X-Git-Url: https://git.mdrn.pl/librarian.git/blobdiff_plain/a34b95aa7ba5fd4838541d1cdcd28358fb808062..26aa2369c47124fb322236e03084324914aa24af:/librarian/dcparser.py diff --git a/librarian/dcparser.py b/librarian/dcparser.py index 21244ef..afcefa0 100644 --- a/librarian/dcparser.py +++ b/librarian/dcparser.py @@ -70,7 +70,8 @@ def as_date(text): except ValueError, e: raise ValueError("Unrecognized date format. Try YYYY-MM-DD or YYYY.") -as_person = Person.from_text +def as_person(text): + return Person.from_text(text) def as_unicode(text): if isinstance(text, unicode): @@ -78,12 +79,15 @@ def as_unicode(text): else: return text.decode('utf-8') +def as_wluri_strict(text): + return WLURI.strict(text) + class Field(object): def __init__(self, uri, attr_name, validator=as_unicode, strict=None, multiple=False, salias=None, **kwargs): self.uri = uri self.name = attr_name - self.validator = lambda x: validator(x) - self.strict = lambda x: strict(x) + self.validator = validator + self.strict = strict self.multiple = multiple self.salias = salias @@ -91,6 +95,7 @@ class Field(object): self.default = kwargs.get('default', [] if multiple else [None]) def validate_value(self, val, strict=False): + val = [v.strip() if v is not None else v for v in val] if strict and self.strict is not None: validator = self.strict else: @@ -111,10 +116,21 @@ 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: @@ -123,7 +139,7 @@ class Field(object): return self.validate_value(f, strict=strict) def __eq__(self, other): - if isinstance(other, Field) and other.name == self.name: + if isinstance(other, Field) and other.name == self.name and other.uri == self.uri: return True return False @@ -148,9 +164,13 @@ class WorkInfo(object): __metaclass__ = DCInfo FIELDS = ( - Field( DCNS('creator'), 'authors', as_person, salias='author', multiple=True), + Field( DCNS('creator.expert'), 'authors_expert', as_person, salias='author', required=False, multiple=True), + Field( DCNS('creator.methodologist'), 'authors_methodologist', as_person, salias='author', required=False, multiple=True), + Field( DCNS('creator.scenario'), 'authors_scenario', as_person, salias='author', required=False, multiple=True), + Field( DCNS('creator.textbook'), 'authors_textbook', as_person, salias='author', required=False, multiple=True), + Field( DCNS('requires'), 'requires', required=False, multiple=True), Field( DCNS('title'), 'title'), - Field( DCNS('type'), 'type', required=False, multiple=True), + Field( DCNS('type'), 'type', required=False), Field( DCNS('contributor.editor'), 'editors', \ as_person, salias='editor', multiple=True, default=[]), @@ -161,12 +181,15 @@ class WorkInfo(object): Field( DCNS('date.pd'), 'released_to_public_domain_at', as_date, required=False), Field( DCNS('publisher'), 'publisher'), + Field( DCNS('subject.competence'), 'competences', multiple=True, required=False), + Field( DCNS('subject.curriculum'), 'curriculum', multiple=True, 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('identifier.url'), 'url', WLURI, strict=WLURI.strict), + Field( DCNS('identifier.url'), 'url', WLURI, strict=as_wluri_strict), Field( DCNS('rights.license'), 'license', required=False), Field( DCNS('rights'), 'license_description'), ) @@ -220,7 +243,7 @@ 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. """ @@ -229,7 +252,10 @@ 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) + if field.multiple: + value = getattr(self, 'prop_' + field.name, []) + value setattr(self, 'prop_' + field.name, value) self.fmap[field.name] = field if field.salias: self.fmap[field.salias] = field @@ -244,7 +270,7 @@ class WorkInfo(object): if not field.multiple: raise "OUCH!! for field %s" % name - return value[0] + return value[0] if value else None except (KeyError, AttributeError): return object.__getattribute__(self, name) @@ -343,20 +369,25 @@ class BookInfo(WorkInfo): Field( DCNS('audience'), 'audiences', salias='audience', multiple=True, required=False), - Field( DCNS('subject.period'), 'epochs', salias='epoch', multiple=True), - Field( DCNS('subject.type'), 'kinds', salias='kind', multiple=True), - Field( DCNS('subject.genre'), 'genres', salias='genre', multiple=True), - + 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('contributor.translator'), 'translators', \ as_person, salias='translator', multiple=True, default=[]), - Field( DCNS('relation.hasPart'), 'parts', - WLURI, strict=WLURI.strict, multiple=True, required=False), - Field( DCNS('relation.isVariantOf'), 'variant_of', - WLURI, strict=WLURI.strict, required=False), - - Field( DCNS('relation.cover_image.url'), 'cover_url', required=False), - Field( DCNS('relation.cover_image.attribution'), 'cover_by', required=False), - Field( DCNS('relation.cover_image.source'), 'cover_source', 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'), 'relations', + WLURI, strict=as_wluri_strict, multiple=True, required=False), + + 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), )