added pdf.to_imgs func

pyproject.toml

@@ -8,7 +8,7 @@ description = "Streamlines the process of preparing documents for LLM's."
 readme = "README.md"
 requires-python = ">=3.8"
 license = { text = "MIT" }
-version = "0.5.7"
+version = "0.5.8"
 authors = [{name = "Sergey Filimonov", email = "[email protected]"}]
 dependencies = [
     "PyMuPDF >= 1.23.2",

src/openparse/pdf.py

@@ -8,6 +8,7 @@
 
 from pdfminer.high_level import extract_pages
 from pdfminer.layout import LTPage
+from PIL import Image
 from pydantic import BaseModel
 from pypdf import PdfReader, PdfWriter
 
@@ -237,3 +238,23 @@ def _flip_coordinates(self, bbox: Bbox) -> Bbox:
             x1=bbox.x1,
             y1=fy1,
         )
+
+    def to_imgs(self, page_numbers: Optional[List[int]] = None) -> List[Image.Image]:
+        doc = self.to_pymupdf_doc()
+        images = []
+
+        if not doc.is_pdf:
+            raise ValueError("The document is not in PDF format.")
+        if doc.needs_pass:
+            raise ValueError("The PDF document is password protected.")
+
+        if page_numbers is None:
+            page_numbers = list(range(doc.page_count))
+
+        for n in page_numbers:
+            page = doc[n]
+            pix = page.get_pixmap()
+            image = Image.frombytes("RGB", (pix.width, pix.height), pix.samples)
+            images.append(image)
+
+        return images

src/openparse/version.py

@@ -1,4 +1,4 @@
-OPEN_PARSE_VERSION = "0.5.7"
+OPEN_PARSE_VERSION = "0.5.8"
 
 
 def version_info() -> str:

src/tests/conftest.py

@@ -0,0 +1,16 @@
+from pathlib import Path
+
+import pytest
+
+
+@pytest.hookimpl(tryfirst=True)
+def pytest_sessionstart(session):
+    expected_dir = Path(__file__).resolve().parent / "src"
+
+    current_dir = Path.cwd()
+
+    if not current_dir == expected_dir:
+        pytest.exit(
+            "Pytest must be run from the project root directory",
+            returncode=1,
+        )

src/tests/test_doc_parser.py

@@ -1,6 +1,7 @@
-import openparse
 import re
 
+import openparse
+
 
 def test_parse_doc():
     basic_doc_path = "src/evals/data/full-pdfs/mock-1-page-lease.pdf"