class ChapterSchema(Schema):
text_id = fields.Nested(TextIdSchema, required=True, data_key="textId")
classification = ValueEnum(Classification, required=True)
stage = ValueEnum(Stage, required=True)
version = fields.String(required=True)
name = fields.String(required=True, validate=validate.Length(min=1))
order = fields.Integer(required=True)
manuscripts = fields.Nested(ManuscriptSchema, many=True, required=True)
uncertain_fragments: fields.Field = fields.Nested(
MuseumNumberSchema,
many=True,
load_default=tuple(),
data_key="uncertainFragments",
)
lines = fields.Nested(LineSchema, many=True, required=True)
signs = fields.List(fields.String(), load_default=tuple())
record = fields.Nested(RecordSchema, load_default=Record())
parser_version = fields.String(load_default="", data_key="parserVersion")
@post_load
def make_chapter(self, data: dict, **kwargs) -> Chapter:
return Chapter(
data["text_id"],
Classification(data["classification"]),
Stage(data["stage"]),
data["version"],
data["name"],
data["order"],
tuple(data["manuscripts"]),
tuple(data["uncertain_fragments"]),
tuple(data["lines"]),
tuple(data["signs"]),
data["record"],
data["parser_version"],
)
class TextSchema(Schema):
genre = ValueEnum(Genre, load_default=Genre.LITERATURE)
category = fields.Integer(required=True, validate=validate.Range(min=0))
index = fields.Integer(required=True, validate=validate.Range(min=0))
name = fields.String(required=True, validate=validate.Length(min=1))
has_doi = fields.Boolean(load_default=False, data_key="hasDoi")
number_of_verses = fields.Integer(required=True,
data_key="numberOfVerses",
validate=validate.Range(min=0))
approximate_verses = fields.Boolean(required=True,
data_key="approximateVerses")
intro = fields.String(load_default="")
chapters = fields.Nested(ChapterListingSchema, many=True, required=True)
references = fields.Nested(ReferenceSchema,
many=True,
load_default=tuple())
@post_load
def make_text(self, data: dict, **kwargs) -> Text:
return Text(
data["genre"],
data["category"],
data["index"],
data["name"],
data["has_doi"],
data["number_of_verses"],
data["approximate_verses"],
data["intro"],
tuple(data["chapters"]),
tuple(data["references"]),
)
class JoinerSchema(BaseTokenSchema):
enum_value = ValueEnum(atf.Joiner, required=True, data_key="value", load_only=True)
@post_load
def make_token(self, data, **kwargs):
return Joiner(
frozenset(data["enclosure_type"]), data["erasure"], data["enum_value"]
)
class TextIdSchema(Schema):
genre = ValueEnum(Genre, load_default=Genre.LITERATURE)
category = fields.Integer(required=True, validate=validate.Range(min=0))
index = fields.Integer(required=True, validate=validate.Range(min=0))
@post_load
def make_id(self, data, **kwargs) -> TextId:
return TextId(data["genre"], data["category"], data["index"])
class ChapterNameSchema(Schema):
stage = ValueEnum(Stage, required=True)
version = fields.String(required=True)
name = fields.String(required=True, validate=validate.Length(min=1))
@post_load
def make_id(self, data, **kwargs) -> ChapterName:
return ChapterName(data["stage"], data["version"], data["name"])
class ChapterIdSchema(Schema):
text_id = fields.Nested(TextIdSchema, required=True, data_key="textId")
stage = ValueEnum(Stage, required=True)
name = fields.String(required=True, validate=validate.Length(min=1))
@post_load
def make_id(self, data, **kwargs) -> ChapterId:
return ChapterId(data["text_id"], data["stage"], data["name"])
class RecordEntrySchema(Schema):
user = fields.String(required=True)
type = ValueEnum(RecordType, required=True)
date = fields.String(required=True)
@post_load
def make_record_entry(self, data, **kwargs):
return RecordEntry(**data)
class NamedSignSchema(BaseTokenSchema):
name = fields.String(required=True)
name_parts = fields.List(
fields.Nested(lambda: OneOfTokenSchema()), required=True, data_key="nameParts"
)
sub_index = fields.Integer(data_key="subIndex", allow_none=True)
modifiers = fields.List(fields.String(), required=True)
flags = fields.List(ValueEnum(Flag), required=True)
sign = fields.Nested(lambda: OneOfTokenSchema(), allow_none=True)
class UnclearSignSchema(BaseTokenSchema):
flags = fields.List(ValueEnum(Flag), required=True)
@post_load
def make_token(self, data, **kwargs):
return (
UnclearSign.of(tuple(data["flags"]))
.set_enclosure_type(frozenset(data["enclosure_type"]))
.set_erasure(data["erasure"])
)
class EgyptianMetricalFeetSeparatorSchema(BaseTokenSchema):
flags = fields.List(ValueEnum(Flag), required=True)
@post_load
def make_token(self, data, **kwargs):
return (
EgyptianMetricalFeetSeparator.of(tuple(data["flags"]))
.set_enclosure_type(frozenset(data["enclosure_type"]))
.set_erasure(data["erasure"])
)
class AnnotationDataSchema(Schema):
id = fields.String(required=True)
type = ValueEnum(AnnotationValueType,
load_default=AnnotationValueType.HAS_SIGN)
sign_name = fields.String(load_default="", data_key="signName")
value = fields.String(required=True)
path = fields.List(fields.Int, required=True)
@post_load
def make_data(self, data, **kwargs):
return AnnotationData(**data)
class GreekLetterSchema(BaseTokenSchema):
letter = fields.String(required=True, validate=validate.Length(1, 1))
flags = fields.List(ValueEnum(Flag), required=True)
@post_load
def make_token(self, data, **kwargs):
return GreekLetter(
frozenset(data["enclosure_type"]),
data["erasure"],
data["letter"],
data["flags"],
)
class AkkadianWordSchema(BaseWordSchema):
modifiers = fields.List(ValueEnum(Flag), required=True)
@post_load
def make_token(self, data, **kwargs):
return AkkadianWord.of(
tuple(data["parts"]),
tuple(data["modifiers"]),
tuple(data["unique_lemma"]),
data["alignment"],
data["variant"],
).set_enclosure_type(frozenset(data["enclosure_type"]))
class GraphemeSchema(BaseTokenSchema):
name = fields.String(required=True)
modifiers = fields.List(fields.String(), required=True)
flags = fields.List(ValueEnum(Flag), required=True)
@post_load
def make_token(self, data, **kwargs):
return (
Grapheme.of(data["name"], data["modifiers"], data["flags"])
.set_enclosure_type(frozenset(data["enclosure_type"]))
.set_erasure(data["erasure"])
)
class DividerSchema(BaseTokenSchema):
divider = fields.String(required=True)
modifiers = fields.List(fields.String(), required=True)
flags = fields.List(ValueEnum(Flag), required=True)
@post_load
def make_token(self, data, **kwargs):
return (
Divider.of(data["divider"], tuple(data["modifiers"]), tuple(data["flags"]))
.set_enclosure_type(frozenset(data["enclosure_type"]))
.set_erasure(data["erasure"])
)
class FragmentSchema(Schema):
number = fields.Nested(MuseumNumberSchema,
required=True,
data_key="museumNumber")
accession = fields.String(required=True)
cdli_number = fields.String(required=True, data_key="cdliNumber")
bm_id_number = fields.String(required=True, data_key="bmIdNumber")
publication = fields.String(required=True)
description = fields.String(required=True)
collection = fields.String(required=True)
script = fields.String(required=True)
museum = fields.String(required=True)
width = fields.Nested(MeasureSchema, required=True)
length = fields.Nested(MeasureSchema, required=True)
thickness = fields.Nested(MeasureSchema, required=True)
joins = fields.Pluck(JoinsSchema, "fragments", load_default=Joins())
record = fields.Pluck(RecordSchema, "entries")
folios = fields.Pluck(FoliosSchema, "entries")
text = fields.Nested(TextSchema)
signs = fields.String(load_default="")
notes = fields.String(required=True)
references = fields.Nested(ReferenceSchema, many=True, required=True)
uncurated_references = fields.Nested(
UncuratedReferenceSchema,
many=True,
data_key="uncuratedReferences",
load_default=None,
)
genres = fields.Nested(GenreSchema, many=True, load_default=tuple())
line_to_vec = fields.List(
fields.List(ValueEnum(LineToVecEncoding)),
load_default=tuple(),
data_key="lineToVec",
)
@post_load
def make_fragment(self, data, **kwargs):
data["references"] = tuple(data["references"])
data["genres"] = tuple(data["genres"])
data["line_to_vec"] = tuple(map(tuple, data["line_to_vec"]))
if data["uncurated_references"] is not None:
data["uncurated_references"] = tuple(data["uncurated_references"])
return Fragment(**data)
@post_dump
def filter_none(self, data, **kwargs):
return pydash.omit_by(data, pydash.is_none)
class ChapterListingSchema(Schema):
stage = ValueEnum(Stage, required=True)
name = fields.String(required=True, validate=validate.Length(min=1))
translation = fields.Nested(TranslationLineSchema,
many=True,
load_default=tuple())
uncertain_fragments = fields.Nested(
UncertainFragmentSchema,
many=True,
load_default=tuple(),
data_key="uncertainFragments",
)
@post_load
def make_chapter_listing(self, data: dict, **kwargs) -> ChapterListing:
return ChapterListing(
Stage(data["stage"]),
data["name"],
tuple(data["translation"]),
tuple(data["uncertain_fragments"]),
)
class _TestSchemaStr(Schema):
value = ValueEnum(_TestEnumStr, required=True, allow_none=True)
name = NameEnum(_TestEnumStr, required=True, allow_none=True)
class ManuscriptSchema(Schema):
id = fields.Integer(required=True)
siglum_disambiguator = fields.String(required=True,
data_key="siglumDisambiguator")
museum_number: fields.Field = fields.Nested(MuseumNumberSchema,
required=True,
allow_none=True,
data_key="museumNumber")
accession = fields.String(required=True)
period_modifier = ValueEnum(PeriodModifier,
required=True,
data_key="periodModifier")
period = fields.Function(
lambda manuscript: manuscript.period.long_name,
lambda value: Period.from_name(value),
required=True,
)
provenance = fields.Function(
lambda manuscript: manuscript.provenance.long_name,
lambda value: Provenance.from_name(value),
required=True,
)
type = fields.Function(
lambda manuscript: manuscript.type.long_name,
lambda value: ManuscriptType.from_name(value),
required=True,
)
notes = fields.String(required=True)
colophon: fields.Field = fields.Nested(TransliterationSchema,
load_default=Transliteration())
unplaced_lines: fields.Field = fields.Nested(
TransliterationSchema,
load_default=Transliteration(),
data_key="unplacedLines")
references = fields.Nested(ReferenceSchema, many=True, required=True)
joins = fields.Pluck(JoinsSchema,
"fragments",
load_default=Joins(),
load_only=True)
is_in_fragmentarium = fields.Boolean(load_default=False,
data_key="isInFragmentarium",
load_only=True)
@validates_schema
def validate_provenance(self, data, **kwargs):
provenace = data["provenance"]
period = data["period"]
type_ = data["type"]
if is_invalid_standard_text(provenace, period, type_):
raise ValidationError(
"period and type must be None if provenance is Standard Text")
elif is_invalid_non_standard_text(provenace, period, type_):
raise ValidationError(
"period and type must not be None if provenance is not Standard Text"
)
@post_load
def make_manuscript(self, data: dict, **kwargs) -> Manuscript:
return Manuscript(
data["id"],
data["siglum_disambiguator"],
data["museum_number"],
data["accession"],
data["period_modifier"],
data["period"],
data["provenance"],
data["type"],
data["notes"],
data["colophon"],
data["unplaced_lines"],
tuple(data["references"]),
data["joins"],
data["is_in_fragmentarium"],
)
class _TestSchemaInt(Schema):
value = ValueEnum(_TestEnumInt, required=True, allow_none=True)
name = NameEnum(_TestEnumInt, required=True, allow_none=True)
