tablassert 7.0.2__tar.gz → 7.1.0__tar.gz

{tablassert-7.0.2 → tablassert-7.1.0}/CHANGELOG.md

@@ -2,6 +2,12 @@
 
 All notable changes to this project are documented in this file.
 
+## Unreleased
+
+### Changes
+- Updated `fullmap` ranking to prioritize case-insensitive exact matches between normalized terms and preferred names.
+- Updated `fullmap` term de-duplication to keep first occurrences, improving deterministic output ordering.
+
 ## 7.0.2 - 2026-03-23
 
 ### Changes

{tablassert-7.0.2 → tablassert-7.1.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: tablassert
-Version: 7.0.2
+Version: 7.1.0
 Summary: Tablassert is a highly performant declarative knowledge graph backend designed to extract knowledge assertions from tabular data while exporting NCATS Translator-compliant Knowledge Graph Exchange (KGX) NDJSON.
 Project-URL: Homepage, https://github.com/SkyeAv/Tablassert
 Project-URL: Source, https://github.com/SkyeAv/Tablassert
@@ -14,6 +14,7 @@ Requires-Python: >=3.13
 Requires-Dist: diskcache>=5.6.3
 Requires-Dist: duckdb>=1.5.0
 Requires-Dist: fastexcel>=0.19.0
+Requires-Dist: lazy-loader>=0.5
 Requires-Dist: loguru>=0.7.3
 Requires-Dist: mkdocs>=1.6.1
 Requires-Dist: onnxruntime>=1.24.3

{tablassert-7.0.2 → tablassert-7.1.0}/docs/api/fullmap.md

@@ -103,10 +103,11 @@ Returns a Polars LazyFrame with these columns added:
 
 The function executes a SQL query that:
 
-1. **Builds an in-memory term table** by collecting distinct terms from both NLP levels and registering them in DuckDB as `PARQUET` via `conn.register("PARQUET", df.to_arrow())`.
+1. **Builds an in-memory term table** by collecting terms from both NLP levels, deduplicating by keeping first occurrences for deterministic ordering, and registering them in DuckDB as `PARQUET` via `conn.register("PARQUET", df.to_arrow())`.
 
 2. **Ranks matches** by:
    - Category priority (if `prioritize` specified)
+   - Preferred-name exactness (case-insensitive exact match of normalized term to preferred name)
    - NLP level (exact case match preferred over normalized)
    - Category frequency (if `column_context=True`)
 

{tablassert-7.0.2 → tablassert-7.1.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "tablassert"
-version = "7.0.2"
+version = "7.1.0"
 description = "Tablassert is a highly performant declarative knowledge graph backend designed to extract knowledge assertions from tabular data while exporting NCATS Translator-compliant Knowledge Graph Exchange (KGX) NDJSON."
 authors = [
     { name = "Skye Lane Goetz", email = "sgoetz@isbscience.org" }
@@ -25,6 +25,7 @@ dependencies = [
     "diskcache>=5.6.3",
     "duckdb>=1.5.0",
     "fastexcel>=0.19.0",
+    "lazy-loader>=0.5",
     "loguru>=0.7.3",
     "mkdocs>=1.6.1",
     "onnxruntime>=1.24.3",

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/downloader.py

@@ -1,10 +1,17 @@
+from __future__ import annotations
+
 from pathlib import Path
 from time import sleep
-from typing import Optional
+from typing import TYPE_CHECKING, Optional
 
-import pyexcel
+import lazy_loader as Lazy
 from playwright.sync_api import sync_playwright
 
+if TYPE_CHECKING:
+    import pyexcel
+else:
+    pyexcel = Lazy.load("pyexcel")
+
 
 def modernize_xls(p: Path) -> Path:
     xlsx: Path = p.with_suffix(".xlsx")

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/enums.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from enum import Enum
 
 

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/fullmap.py

@@ -1,11 +1,18 @@
+from __future__ import annotations
+
 from operator import add
-from typing import Optional
+from typing import TYPE_CHECKING, Optional
 
-import polars as pl
+import lazy_loader as Lazy
 
 from tablassert.enums import Categories
 from tablassert.log import logger
 
+if TYPE_CHECKING:
+    import polars as pl
+else:
+    pl = Lazy.load("polars")
+
 
 def distinct(lf: pl.LazyFrame, l0: str, l1: str) -> pl.LazyFrame:
     # ? Extract Unique Terms From Two Text Normalization Columns As LazyFrame
@@ -15,7 +22,7 @@ def distinct(lf: pl.LazyFrame, l0: str, l1: str) -> pl.LazyFrame:
     t1: pl.LazyFrame = lf.select(pl.col(l1).alias("term")).unique()
     t1 = t1.with_columns(pl.lit(1).alias("nlp level"))
 
-    terms: pl.LazyFrame = pl.concat([t0, t1]).unique(subset=["term"])
+    terms: pl.LazyFrame = pl.concat([t0, t1]).unique(subset=["term"], keep="first")
 
     bad: str = r"^\d+$|^(none|nan|na|null|unknown)$|^$"
     return terms.filter(~pl.col("term").str.contains(bad))
@@ -38,6 +45,9 @@ def query_builder(
         CASE
             {priority_case}
             ELSE 50
+        END * CASE
+            WHEN LOWER(CU.PREFERRED_NAME) = PA.term THEN 1
+            ELSE 10
         END AS PR
     FROM SYNONYMS SY
     JOIN SOURCES SO ON SY.SOURCE_ID = SO.SOURCE_ID
@@ -76,13 +86,16 @@ def query_distinct(
     results: pl.DataFrame = conn.execute(query).pl()  # pyright: ignore
 
     sort_by: list[str] = ["term", "PR", "NLP_LEVEL"]
+    descending: list[bool] = [False, False, False]
 
     if column_context:
         frequency: pl.DataFrame = results.group_by("CATEGORY_NAME").agg(pl.len().alias("FREQUENCY"))
         results = results.join(frequency, on="CATEGORY_NAME", how="left")
+
         sort_by += ["FREQUENCY"]
+        descending += [True]
 
-    results = results.sort(sort_by, descending=[False, False, False, True])
+    results = results.sort(sort_by, descending=descending)
     results = results.unique(subset=["term"], keep="first")
 
     return results

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/ingests.py

@@ -1,10 +1,17 @@
+from __future__ import annotations
+
 from copy import deepcopy
 from pathlib import Path
-from typing import Any, Union
+from typing import TYPE_CHECKING, Any, Union
 
-import yaml
+import lazy_loader as Lazy
 from yaml import CLoader
 
+if TYPE_CHECKING:
+    import yaml
+else:
+    yaml = Lazy.load("yaml")
+
 
 def fastmerge(a: Union[list[Any], dict[str, Any]], b: Union[list[Any], dict[str, Any]]) -> Any:
     # ? Streamlined (Fast) Implementation Of Deepmerge Config

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/lib.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 import math
 import operator
 from functools import reduce
@@ -6,13 +8,9 @@ from multiprocessing import Pool
 from operator import add, eq, le
 from os.path import basename
 from pathlib import Path
-from typing import Any, Callable, Literal, Optional, Self, Union
+from typing import TYPE_CHECKING, Any, Callable, Literal, Optional, Self, Union
 
-import duckdb
-import orjson
-import polars as pl
-import typer
-import xxhash
+import lazy_loader as Lazy
 from pydantic import Field, NonNegativeInt, PositiveInt
 from rich.progress import BarColumn, Progress, SpinnerColumn, TaskProgressColumn, TextColumn, TimeElapsedColumn
 from sqlite_utils import Database
@@ -26,6 +24,19 @@ from tablassert.models import Encoding, Graph, NodeEncoding, Section
 from tablassert.qc import fullmap_audit
 from tablassert.utils import STORE, mkhash, namespace_uuid
 
+if TYPE_CHECKING:
+    import duckdb
+    import orjson
+    import polars as pl
+    import typer
+    import xxhash
+else:
+    duckdb = Lazy.load("duckdb")
+    orjson = Lazy.load("orjson")
+    pl = Lazy.load("polars")
+    typer = Lazy.load("typer")
+    xxhash = Lazy.load("xxhash")
+
 # ? Newline To Make Progress Bar More Readable
 print("\n")
 

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/log.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from pathlib import Path
 
 from loguru import logger

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/models.py

@@ -1,3 +1,5 @@
+from __future__ import annotations
+
 from pathlib import Path
 from typing import Literal, Optional, Union
 

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/qc.py

@@ -1,13 +1,22 @@
+from __future__ import annotations
+
 from operator import add, eq, ge
 from pathlib import Path
-from typing import Literal, Optional
+from typing import TYPE_CHECKING, Literal, Optional
 
-import onnxruntime as ort
-import polars as pl
+import lazy_loader as Lazy
 from rapidfuzz import fuzz
-from sentence_transformers import SentenceTransformer
 from sklearn.metrics.pairwise import cosine_similarity
 
+if TYPE_CHECKING:
+    import onnxruntime as ort
+    import polars as pl
+    import sentence_transformers
+else:
+    ort = Lazy.load("onnxruntime")
+    sentence_transformers = Lazy.load("sentence_transformers")
+    pl = Lazy.load("polars")
+
 from tablassert.log import logger
 from tablassert.utils import DISKCACHE
 
@@ -28,9 +37,11 @@ def get_biobert() -> object:
     if BIOBERT:
         return BIOBERT
     elif not BIOBERT and MODEL.exists():
-        BIOBERT = SentenceTransformer(str(MODEL), backend=MODEL_BACKEND, model_kwargs=MODEL_KWARGS)  # pyright: ignore
+        BIOBERT = sentence_transformers.SentenceTransformer(
+            str(MODEL), backend=MODEL_BACKEND, model_kwargs=MODEL_KWARGS
+        )  # pyright: ignore
     else:
-        BIOBERT = SentenceTransformer(
+        BIOBERT = sentence_transformers.SentenceTransformer(
             "pritamdeka/BioBERT-mnli-snli-scinli-scitail-mednli-stsb", backend=MODEL_BACKEND, model_kwargs=MODEL_KWARGS
         )  # pyright: ignore
         MODEL.mkdir(parents=True, exist_ok=True)

{tablassert-7.0.2 → tablassert-7.1.0}/src/tablassert/utils.py

@@ -1,12 +1,20 @@
+from __future__ import annotations
+
 from functools import cache
 from pathlib import Path
-from typing import Any
+from typing import TYPE_CHECKING, Any
 from uuid import UUID, uuid3
 
-import polars as pl
-import xxhash
+import lazy_loader as Lazy
 from diskcache import Cache
 
+if TYPE_CHECKING:
+    import polars as pl
+    import xxhash
+else:
+    pl = Lazy.load("polars")
+    xxhash = Lazy.load("xxhash")
+
 STORE: Path = Path("./.storassert")
 STORE.mkdir(parents=True, exist_ok=True)
 

{tablassert-7.0.2 → tablassert-7.1.0}/uv.lock

@@ -363,6 +363,18 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/7b/91/984aca2ec129e2757d1e4e3c81c3fcda9d0f85b74670a094cc443d9ee949/joblib-1.5.3-py3-none-any.whl", hash = "sha256:5fc3c5039fc5ca8c0276333a188bbd59d6b7ab37fe6632daa76bc7f9ec18e713", size = 309071, upload-time = "2025-12-15T08:41:44.973Z" },
 ]
 
+[[package]]
+name = "lazy-loader"
+version = "0.5"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "packaging" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/49/ac/21a1f8aa3777f5658576777ea76bfb124b702c520bbe90edf4ae9915eafa/lazy_loader-0.5.tar.gz", hash = "sha256:717f9179a0dbed357012ddad50a5ad3d5e4d9a0b8712680d4e687f5e6e6ed9b3", size = 15294, upload-time = "2026-03-06T15:45:09.054Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/8a/a1/8d812e53a5da1687abb10445275d41a8b13adb781bbf7196ddbcf8d88505/lazy_loader-0.5-py3-none-any.whl", hash = "sha256:ab0ea149e9c554d4ffeeb21105ac60bed7f3b4fd69b1d2360a4add51b170b005", size = 8044, upload-time = "2026-03-06T15:45:07.668Z" },
+]
+
 [[package]]
 name = "lml"
 version = "0.2.0"
@@ -1649,12 +1661,13 @@ wheels = [
 
 [[package]]
 name = "tablassert"
-version = "7.0.2"
+version = "7.1.0"
 source = { editable = "." }
 dependencies = [
     { name = "diskcache" },
     { name = "duckdb" },
     { name = "fastexcel" },
+    { name = "lazy-loader" },
     { name = "loguru" },
     { name = "mkdocs" },
     { name = "onnxruntime" },
@@ -1693,6 +1706,7 @@ requires-dist = [
     { name = "diskcache", specifier = ">=5.6.3" },
     { name = "duckdb", specifier = ">=1.5.0" },
     { name = "fastexcel", specifier = ">=0.19.0" },
+    { name = "lazy-loader", specifier = ">=0.5" },
     { name = "loguru", specifier = ">=0.7.3" },
     { name = "mkdocs", specifier = ">=1.6.1" },
     { name = "onnxruntime", specifier = ">=1.24.3" },