diff --git a/.github/workflows/publish.yml b/.github/workflows/publish.yml index 09418c5..3b18a33 100644 --- a/.github/workflows/publish.yml +++ b/.github/workflows/publish.yml @@ -36,7 +36,7 @@ jobs: - name: Build and push Docker image uses: docker/build-push-action@v5.3.0 with: - context: . + context: ./src push: true tags: ${{ steps.meta.outputs.tags }} labels: ${{ steps.meta.outputs.labels }} diff --git a/src/main.py b/src/main.py index ea53359..933f21e 100644 --- a/src/main.py +++ b/src/main.py @@ -27,6 +27,7 @@ def predict(input_file: Path, output_file: Path): print(f"Created new PDF file: {output_file}") # Perform OCR on the images for image in convert_from_path(input_file, dpi=300, fmt="jpeg"): + print("Converted PDF page to image") prediction = pytesseract.image_to_pdf_or_hocr(image, extension="pdf") print("Extracted text from image") doc.insert_pdf(fitz.open("pdf", prediction))