1 # -*- coding: utf-8 -*-
8 from librarian.parser import WLDocument
9 from librarian import XHTMLNS, ParseError
11 from lxml.etree import XMLSyntaxError, XSLTApplyError
13 ENTITY_SUBSTITUTIONS = [
22 'legacy': 'book2html.xslt',
23 'full': 'wl2html_full.xslt',
24 'partial': 'wl2html_partial.xslt'
27 def get_stylesheet(name):
28 return os.path.join(os.path.dirname(__file__), STYLESHEETS[name])
30 def substitute_entities(context, text):
31 """XPath extension function converting all entites in passed text."""
32 if isinstance(text, list):
34 for entity, substitutution in ENTITY_SUBSTITUTIONS:
35 text = text.replace(entity, substitutution)
38 # Register substitute_entities function with lxml
39 ns = etree.FunctionNamespace('http://wolnelektury.pl/functions')
40 ns['substitute_entities'] = substitute_entities
42 def transform(input, output_filename=None, is_file=True, \
43 parse_dublincore=True, stylesheet='legacy', options={}):
44 """Transforms file input_filename in XML to output_filename in XHTML."""
47 style_filename = get_stylesheet(stylesheet)
48 style = etree.parse(style_filename)
51 document = WLDocument.from_file(input, True, \
52 parse_dublincore=parse_dublincore)
54 document = WLDocument.from_string(input, True, \
55 parse_dublincore=parse_dublincore)
57 result = document.transform(style, **options)
58 del document # no longer needed large object :)
60 if etree.ETXPath('//p|//{%s}p' % str(XHTMLNS))(result) is not None:
61 add_anchors(result.getroot())
62 add_table_of_contents(result.getroot())
64 if output_filename is not None:
65 result.write(output_filename, xml_declaration=False, pretty_print=True, encoding='utf-8')
70 print "[Librarian] didn't find any paragraphs"
73 raise ValueError("'%s' is not a valid stylesheet.")
74 except (XMLSyntaxError, XSLTApplyError), e:
77 class Fragment(object):
78 def __init__(self, id, themes):
79 super(Fragment, self).__init__()
84 def append(self, event, element):
85 self.events.append((event, element))
87 def closed_events(self):
89 for event, element in self.events:
91 stack.append(('end', element))
96 print 'CLOSED NON-OPEN TAG:', element
99 return self.events + stack
103 for event, element in self.closed_events():
105 result.append(u'<%s %s>' % (element.tag, ' '.join('%s="%s"' % (k, v) for k, v in element.attrib.items())))
107 result.append(element.text)
109 result.append(u'</%s>' % element.tag)
111 result.append(element.tail)
113 result.append(element)
115 return ''.join(result)
117 def __unicode__(self):
118 return self.to_string()
121 def extract_fragments(input_filename):
122 """Extracts theme fragments from input_filename."""
124 closed_fragments = {}
126 for event, element in etree.iterparse(input_filename, events=('start', 'end')):
127 # Process begin and end elements
128 if element.get('class', '') in ('theme-begin', 'theme-end'):
129 if not event == 'end': continue # Process elements only once, on end event
132 if element.get('class', '') == 'theme-begin':
133 fragment = Fragment(id=element.get('fid'), themes=element.text)
136 if element.getparent().get('id', None) != 'book-text':
137 parents = [element.getparent()]
138 while parents[-1].getparent().get('id', None) != 'book-text':
139 parents.append(parents[-1].getparent())
142 for parent in parents:
143 fragment.append('start', parent)
145 open_fragments[fragment.id] = fragment
147 # Close existing fragment
150 fragment = open_fragments[element.get('fid')]
152 print '%s:closed not open fragment #%s' % (input_filename, element.get('fid'))
154 closed_fragments[fragment.id] = fragment
155 del open_fragments[fragment.id]
157 # Append element tail to lost_text (we don't want to lose any text)
159 for fragment_id in open_fragments:
160 open_fragments[fragment_id].append('text', element.tail)
163 # Process all elements except begin and end
165 # Omit annotation tags
166 if len(element.get('name', '')) or element.get('class', '') == 'annotation':
167 if event == 'end' and element.tail:
168 for fragment_id in open_fragments:
169 open_fragments[fragment_id].append('text', element.tail)
171 for fragment_id in open_fragments:
172 open_fragments[fragment_id].append(event, copy.copy(element))
174 return closed_fragments, open_fragments
177 def add_anchor(element, prefix, with_link=True, with_target=True, link_text=None):
179 if link_text is None:
181 anchor = etree.Element('a', href='#%s' % prefix)
182 anchor.set('class', 'anchor')
183 anchor.text = unicode(link_text)
185 anchor.tail = element.text
187 element.insert(0, anchor)
190 anchor_target = etree.Element('a', name='%s' % prefix)
191 anchor_target.set('class', 'target')
192 anchor_target.text = u' '
194 anchor_target.tail = element.text
196 element.insert(0, anchor_target)
199 def any_ancestor(element, test):
200 for ancestor in element.iterancestors():
206 def add_anchors(root):
208 for element in root.iterdescendants():
209 if any_ancestor(element, lambda e: e.get('class') in ('note', 'motto', 'motto_podpis', 'dedication')
210 or e.tag == 'blockquote'):
213 if element.tag == 'p' and 'verse' in element.get('class', ''):
214 if counter == 1 or counter % 5 == 0:
215 add_anchor(element, "f%d" % counter, link_text=counter)
217 elif 'paragraph' in element.get('class', ''):
218 add_anchor(element, "f%d" % counter, link_text=counter)
222 def add_table_of_contents(root):
225 for element in root.iterdescendants():
226 if element.tag in ('h2', 'h3'):
227 if any_ancestor(element, lambda e: e.get('id') in ('footnotes',) or e.get('class') in ('person-list',)):
230 if element.tag == 'h3' and len(sections) and sections[-1][1] == 'h2':
231 sections[-1][3].append((counter, element.tag, ''.join(element.xpath('text()')), []))
233 sections.append((counter, element.tag, ''.join(element.xpath('text()')), []))
234 add_anchor(element, "s%d" % counter, with_link=False)
237 toc = etree.Element('div')
239 toc_header = etree.SubElement(toc, 'h2')
240 toc_header.text = u'Spis treści'
241 toc_list = etree.SubElement(toc, 'ol')
243 for n, section, text, subsections in sections:
244 section_element = etree.SubElement(toc_list, 'li')
245 add_anchor(section_element, "s%d" % n, with_target=False, link_text=text)
248 subsection_list = etree.SubElement(section_element, 'ol')
249 for n, subsection, text, _ in subsections:
250 subsection_element = etree.SubElement(subsection_list, 'li')
251 add_anchor(subsection_element, "s%d" % n, with_target=False, link_text=text)