src/librarian/builders/html.py

   1 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
   2 # Copyright © Fundacja Wolne Lektury. See NOTICE for more information.
   3 #
   4 from collections import defaultdict
   5 import os
   6 from urllib.request import urlopen
   7 from lxml import etree
   8 from librarian.html import add_table_of_contents, add_table_of_themes, add_image_sizes
   9 from librarian import OutputFile
  10
  11
  12 class HtmlBuilder:
  13     file_extension = "html"
  14     with_themes = True
  15     with_toc = True
  16     with_footnotes = True
  17     with_nota_red = True
  18     with_ids = True
  19     with_numbering = True
  20     no_externalities = False
  21     orphans = True
  22
  23     root_tag = 'div'
  24     root_attrib = {'id': 'book-text'}
  25
  26     def __init__(self, gallery_path=None, gallery_url=None, base_url=None):
  27         self._base_url = base_url
  28         self.gallery_path = gallery_path
  29         self.gallery_url = gallery_url
  30
  31         self.tree = text = etree.Element(self.root_tag, **self.root_attrib)
  32         self.header = etree.Element('h1')
  33
  34         self.footnotes = etree.Element('div', id='footnotes')
  35
  36         self.nota_red = etree.Element('div', id='nota_red')
  37
  38         self.cursors = {
  39             None: text,
  40             'header': self.header,
  41             'footnotes': self.footnotes,
  42             'nota_red': self.nota_red,
  43         }
  44         self.current_cursors = [text]
  45
  46     @property
  47     def base_url(self):
  48         if self._base_url is not None:
  49             return self._base_url
  50         else:
  51             return 'https://wolnelektury.pl/media/book/pictures/{}/'.format(self.document.meta.url.slug)
  52
  53     @property
  54     def cursor(self):
  55         return self.current_cursors[-1]
  56
  57     def enter_fragment(self, fragment):
  58         cursor = self.cursors.get(fragment, self.cursor)
  59         self.current_cursors.append(cursor)
  60
  61     def exit_fragment(self):
  62         self.current_cursors.pop()
  63
  64     def create_fragment(self, name, element):
  65         assert name not in self.cursors
  66         self.cursors[name] = element
  67
  68     def forget_fragment(self, name):
  69         del self.cursors[name]
  70
  71     def build(self, document, element=None, **kwargs):
  72         self.document = document
  73         self.document.assign_ids()
  74         self.prepare_images()
  75
  76         if element is None:
  77             element = document.tree.getroot()
  78
  79         element.html_build(self)
  80         self.postprocess(document)
  81         return self.output()
  82
  83     def prepare_images(self):
  84         # Temporarily use the legacy method, before transitioning to external generators.
  85         if self.gallery_path is None:
  86             return
  87         try:
  88             os.makedirs(self.gallery_path)
  89         except OSError:
  90             pass
  91         add_image_sizes(self.document.tree, self.gallery_path, self.gallery_url, self.base_url)
  92
  93     def output(self):
  94         if not len(self.tree):
  95             return None
  96         return OutputFile.from_bytes(
  97             etree.tostring(
  98                 self.tree,
  99                 method='html',
 100                 encoding='utf-8',
 101                 pretty_print=True
 102             )
 103         )
 104
 105     def postprocess(self, document):
 106         _ = document.tree.getroot().gettext
 107
 108         if document.meta.translators:
 109             self.enter_fragment('header')
 110             self.start_element('span', {'class': 'translator'})
 111             self.push_text(_("translated by") + " ")
 112             self.push_text(
 113                 ", ".join(
 114                     translator.readable()
 115                     for translator in document.meta.translators
 116                 )
 117             )
 118             self.exit_fragment()
 119
 120         if len(self.header):
 121             self.tree.insert(0, self.header)
 122
 123         if self.with_nota_red and len(self.nota_red):
 124             self.tree.append(self.nota_red)
 125         if self.with_themes:
 126             add_table_of_themes(self.tree)
 127         if self.with_toc:
 128             add_table_of_contents(self.tree)
 129
 130         if len(self.footnotes):
 131             fnheader = etree.Element("h3")
 132             fnheader.text = _("Footnotes")
 133             self.footnotes.insert(0, fnheader)
 134             self.tree.append(self.footnotes)
 135
 136     def start_element(self, tag, attrib=None):
 137         self.current_cursors.append(etree.SubElement(
 138             self.cursor,
 139             tag,
 140             **(attrib or {})
 141         ))
 142
 143     def end_element(self):
 144         self.current_cursors.pop()
 145
 146     def push_text(self, text):
 147         cursor = self.cursor
 148         if len(cursor):
 149             cursor[-1].tail = (cursor[-1].tail or '') + text
 150         else:
 151             cursor.text = (cursor.text or '') + text
 152
 153     def add_visible_number(self, element):
 154         assert '_id' in element.attrib, etree.tostring(element)
 155         self.start_element('a', {
 156             'href': f'#{element.attrib["_id"]}',
 157             'class': 'wl-num',
 158         })
 159         self.push_text(element.attrib['_visible_numbering'])
 160         self.end_element()
 161
 162
 163 class StandaloneHtmlBuilder(HtmlBuilder):
 164     css_url = "https://static.wolnelektury.pl/css/compressed/book_text.css"
 165
 166     def postprocess(self, document):
 167         super(StandaloneHtmlBuilder, self).postprocess(document)
 168
 169         tree = etree.Element('html')
 170         body = etree.SubElement(tree, 'body')
 171         body.append(self.tree)
 172         self.tree = tree
 173
 174         head = etree.Element('head')
 175         tree.insert(0, head)
 176
 177         etree.SubElement(head, 'meta', charset='utf-8')
 178         etree.SubElement(head, 'title').text = document.meta.title
 179
 180         etree.SubElement(
 181             head,
 182             'meta',
 183             name="viewport",
 184             content="width=device-width, initial-scale=1, maximum-scale=1"
 185         )
 186
 187         if self.no_externalities:
 188             etree.SubElement(
 189                 head, 'style',
 190             ).text = urlopen(self.css_url).read().decode('utf-8')
 191         else:
 192             etree.SubElement(
 193                 head,
 194                 'link',
 195                 href=self.css_url,
 196                 rel="stylesheet",
 197                 type="text/css",
 198             )
 199
 200             etree.SubElement(
 201                 body, 'script',
 202                 src="https://ajax.googleapis.com/ajax/libs/jquery/1/jquery.min.js"
 203             )
 204
 205             etree.SubElement(
 206                 body,
 207                 "script",
 208                 src="http://malsup.github.io/min/jquery.cycle2.min.js"
 209             )
 210
 211
 212 class SnippetHtmlBuilder(HtmlBuilder):
 213     with_themes = False
 214     with_toc = False
 215     with_footnotes = False
 216     with_nota_red = False
 217     with_ids = False
 218     with_numbering = False
 219
 220
 221 class AbstraktHtmlBuilder(HtmlBuilder):
 222     with_themes = False
 223     with_toc = False
 224     with_footnotes = False
 225     with_nota_red = False
 226     with_ids = False
 227     with_numbering = False
 228
 229     root_tag = 'blockquote'
 230     root_attrib = {}
 231
 232     def build(self, document, element=None, **kwargs):
 233         if element is None:
 234             element = document.tree.find('//abstrakt')
 235         if element is None:
 236             return OutputFile.from_bytes(b'')
 237         element.attrib['_force'] = '1'
 238         return super().build(document, element, **kwargs)
 239
 240
 241 class DaisyHtmlBuilder(StandaloneHtmlBuilder):
 242     file_extension = 'xhtml'
 243     with_themes = False
 244     with_toc = False
 245     with_footnotes = False
 246     with_nota_red = False
 247     with_deep_identifiers = False
 248     no_externalities = True
 249     with_numbering = False
 250
 251     def output(self):
 252         tree = etree.ElementTree(self.tree)
 253         tree.docinfo.public_id = '-//W3C//DTD XHTML 1.0 Transitional//EN'
 254         tree.docinfo.system_url = 'http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd'
 255         return OutputFile.from_bytes(
 256             etree.tostring(
 257                 tree,
 258                 encoding='utf-8',
 259                 pretty_print=True,
 260                 xml_declaration=True
 261             )
 262         )
 263