filter_red_titles.py

"""Filter slides with red titles from powerpoints, and save the filtered slides as one PDF file.

This script can only be run on Windows because it requires some Windows-specific libraries.

How to use
----------
- Install comtypes, pypdf2, and python-pptx using pip. Powerpoint must also be installed.
- Put all of the powerpoints to be filtered in one directory.
- Navigate to that directory in a terminal.
- Run this script. After some time, a file will be created named "output.pdf" with the filtered slides.
"""

from pathlib import Path
import sys
import tempfile

import comtypes.client
from PyPDF2 import PdfFileMerger
import pptx


def slide_has_red_title(slide):
    for shape in slide.shapes:
        try:
            is_red = shape.fill.fore_color.rgb == (255, 0, 0)
            is_title = shape.top < 50000
            if is_red and is_title:
                return True
        except (AttributeError, TypeError):
            pass
    return False


def delete_slide(presentation, slide):
    id_dict = {slide.id: [i, slide.rId] for i,slide in enumerate(presentation.slides._sldIdLst)}
    slide_id = slide.slide_id
    presentation.part.drop_rel(id_dict[slide_id][1])
    del presentation.slides._sldIdLst[id_dict[slide_id][0]]


def filter_red_title_slides(presentation, keep_first_slide=False):
    for i, s in enumerate(presentation.slides):
        if keep_first_slide and i == 0:
            continue
        if not slide_has_red_title(s):
            delete_slide(presentation, s)


def batch_convert_ppt_to_pdf(infiles, format_type=32):
    powerpoint = comtypes.client.CreateObject("Powerpoint.Application")
    powerpoint.Visible = 1
    m = len(infiles)
    for j, infile in enumerate(infiles):
        deck = powerpoint.Presentations.Open(str(infile))
        deck.SaveAs(str(infile.with_suffix('.pdf')), format_type)
        deck.Close()
        print("    {} / {}".format(j + 1, m), end='\r')
    powerpoint.Quit()


def merge_pdfs(infiles, outfile):
    merger = PdfFileMerger()
    m = len(infiles)
    for j, pdf in enumerate(infiles):
        merger.append(str(pdf))
        print("{} / {}".format(j + 1, m), end='\r')
    merger.write(str(outfile))
    merger.close()


if __name__ == '__main__':

    with tempfile.TemporaryDirectory() as tmpdirname:

        ppt_files = Path('.').glob('*.pptx')
       
        # Filter all of the presentations, and save to a temporary directory.
        print("++ Filtering presentations ...")
        for j, filename in enumerate(ppt_files):
            p = pptx.Presentation(filename)
            filter_red_title_slides(p, keep_first_slide=True)
            ptmp = Path(tmpdirname)
            outpath = ptmp / '{}_filtered.pptx'.format(Path(filename).name)
            p.save(outpath)

        # Convert all the filtered presentations to PDF.
        filtered = list(ptmp.glob('*_filtered.pptx'))
        if not len(filtered):
            raise ValueError("No filtered powerpoint files found.")
        print("++ Converting filtered presentations to PDF ...")
        batch_convert_ppt_to_pdf(infiles=filtered)

        # Merge the PDFs, and save.
        pdf_files = list(ptmp.glob('*_filtered.pdf'))
        if not len(pdf_files):
            raise ValueError("No PDF files found.")
        print("++ Merging and saving to PDF ...")
        merge_pdfs(pdf_files, "output.pdf")