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.
9 from librarian.dcparser import Person
11 def _register_function(f):
12 """ Register extension function with lxml """
13 ns = etree.FunctionNamespace('http://wolnelektury.pl/functions')
17 def reg_substitute_entities():
18 ENTITY_SUBSTITUTIONS = [
26 def substitute_entities(context, text):
27 """XPath extension function converting all entites in passed text."""
28 if isinstance(text, list):
30 for entity, substitutution in ENTITY_SUBSTITUTIONS:
31 text = text.replace(entity, substitutution)
34 _register_function(substitute_entities)
38 def strip(context, text):
39 """Remove unneeded whitespace from beginning and end"""
40 if isinstance(text, list):
42 return re.sub(r'\s+', ' ', text).strip()
43 _register_function(strip)
46 def reg_starts_white():
47 def starts_white(context, text):
48 if isinstance(text, list):
52 return text[0].isspace()
53 _register_function(starts_white)
57 def ends_white(context, text):
58 if isinstance(text, list):
62 return text[-1].isspace()
63 _register_function(ends_white)
67 def wrap_words(context, text, wrapping):
68 """XPath extension function automatically wrapping words in passed text"""
69 if isinstance(text, list):
74 words = re.split(r'\s', text)
79 line_length += len(word) + 1
80 if line_length > wrapping:
81 # Max line length was exceeded. We create new line
83 line_length = len(word)
84 lines[-1].append(word)
85 return '\n'.join(' '.join(line) for line in lines)
86 _register_function(wrap_words)
89 def reg_person_name():
90 def person_name(context, text):
91 """ Converts "Name, Forename" to "Forename Name" """
92 if isinstance(text, list):
94 return Person.from_text(text).readable()
95 _register_function(person_name)
99 def texcommand(context, text):
100 """Remove non-letters"""
101 if isinstance(text, list):
103 return re.sub(r'[^a-zA-Z]', '', text).strip()
104 _register_function(texcommand)