xppl/xppl/cover.py

import io

import PyPDF2
from wand.color import Color
from wand.image import Image


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.background_color = Color('white')
    img.alpha_channel = 'remove'
    img.convert("jpeg")

    return img


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

    try:
        src_pdf = PyPDF2.PdfFileReader(open(src_filename, "rb"))
    except Exception:
        print("couln't open PDF")
        return

    if src_pdf.isEncrypted:
        try:
            src_pdf.decrypt('')
        except Exception:
            print("couln't decrypt")
            return

    # 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 = "xppl/cover/"+page["filename"] + "_cover.jpeg"

        img = pdf_page_to_png(src_pdf, pagenum=page["pagenum"], resolution=130)

        # Ensmallen
        img.transform("", "250")
        img.save(filename=big_filename)

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