xppl/app/cover.py

import PyPDF2
from wand.image import Image
import io
import os


def pdf_page_to_png(src_pdf, pagenum = 0, resolution = 72,):
    """
    Returns specified PDF page as wand.image.Image png.
    :param PyPDF2.PdfFileReader src_pdf: PDF from which to take pages.
    :param int pagenum: Page number to take.
    :param int resolution: Resolution for resulting png in DPI.
    """
    dst_pdf = PyPDF2.PdfFileWriter()
    dst_pdf.addPage(src_pdf.getPage(pagenum))

    pdf_bytes = io.BytesIO()
    dst_pdf.write(pdf_bytes)
    pdf_bytes.seek(0)

    img = Image(file = pdf_bytes, resolution = resolution)
    img.convert("png")

    return img


def get_cover(file_path, filename):
    # Main
    # ====
    print(file_path)
    src_filename = file_path

    src_pdf = PyPDF2.PdfFileReader(open(src_filename, "rb"))

    # What follows is a lookup table of page numbers within sample_log.pdf and the corresponding filenames.
    pages = [{"pagenum": 0,  "filename": filename}]

    # Convert each page to a png image.
    for page in pages:
        big_filename = "app/cover/"+page["filename"] + "_cover.png"
        small_filename = "app/cover/"+page["filename"] + "cover_small" + ".png"

        img = pdf_page_to_png(src_pdf, pagenum = page["pagenum"], resolution = 200)
        img.save(filename = big_filename)

        # Ensmallen
        img.transform("", "200")
        img.save(filename = small_filename)

    return page["filename"] + "_cover.png"

#---
#epub
#https://ebooks.stackexchange.com/questions/6517/command-line-extraction-of-metadata-title-author-from-epub-file
#https://hackage.haskell.org/package/epub-tools
#http://stackoverflow.com/questions/9751475/extract-cover-image-from-chm-and-epub-files