books: updates to serials and multiparts (#251)

* Store serials in _migration
* Store keywords in _migration
* Change `role` to `roles`
* Use `has_keyword`, `is_multipart` and `has_serial` to indicate if the
record has any keywords, multipart or serials

cds_dojson/marc21/fields/books/base.py

@@ -44,6 +44,18 @@
 from .utils import extract_parts, is_excluded
 
 
+@model.over('legacy_recid', '^001')
+def recid(self, key, value):
+    """Record Identifier."""
+    return int(value)
+
+
+@model.over('agency_code', '^003')
+def agency_code(self, key, value):
+    """Control number identifier."""
+    return 'SzGeCERN'
+
+
 @model.over('acquisition_source', '(^916__)|(^859__)|(^595__)')
 @filter_values
 def acquisition_source(self, key, value):
@@ -547,7 +559,7 @@ def subject_classification(self, key, value):
     elif key == '084__':
         sub_2 = clean_val('2', value, str)
         if sub_2 and sub_2.upper() in SUBJECT_CLASSIFICATION_EXCEPTIONS:
-            self['keywords'] = keywords(self, key, value)
+            keywords(self, key, value)
             raise IgnoreKey('subject_classification')
         else:
             _subject_classification.update({'schema': 'ICS'})
@@ -560,25 +572,28 @@ def subject_classification(self, key, value):
 @filter_list_values
 def keywords(self, key, value):
     """Keywords."""
-    _keywords = self.get('keywords', [])
+    _migration = self.get('_migration', {})
+    if 'keywords' not in _migration:
+        _migration['keywords'] = []
     for v in force_list(value):
         if key == '084__':
             sub_2 = clean_val('2', value, str)
             if sub_2 and sub_2 == 'PACS':
-                _keywords.append({
+                _migration['keywords'].append({
                     'name': clean_val('a', v, str, req=True),
                     'provenance': 'PACS',
                 })
             else:
                 raise IgnoreKey('keywords')
         elif key == '6531_':
-            _keywords.append({
+            _migration['keywords'].append({
                 'name': clean_val('a', value, str),
                 'provenance': value.get('9') or value.get('g'),
                 # Easier to solve here
             })
-
-    return _keywords
+    _migration['has_keywords'] = True
+    self['_migration'] = _migration
+    raise IgnoreKey('keywords')
 
 
 @model.over('conference_info', '(^111__)|(^270__)|(^711__)')
@@ -666,14 +681,22 @@ def imprints(self, key, value):
 
 @model.over('book_series', '^490__')
 @for_each_value
-@filter_values
 def book_series(self, key, value):
-    """Translates book series field."""
-    return {
+    """Match barcodes to volumes."""
+    val_n = clean_val('n', value, str)
+    val_x = clean_val('x', value, str)
+
+    _migration = self.get('_migration', {})
+    if 'serials' not in _migration:
+        _migration['serials'] = []
+    _migration['serials'].append({
         'title': clean_val('a', value, str),
         'volume': clean_val('v', value, str),
         'issn': clean_val('x', value, str),
-    }
+    })
+    _migration['has_serial'] = True
+    self['_migration'] = _migration
+    raise IgnoreKey('book_series')
 
 
 @model.over('public_notes', '^500__')

cds_dojson/marc21/fields/books/book.py

@@ -24,8 +24,9 @@
 from dojson.utils import filter_values, for_each_value
 
 from cds_dojson.marc21.fields.books.errors import ManualMigrationRequired, \
-    UnexpectedValue
-from cds_dojson.marc21.fields.books.utils import is_excluded, extract_parts
+    MissingRequiredField, UnexpectedValue
+from cds_dojson.marc21.fields.books.utils import is_excluded, extract_parts, \
+    extract_volume_number
 from cds_dojson.marc21.fields.utils import clean_val
 from cds_dojson.marc21.models.books.book import model
 
@@ -35,18 +36,24 @@
 @filter_values
 def alternative_titles(self, key, value):
     """Alternative titles."""
-    if 'n' in value:
-        self['volume'] = volume(self, key, value)
+    if ('n' in value and 'p' not in value) or \
+       ('n' not in value and 'p' in value):
+        raise MissingRequiredField(subfield='n or p')
+
     if 'p' in value:
-        # if series detected
-        if self.get('volumes', None):
-            val_p = clean_val('p', value, str)
-            self['volumes_titles'].append(
-                {'title': val_p, 'volume': volume(self, key, value)}
-            ) if val_p else None
-        else:
-            self['volumes_titles'] = []
-        return {}
+        _migration = self.get('_migration', {})
+        if 'volumes' not in _migration:
+            _migration['volumes'] = []
+
+        val_n = clean_val('n', value, str)
+        _migration['volumes'].append({
+            'volume': extract_volume_number(val_n, raise_exception=True),
+            'title': clean_val('p', value, str),
+        })
+        _migration['is_multipart'] = True
+        _migration['record_type'] = 'multipart'
+        self['_migration'] = _migration
+        raise IgnoreKey('alternative_titles')
     else:
         return {
             'title': clean_val('a', value, str, req=True),
@@ -55,17 +62,6 @@ def alternative_titles(self, key, value):
         }
 
 
-@model.over('volume', '^246__')
-@for_each_value
-def volume(self, key, value):
-    """Translates volumes index in series."""
-    _volume = self.get('volume', None)
-    val_n = clean_val('n', value, str, req=True)
-    if _volume and _volume != val_n:
-        raise ManualMigrationRequired(subfield='n')
-    return val_n
-
-
 @model.over('number_of_pages', '^300__')   # item
 def number_of_pages(self, key, value):
     """Translates number_of_pages fields."""

cds_dojson/marc21/fields/books/multipart.py

@@ -22,6 +22,7 @@
 from dojson.errors import IgnoreKey
 from dojson.utils import for_each_value, filter_values, force_list
 
+from cds_dojson.marc21.fields.books.base import book_series as base_book_series
 from cds_dojson.marc21.fields.books.errors import UnexpectedValue, \
     ManualMigrationRequired, MissingRequiredField
 from cds_dojson.marc21.fields.books.utils import extract_parts, \
@@ -100,7 +101,9 @@ def migration(self, key, value):
     _series_title = self.get('title', None)
 
     # I added this in the model, I'm sure it's there
-    _migration = self.get('_migration', {'volumes': []})
+    _migration = self.get('_migration', {})
+    if 'volumes' not in _migration:
+        _migration['volumes'] = []
 
     for v in force_list(value):
         # check if it is a multipart monograph
@@ -158,3 +161,10 @@ def number_of_volumes(self, key, value):
         if _volumes:
             return _volumes[0]
     raise IgnoreKey('number_of_volumes')
+
+
+@model.over('book_series', '^490__')
+@for_each_value
+def book_series(self, key, value):
+    """Match barcodes to volumes."""
+    base_book_series(self, key, value)

cds_dojson/marc21/fields/utils.py

@@ -371,8 +371,10 @@ def build_contributor_books(value):
         'ids': _extract_json_ids(value, 'schema') or None,
         'full_name': value.get('name') or clean_val('a', value, str),
         'email': clean_email(value.get('email')),
-        'role': _get_correct_books_contributor_role(
-            'e', value.get('e', 'author')),
+        'roles': [
+            _get_correct_books_contributor_role(
+                'e', value.get('e', 'author')).lower()
+        ],
         'curated_relation': True if value_9 == '#BEARD#' else None
     }
 

cds_dojson/marc21/models/books/base.py

@@ -164,5 +164,5 @@ class BooksBase(OverdoJSONSchema):
     """Base model conversion MARC21 to JSON."""
 
 
-model = BooksBase(bases=(cds_base, ),
+model = BooksBase(bases=(),
                   entry_point_group='cds_dojson.marc21.books')

cds_dojson/marc21/models/books/book.py

@@ -31,7 +31,7 @@ class CDSBook(CDSOverdoBookBase):
                 '690C_:BOOKSUGGESTION OR 980__:PROCEEDINGS OR 980__:PERI OR ' \
                 '697C_:LEGSERLIB OR 697C_:"ENGLISH BOOK CLUB" -980__:DELETED'
 
-    __schema__ = 'records/books/book/book-v.0.0.1.json'
+    __schema__ = 'https://127.0.0.1:5000/schemas/documents/document-v1.0.0.json'
 
     __ignore_keys__ = COMMON_IGNORE_FIELDS
 
@@ -44,24 +44,28 @@ def do(self, blob, ignore_missing=True, exception_handlers=None):
         json['$schema'] = self.__class__.__schema__
 
         if '_migration' in json:
-            json['_migration'].update({'record_type': 'document'})
+            json['_migration'].setdefault('record_type', 'document')
+            json['_migration'].setdefault('volumes', [])
+            json['_migration'].setdefault('serials', [])
             json['_migration'].setdefault('has_serial', False)
-            json['_migration'].setdefault('has_multipart', False)
+            json['_migration'].setdefault('is_multipart', False)
             json['_migration'].setdefault('has_keywords', False)
             json['_migration'].setdefault('has_related', False)
+
         else:
             json['_migration'] = {
                 'record_type': 'document',
                 'has_serial': False,
-                'has_multipart': False,
+                'is_multipart': False,
                 'has_keywords': False,
                 'has_related': False,
-                'volumes': []
+                'volumes': [],
+                'serials': [],
             }
 
         return json
 
 
 model = CDSBook(
-    bases=(books_base, cds_base, ),
+    bases=(books_base, ),
     entry_point_group='cds_dojson.marc21.book')

cds_dojson/marc21/models/books/standard.py

@@ -44,7 +44,7 @@ def do(self, blob, ignore_missing=True, exception_handlers=None):
         json['_migration'] = {
             'record_type': 'document',
             'has_serial': False,
-            'has_multipart': False,
+            'is_multipart': False,
             'has_keywords': False,
             'has_related': False,
             'volumes': []