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         self.counters = defaultdict(lambda: 1)
  36
  37         self.nota_red = etree.Element('div', id='nota_red')
  38
  39         self.cursors = {
  40             None: text,
  41             'header': self.header,
  42             'footnotes': self.footnotes,
  43             'nota_red': self.nota_red,
  44         }
  45         self.current_cursors = [text]
  46
  47     @property
  48     def base_url(self):
  49         if self._base_url is not None:
  50             return self._base_url
  51         else:
  52             return 'https://wolnelektury.pl/media/book/pictures/{}/'.format(self.document.meta.url.slug)
  53
  54     @property
  55     def cursor(self):
  56         return self.current_cursors[-1]
  57
  58     def enter_fragment(self, fragment):
  59         cursor = self.cursors.get(fragment, self.cursor)
  60         self.current_cursors.append(cursor)
  61
  62     def exit_fragment(self):
  63         self.current_cursors.pop()
  64
  65     def create_fragment(self, name, element):
  66         assert name not in self.cursors
  67         self.cursors[name] = element
  68
  69     def forget_fragment(self, name):
  70         del self.cursors[name]
  71
  72     def build(self, document, element=None, **kwargs):
  73         self.document = document
  74
  75         self.assign_ids(self.document.tree)
  76         self.prepare_images()
  77
  78         if element is None:
  79             element = document.tree.getroot()
  80
  81         element.html_build(self)
  82         self.postprocess(document)
  83         return self.output()
  84
  85     def assign_ids(self, tree):
  86         # Assign IDs depth-first, to account for any <numeracja> inside.
  87         for _e, elem in etree.iterwalk(tree, events=('end',)):
  88             if getattr(elem, 'NUMBERING', None):
  89                 elem.assign_id(self)
  90
  91     def prepare_images(self):
  92         # Temporarily use the legacy method, before transitioning to external generators.
  93         if self.gallery_path is None:
  94             return
  95         try:
  96             os.makedirs(self.gallery_path)
  97         except OSError:
  98             pass
  99         add_image_sizes(self.document.tree, self.gallery_path, self.gallery_url, self.base_url)
 100
 101     def output(self):
 102         if not len(self.tree):
 103             return None
 104         return OutputFile.from_bytes(
 105             etree.tostring(
 106                 self.tree,
 107                 method='html',
 108                 encoding='utf-8',
 109                 pretty_print=True
 110             )
 111         )
 112
 113     def postprocess(self, document):
 114         _ = document.tree.getroot().gettext
 115
 116         if document.meta.translators:
 117             self.enter_fragment('header')
 118             self.start_element('span', {'class': 'translator'})
 119             self.push_text(_("translated by") + " ")
 120             self.push_text(
 121                 ", ".join(
 122                     translator.readable()
 123                     for translator in document.meta.translators
 124                 )
 125             )
 126             self.exit_fragment()
 127
 128         if len(self.header):
 129             self.tree.insert(0, self.header)
 130
 131         if self.with_nota_red and len(self.nota_red):
 132             self.tree.append(self.nota_red)
 133         if self.with_themes:
 134             add_table_of_themes(self.tree)
 135         if self.with_toc:
 136             add_table_of_contents(self.tree)
 137
 138         if self.counters['fn'] > 1:
 139             fnheader = etree.Element("h3")
 140             fnheader.text = _("Footnotes")
 141             self.footnotes.insert(0, fnheader)
 142             self.tree.append(self.footnotes)
 143
 144     def start_element(self, tag, attrib=None):
 145         self.current_cursors.append(etree.SubElement(
 146             self.cursor,
 147             tag,
 148             **(attrib or {})
 149         ))
 150
 151     def end_element(self):
 152         self.current_cursors.pop()
 153
 154     def push_text(self, text):
 155         cursor = self.cursor
 156         if len(cursor):
 157             cursor[-1].tail = (cursor[-1].tail or '') + text
 158         else:
 159             cursor.text = (cursor.text or '') + text
 160
 161     def add_visible_number(self, element):
 162         assert '_id' in element.attrib, etree.tostring(element)
 163         self.start_element('a', {
 164             'href': f'#{element.attrib["_id"]}',
 165             'class': 'wl-num',
 166         })
 167         self.push_text(element.attrib['_visible_numbering'])
 168         self.end_element()
 169
 170
 171 class StandaloneHtmlBuilder(HtmlBuilder):
 172     css_url = "https://static.wolnelektury.pl/css/compressed/book_text.css"
 173
 174     def postprocess(self, document):
 175         super(StandaloneHtmlBuilder, self).postprocess(document)
 176
 177         tree = etree.Element('html')
 178         body = etree.SubElement(tree, 'body')
 179         body.append(self.tree)
 180         self.tree = tree
 181
 182         head = etree.Element('head')
 183         tree.insert(0, head)
 184
 185         etree.SubElement(head, 'meta', charset='utf-8')
 186         etree.SubElement(head, 'title').text = document.meta.title
 187
 188         etree.SubElement(
 189             head,
 190             'meta',
 191             name="viewport",
 192             content="width=device-width, initial-scale=1, maximum-scale=1"
 193         )
 194
 195         if self.no_externalities:
 196             etree.SubElement(
 197                 head, 'style',
 198             ).text = urlopen(self.css_url).read().decode('utf-8')
 199         else:
 200             etree.SubElement(
 201                 head,
 202                 'link',
 203                 href=self.css_url,
 204                 rel="stylesheet",
 205                 type="text/css",
 206             )
 207
 208             etree.SubElement(
 209                 body, 'script',
 210                 src="https://ajax.googleapis.com/ajax/libs/jquery/1/jquery.min.js"
 211             )
 212
 213             etree.SubElement(
 214                 body,
 215                 "script",
 216                 src="http://malsup.github.io/min/jquery.cycle2.min.js"
 217             )
 218
 219
 220 class SnippetHtmlBuilder(HtmlBuilder):
 221     with_themes = False
 222     with_toc = False
 223     with_footnotes = False
 224     with_nota_red = False
 225     with_ids = False
 226     with_numbering = False
 227
 228
 229 class AbstraktHtmlBuilder(HtmlBuilder):
 230     with_themes = False
 231     with_toc = False
 232     with_footnotes = False
 233     with_nota_red = False
 234     with_ids = False
 235     with_numbering = False
 236
 237     root_tag = 'blockquote'
 238     root_attrib = {}
 239
 240     def build(self, document, element=None, **kwargs):
 241         if element is None:
 242             element = document.tree.find('//abstrakt')
 243         if element is None:
 244             return OutputFile.from_bytes(b'')
 245         element.attrib['_force'] = '1'
 246         return super().build(document, element, **kwargs)
 247
 248
 249 class DaisyHtmlBuilder(StandaloneHtmlBuilder):
 250     file_extension = 'xhtml'
 251     with_themes = False
 252     with_toc = False
 253     with_footnotes = False
 254     with_nota_red = False
 255     with_deep_identifiers = False
 256     no_externalities = True
 257     with_numbering = False
 258
 259     def output(self):
 260         tree = etree.ElementTree(self.tree)
 261         tree.docinfo.public_id = '-//W3C//DTD XHTML 1.0 Transitional//EN'
 262         tree.docinfo.system_url = 'http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd'
 263         return OutputFile.from_bytes(
 264             etree.tostring(
 265                 tree,
 266                 encoding='utf-8',
 267                 pretty_print=True,
 268                 xml_declaration=True
 269             )
 270         )
 271