From 420e20a8cf56ffe3ba4ef6b0d0000be9c8dd5c7c Mon Sep 17 00:00:00 2001
From: Alex-Gilbert <alexp.gilbert@pm.me>
Date: Sun, 15 Feb 2026 18:38:50 -0600
Subject: [PATCH] feat: PDF rasterization module

Add rasterize_pdf() that converts PDF pages to PIL images using
pdf2image. Supports configurable DPI and page range selection.
Includes tests verifying multi-page output, high-DPI dimensions,
and single-page extraction.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
---
 tests/test_rasterize.py        | 24 ++++++++++++++++++++++++
 unredact/pipeline/rasterize.py | 19 +++++++++++++++++++
 2 files changed, 43 insertions(+)
 create mode 100644 tests/test_rasterize.py
 create mode 100644 unredact/pipeline/rasterize.py

diff --git a/tests/test_rasterize.py b/tests/test_rasterize.py
new file mode 100644
index 0000000..9ddfa51
--- /dev/null
+++ b/tests/test_rasterize.py
@@ -0,0 +1,24 @@
+from pathlib import Path
+
+from PIL import Image
+
+from unredact.pipeline.rasterize import rasterize_pdf
+
+
+def test_rasterize_returns_list_of_images(sample_pdf: Path):
+    pages = rasterize_pdf(sample_pdf)
+    assert len(pages) == 2  # EFTA00554620.pdf has 2 pages
+    assert all(isinstance(p, Image.Image) for p in pages)
+
+
+def test_rasterize_high_dpi(sample_pdf: Path):
+    pages = rasterize_pdf(sample_pdf, dpi=300)
+    # At 300 DPI, a letter-size page is roughly 2550x3300
+    w, h = pages[0].size
+    assert w > 2000
+    assert h > 3000
+
+
+def test_rasterize_single_page(sample_pdf: Path):
+    pages = rasterize_pdf(sample_pdf, first_page=1, last_page=1)
+    assert len(pages) == 1
diff --git a/unredact/pipeline/rasterize.py b/unredact/pipeline/rasterize.py
new file mode 100644
index 0000000..4c59452
--- /dev/null
+++ b/unredact/pipeline/rasterize.py
@@ -0,0 +1,19 @@
+from pathlib import Path
+
+from pdf2image import convert_from_path
+from PIL import Image
+
+
+def rasterize_pdf(
+    pdf_path: Path,
+    dpi: int = 300,
+    first_page: int | None = None,
+    last_page: int | None = None,
+) -> list[Image.Image]:
+    """Convert PDF pages to PIL images at the given DPI."""
+    kwargs: dict = {"dpi": dpi}
+    if first_page is not None:
+        kwargs["first_page"] = first_page
+    if last_page is not None:
+        kwargs["last_page"] = last_page
+    return convert_from_path(str(pdf_path), **kwargs)