1 # -*- coding: utf-8 -*-
3 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
4 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
6 from __future__ import unicode_literals
8 from librarian import ValidationError, NoDublinCore, ParseError, NoProvider
9 from librarian import RDFNS
10 from librarian.cover import make_cover
11 from librarian import dcparser
13 from xml.parsers.expat import ExpatError
14 from lxml import etree
15 from lxml.etree import XMLSyntaxError, XSLTApplyError
22 from .elements import WL_ELEMENTS
25 class WLElementLookup(etree.CustomElementClassLookup):
26 def lookup(self, node_type, document, namespace, name):
27 if node_type != 'element':
31 return WL_ELEMENTS[name]
34 parser = etree.XMLParser()
35 parser.set_element_class_lookup(
41 class WLDocument(object):
42 """Legacy class, to be replaced with documents.WLDocument."""
43 LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE)
46 def __init__(self, edoc, parse_dublincore=True, provider=None,
47 strict=False, meta_fallbacks=None):
49 self.provider = provider
51 root_elem = edoc.getroot()
53 dc_path = './/' + RDFNS('RDF')
55 if root_elem.tag != 'utwor':
56 raise ValidationError(
57 "Invalid root element. Found '%s', should be 'utwor'"
62 self.rdf_elem = root_elem.find(dc_path)
64 if self.rdf_elem is None:
66 "Document must have a '%s' element." % RDFNS('RDF')
69 self.book_info = dcparser.BookInfo.from_element(
70 self.rdf_elem, fallbacks=meta_fallbacks, strict=strict)
75 def from_bytes(cls, xml, *args, **kwargs):
76 return cls.from_file(six.BytesIO(xml), *args, **kwargs)
79 def from_file(cls, xmlfile, *args, **kwargs):
81 # first, prepare for parsing
82 if isinstance(xmlfile, six.text_type):
83 file = open(xmlfile, 'rb')
91 if not isinstance(data, six.text_type):
92 data = data.decode('utf-8')
94 data = data.replace(u'\ufeff', '')
97 parser = etree.XMLParser(remove_blank_text=False)
98 tree = etree.parse(six.BytesIO(data.encode('utf-8')), parser)
100 return cls(tree, *args, **kwargs)
101 except (ExpatError, XMLSyntaxError, XSLTApplyError) as e:
104 def swap_endlines(self):
105 """Converts line breaks in stanzas into <br/> tags."""
106 # only swap inside stanzas
107 for elem in self.edoc.iter('strofa'):
108 for child in list(elem):
110 chunks = self.LINE_SWAP_EXPR.split(child.tail)
111 ins_index = elem.index(child) + 1
112 while len(chunks) > 1:
113 ins = etree.Element('br')
114 ins.tail = chunks.pop()
115 elem.insert(ins_index, ins)
116 child.tail = chunks.pop(0)
118 chunks = self.LINE_SWAP_EXPR.split(elem.text)
119 while len(chunks) > 1:
120 ins = etree.Element('br')
121 ins.tail = chunks.pop()
123 elem.text = chunks.pop(0)
126 if self.provider is None:
127 raise NoProvider('No document provider supplied.')
128 if self.book_info is None:
129 raise NoDublinCore('No Dublin Core in document.')
130 for part_uri in self.book_info.parts:
131 yield self.from_file(
132 self.provider.by_uri(part_uri), provider=self.provider
135 def chunk(self, path):
136 # convert the path to XPath
137 expr = self.path_to_xpath(path)
138 elems = self.edoc.xpath(expr)
145 def path_to_xpath(self, path):
148 for part in path.split('/'):
149 match = re.match(r'([^\[]+)\[(\d+)\]', part)
153 tag, n = match.groups()
154 parts.append("*[%d][name() = '%s']" % (int(n)+1, tag))
159 return '/'.join(parts)
161 def transform(self, stylesheet, **options):
162 return self.edoc.xslt(stylesheet, **options)
166 parent = self.rdf_elem.getparent()
167 parent.replace(self.rdf_elem, self.book_info.to_etree(parent))
171 return etree.tostring(self.edoc, encoding='unicode', pretty_print=True)
173 def merge_chunks(self, chunk_dict):
176 for key, data in chunk_dict.iteritems():
178 xpath = self.path_to_xpath(key)
179 node = self.edoc.xpath(xpath)[0]
180 repl = etree.fromstring(
181 "<%s>%s</%s>" % (node.tag, data, node.tag)
183 node.getparent().replace(node, repl)
184 except Exception as e:
185 unmerged.append(repr((key, xpath, e)))
189 def clean_ed_note(self, note_tag='nota_red'):
190 """ deletes forbidden tags from nota_red """
192 for node in self.edoc.xpath('|'.join(
193 '//%s//%s' % (note_tag, tag) for tag in
194 ('pa', 'pe', 'pr', 'pt', 'begin', 'end', 'motyw'))):
201 """Returns a set of all editors for book and its children.
203 :returns: set of dcparser.Person objects
205 if self.book_info is None:
206 raise NoDublinCore('No Dublin Core in document.')
207 persons = set(self.book_info.editors
208 + self.book_info.technical_editors)
209 for child in self.parts():
210 persons.update(child.editors())
217 def as_html(self, *args, **kwargs):
218 from librarian import html
219 return html.transform(self, *args, **kwargs)
221 def as_text(self, *args, **kwargs):
222 from librarian import text
223 return text.transform(self, *args, **kwargs)
225 def as_epub(self, *args, **kwargs):
226 from librarian import epub
227 return epub.transform(self, *args, **kwargs)
229 def as_pdf(self, *args, **kwargs):
230 from librarian import pdf
231 return pdf.transform(self, *args, **kwargs)
233 def as_mobi(self, *args, **kwargs):
234 from librarian import mobi
235 return mobi.transform(self, *args, **kwargs)
237 def as_fb2(self, *args, **kwargs):
238 from librarian import fb2
239 return fb2.transform(self, *args, **kwargs)
241 def as_cover(self, cover_class=None, *args, **kwargs):
242 if cover_class is None:
243 cover_class = make_cover
244 return cover_class(self.book_info, *args, **kwargs).output_file()
247 def latex_dir(self, *args, **kwargs):
248 kwargs['latex_dir'] = True
249 from librarian import pdf
250 return pdf.transform(self, *args, **kwargs)
252 def save_output_file(self, output_file, output_path=None,
253 output_dir_path=None, make_author_dir=False,
256 save_path = output_dir_path
258 save_path = os.path.join(
260 six.text_type(self.book_info.author).encode('utf-8')
262 save_path = os.path.join(save_path, self.book_info.url.slug)
264 save_path += '.%s' % ext
266 save_path = output_path
268 output_file.save_as(save_path)