and `x.qualifier` (in the qualifier loop) are now prefixed via `add("biolink:", ...)`, ensuring all output edges carry fully-qualified Biolink CURIEs rather than bare predicate/qualifier names.
+
+## 7.4.9 - 2026-05-26
+
+### Bug Fixes
+- Fixed `OSError: Too many open files` during subgraph build at large scales (700+ sections). `with_mesh()` and `with_captions()` in `lib.py` opened a `sqlite_utils.Database` per section but never closed the underlying SQLite connection, leaving FD release to GC. In the tight sequential `compile_subgraph` loop the leaked FDs accumulated past the OS soft limit, causing the next `to_store()` → `df.write_parquet()` (which polars 1.39 routes through `sink_parquet`) to fail opening its target `.storassert/*.parquet`. Both functions now wrap their query bodies in `try:` / `finally: db.conn.close()`.
+
 ## 7.4.8 - 2026-05-12
 
 ### Changes

{tablassert-7.4.8 → tablassert-7.4.10}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tablassert
-Version: 7.4.8
+Version: 7.4.10
 Summary: Extract knowledge assertions from tabular data into NCATS Translator-compliant KGX NDJSON — declaratively, with entity resolution and quality control built in.
 Project-URL: Homepage, https://github.com/SkyeAv/Tablassert
 Project-URL: Source, https://github.com/SkyeAv/Tablassert

tablassert-7.4.10/docs/changelog.md

@@ -0,0 +1,11 @@
+# Changelog
+
+The canonical release history lives in the repository root at [`CHANGELOG.md`](https://github.com/SkyeAv/Tablassert/blob/main/CHANGELOG.md).
+
+## Current Release Notes
+
+## 7.4.10 - 2026-05-29
+
+### Changes
+
+- Enforced explicit `biolink:` namespace prefix on predicates and qualifiers emitted by `compile_subgraph()` in `lib.py`. Both `self.statement.predicate` and `x.qualifier` (in the qualifier loop) are now prefixed via `add("biolink:", ...)`, ensuring all output edges carry fully-qualified Biolink CURIEs rather than bare predicate/qualifier names.

{tablassert-7.4.8 → tablassert-7.4.10}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "tablassert"
-version = "7.4.8"
+version = "7.4.10"
 description = "Extract knowledge assertions from tabular data into NCATS Translator-compliant KGX NDJSON — declaratively, with entity resolution and quality control built in."
 authors = [
     { name = "Skye Lane Goetz", email = "sgoetz@isbscience.org" }

{tablassert-7.4.8 → tablassert-7.4.10}/src/tablassert/fullmap.py

@@ -150,15 +150,15 @@ def log_unmatched(
 ) -> None:
     # * Log Unmatched Entities
     level_one: pl.LazyFrame = terms.filter(pl.col("nlp level") == 1)
-    antimatches: pl.LazyFrame = level_one.join(matches.lazy().select("term"), left_on="term", right_on="term", how="anti")
+    antimatches: pl.LazyFrame = level_one.join(
+        matches.lazy().select("term"), left_on="term", right_on="term", how="anti"
+    )
 
     # ! Collection Point: Requires Eager
     unnmatched: pl.DataFrame = antimatches.select("term").unique().collect()
     if unnmatched.height > 0:
         for term in unnmatched.get_column("term").to_list():
-            logger.info(
-                f"FAILED | STORE: {section_hash} | CONFIG: {config_file} | COL: {col} | VALUE: {term!r}"
-            )
+            logger.info(f"FAILED | STORE: {section_hash} | CONFIG: {config_file} | COL: {col} | VALUE: {term!r}")
 
 
 def resolve(

{tablassert-7.4.8 → tablassert-7.4.10}/src/tablassert/lib.py

@@ -195,7 +195,8 @@ def with_mesh(lf: pl.LazyFrame, pubmed_db: Path, curie: str) -> pl.LazyFrame:
 
     df: pl.DataFrame = lf.collect()
     db: object = Database(pubmed_db)
-    query: str = """
+    try:
+        query: str = """
 SELECT
   mesh.mesh_major,
   mesh.mesh,
@@ -209,7 +210,9 @@ INNER JOIN info ON ids.pmid = info.pmid
 WHERE ids.alt = :curie OR ids.pmid = :curie
 LIMIT 1
 """
-    rows: list[dict[str, str]] = list(db.query(query, {"curie": curie})) or []
+        rows: list[dict[str, str]] = list(db.query(query, {"curie": curie})) or []
+    finally:
+        db.conn.close()  # pyright: ignore
     all_ids: list[str] = [add("MESH:", x["mesh"]) for x in rows if x]
     is_major: list[bool] = [eq(x["mesh_major"], "Y") for x in rows]
     domain: list[str] = [x for x, y in zip(all_ids, is_major) if y]
@@ -244,14 +247,17 @@ def with_captions(lf: pl.LazyFrame, pmc_db: Path, curie: str, url: str) -> pl.La
 
     df: pl.DataFrame = lf.collect()
     db: object = Database(pmc_db)
-    filename: str = basename(url)
-    query: str = """
+    try:
+        filename: str = basename(url)
+        query: str = """
 SELECT caption
 FROM captions
 WHERE pmc = :curie AND file = :filename
 LIMIT 1
 """
-    rows: list[dict[str, str]] = list(db.query(query, {"curie": curie, "filename": filename})) or []
+        rows: list[dict[str, str]] = list(db.query(query, {"curie": curie, "filename": filename})) or []
+    finally:
+        db.conn.close()  # pyright: ignore
     row: dict[str, str] = rows[0] if rows else {}
 
     caption: Optional[str] = row.get("caption")
@@ -336,8 +342,8 @@ class Tcode(Section):
                 [op for x in self.annotations for op in self.encoding(x, x.annotation)] if self.annotations else None,
                 self.node(self.statement.subject, "subject", conns),
                 self.node(self.statement.object, "object", conns),
-                (value, ("predicate", self.statement.predicate)),
-                [op for x in self.statement.qualifiers for op in self.node(x, x.qualifier, conns)]
+                (value, ("predicate", add("biolink:", self.statement.predicate))),
+                [op for x in self.statement.qualifiers for op in self.node(x, add("biolink:", x.qualifier), conns)]
                 if self.statement.qualifiers
                 else None,
                 (value, ("syntax", self.syntax)),

{tablassert-7.4.8 → tablassert-7.4.10}/uv.lock

@@ -2360,7 +2360,7 @@ wheels = [
 
 [[package]]
 name = "tablassert"
-version = "7.4.8"
+version = "7.4.10"
 source = { editable = "." }
 dependencies = [
     { name = "cyclopts" },

tablassert-7.4.8/docs/changelog.md

@@ -1,13 +0,0 @@
-# Changelog
-
-The canonical release history lives in the repository root at [`CHANGELOG.md`](https://github.com/SkyeAv/Tablassert/blob/main/CHANGELOG.md).
-
-## Current Release Notes
-
-## 7.4.8 - 2026-05-12
-
-### Changes
-
-- Expanded `fullmap_audit()` failure logging in `qc.py` so each rejected CURIE log line now carries the underlying `FUZZ_RATIO`, `FUZZ_PARTIAL`, and (when the BERT stage ran) `BERT_SIMILARITY` score values, making it easier to diagnose why a term was dropped during QC.
-
-For older releases and the full project history, open the root `CHANGELOG.md` in the repository.