|
| 1 | +#!/usr/bin/env python3 |
| 2 | +# Author: Antonios Tsolis (Alamot) |
| 3 | +import os |
| 4 | +import sys |
| 5 | +import zipfile |
| 6 | +from lxml import etree |
| 7 | +from PIL import Image |
| 8 | + |
| 9 | + |
| 10 | +# Let's define the required XML namespaces |
| 11 | +namespaces = { |
| 12 | + "calibre":"http://calibre.kovidgoyal.net/2009/metadata", |
| 13 | + "dc":"http://purl.org/dc/elements/1.1/", |
| 14 | + "dcterms":"http://purl.org/dc/terms/", |
| 15 | + "opf":"http://www.idpf.org/2007/opf", |
| 16 | + "u":"urn:oasis:names:tc:opendocument:xmlns:container", |
| 17 | + "xsi":"http://www.w3.org/2001/XMLSchema-instance", |
| 18 | +} |
| 19 | + |
| 20 | + |
| 21 | +def get_epub_cover(epub_path): |
| 22 | + ''' Return the cover image file from an epub archive. ''' |
| 23 | + |
| 24 | + # We open the epub archive using zipfile.ZipFile(): |
| 25 | + with zipfile.ZipFile(epub_path) as z: |
| 26 | + |
| 27 | + # We load "META-INF/container.xml" using lxml.etree.fromString(): |
| 28 | + t = etree.fromstring(z.read("META-INF/container.xml")) |
| 29 | + # We use xpath() to find the attribute "full-path": |
| 30 | + ''' |
| 31 | + <container version="1.0" xmlns="urn:oasis:names:tc:opendocument:xmlns:container"> |
| 32 | + <rootfiles> |
| 33 | + <rootfile full-path="OEBPS/content.opf" ... /> |
| 34 | + </rootfiles> |
| 35 | + </container> |
| 36 | + ''' |
| 37 | + rootfile_path = t.xpath("/u:container/u:rootfiles/u:rootfile", |
| 38 | + namespaces=namespaces)[0].get("full-path") |
| 39 | + print("Path of root file found: " + rootfile_path) |
| 40 | + |
| 41 | + # We load the "root" file, indicated by the "full_path" attribute of "META-INF/container.xml", using lxml.etree.fromString(): |
| 42 | + t = etree.fromstring(z.read(rootfile_path)) |
| 43 | + # We use xpath() to find the attribute "content": |
| 44 | + ''' |
| 45 | + <metadata xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:opf="http://www.idpf.org/2007/opf"> |
| 46 | + ... |
| 47 | + <meta content="my-cover-image" name="cover"/> |
| 48 | + ... |
| 49 | + </metadata> |
| 50 | + ''' |
| 51 | + cover_id = t.xpath("//opf:metadata/opf:meta[@name='cover']", |
| 52 | + namespaces=namespaces)[0].get("content") |
| 53 | + print("ID of cover image found: " + cover_id) |
| 54 | + |
| 55 | + # We use xpath() to find the attribute "href": |
| 56 | + ''' |
| 57 | + <manifest> |
| 58 | + ... |
| 59 | + <item id="my-cover-image" href="images/978.jpg" ... /> |
| 60 | + ... |
| 61 | + </manifest> |
| 62 | + ''' |
| 63 | + cover_href = t.xpath("//opf:manifest/opf:item[@id='" + cover_id + "']", |
| 64 | + namespaces=namespaces)[0].get("href") |
| 65 | + # In order to get the full path for the cover image, we have to join rootfile_path and cover_href: |
| 66 | + cover_path = os.path.join(os.path.dirname(rootfile_path), cover_href) |
| 67 | + print("Path of cover image found: " + cover_path) |
| 68 | + |
| 69 | + # We return the image |
| 70 | + return z.open(cover_path) |
| 71 | + |
| 72 | + |
| 73 | +if len(sys.argv) < 2: |
| 74 | + print("Usage: " + sys.argv[0] + " filename.epub") |
| 75 | + exit() |
| 76 | + |
| 77 | +epubfile = sys.argv[1] |
| 78 | +if not os.path.isfile(epubfile): |
| 79 | + print("File not found: " + epubfile) |
| 80 | + exit() |
| 81 | + |
| 82 | +image = Image.open(get_epub_cover(epubfile)) |
| 83 | +image.show() |
| 84 | + |
| 85 | + |
0 commit comments