From 58a4dfa335aaf9212a94e1219f57d5fed19a8f23 Mon Sep 17 00:00:00 2001 From: Roland Date: Mon, 19 Dec 2022 12:52:02 -0700 Subject: [PATCH] Creating model_pdf which attaches a PDF's title to it's File. --- wrolpi/files/__init__.py | 1 + wrolpi/files/pdfs.py | 37 ++++++++++++++++++++++++++++++++++ wrolpi/files/test/test_pdfs.py | 13 ++++++++++++ 3 files changed, 51 insertions(+) create mode 100644 wrolpi/files/pdfs.py create mode 100644 wrolpi/files/test/test_pdfs.py diff --git a/wrolpi/files/__init__.py b/wrolpi/files/__init__.py index c6029615..4347d483 100644 --- a/wrolpi/files/__init__.py +++ b/wrolpi/files/__init__.py @@ -1,2 +1,3 @@ from . import ebooks # noqa +from . import pdfs # noqa from .api import bp diff --git a/wrolpi/files/pdfs.py b/wrolpi/files/pdfs.py new file mode 100644 index 00000000..26986816 --- /dev/null +++ b/wrolpi/files/pdfs.py @@ -0,0 +1,37 @@ +import pathlib +from typing import List, Dict, Optional + +from sqlalchemy.orm import Session + +from wrolpi.common import register_modeler, logger +from wrolpi.files.models import File + +try: + from PyPDF2 import PdfReader +except ImportError: + PdfReader = None + +logger = logger.getChild(__name__) + + +def get_pdf_title(path: pathlib.Path) -> Optional[str]: + reader = PdfReader(path) + if reader.metadata and reader.metadata.title: + return reader.metadata.title.strip() + + +@register_modeler +def model_pdf(groups: Dict[str, List[File]], _: Session): + """Reads the contents of a PDF and extracts the title, attaches the title to the File. + + This modeler does not claim any groups passed to it. Does NOT mark the PDFs as indexed.""" + if not PdfReader: + logger.warning(f'Cannot get title from PDF without PyPDF2') + return + + pdf_files = (file for group in groups.values() for file in group if file.mimetype == 'application/pdf') + + for file in pdf_files: + path: pathlib.Path = file.path.path if hasattr(file.path, 'path') else file.path + if title := get_pdf_title(path): + file.title = title diff --git a/wrolpi/files/test/test_pdfs.py b/wrolpi/files/test/test_pdfs.py new file mode 100644 index 00000000..264153b6 --- /dev/null +++ b/wrolpi/files/test/test_pdfs.py @@ -0,0 +1,13 @@ +import pytest + +from wrolpi.files.lib import refresh_files +from wrolpi.files.models import File + + +@pytest.mark.asyncio +async def test_pdf_title(test_session, example_pdf): + """PDF modeler extracts the PDF's title.""" + await refresh_files() + + pdf: File = test_session.query(File).one() + assert pdf.title == 'WROLPi Test PDF'