diff options
| -rw-r--r-- | pyblackbird_cc/resources/services.py | 36 | ||||
| -rw-r--r-- | requirements/base.txt | 4 |
2 files changed, 21 insertions, 19 deletions
diff --git a/pyblackbird_cc/resources/services.py b/pyblackbird_cc/resources/services.py index b33afcf..441f623 100644 --- a/pyblackbird_cc/resources/services.py +++ b/pyblackbird_cc/resources/services.py @@ -2,7 +2,8 @@ import dataclasses import os.path import tempfile -import pypdfium2 as pdfium +from pdf2image import convert_from_path +from PyPDF2 import PdfReader @dataclasses.dataclass @@ -21,11 +22,11 @@ def get_pdf_metadata_from_path(file_path: str) -> PDFMetadata: if file_path is None: raise ValueError("file_path cannot be None") if not os.path.isfile(file_path): - raise ValueError(f"file_path must be a file. {file_path} is not a file.") - pdf = pdfium.PdfDocument(file_path) - n_pages = len(pdf) + raise ValueError("file_path must be a file. {file_path} is not a file.") + reader = PdfReader(file_path) + n_pages = len(reader.pages) file_size = os.path.getsize(file_path) - pdf.close() + #pdf.close() return PDFMetadata(file_name=file_path, file_size=file_size, n_pages=n_pages) @@ -38,19 +39,20 @@ def export_pages_as_images(file_path: str) -> list[str]: if file_path is None: raise ValueError("file_path cannot be None") output_dir = tempfile.mkdtemp() # Create a temporary directory + reader = PdfReader(file_path) + n_pages = len(reader.pages) try: - pdf = pdfium.PdfDocument(file_path) - # get the file_name of this PDF file at file_path - file_name = os.path.basename(file_path) - image_paths = [] - for i in range(len(pdf)): - page = pdf[i] - image = page.render(scale=0.6).to_pil() - image_path = os.path.join(output_dir, f"{file_name}_{i:03d}.jpg") - image.save(image_path) - image_paths.append(image_path) - pdf.close() - return image_paths + with tempfile.TemporaryDirectory() as path: + images_from_path = convert_from_path(file_path, 56, size=300, output_folder=path) + # get the file_name of this PDF file at file_path + file_name = os.path.basename(file_path) + image_paths = [] + for i in range(n_pages): + image = images_from_path[i] + image_path = os.path.join(output_dir, f"{file_name}_{i:03d}.jpg") + image.save(image_path) + image_paths.append(image_path) + return image_paths finally: # Optionally handle cleanup later or elsewhere in your code # Remove later with shutil.rmtree(output_dir) diff --git a/requirements/base.txt b/requirements/base.txt index fdf7dd1..51b046e 100644 --- a/requirements/base.txt +++ b/requirements/base.txt @@ -9,8 +9,8 @@ django-celery-beat==2.6.0 # https://github.com/celery/django-celery-beat flower==2.0.1 # https://github.com/mher/flower # migrated stuff from pyblackbird -pypdf==4.2.0 -pypdfium2==4.29.0 +pdf2image==1.17.0 +pypdf2==3.0.1 python-magic==0.4.27 python-dotenv==1.0.1 boto3==1.34.89 |