-# coding: utf-8
-from __future__ import unicode_literals
-
-try:
- from urllib.request import urlopen
-except ImportError:
- from urllib2 import urlopen
+# This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
+# Copyright © Fundacja Wolne Lektury. See NOTICE for more information.
+#
+from collections import defaultdict
+import os
+from urllib.request import urlopen
from lxml import etree
-from librarian.html import add_anchors, add_table_of_contents, add_table_of_themes
+from librarian.html import add_table_of_contents, add_table_of_themes, add_image_sizes
from librarian import OutputFile
class HtmlBuilder:
file_extension = "html"
- with_anchors = True
with_themes = True
with_toc = True
with_footnotes = True
with_nota_red = True
+ with_ids = True
+ with_numbering = True
no_externalities = False
+ orphans = True
+
+ root_tag = 'div'
+ root_attrib = {'id': 'book-text'}
- def __init__(self, image_location='https://wolnelektury.pl/media/book/pictures/marcos-historia-kolorow/'):
- self.image_location = image_location
+ def __init__(self, gallery_path=None, gallery_url=None, base_url=None):
+ self._base_url = base_url
+ self.gallery_path = gallery_path
+ self.gallery_url = gallery_url
- self.tree = text = etree.Element('div', **{'id': 'book-text'})
- self.header = etree.SubElement(text, 'h1')
+ self.tree = text = etree.Element(self.root_tag, **self.root_attrib)
+ self.header = etree.Element('h1')
self.footnotes = etree.Element('div', id='footnotes')
- self.footnote_counter = 0
+ self.counters = defaultdict(lambda: 1)
self.nota_red = etree.Element('div', id='nota_red')
}
self.current_cursors = [text]
+ @property
+ def base_url(self):
+ if self._base_url is not None:
+ return self._base_url
+ else:
+ return 'https://wolnelektury.pl/media/book/pictures/{}/'.format(self.document.meta.url.slug)
+
@property
def cursor(self):
return self.current_cursors[-1]
def enter_fragment(self, fragment):
- self.current_cursors.append(self.cursors[fragment])
+ cursor = self.cursors.get(fragment, self.cursor)
+ self.current_cursors.append(cursor)
def exit_fragment(self):
self.current_cursors.pop()
def forget_fragment(self, name):
del self.cursors[name]
- def preprocess(self, document):
- document._compat_assign_ordered_ids()
- document._compat_assign_section_ids()
+ def build(self, document, element=None, **kwargs):
+ self.document = document
+
+ self.assign_ids(self.document.tree)
+ self.prepare_images()
- def build(self, document, **kwargs):
- self.preprocess(document)
- document.tree.getroot().html_build(self)
+ if element is None:
+ element = document.tree.getroot()
+
+ element.html_build(self)
self.postprocess(document)
return self.output()
+ def assign_ids(self, tree):
+ # Assign IDs depth-first, to account for any <numeracja> inside.
+ for _e, elem in etree.iterwalk(tree, events=('end',)):
+ if getattr(elem, 'NUMBERING', None):
+ elem.assign_id(self)
+
+ def prepare_images(self):
+ # Temporarily use the legacy method, before transitioning to external generators.
+ if self.gallery_path is None:
+ return
+ try:
+ os.makedirs(self.gallery_path)
+ except OSError:
+ pass
+ add_image_sizes(self.document.tree, self.gallery_path, self.gallery_url, self.base_url)
+
def output(self):
+ if not len(self.tree):
+ return None
return OutputFile.from_bytes(
etree.tostring(
self.tree,
)
def postprocess(self, document):
- _ = document.tree.getroot().master.gettext
+ _ = document.tree.getroot().gettext
if document.meta.translators:
self.enter_fragment('header')
)
self.exit_fragment()
- if self.with_anchors:
- add_anchors(self.tree)
+ if len(self.header):
+ self.tree.insert(0, self.header)
+
if self.with_nota_red and len(self.nota_red):
self.tree.append(self.nota_red)
if self.with_themes:
if self.with_toc:
add_table_of_contents(self.tree)
- if self.footnote_counter:
+ if self.counters['fn'] > 1:
fnheader = etree.Element("h3")
fnheader.text = _("Footnotes")
self.footnotes.insert(0, fnheader)
else:
cursor.text = (cursor.text or '') + text
+ def add_visible_number(self, element):
+ assert '_id' in element.attrib, etree.tostring(element)
+ self.start_element('a', {
+ 'href': f'#{element.attrib["_id"]}',
+ 'class': 'wl-num',
+ })
+ self.push_text(element.attrib['_visible_numbering'])
+ self.end_element()
+
class StandaloneHtmlBuilder(HtmlBuilder):
css_url = "https://static.wolnelektury.pl/css/compressed/book_text.css"
head = etree.Element('head')
tree.insert(0, head)
-
etree.SubElement(head, 'meta', charset='utf-8')
etree.SubElement(head, 'title').text = document.meta.title
)
+class SnippetHtmlBuilder(HtmlBuilder):
+ with_themes = False
+ with_toc = False
+ with_footnotes = False
+ with_nota_red = False
+ with_ids = False
+ with_numbering = False
+
+
+class AbstraktHtmlBuilder(HtmlBuilder):
+ with_themes = False
+ with_toc = False
+ with_footnotes = False
+ with_nota_red = False
+ with_ids = False
+ with_numbering = False
+
+ root_tag = 'blockquote'
+ root_attrib = {}
+
+ def build(self, document, element=None, **kwargs):
+ if element is None:
+ element = document.tree.find('//abstrakt')
+ element.attrib['_force'] = '1'
+ return super().build(document, element, **kwargs)
+
+
class DaisyHtmlBuilder(StandaloneHtmlBuilder):
file_extension = 'xhtml'
- with_anchors = False
with_themes = False
with_toc = False
with_footnotes = False
with_nota_red = False
with_deep_identifiers = False
no_externalities = True
+ with_numbering = False
def output(self):
tree = etree.ElementTree(self.tree)
fnp / librarian.git / blobdiff