X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/6642c1c71c5c6ce6ef3401c8c9da84cf076b018b..e82ceed3d3d72dc75697449285a67e0a42a9cc56:/librarian/text.py diff --git a/librarian/text.py b/librarian/text.py index 0754a99..d54c36b 100644 --- a/librarian/text.py +++ b/librarian/text.py @@ -1,12 +1,14 @@ # -*- coding: utf-8 -*- -import os +# +# This file is part of Librarian, licensed under GNU Affero GPLv3 or later. +# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information. +# +from librarian import dcparser, parser +from lxml import etree import cStringIO -import re import codecs - -from lxml import etree - -from librarian import dcparser +import os +import re ENTITY_SUBSTITUTIONS = [ @@ -78,22 +80,27 @@ ns['substitute_entities'] = substitute_entities ns['wrap_words'] = wrap_words -def transform(input_filename, output_filename, **options): +def transform(input_filename, output_filename, is_file=True, parse_dublincore=True, **options): """Transforms file input_filename in XML to output_filename in TXT.""" # Parse XSLT - style_filename = os.path.join(os.path.dirname(__file__), 'book2txt.xslt') + style_filename = os.path.join(os.path.dirname(__file__), 'xslt/book2txt.xslt') style = etree.parse(style_filename) if is_file: - document = WLDocument.from_file(input, True) + document = parser.WLDocument.from_file(input_filename, True, parse_dublincore=parse_dublincore) else: - document = WLDocument.from_string(input, True) + document = parser.WLDocument.from_string(input_filename, True, parse_dublincore=parse_dublincore) result = document.transform(style, **options) output_file = codecs.open(output_filename, 'wb', encoding='utf-8') + + if parse_dublincore: + url = dcparser.parse(input_filename).url + else: + url = '*' * 10 output_file.write(TEMPLATE % { - 'url': dcparser.parse(input_filename).url, + 'url': url, 'text': unicode(result), })