doc-redaction 2.2.7__tar.gz → 2.2.8__tar.gz

{doc_redaction-2.2.7/doc_redaction.egg-info → doc_redaction-2.2.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: doc_redaction
-Version: 2.2.7
+Version: 2.2.8
 Summary: Redact PDF/image-based documents, Word, or CSV/XLSX files using a Gradio-based GUI interface
 Author-email: Sean Pedrick-Case <spedrickcase@lambeth.gov.uk>
 Maintainer-email: Sean Pedrick-Case <spedrickcase@lambeth.gov.uk>
@@ -8,7 +8,6 @@ License-Expression: AGPL-3.0-only
 Project-URL: Homepage, https://seanpedrick-case.github.io/doc_redaction/
 Project-URL: Repository, https://github.com/seanpedrick-case/doc_redaction
 Keywords: redaction,pdf,nlp,documents,document-processing,gradio,pii,pii-detection
-Platform: UNKNOWN
 Classifier: Development Status :: 5 - Production/Stable
 Classifier: Intended Audience :: Developers
 Classifier: Intended Audience :: Legal Industry
@@ -48,7 +47,7 @@ Requires-Dist: python-docx<=1.2.0
 Requires-Dist: polars<=1.38.1
 Requires-Dist: defusedxml<=0.7.1
 Requires-Dist: numpy<=2.4.4
-Requires-Dist: spaces<=0.48.3
+Requires-Dist: spaces<=0.49.0
 Requires-Dist: google-genai<=1.73.0
 Requires-Dist: openai<=2.31.0
 Requires-Dist: markdown<=3.10.2

{doc_redaction-2.2.7 → doc_redaction-2.2.8/doc_redaction.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: doc_redaction
-Version: 2.2.7
+Version: 2.2.8
 Summary: Redact PDF/image-based documents, Word, or CSV/XLSX files using a Gradio-based GUI interface
 Author-email: Sean Pedrick-Case <spedrickcase@lambeth.gov.uk>
 Maintainer-email: Sean Pedrick-Case <spedrickcase@lambeth.gov.uk>
@@ -8,7 +8,6 @@ License-Expression: AGPL-3.0-only
 Project-URL: Homepage, https://seanpedrick-case.github.io/doc_redaction/
 Project-URL: Repository, https://github.com/seanpedrick-case/doc_redaction
 Keywords: redaction,pdf,nlp,documents,document-processing,gradio,pii,pii-detection
-Platform: UNKNOWN
 Classifier: Development Status :: 5 - Production/Stable
 Classifier: Intended Audience :: Developers
 Classifier: Intended Audience :: Legal Industry
@@ -48,7 +47,7 @@ Requires-Dist: python-docx<=1.2.0
 Requires-Dist: polars<=1.38.1
 Requires-Dist: defusedxml<=0.7.1
 Requires-Dist: numpy<=2.4.4
-Requires-Dist: spaces<=0.48.3
+Requires-Dist: spaces<=0.49.0
 Requires-Dist: google-genai<=1.73.0
 Requires-Dist: openai<=2.31.0
 Requires-Dist: markdown<=3.10.2

{doc_redaction-2.2.7 → doc_redaction-2.2.8}/doc_redaction.egg-info/requires.txt

@@ -25,7 +25,7 @@ python-docx<=1.2.0
 polars<=1.38.1
 defusedxml<=0.7.1
 numpy<=2.4.4
-spaces<=0.48.3
+spaces<=0.49.0
 google-genai<=1.73.0
 openai<=2.31.0
 markdown<=3.10.2

{doc_redaction-2.2.7 → doc_redaction-2.2.8}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "doc_redaction"
-version = "2.2.7"
+version = "2.2.8"
 description = "Redact PDF/image-based documents, Word, or CSV/XLSX files using a Gradio-based GUI interface"
 readme = "README_PYPI.md"
 authors = [
@@ -65,7 +65,7 @@ dependencies = [
     "polars<=1.38.1",
     "defusedxml<=0.7.1",
     "numpy<=2.4.4",
-    "spaces<=0.48.3",
+    "spaces<=0.49.0",
     "google-genai<=1.73.0",
     "openai<=2.31.0",
     "markdown<=3.10.2",

{doc_redaction-2.2.7 → doc_redaction-2.2.8}/test/test_gradio_upload_staging.py

@@ -3,10 +3,12 @@
 from __future__ import annotations
 
 from pathlib import Path
+from typing import Any
 from unittest.mock import patch
 
 from tools.simplified_api import (
     _is_gradio_ephemeral_upload_path,
+    preview_boxes_api,
     stage_gradio_upload_if_ephemeral,
 )
 
@@ -42,3 +44,37 @@ def test_stage_gradio_upload_skips_normal_paths(tmp_path: Path) -> None:
         out = stage_gradio_upload_if_ephemeral(str(p))
     mock_staging.assert_not_called()
     assert out == str(p)
+
+
+def test_preview_boxes_api_staging_uses_single_path_arg(tmp_path: Path) -> None:
+    """Regression: preview_boxes must not pass a second arg to stage_gradio_upload_if_ephemeral."""
+    pdf = tmp_path / "doc.pdf"
+    review = tmp_path / "plan_review_file.csv"
+    pdf.write_bytes(b"%PDF-1.4")
+    review.write_bytes(b"x")
+    fake_png = tmp_path / "p1.png"
+    fake_png.write_bytes(b"png")
+    out_root = tmp_path / "outputs"
+    out_root.mkdir()
+
+    staging_calls: list[tuple[Any, ...]] = []
+
+    def recording_stage(src: str) -> str:
+        staging_calls.append((src,))
+        return src
+
+    with patch("tools.simplified_api.OUTPUT_FOLDER", str(out_root)):
+        with patch(
+            "tools.preview_redaction_boxes.preview_redaction_boxes",
+            return_value=[str(fake_png)],
+        ):
+            with patch(
+                "tools.simplified_api.stage_gradio_upload_if_ephemeral",
+                side_effect=recording_stage,
+            ):
+                zip_path, msg = preview_boxes_api(str(pdf), str(review))
+
+    assert len(staging_calls) == 2
+    assert all(len(c) == 1 for c in staging_calls)
+    assert zip_path.endswith("preview_boxes.zip")
+    assert "Preview complete" in msg

{doc_redaction-2.2.7 → doc_redaction-2.2.8}/tools/simplified_api.py

@@ -1130,8 +1130,8 @@ def preview_boxes_api(
     if not pdf_path or not csv_path:
         return "", "Error: both pdf_file and review_csv_file are required."
 
-    pdf_path = stage_gradio_upload_if_ephemeral(pdf_path, INPUT_FOLDER)
-    csv_path = stage_gradio_upload_if_ephemeral(csv_path, INPUT_FOLDER)
+    pdf_path = stage_gradio_upload_if_ephemeral(pdf_path)
+    csv_path = stage_gradio_upload_if_ephemeral(csv_path)
 
     page_list: list[int] | None = None
     if pages: