X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/a553cea6951d2e226fa5863918574f2f4e498bee..9868276112ecaa2081f36b9304f6d10492a2baed:/librarian/dcparser.py?ds=sidebyside

diff --git a/librarian/dcparser.py b/librarian/dcparser.py
index 1df1e5a..12bb24f 100644
--- a/librarian/dcparser.py
+++ b/librarian/dcparser.py
@@ -10,12 +10,19 @@ import re
 from librarian.util import roman_to_int
 
 from librarian import (ValidationError, NoDublinCore, ParseError, DCNS, RDFNS,
-                       WLURI)
+                       XMLNS, WLURI, WLNS, PLMETNS)
 
 import lxml.etree as etree # ElementTree API using libxml2
 from lxml.etree import XMLSyntaxError
 
 
+class TextPlus(unicode):
+    pass
+
+class DatePlus(date):
+    pass
+
+
 # ==============
 # = Converters =
 # ==============
@@ -75,8 +82,9 @@ for now we will translate this to some single date losing information of course.
     try:
         # check out the "N. poł X w." syntax
         if isinstance(text, str): text = text.decode("utf-8")
+
         century_format = u"(?:([12]) *poł[.]? +)?([MCDXVI]+) *w[.,]*(?: *l[.]? *([0-9]+))?"
-        vague_format = u"(?:po *|ok. *)([0-9]+)"
+        vague_format = u"(?:po *|ok. *)?([0-9]{4})(-[0-9]{2}-[0-9]{2})?"
 
         m = re.match(century_format, text)
         m2 = re.match(vague_format, text)
@@ -93,15 +101,16 @@ for now we will translate this to some single date losing information of course.
                 decade = int(decade or 0)
                 t = ((century*100 + decade), 1, 1)
         elif m2:
-            year = int(m2.group(1))
-            t = (year, 1, 1)
+            year = m2.group(1)
+            mon_day = m2.group(2)
+            if mon_day:
+                t = time.strptime(year + mon_day, "%Y-%m-%d")
+            else:
+                t = time.strptime(year, '%Y')
         else:
-            try:
-                t = time.strptime(text, '%Y-%m-%d')
-            except ValueError:
-                t = time.strptime(text, '%Y')
+            raise ValueError
 
-        return date(t[0], t[1], t[2])
+        return DatePlus(t[0], t[1], t[2])
     except ValueError, e:
         raise ValueError("Unrecognized date format. Try YYYY-MM-DD or YYYY.")
 
@@ -112,7 +121,7 @@ def as_unicode(text):
     if isinstance(text, unicode):
         return text
     else:
-        return text.decode('utf-8')
+        return TextPlus(text.decode('utf-8'))
 
 def as_wluri_strict(text):
     return WLURI.strict(text)
@@ -138,7 +147,15 @@ class Field(object):
             if self.multiple:
                 if validator is None:
                     return val
-                return [ validator(v) if v is not None else v for v in val ]
+                new_values = []
+                for v in val:
+                    nv = v
+                    if v is not None:
+                        nv = validator(v)
+                        if hasattr(v, 'lang'):
+                            setattr(nv, 'lang', v.lang)
+                    new_values.append(nv)
+                return new_values
             elif len(val) > 1:
                 raise ValidationError("Multiple values not allowed for field '%s'" % self.uri)
             elif len(val) == 0:
@@ -146,7 +163,10 @@ class Field(object):
             else:
                 if validator is None or val[0] is None:
                     return val[0]
-                return validator(val[0])
+                nv = validator(val[0])
+                if hasattr(val[0], 'lang'):
+                    setattr(nv, 'lang', val[0].lang)
+                return nv
         except ValueError, e:
             raise ValidationError("Field '%s' - invald value: %s" % (self.uri, e.message))
 
@@ -222,6 +242,9 @@ class WorkInfo(object):
         Field( DCNS('identifier.url'), 'url', WLURI, strict=as_wluri_strict),
         Field( DCNS('rights.license'), 'license', required=False),
         Field( DCNS('rights'), 'license_description'),
+
+        Field( PLMETNS('digitisationSponsor'), 'sponsors', multiple=True, default=[]),
+        Field( WLNS('digitisationSponsorNote'), 'sponsor_note', required=False),
     )
 
     @classmethod
@@ -266,9 +289,23 @@ class WorkInfo(object):
         if desc is None:
             raise NoDublinCore("No DublinCore section found.")
 
+        lang = None
+        p = desc
+        while p is not None and lang is None:
+            lang = p.attrib.get(XMLNS('lang'))
+            p = p.getparent()
+
         for e in desc.getchildren():
             fv = field_dict.get(e.tag, [])
-            fv.append(e.text)
+            if e.text is not None:
+                text = e.text
+                if not isinstance(text, unicode):
+                    text = text.decode('utf-8')
+                val = TextPlus(text)
+                val.lang = e.attrib.get(XMLNS('lang'), lang)
+            else:
+                val = e.text
+            fv.append(val)
             field_dict[e.tag] = fv
 
         return cls(desc.attrib, field_dict, *args, **kwargs)
@@ -414,6 +451,9 @@ class BookInfo(WorkInfo):
         Field( DCNS('relation.coverImage.url'), 'cover_url', required=False),
         Field( DCNS('relation.coverImage.attribution'), 'cover_by', required=False),
         Field( DCNS('relation.coverImage.source'), 'cover_source', required=False),
+        # WLCover-specific.
+        Field( WLNS('coverBarColor'), 'cover_bar_color', required=False),
+        Field( WLNS('coverBoxPosition'), 'cover_box_position', required=False),
     )