Add swappable CSS for HTML previews.

[librarian.git] / librarian / dcparser.py

diff --git a/librarian/dcparser.py b/librarian/dcparser.py
index fd3eec5..92afc01 100644 (file)
--- a/librarian/dcparser.py
+++ b/librarian/dcparser.py
@@ -3,10 +3,14 @@
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #
 # This file is part of Librarian, licensed under GNU Affero GPLv3 or later.
 # Copyright © Fundacja Nowoczesna Polska. See NOTICE for more information.
 #

+from __future__ import unicode_literals
+

 from xml.parsers.expat import ExpatError
 from datetime import date
 from xml.parsers.expat import ExpatError
 from datetime import date

+from functools import total_ordering

 import time
 import re
 import time
 import re

+import six

 from librarian.util import roman_to_int
 
 from librarian import (ValidationError, NoDublinCore, ParseError, DCNS, RDFNS,
 from librarian.util import roman_to_int
 
 from librarian import (ValidationError, NoDublinCore, ParseError, DCNS, RDFNS,


@@ -16,7 +20,7 @@ import lxml.etree as etree  # ElementTree API using libxml2
 from lxml.etree import XMLSyntaxError
 
 
 from lxml.etree import XMLSyntaxError
 
 

-class TextPlus(unicode):
+class TextPlus(six.text_type):

     pass
 
 
     pass
 
 


@@ -27,6 +31,8 @@ class DatePlus(date):
 # ==============
 # = Converters =
 # ==============
 # ==============
 # = Converters =
 # ==============

+@six.python_2_unicode_compatible
+@total_ordering

 class Person(object):
     """Single person with last name and a list of first names."""
     def __init__(self, last_name, *first_names):
 class Person(object):
     """Single person with last name and a list of first names."""
     def __init__(self, last_name, *first_names):


@@ -40,13 +46,13 @@ class Person(object):
             surname = parts[0]
             names = []
         elif len(parts) != 2:
             surname = parts[0]
             names = []
         elif len(parts) != 2:

-            raise ValueError("Invalid person name. There should be at most one comma: \"%s\"." % text)
+            raise ValueError("Invalid person name. There should be at most one comma: \"%s\"." % text.encode('utf-8'))

         else:
             surname = parts[0]
             if len(parts[1]) == 0:
                 # there is no non-whitespace data after the comma
                 raise ValueError("Found a comma, but no names given: \"%s\" -> %r." % (text, parts))
         else:
             surname = parts[0]
             if len(parts[1]) == 0:
                 # there is no non-whitespace data after the comma
                 raise ValueError("Found a comma, but no names given: \"%s\" -> %r." % (text, parts))

-            names = [name for name in parts[1].split() if len(name)]  # all non-whitespace tokens
+            names = parts[1].split()

         return cls(surname, *names)
 
     def readable(self):
         return cls(surname, *names)
 
     def readable(self):


@@ -55,13 +61,13 @@ class Person(object):
     def __eq__(self, right):
         return self.last_name == right.last_name and self.first_names == right.first_names
 
     def __eq__(self, right):
         return self.last_name == right.last_name and self.first_names == right.first_names
 

-    def __cmp__(self, other):
-        return cmp((self.last_name, self.first_names), (other.last_name, other.first_names))
+    def __lt__(self, other):
+        return (self.last_name, self.first_names) < (other.last_name, other.first_names)

 
     def __hash__(self):
         return hash((self.last_name, self.first_names))
 
 
     def __hash__(self):
         return hash((self.last_name, self.first_names))
 

-    def __unicode__(self):
+    def __str__(self):

         if len(self.first_names) > 0:
             return '%s, %s' % (self.last_name, ' '.join(self.first_names))
         else:
         if len(self.first_names) > 0:
             return '%s, %s' % (self.last_name, ' '.join(self.first_names))
         else:


@@ -83,7 +89,7 @@ for now we will translate this to some single date losing information of course.
     """
     try:
         # check out the "N. poł X w." syntax
     """
     try:
         # check out the "N. poł X w." syntax

-        if isinstance(text, str):
+        if isinstance(text, six.binary_type):

             text = text.decode("utf-8")
 
         century_format = u"(?:([12]) *poł[.]? +)?([MCDXVI]+) *w[.,]*(?: *l[.]? *([0-9]+))?"
             text = text.decode("utf-8")
 
         century_format = u"(?:([12]) *poł[.]? +)?([MCDXVI]+) *w[.,]*(?: *l[.]? *([0-9]+))?"


@@ -94,7 +100,7 @@ for now we will translate this to some single date losing information of course.
         if m:
             half = m.group(1)
             decade = m.group(3)
         if m:
             half = m.group(1)
             decade = m.group(3)

-            century = roman_to_int(str(m.group(2)))
+            century = roman_to_int(m.group(2))

             if half is not None:
                 if decade is not None:
                     raise ValueError("Bad date format. Cannot specify both half and decade of century")
             if half is not None:
                 if decade is not None:
                     raise ValueError("Bad date format. Cannot specify both half and decade of century")


@@ -114,7 +120,7 @@ for now we will translate this to some single date losing information of course.
             raise ValueError
 
         return DatePlus(t[0], t[1], t[2])
             raise ValueError
 
         return DatePlus(t[0], t[1], t[2])

-    except ValueError, e:
+    except ValueError as e:

         raise ValueError("Unrecognized date format. Try YYYY-MM-DD or YYYY.")
 
 
         raise ValueError("Unrecognized date format. Try YYYY-MM-DD or YYYY.")
 
 


@@ -123,7 +129,7 @@ def as_person(text):
 
 
 def as_unicode(text):
 
 
 def as_unicode(text):

-    if isinstance(text, unicode):
+    if isinstance(text, six.text_type):

         return text
     else:
         return TextPlus(text.decode('utf-8'))
         return text
     else:
         return TextPlus(text.decode('utf-8'))


@@ -174,7 +180,7 @@ class Field(object):
                 if hasattr(val[0], 'lang'):
                     setattr(nv, 'lang', val[0].lang)
                 return nv
                 if hasattr(val[0], 'lang'):
                     setattr(nv, 'lang', val[0].lang)
                 return nv

-        except ValueError, e:
+        except ValueError as e:

             raise ValidationError("Field '%s' - invald value: %s" % (self.uri, e.message))
 
     def validate(self, fdict, fallbacks=None, strict=False):
             raise ValidationError("Field '%s' - invald value: %s" % (self.uri, e.message))
 
     def validate(self, fdict, fallbacks=None, strict=False):


@@ -221,43 +227,40 @@ class DCInfo(type):
         return super(DCInfo, mcs).__new__(mcs, classname, bases, class_dict)
 
 
         return super(DCInfo, mcs).__new__(mcs, classname, bases, class_dict)
 
 

-class WorkInfo(object):
-    __metaclass__ = DCInfo
-
+class WorkInfo(six.with_metaclass(DCInfo, object)):

     FIELDS = (
         Field(DCNS('creator'), 'authors', as_person, salias='author', multiple=True),
         Field(DCNS('title'), 'title'),
         Field(DCNS('type'), 'type', required=False, multiple=True),
 
         Field(DCNS('contributor.editor'), 'editors',
     FIELDS = (
         Field(DCNS('creator'), 'authors', as_person, salias='author', multiple=True),
         Field(DCNS('title'), 'title'),
         Field(DCNS('type'), 'type', required=False, multiple=True),
 
         Field(DCNS('contributor.editor'), 'editors',

-              as_person, salias='editor', multiple=True, default=[]),
+              as_person, salias='editor', multiple=True, required=False),

         Field(DCNS('contributor.technical_editor'), 'technical_editors',
         Field(DCNS('contributor.technical_editor'), 'technical_editors',

-              as_person, salias='technical_editor', multiple=True, default=[]),
-        Field(DCNS('contributor.funding'), 'funders', salias='funder', multiple=True, default=[]),
+              as_person, salias='technical_editor', multiple=True, required=False),
+        Field(DCNS('contributor.funding'), 'funders', salias='funder', multiple=True, required=False),

         Field(DCNS('contributor.thanks'), 'thanks', required=False),
 
         Field(DCNS('date'), 'created_at'),
         Field(DCNS('date.pd'), 'released_to_public_domain_at', as_date, required=False),
         Field(DCNS('contributor.thanks'), 'thanks', required=False),
 
         Field(DCNS('date'), 'created_at'),
         Field(DCNS('date.pd'), 'released_to_public_domain_at', as_date, required=False),

-        Field(DCNS('publisher'), 'publisher'),
+        Field(DCNS('publisher'), 'publisher', multiple=True),

 
         Field(DCNS('language'), 'language'),
         Field(DCNS('description'), 'description', required=False),
 
         Field(DCNS('source'), 'source_name', required=False),
 
         Field(DCNS('language'), 'language'),
         Field(DCNS('description'), 'description', required=False),
 
         Field(DCNS('source'), 'source_name', required=False),

-        Field(DCNS('source.URL'), 'source_url', required=False),
+        Field(DCNS('source.URL'), 'source_urls', salias='source_url', multiple=True, required=False),

         Field(DCNS('identifier.url'), 'url', WLURI, strict=as_wluri_strict),
         Field(DCNS('rights.license'), 'license', required=False),
         Field(DCNS('rights'), 'license_description'),
 
         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(PLMETNS('digitisationSponsor'), 'sponsors', multiple=True, required=False),

         Field(WLNS('digitisationSponsorNote'), 'sponsor_note', required=False),
         Field(WLNS('developmentStage'), 'stage', required=False),
     )
 
     @classmethod
         Field(WLNS('digitisationSponsorNote'), 'sponsor_note', required=False),
         Field(WLNS('developmentStage'), 'stage', required=False),
     )
 
     @classmethod

-    def from_string(cls, xml, *args, **kwargs):
-        from StringIO import StringIO
-        return cls.from_file(StringIO(xml), *args, **kwargs)
+    def from_bytes(cls, xml, *args, **kwargs):
+        return cls.from_file(six.BytesIO(xml), *args, **kwargs)

 
     @classmethod
     def from_file(cls, xmlfile, *args, **kwargs):
 
     @classmethod
     def from_file(cls, xmlfile, *args, **kwargs):


@@ -282,9 +285,9 @@ class WorkInfo(object):
 
             # extract data from the element and make the info
             return cls.from_element(desc_tag, *args, **kwargs)
 
             # extract data from the element and make the info
             return cls.from_element(desc_tag, *args, **kwargs)

-        except XMLSyntaxError, e:
+        except XMLSyntaxError as e:

             raise ParseError(e)
             raise ParseError(e)

-        except ExpatError, e:
+        except ExpatError as e:

             raise ParseError(e)
 
     @classmethod
             raise ParseError(e)
 
     @classmethod


@@ -294,7 +297,7 @@ class WorkInfo(object):
         desc = rdf_tag.find(".//" + RDFNS('Description'))
 
         if desc is None:
         desc = rdf_tag.find(".//" + RDFNS('Description'))
 
         if desc is None:

-            raise NoDublinCore("No DublinCore section found.")
+            raise NoDublinCore("There must be a '%s' element inside the RDF." % RDFNS('Description'))

 
         lang = None
         p = desc
 
         lang = None
         p = desc


@@ -306,10 +309,14 @@ class WorkInfo(object):
             fv = field_dict.get(e.tag, [])
             if e.text is not None:
                 text = e.text
             fv = field_dict.get(e.tag, [])
             if e.text is not None:
                 text = e.text

-                if not isinstance(text, unicode):
+                if not isinstance(text, six.text_type):

                     text = text.decode('utf-8')
                 val = TextPlus(text)
                 val.lang = e.attrib.get(XMLNS('lang'), lang)
                     text = text.decode('utf-8')
                 val = TextPlus(text)
                 val.lang = e.attrib.get(XMLNS('lang'), lang)

+                if e.tag == 'meta':
+                    meta_id = e.attrib.get('id')
+                    if meta_id and meta_id.endswith('-id'):
+                        field_dict[meta_id] = [val.replace('ISBN-', 'ISBN ')]

             else:
                 val = e.text
             fv.append(val)
             else:
                 val = e.text
             fv.append(val)


@@ -390,11 +397,11 @@ class WorkInfo(object):
                     for x in v:
                         e = etree.Element(field.uri)
                         if x is not None:
                     for x in v:
                         e = etree.Element(field.uri)
                         if x is not None:

-                            e.text = unicode(x)
+                            e.text = six.text_type(x)

                         description.append(e)
                 else:
                     e = etree.Element(field.uri)
                         description.append(e)
                 else:
                     e = etree.Element(field.uri)

-                    e.text = unicode(v)
+                    e.text = six.text_type(v)

                     description.append(e)
 
         return root
                     description.append(e)
 
         return root


@@ -409,9 +416,9 @@ class WorkInfo(object):
                 if field.multiple:
                     if len(v) == 0:
                         continue
                 if field.multiple:
                     if len(v) == 0:
                         continue

-                    v = [unicode(x) for x in v if x is not None]
+                    v = [six.text_type(x) for x in v if x is not None]

                 else:
                 else:

-                    v = unicode(v)
+                    v = six.text_type(v)

 
                 dc[field.name] = {'uri': field.uri, 'value': v}
         rdf['fields'] = dc
 
                 dc[field.name] = {'uri': field.uri, 'value': v}
         rdf['fields'] = dc


@@ -426,15 +433,15 @@ class WorkInfo(object):
                 if field.multiple:
                     if len(v) == 0:
                         continue
                 if field.multiple:
                     if len(v) == 0:
                         continue

-                    v = [unicode(x) for x in v if x is not None]
+                    v = [six.text_type(x) for x in v if x is not None]

                 else:
                 else:

-                    v = unicode(v)
+                    v = six.text_type(v)

                 result[field.name] = v
 
             if field.salias:
                 v = getattr(self, field.salias)
                 if v is not None:
                 result[field.name] = v
 
             if field.salias:
                 v = getattr(self, field.salias)
                 if v is not None:

-                    result[field.salias] = unicode(v)
+                    result[field.salias] = six.text_type(v)

 
         return result
 
 
         return result
 


@@ -446,9 +453,11 @@ class BookInfo(WorkInfo):
         Field(DCNS('subject.period'), 'epochs', salias='epoch', multiple=True, required=False),
         Field(DCNS('subject.type'), 'kinds', salias='kind', multiple=True, required=False),
         Field(DCNS('subject.genre'), 'genres', salias='genre', multiple=True, required=False),
         Field(DCNS('subject.period'), 'epochs', salias='epoch', multiple=True, required=False),
         Field(DCNS('subject.type'), 'kinds', salias='kind', multiple=True, required=False),
         Field(DCNS('subject.genre'), 'genres', salias='genre', multiple=True, required=False),

-                
+
+        Field(DCNS('subject.location'), 'location', required=False),
+

         Field(DCNS('contributor.translator'), 'translators',
         Field(DCNS('contributor.translator'), 'translators',

-              as_person,  salias='translator', multiple=True, default=[]),
+              as_person,  salias='translator', multiple=True, required=False),

         Field(DCNS('relation.hasPart'), 'parts', WLURI, strict=as_wluri_strict, multiple=True, required=False),
         Field(DCNS('relation.isVariantOf'), 'variant_of', WLURI, strict=as_wluri_strict, required=False),
 
         Field(DCNS('relation.hasPart'), 'parts', WLURI, strict=as_wluri_strict, multiple=True, required=False),
         Field(DCNS('relation.isVariantOf'), 'variant_of', WLURI, strict=as_wluri_strict, required=False),
 


@@ -458,6 +467,14 @@ class BookInfo(WorkInfo):
         # WLCover-specific.
         Field(WLNS('coverBarColor'), 'cover_bar_color', required=False),
         Field(WLNS('coverBoxPosition'), 'cover_box_position', required=False),
         # WLCover-specific.
         Field(WLNS('coverBarColor'), 'cover_bar_color', required=False),
         Field(WLNS('coverBoxPosition'), 'cover_box_position', required=False),

+        Field(WLNS('coverClass'), 'cover_class', default=['default']),
+        Field(WLNS('coverLogoUrl'), 'cover_logo_urls', multiple=True, required=False),
+
+        Field('pdf-id',  'isbn_pdf',  required=False),
+        Field('epub-id', 'isbn_epub', required=False),
+        Field('mobi-id', 'isbn_mobi', required=False),
+        Field('txt-id',  'isbn_txt',  required=False),
+        Field('html-id', 'isbn_html', required=False),

     )
 
 
     )