X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/12b5230d8fdb3ad995e867fb5d58a69e8a627e68..2c15db814c7f40406b6a86383e3e4bc8825b7faf:/librarian/html.py diff --git a/librarian/html.py b/librarian/html.py index 6a6e3db..ed7b4d6 100644 --- a/librarian/html.py +++ b/librarian/html.py @@ -3,9 +3,10 @@ # This file is part of Librarian, licensed under GNU Affero GPLv3 or later. # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. # +from __future__ import print_function, unicode_literals + import os import re -import cStringIO import copy from lxml import etree @@ -13,6 +14,8 @@ from librarian import XHTMLNS, ParseError, OutputFile from librarian import functions from lxml.etree import XMLSyntaxError, XSLTApplyError +import six + functions.reg_substitute_entities() functions.reg_person_name() @@ -32,6 +35,16 @@ def html_has_content(text): return etree.ETXPath('//p|//{%(ns)s}p|//h1|//{%(ns)s}h1' % {'ns': str(XHTMLNS)})(text) +def transform_abstrakt(abstrakt_element): + style_filename = get_stylesheet('legacy') + style = etree.parse(style_filename) + xml = etree.tostring(abstrakt_element, encoding='unicode') + document = etree.parse(six.StringIO(xml.replace('abstrakt', 'dlugi_cytat'))) # HACK + result = document.xslt(style) + html = re.sub('', '', etree.tostring(result, encoding='unicode')) + return re.sub(']*>', '', html) + + def transform(wldoc, stylesheet='legacy', options=None, flags=None): """Transforms the WL document to XHTML. @@ -57,6 +70,7 @@ def transform(wldoc, stylesheet='legacy', options=None, flags=None): if not options: options = {} + options.setdefault('gallery', "''") result = document.transform(style, **options) del document # no longer needed large object :) @@ -65,16 +79,17 @@ def transform(wldoc, stylesheet='legacy', options=None, flags=None): add_table_of_themes(result.getroot()) add_table_of_contents(result.getroot()) - return OutputFile.from_string(etree.tostring( + return OutputFile.from_bytes(etree.tostring( result, method='html', xml_declaration=False, pretty_print=True, encoding='utf-8')) else: return None except KeyError: raise ValueError("'%s' is not a valid stylesheet.") - except (XMLSyntaxError, XSLTApplyError), e: + except (XMLSyntaxError, XSLTApplyError) as e: raise ParseError(e) +@six.python_2_unicode_compatible class Fragment(object): def __init__(self, id, themes): super(Fragment, self).__init__() @@ -94,7 +109,7 @@ class Fragment(object): try: stack.pop() except IndexError: - print 'CLOSED NON-OPEN TAG:', element + print('CLOSED NON-OPEN TAG:', element) stack.reverse() return self.events + stack @@ -116,7 +131,7 @@ class Fragment(object): return ''.join(result) - def __unicode__(self): + def __str__(self): return self.to_string() @@ -127,7 +142,7 @@ def extract_fragments(input_filename): # iterparse would die on a HTML document parser = etree.HTMLParser(encoding='utf-8') - buf = cStringIO.StringIO() + buf = six.BytesIO() buf.write(etree.tostring(etree.parse(input_filename, parser).getroot()[0][0], encoding='utf-8')) buf.seek(0) @@ -161,7 +176,7 @@ def extract_fragments(input_filename): try: fragment = open_fragments[element.get('fid')] except KeyError: - print '%s:closed not open fragment #%s' % (input_filename, element.get('fid')) + print('%s:closed not open fragment #%s' % (input_filename, element.get('fid'))) else: closed_fragments[fragment.id] = fragment del open_fragments[fragment.id] @@ -195,7 +210,7 @@ def add_anchor(element, prefix, with_link=True, with_target=True, link_text=None link_text = prefix anchor = etree.Element('a', href='#%s' % prefix) anchor.set('class', 'anchor') - anchor.text = unicode(link_text) + anchor.text = six.text_type(link_text) parent.insert(index, anchor) if with_target: @@ -216,7 +231,7 @@ def add_anchors(root): counter = 1 for element in root.iterdescendants(): def f(e): - return e.get('class') in ('note', 'motto', 'motto_podpis', 'dedication') or \ + return e.get('class') in ('note', 'motto', 'motto_podpis', 'dedication', 'frame') or \ e.get('id') == 'nota_red' or e.tag == 'blockquote' if any_ancestor(element, f): continue @@ -235,7 +250,7 @@ def raw_printable_text(element): for e in working.findall('a'): if e.get('class') in ('annotation', 'theme-begin'): e.text = '' - return etree.tostring(working, method='text', encoding=unicode).strip() + return etree.tostring(working, method='text', encoding='unicode').strip() def add_table_of_contents(root): @@ -288,7 +303,7 @@ def add_table_of_themes(root): theme_names = [s.strip() for s in fragment.text.split(',')] for theme_name in theme_names: book_themes.setdefault(theme_name, []).append(fragment.get('name')) - book_themes = book_themes.items() + book_themes = list(book_themes.items()) book_themes.sort(key=lambda s: sortify(s[0])) themes_div = etree.Element('div', id="themes") themes_ol = etree.SubElement(themes_div, 'ol') @@ -314,7 +329,7 @@ def extract_annotations(html_path): parser = etree.HTMLParser(encoding='utf-8') tree = etree.parse(html_path, parser) footnotes = tree.find('//*[@id="footnotes"]') - re_qualifier = re.compile(ur'[^\u2014]+\s+\(([^\)]+)\)\s+\u2014') + re_qualifier = re.compile(r'[^\u2014]+\s+\(([^\)]+)\)\s+\u2014') if footnotes is not None: for footnote in footnotes.findall('div'): fn_type = footnote.get('class').split('-')[1] @@ -323,8 +338,8 @@ def extract_annotations(html_path): footnote.text = None if len(footnote) and footnote[-1].tail == '\n': footnote[-1].tail = None - text_str = etree.tostring(footnote, method='text', encoding=unicode).strip() - html_str = etree.tostring(footnote, method='html', encoding=unicode).strip() + text_str = etree.tostring(footnote, method='text', encoding='unicode').strip() + html_str = etree.tostring(footnote, method='html', encoding='unicode').strip() match = re_qualifier.match(text_str) if match: