good pdf, some general updates

[librarian.git] / librarian / text.py

diff --git a/librarian/text.py b/librarian/text.py
index db0d2b2..b7fbad6 100644 (file)
--- a/librarian/text.py
+++ b/librarian/text.py
@@ -1,89 +1,66 @@
 # -*- 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, functions
+from lxml import etree
 import cStringIO
-import re
 import codecs
-
-from lxml import etree
-
-from librarian import dcparser
-
-
-ENTITY_SUBSTITUTIONS = [
-    (u'---', u'—'),
-    (u'--', u'–'),
-    (u'...', u'…'),
-    (u',,', u'„'),
-    (u'"', u'”'),
-]
-
-
-MAX_LINE_LENGTH = 80
-
-
-def strip(context, text):
-    """Remove unneeded whitespace from beginning and end"""
-    if isinstance(text, list):
-        text = ''.join(text)
-    return re.sub(r'\s+', ' ', text).strip()
+import os
+import re
 
 
-def substitute_entities(context, text):
-    """XPath extension function converting all entites in passed text."""
-    if isinstance(text, list):
-        text = ''.join(text)
-    for entity, substitutution in ENTITY_SUBSTITUTIONS:
-        text = text.replace(entity, substitutution)
-    return text
+functions.reg_substitute_entities()
+functions.reg_wrap_words()
+functions.reg_split()
 
+TEMPLATE = u"""\
+Kodowanie znaków w dokumencie: UTF-8.
+-----
+Publikacja zrealizowana w ramach projektu Wolne Lektury (http://wolnelektury.pl/). Reprodukcja cyfrowa wykonana przez
+Bibliotekę Narodową z egzemplarza pochodzącego ze zbiorów BN. 
+\n%(license_description)s.
 
-def wrap_words(context, text):
-    """XPath extension function automatically wrapping words in passed text"""
-    if isinstance(text, list):
-        text = ''.join(text)
-    words = re.split(r'\s', text)
-    
-    line_length = 0
-    lines = [[]]
-    for word in words:
-        line_length += len(word) + 1
-        if line_length > MAX_LINE_LENGTH:
-            # Max line length was exceeded. We create new line
-            lines.append([])
-            line_length = len(word)
-        lines[-1].append(word)
-    return '\n'.join(' '.join(line) for line in lines)
+Wersja lektury w opracowaniu merytorycznym i krytycznym (przypisy i motywy) dostępna jest na stronie %(url)s.
+-----
 
 
-# Register substitute_entities function with lxml
-ns = etree.FunctionNamespace('http://wolnelektury.pl/functions')
-ns['strip'] = strip
-ns['substitute_entities'] = substitute_entities
-ns['wrap_words'] = wrap_words
 
+%(text)s
+"""
 
-def transform(input_filename, output_filename):
+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)
 
-    doc_file = cStringIO.StringIO()
-    expr = re.compile(r'/\s', re.MULTILINE | re.UNICODE);
-    
-    f = open(input_filename, 'r')
-    for line in f:
-        line = line.decode('utf-8')
-        line = expr.sub(u'<br/>\n', line)
-        doc_file.write(line.encode('utf-8'))
-    f.close()
+    if is_file:
+        document = parser.WLDocument.from_file(input_filename, True, parse_dublincore=parse_dublincore)
+    else:
+        document = parser.WLDocument.from_string(input_filename, True, parse_dublincore=parse_dublincore)
 
-    doc_file.seek(0)
+    result = document.transform(style, **options)
 
-    parser = etree.XMLParser(remove_blank_text=True)
-    doc = etree.parse(doc_file, parser)
-    
-    result = doc.xslt(style)
     output_file = codecs.open(output_filename, 'wb', encoding='utf-8')
-    output_file.write(unicode(result) % dcparser.parse(input_filename).url)
+
+    if parse_dublincore:
+        parsed_dc = dcparser.parse(input_filename)
+        url = parsed_dc.url
+        license_description = parsed_dc.license_description
+       license = parsed_dc.license
+        if license:
+            license_description = u"Ten utwór jest udostepniony na licencji %s: \n%s" % (license_description, license)        
+        else:
+            license_description = u"Ten utwór nie jest chroniony prawem autorskim i znajduje się w domenie publicznej, co oznacza, że możesz go swobodnie wykorzystywać, publikować i rozpowszechniać" 
+    else:
+        url = '*' * 10
+        license = ""
+        license_description = ""
+    output_file.write(TEMPLATE % {
+        'url': url,
+       'license_description': license_description,
+        'text': unicode(result),
+    })