librarian/packagers.py

   1 # -*- coding: utf-8 -*-
   2 #
   3 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
   4 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
   5 #
   6 import os
   7 from copy import deepcopy
   8 from lxml import etree
   9 from librarian import pdf, epub, DirDocProvider, ParseError, cover
  10 from librarian.parser import WLDocument
  11
  12
  13 class Packager(object):
  14     cover = None
  15     flags = None
  16
  17     @classmethod
  18     def prepare_file(cls, main_input, output_dir, verbose=False):
  19         path, fname = os.path.realpath(main_input).rsplit('/', 1)
  20         provider = DirDocProvider(path)
  21         slug, ext = os.path.splitext(fname)
  22
  23         if output_dir != '':
  24             try:
  25                 os.makedirs(output_dir)
  26             except:
  27                 pass
  28         outfile = os.path.join(output_dir, slug + '.' + cls.ext)
  29
  30         doc = WLDocument.from_file(main_input, provider=provider)
  31         output_file = cls.converter.transform(doc,
  32                 cover=cls.cover, flags=cls.flags)
  33         doc.save_output_file(output_file, output_path=outfile)
  34
  35
  36     @classmethod
  37     def prepare(cls, input_filenames, output_dir='', verbose=False):
  38         try:
  39             for main_input in input_filenames:
  40                 if verbose:
  41                     print main_input
  42                 cls.prepare_file(main_input, output_dir, verbose)
  43         except ParseError, e:
  44             print '%(file)s:%(name)s:%(message)s' % {
  45                 'file': main_input,
  46                 'name': e.__class__.__name__,
  47                 'message': e.message
  48             }
  49
  50
  51 class EpubPackager(Packager):
  52     converter = epub
  53     ext = 'epub'
  54
  55 class PdfPackager(Packager):
  56     converter = pdf
  57     ext = 'pdf'
  58
  59
  60 class GandalfEpubPackager(EpubPackager):
  61     cover = cover.GandalfCover
  62
  63 class GandalfPdfPackager(PdfPackager):
  64     cover = cover.GandalfCover
  65
  66 class BookotekaEpubPackager(EpubPackager):
  67     cover = cover.BookotekaCover
  68
  69 class PrestigioEpubPackager(EpubPackager):
  70     cover = cover.PrestigioCover
  71     flags = ('less-advertising',)
  72
  73 class PrestigioPdfPackager(PdfPackager):
  74     cover = cover.PrestigioCover
  75     flags = ('less-advertising',)
  76
  77
  78 class VirtualoEpubPackager(Packager):
  79     @staticmethod
  80     def utf_trunc(text, limit):
  81         """ truncates text to at most `limit' bytes in utf-8 """
  82         if text is None:
  83             return text
  84         if len(text.encode('utf-8')) > limit:
  85             newlimit = limit - 3
  86             while len(text.encode('utf-8')) > newlimit:
  87                 text = text[:(newlimit - len(text.encode('utf-8'))) / 4]
  88             text += '...'
  89         return text
  90
  91     @classmethod
  92     def prepare(cls, input_filenames, output_dir='', verbose=False):
  93         xml = etree.fromstring("""<?xml version="1.0" encoding="utf-8"?>
  94             <products xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"></products>""")
  95         product = etree.fromstring("""<product>
  96                 <publisherProductId></publisherProductId>
  97                 <title></title>
  98                 <info></info>
  99                 <description></description>
 100                 <authors>
 101                     <author>
 102                         <names>Jan</names>
 103                         <lastName>Kowalski</lastName>
 104                     </author>
 105                 </authors>
 106                 <price>0.0</price>
 107                 <language>PL</language>
 108             </product>""")
 109
 110         try:
 111             for main_input in input_filenames:
 112                 if verbose:
 113                     print main_input
 114                 path, fname = os.path.realpath(main_input).rsplit('/', 1)
 115                 provider = DirDocProvider(path)
 116                 slug, ext = os.path.splitext(fname)
 117
 118                 outfile_dir = os.path.join(output_dir, slug)
 119                 os.makedirs(os.path.join(output_dir, slug))
 120
 121                 doc = WLDocument.from_file(main_input, provider=provider)
 122                 info = doc.book_info
 123
 124                 product_elem = deepcopy(product)
 125                 product_elem[0].text = cls.utf_trunc(slug, 100)
 126                 product_elem[1].text = cls.utf_trunc(info.title, 255)
 127                 product_elem[2].text = cls.utf_trunc(info.description, 255)
 128                 product_elem[3].text = cls.utf_trunc(info.source_name, 3000)
 129                 product_elem[4][0][0].text = cls.utf_trunc(u' '.join(info.author.first_names), 100)
 130                 product_elem[4][0][1].text = cls.utf_trunc(info.author.last_name, 100)
 131                 xml.append(product_elem)
 132
 133                 cover.VirtualoCover(info).save(os.path.join(outfile_dir, slug+'.jpg'))
 134                 outfile = os.path.join(outfile_dir, '1.epub')
 135                 outfile_sample = os.path.join(outfile_dir, '1.sample.epub')
 136                 doc.save_output_file(epub.transform(doc),
 137                         output_path=outfile)
 138                 doc.save_output_file(epub.transform(doc, sample=25),
 139                         output_path=outfile_sample)
 140         except ParseError, e:
 141             print '%(file)s:%(name)s:%(message)s' % {
 142                 'file': main_input,
 143                 'name': e.__class__.__name__,
 144                 'message': e.message
 145             }
 146
 147         xml_file = open(os.path.join(output_dir, 'import_products.xml'), 'w')
 148         xml_file.write(etree.tostring(xml, pretty_print=True, encoding=unicode).encode('utf-8'))
 149         xml_file.close()