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