+ return unicode(etree.tostring(e, encoding="utf-8"), 'utf-8')
+
+
+def split_xml(text):
+ """Splits text into chapters.
+
+ All this stuff really must go somewhere else.
+
+ """
+ src = etree.fromstring(text)
+ chunks = []
+
+ splitter = u'naglowek_rozdzial'
+ parts = src.findall('.//naglowek_rozdzial')
+ while parts:
+ # copy the document
+ copied = deepcopy(src)
+
+ element = parts[-1]
+
+ # find the chapter's title
+ name_elem = deepcopy(element)
+ for tag in 'extra', 'motyw', 'pa', 'pe', 'pr', 'pt', 'uwaga':
+ for a in name_elem.findall('.//' + tag):
+ a.text = ''
+ del a[:]
+ name = etree.tostring(name_elem, method='text', encoding='utf-8').strip()
+
+ # in the original, remove everything from the start of the last chapter
+ parent = element.getparent()
+ del parent[parent.index(element):]
+ element, parent = parent, parent.getparent()
+ while parent is not None:
+ del parent[parent.index(element) + 1:]
+ element, parent = parent, parent.getparent()
+
+ # in the copy, remove everything before the last chapter
+ element = copied.findall('.//naglowek_rozdzial')[-1]
+ parent = element.getparent()
+ while parent is not None:
+ parent.text = None
+ while parent[0] is not element:
+ del parent[0]
+ element, parent = parent, parent.getparent()
+ chunks[:0] = [[name, unicode(etree.tostring(copied, encoding='utf-8'), 'utf-8')]]
+
+ parts = src.findall('.//naglowek_rozdzial')
+
+ chunks[:0] = [[u'początek', unicode(etree.tostring(src, encoding='utf-8'), 'utf-8')]]
+
+ for ch in chunks[1:]:
+ ch[1] = add_trim_begin(ch[1])
+ for ch in chunks[:-1]:
+ ch[1] = add_trim_end(ch[1])
+
+ return chunks
+
+
+def wl2_to_wl1(wl2_xml, slug):
+ from lxml import etree
+ import re
+ from StringIO import StringIO
+ from urllib import unquote
+ import os.path
+ from django.conf import settings
+ from fnpdjango.utils.text.slughifi import slughifi
+ from librarian import ParseError, DCNS
+
+ def _register_function(f):
+ """ Register extension function with lxml """
+ ns = etree.FunctionNamespace('http://wolnelektury.pl/functions')
+ ns[f.__name__] = f
+ return f
+
+ @_register_function
+ def slugify(context, text):
+ """Remove unneeded whitespace from beginning and end"""
+ if isinstance(text, list):
+ text = ''.join(text)
+ return slughifi(text)
+
+ @_register_function
+ def rmext(context, text):
+ if isinstance(text, list):
+ text = ''.join(text)
+ text = unquote(text)
+ if '.' in text:
+ name, ext = text.rsplit('.', 1)
+ if ext.lower() in ('doc', 'docx', 'odt', 'pdf', 'jpg', 'jpeg'):
+ text = name
+ return text
+
+ t = etree.parse(os.path.join(settings.PROJECT_ROOT, 'xslt/wl2to1.xslt'))
+ ft = wl2_xml.replace(' ', ' ')
+ f2 = StringIO(ft)
+ i1 = etree.parse(f2)
+
+ for sect in i1.findall('//section'):
+ if sect[0].text and sect[0].text.strip() == u'Przebieg zajęć':
+ # Prostujemy.
+ first = sect.find('section')
+ subs = first.findall('.//section')
+ for sub in subs:
+ sect.append(sub)
+ break
+ else:
+ dc_type = i1.findall('//dc:type', namespaces={'dc': DCNS.uri})
+ if dc_type and dc_type[0] in ('course', 'synthetic'):
+ raise ParseError('Brak przebiegu')
+
+ i1.getroot().attrib['redslug'] = slug
+ i1.getroot().attrib['wlslug'] = slug # THIS!
+ w1t = i1.xslt(t)
+ for h in w1t.findall('//aktywnosc/opis'):
+ if len(h) == 0:
+ raise ParseError('Pusty element aktywnosc/opis')
+ # FIXME assumption that every lesson has at most 9 parts
+ if not h[0].text or not re.match(r'\d\.\s', h[0].text):
+ raise ParseError('Niepoprawny nagłówek (aktywnosc/opis): %s' % repr(h[0].text))
+ h[0].text = h[0].text[3:]
+ return etree.tostring(w1t, encoding='utf-8')
\ No newline at end of file
fnp / redakcja.git / blobdiff