encode unicode for lxml

[librarian.git] / librarian / __init__.py

diff --git a/librarian/__init__.py b/librarian/__init__.py
index 5997a4e..5c4a355 100644 (file)
--- a/librarian/__init__.py
+++ b/librarian/__init__.py
@@ -1,14 +1,12 @@
 # -*- coding: utf-8 -*-
-# exception classes
+#
+# This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
+# Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
+#
+import os
 
 class ParseError(Exception):
-    
-    def __init__(self, cause, message=None):
-        self.cause = cause
-        try:
-            self.message = message or self.cause.message
-        except:
-            self.message = "No message."
+    pass
 
 class ValidationError(Exception):
     pass
@@ -26,7 +24,7 @@ class XMLNamespace(object):
         return '{%s}%s' % (self.uri, tag)
 
     def __contains__(self, tag):
-        return tag.startswith('{'+str(self)+'}')
+        return tag.startswith('{' + str(self) + '}')
 
     def __repr__(self):
         return 'XMLNamespace(%r)' % self.uri
@@ -46,14 +44,43 @@ RDFNS = XMLNamespace('http://www.w3.org/1999/02/22-rdf-syntax-ns#')
 DCNS = XMLNamespace('http://purl.org/dc/elements/1.1/')
 XINS = XMLNamespace("http://www.w3.org/2001/XInclude")
 XHTMLNS = XMLNamespace("http://www.w3.org/1999/xhtml")
+NCXNS = XMLNamespace("http://www.daisy.org/z3986/2005/ncx/")
+OPFNS = XMLNamespace("http://www.idpf.org/2007/opf")
 
 WLNS = EmptyNamespace()
 
+
+class DocProvider(object):
+    """ Base class for a repository of XML files.
+        Used for generating joined files, like EPUBs
+    """
+
+    def by_slug(self, slug):
+        raise NotImplemented
+
+    def __getitem__(self, slug):
+        return self.by_slug(slug)
+
+    def by_uri(self, uri):
+        return self.by_slug(uri.rsplit('/', 1)[1])
+
+
+class DirDocProvider(DocProvider):
+    """ Serve docs from a directory of files in form <slug>.xml """
+
+    def __init__(self, dir):
+        self.dir = dir
+        self.files = {}
+
+    def by_slug(self, slug):
+        return open(os.path.join(self.dir, '%s.xml' % slug))
+
+
 import lxml.etree as etree
 import dcparser
 
 DEFAULT_BOOKINFO = dcparser.BookInfo(
-        { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'},\
+        { RDFNS('about'): u'http://wiki.wolnepodreczniki.pl/Lektury:Template'}, \
         { DCNS('creator'): [u'Some, Author'],
           DCNS('title'): [u'Some Title'],
           DCNS('subject.period'): [u'Unknown'],
@@ -73,16 +100,37 @@ DEFAULT_BOOKINFO = dcparser.BookInfo(
             [u"Domena publiczna - zm. [OPIS STANU PRAWNEGO TEKSTU]"] })
 
 def xinclude_forURI(uri):
-    e = etree.Element( XINS("include") )
+    e = etree.Element(XINS("include"))
     e.set("href", uri)
     return etree.tostring(e, encoding=unicode)
-    
+
 def wrap_text(ocrtext, creation_date, bookinfo=DEFAULT_BOOKINFO):
     """Wrap the text within the minimal XML structure with a DC template."""
     bookinfo.created_at = creation_date
-    
-    dcstring = etree.tostring(bookinfo.to_etree(),\
+
+    dcstring = etree.tostring(bookinfo.to_etree(), \
         method='xml', encoding=unicode, pretty_print=True)
 
-    return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext +\
-        u'\n</plain-text>\n</utwor>';
\ No newline at end of file
+    return u'<utwor>\n' + dcstring + u'\n<plain-text>\n' + ocrtext + \
+        u'\n</plain-text>\n</utwor>';
+
+
+def serialize_raw(element):
+    b = u'' + (element.text or '')
+
+    for child in element.iterchildren():
+        e = etree.tostring(child, method='xml', encoding=unicode, pretty_print=True)
+        b += e
+
+    return b
+
+SERIALIZERS = {
+    'raw': serialize_raw,
+}
+
+def serialize_children(element, format='raw'):
+    return SERIALIZERS[format](element)
+
+def get_resource(path):
+    return os.path.join(os.path.dirname(__file__), path)
+