Allow specyfing fallback values for BookInfo fields when creating a WLDocument.

diff --git a/librarian/dcparser.py b/librarian/dcparser.py
index 5a571ec..eddd8e5 100644 (file)
--- a/librarian/dcparser.py
+++ b/librarian/dcparser.py
@@ -115,10 +115,21 @@ class Field(object):
         except ValueError, e:
             raise ValidationError("Field '%s' - invald value: %s" % (self.uri, e.message))
 
-    def validate(self, fdict, strict=False):
+    def validate(self, fdict, fallbacks=None, strict=False):
+        if fallbacks is None:
+            fallbacks = {}
         if not fdict.has_key(self.uri):
             if not self.required:
-                f = self.default
+                # Accept single value for single fields and saliases.
+                if self.name in fallbacks:
+                    if self.multiple:
+                        f = fallbacks[self.name]
+                    else:
+                        f = [fallbacks[self.name]]
+                elif self.salias and self.salias in fallbacks:
+                    f = [fallbacks[self.salias]]
+                else:
+                    f = self.default
             else:
                 raise ValidationError("Required field %s not found" % self.uri)
         else:
@@ -224,7 +235,7 @@ class WorkInfo(object):
 
         return cls(desc.attrib, field_dict, *args, **kwargs)
 
-    def __init__(self, rdf_attrs, dc_fields, strict=False):
+    def __init__(self, rdf_attrs, dc_fields, fallbacks=None, strict=False):
         """rdf_attrs should be a dictionary-like object with any attributes of the RDF:Description.
         dc_fields - dictionary mapping DC fields (with namespace) to list of text values for the
         given field. """
@@ -233,7 +244,8 @@ class WorkInfo(object):
         self.fmap = {}
 
         for field in self.FIELDS:
-            value = field.validate(dc_fields, strict=strict)
+            value = field.validate(dc_fields, fallbacks=fallbacks,
+                            strict=strict)
             setattr(self, 'prop_' + field.name, value)
             self.fmap[field.name] = field
             if field.salias: self.fmap[field.salias] = field

diff --git a/librarian/parser.py b/librarian/parser.py
index b5145a6..e605dd9 100644 (file)
--- a/librarian/parser.py
+++ b/librarian/parser.py
@@ -19,7 +19,8 @@ class WLDocument(object):
     LINE_SWAP_EXPR = re.compile(r'/\s', re.MULTILINE | re.UNICODE)
     provider = None
 
-    def __init__(self, edoc, parse_dublincore=True, provider=None, strict=False):
+    def __init__(self, edoc, parse_dublincore=True, provider=None, 
+                    strict=False, meta_fallbacks=None):
         self.edoc = edoc
         self.provider = provider
 
@@ -37,7 +38,7 @@ class WLDocument(object):
                 raise NoDublinCore('Document has no DublinCore - which is required.')
 
             self.book_info = dcparser.BookInfo.from_element(
-                    self.rdf_elem, strict=strict)
+                    self.rdf_elem, fallbacks=meta_fallbacks, strict=strict)
         else:
             self.book_info = None
 
@@ -46,7 +47,7 @@ class WLDocument(object):
         return cls.from_file(StringIO(xml), *args, **kwargs)
 
     @classmethod
-    def from_file(cls, xmlfile, parse_dublincore=True, provider=None):
+    def from_file(cls, xmlfile, *args, **kwargs):
 
         # first, prepare for parsing
         if isinstance(xmlfile, basestring):
@@ -67,7 +68,7 @@ class WLDocument(object):
             parser = etree.XMLParser(remove_blank_text=False)
             tree = etree.parse(StringIO(data.encode('utf-8')), parser)
 
-            return cls(tree, parse_dublincore=parse_dublincore, provider=provider)
+            return cls(tree, *args, **kwargs)
         except (ExpatError, XMLSyntaxError, XSLTApplyError), e:
             raise ParseError(e)