class ManuscriptLineFactory(factory.Factory):
class Meta:
model = ManuscriptLine
manuscript_id = factory.Sequence(lambda n: n)
labels = (
SurfaceLabel.from_label(Surface.OBVERSE),
ColumnLabel.from_label("iii", [Status.COLLATION, Status.CORRECTION]),
)
line = factory.Sequence(
lambda n: TextLine.of_iterable(
LineNumber(n),
(
Word.of(
[
Reading.of_name("ku"),
Joiner.hyphen(),
BrokenAway.open(),
Reading.of_name("nu"),
Joiner.hyphen(),
Reading.of_name("ši"),
BrokenAway.close(),
]
),
),
)
)
paratext = (NoteLine((StringPart("note"),)), RulingDollarLine(Ruling.SINGLE))
omitted_words = (1,)
def test_labels(text_with_labels) -> None:
assert text_with_labels.labels == [
LineLabel(None, None, None, LineNumber(1)),
LineLabel(
ColumnLabel.from_int(1),
SurfaceLabel([], atf.Surface.SURFACE, "Stone wig"),
ObjectLabel([], atf.Object.OBJECT, "Stone wig"),
LineNumber(2),
),
]
def text_with_labels():
return Text.of_iterable([
TextLine.of_iterable(LineNumber(1),
[Word.of([Reading.of_name("bu")])]),
ColumnAtLine(ColumnLabel.from_int(1)),
SurfaceAtLine(SurfaceLabel([], atf.Surface.SURFACE, "Stone wig")),
ObjectAtLine(ObjectLabel([], atf.Object.OBJECT, "Stone wig")),
TextLine.of_iterable(LineNumber(2),
[Word.of([Reading.of_name("bu")])]),
])
("l.e.", "", "@left", SurfaceLabel(tuple(), Surface.LEFT)),
("r.e.", "", "@right", SurfaceLabel(tuple(), Surface.RIGHT)),
("t.e.", "", "@top", SurfaceLabel(tuple(), Surface.TOP)),
("o", "'", "@obverse", SurfaceLabel((Status.PRIME, ), Surface.OBVERSE)),
("r", "?", "@reverse", SurfaceLabel((Status.UNCERTAIN, ),
Surface.REVERSE)),
("b.e.", "!", "@bottom", SurfaceLabel((Status.CORRECTION, ),
Surface.BOTTOM)),
("e.", "*", "@edge", SurfaceLabel((Status.COLLATION, ), Surface.EDGE)),
(
"l.e.",
"*!",
"@left",
SurfaceLabel((Status.COLLATION, Status.CORRECTION), Surface.LEFT),
),
("i", "", "@column 1", ColumnLabel(tuple(), 1)),
("ii", "", "@column 2", ColumnLabel(tuple(), 2)),
("iii", "", "@column 3", ColumnLabel(tuple(), 3)),
("iv", "", "@column 4", ColumnLabel(tuple(), 4)),
("v", "", "@column 5", ColumnLabel(tuple(), 5)),
("vi", "", "@column 6", ColumnLabel(tuple(), 6)),
("vii", "", "@column 7", ColumnLabel(tuple(), 7)),
("viii", "", "@column 8", ColumnLabel(tuple(), 8)),
("ix", "", "@column 9", ColumnLabel(tuple(), 9)),
("x", "", "@column 10", ColumnLabel(tuple(), 10)),
("i", "'", "@column 1", ColumnLabel((Status.PRIME, ), 1)),
("ii", "?", "@column 2", ColumnLabel((Status.UNCERTAIN, ), 2)),
("iii", "!", "@column 3", ColumnLabel((Status.CORRECTION, ), 3)),
("iv", "*", "@column 4", ColumnLabel((Status.COLLATION, ), 4)),
("v", "'?", "@column 5", ColumnLabel((Status.PRIME, Status.UNCERTAIN), 5)),
]
def test_at_line_column_no_status() -> None:
at_line = ColumnAtLine(ColumnLabel.from_int(1))
assert at_line.lemmatization == (LemmatizationToken("column 1"), )
assert at_line.display_value == "column 1"
def test_at_line_column() -> None:
at_line = ColumnAtLine(ColumnLabel.from_int(1, (atf.Status.COLLATION, )))
assert at_line.lemmatization == (LemmatizationToken("column 1*"), )
assert at_line.display_value == "column 1*"
class LemmatizedFragmentFactory(TransliteratedFragmentFactory):
text = Text((
TextLine.of_iterable(
LineNumber(1, True),
(
Word.of([UnidentifiedSign.of()]),
Word.of([
Logogram.of_name(
"BA",
surrogate=[
Reading.of_name("ku"),
Joiner.hyphen(),
Reading.of_name("u", 4),
],
)
]),
Column.of(),
Tabulation.of(),
Word.of([
BrokenAway.open(),
UnknownNumberOfSigns.of(),
Joiner.hyphen(),
Reading.of_name("ku"),
BrokenAway.close(),
Joiner.hyphen(),
Reading.of_name("nu"),
Joiner.hyphen(),
Reading.of_name("ši"),
]),
Variant.of(Divider.of(":"), Reading.of_name("ku")),
Word.of([
BrokenAway.open(),
UnknownNumberOfSigns.of(),
BrokenAway.close(),
]),
Column.of(2),
Divider.of(":", ("@v", ), (Flag.DAMAGE, )),
CommentaryProtocol.of("!qt"),
Word.of([Number.of_name("10", flags=[Flag.DAMAGE])]),
),
),
TextLine.of_iterable(
LineNumber(2, True),
(
Word.of([BrokenAway.open(),
UnknownNumberOfSigns.of()]),
Word.of([Logogram.of_name("GI", 6)],
unique_lemma=(WordId("ginâ I"), )),
Word.of([Reading.of_name("ana")],
unique_lemma=(WordId("ana I"), )),
Word.of(
[
Reading.of_name("u₄"),
Joiner.hyphen(),
Reading.of_name("š[u"),
],
unique_lemma=(WordId("ūsu I"), ),
),
Word.of([UnknownNumberOfSigns.of(),
BrokenAway.close()]),
),
),
TextLine.of_iterable(
LineNumber(3, True),
(
Word.of([BrokenAway.open(),
UnknownNumberOfSigns.of()]),
Word.of(
unique_lemma=(WordId("kīdu I"), ),
parts=[
Reading.of((
ValueToken.of("k"),
BrokenAway.close(),
ValueToken.of("i"),
)),
Joiner.hyphen(),
Reading.of_name("du"),
],
),
Word.of(unique_lemma=(WordId("u I"), ),
parts=[Reading.of_name("u")]),
Word.of(
unique_lemma=(WordId("bamātu I"), ),
parts=[
Reading.of_name("ba"),
Joiner.hyphen(),
Reading.of_name("ma"),
Joiner.hyphen(),
Reading.of((
ValueToken.of("t"),
BrokenAway.open(),
ValueToken.of("i"),
)),
],
),
Word.of([UnknownNumberOfSigns.of(),
BrokenAway.close()]),
),
),
TextLine.of_iterable(
LineNumber(6, True),
(
Word.of([
BrokenAway.open(),
UnknownNumberOfSigns.of(),
BrokenAway.close(),
]),
Word.of([UnclearSign.of([Flag.DAMAGE])]),
Word.of(unique_lemma=(WordId("mu I"), ),
parts=[Reading.of_name("mu")]),
Word.of(
unique_lemma=(WordId("tamalāku I"), ),
parts=[
Reading.of_name("ta"),
Joiner.hyphen(),
Reading.of_name("ma"),
InWordNewline.of(),
Joiner.hyphen(),
Reading.of_name("tu", 2),
],
),
),
),
TextLine.of_iterable(
LineNumber(7, True),
(
Word.of([
Variant.of(Reading.of_name("šu"),
CompoundGrapheme.of(["BI×IS"]))
]),
LanguageShift.normalized_akkadian(),
AkkadianWord.of([ValueToken.of("kur")],
unique_lemma=(WordId("normalized I"), )),
),
),
StateDollarLine(
atf.Qualification.AT_LEAST,
1,
ScopeContainer(atf.Surface.OBVERSE, ""),
atf.State.MISSING,
None,
),
ImageDollarLine("1", None, "numbered diagram of triangle"),
RulingDollarLine(atf.Ruling.SINGLE),
LooseDollarLine("this is a loose line"),
SealDollarLine(1),
SealAtLine(1),
HeadingAtLine(1),
ColumnAtLine(ColumnLabel([atf.Status.COLLATION], 1)),
SurfaceAtLine(
SurfaceLabel([atf.Status.COLLATION], atf.Surface.SURFACE,
"stone wig")),
ObjectAtLine(
ObjectLabel([atf.Status.COLLATION], atf.Object.OBJECT,
"stone wig")),
DiscourseAtLine(atf.Discourse.DATE),
DivisionAtLine("paragraph", 5),
CompositeAtLine(atf.Composite.DIV, "part", 1),
NoteLine((
StringPart("a note "),
EmphasisPart("italic"),
LanguagePart.of_transliteration(
Language.AKKADIAN, (Word.of([Reading.of_name("bu")]), )),
)),
ParallelComposition(False, "my name", LineNumber(1)),
ParallelText(
True,
TextId(CorpusGenre.LITERATURE, 1, 1),
ChapterName(Stage.OLD_BABYLONIAN, "", "my name"),
LineNumber(1),
False,
),
ParallelFragment(False, MuseumNumber.of("K.1"), True, Labels(),
LineNumber(1), False),
))
class TransliteratedFragmentFactory(FragmentFactory):
text = Text((
TextLine.of_iterable(
LineNumber(1, True),
(
Word.of([UnidentifiedSign.of()]),
Word.of([
Logogram.of_name(
"BA",
surrogate=[
Reading.of_name("ku"),
Joiner.hyphen(),
Reading.of_name("u", 4),
],
)
]),
Column.of(),
Tabulation.of(),
Word.of([
BrokenAway.open(),
UnknownNumberOfSigns.of(),
Joiner.hyphen(),
Reading.of_name("ku"),
BrokenAway.close(),
Joiner.hyphen(),
Reading.of_name("nu"),
Joiner.hyphen(),
Reading.of_name("ši"),
]),
Variant.of(Divider.of(":"), Reading.of_name("ku")),
Word.of([
BrokenAway.open(),
UnknownNumberOfSigns.of(),
BrokenAway.close(),
]),
Column.of(2),
Divider.of(":", ("@v", ), (Flag.DAMAGE, )),
CommentaryProtocol.of("!qt"),
Word.of([Number.of_name("10", flags=[Flag.DAMAGE])]),
),
),
TextLine.of_iterable(
LineNumber(2, True),
(
Word.of([
BrokenAway.open(),
UnknownNumberOfSigns.of(),
BrokenAway.close(),
]),
Word.of([Logogram.of_name("GI", 6)]),
Word.of([Reading.of_name("ana")]),
Word.of([
Reading.of_name("u", 4),
Joiner.hyphen(),
Reading.of((
ValueToken.of("š"),
BrokenAway.open(),
ValueToken.of("u"),
)),
]),
Word.of([UnknownNumberOfSigns.of(),
BrokenAway.close()]),
),
),
TextLine.of_iterable(
LineNumber(3, True),
(
Word.of([BrokenAway.open(),
UnknownNumberOfSigns.of()]),
Word.of([
Reading.of((
ValueToken.of("k"),
BrokenAway.close(),
ValueToken.of("i"),
)),
Joiner.hyphen(),
Reading.of_name("du"),
]),
Word.of([Reading.of_name("u")]),
Word.of([
Reading.of_name("ba"),
Joiner.hyphen(),
Reading.of_name("ma"),
Joiner.hyphen(),
Reading.of((
ValueToken.of("t"),
BrokenAway.open(),
ValueToken.of("i"),
)),
]),
Word.of([UnknownNumberOfSigns.of(),
BrokenAway.close()]),
),
),
TextLine.of_iterable(
LineNumber(6, True),
(
Word.of([
BrokenAway.open(),
UnknownNumberOfSigns.of(),
BrokenAway.close(),
]),
Word.of([UnclearSign.of([Flag.DAMAGE])]),
Word.of([Reading.of_name("mu")]),
Word.of([
Reading.of_name("ta"),
Joiner.hyphen(),
Reading.of_name("ma"),
InWordNewline.of(),
Joiner.hyphen(),
Reading.of_name("tu", 2),
]),
),
),
TextLine.of_iterable(
LineNumber(7, True),
(
Word.of([
Variant.of(Reading.of_name("šu"),
CompoundGrapheme.of(["BI×IS"]))
]),
LanguageShift.normalized_akkadian(),
AkkadianWord.of([ValueToken.of("kur")]),
),
),
StateDollarLine(
atf.Qualification.AT_LEAST,
1,
ScopeContainer(atf.Surface.OBVERSE, ""),
atf.State.MISSING,
None,
),
ImageDollarLine("1", None, "numbered diagram of triangle"),
RulingDollarLine(atf.Ruling.SINGLE),
LooseDollarLine("this is a loose line"),
SealDollarLine(1),
SealAtLine(1),
HeadingAtLine(1),
ColumnAtLine(ColumnLabel([atf.Status.COLLATION], 1)),
SurfaceAtLine(
SurfaceLabel([atf.Status.COLLATION], atf.Surface.SURFACE,
"stone wig")),
ObjectAtLine(
ObjectLabel([atf.Status.COLLATION], atf.Object.OBJECT,
"stone wig")),
DiscourseAtLine(atf.Discourse.DATE),
DivisionAtLine("paragraph", 5),
CompositeAtLine(atf.Composite.DIV, "part", 1),
NoteLine((
StringPart("a note "),
EmphasisPart("italic"),
LanguagePart.of_transliteration(
Language.AKKADIAN, (Word.of([Reading.of_name("bu")]), )),
)),
ParallelComposition(False, "my name", LineNumber(1)),
ParallelText(
True,
TextId(CorpusGenre.LITERATURE, 1, 1),
ChapterName(Stage.OLD_BABYLONIAN, "", "my name"),
LineNumber(1),
False,
),
ParallelFragment(False, MuseumNumber.of("K.1"), True, Labels(),
LineNumber(1), False),
))
signs = (
"X BA KU ABZ075 ABZ207a\\u002F207b\\u0020X ABZ377n1/KU ABZ377n1 ABZ411\n"
"MI DIŠ UD ŠU\n"
"KI DU ABZ411 BA MA TI\n"
"X MU TA MA UD\n"
"ŠU/|BI×IS|")
folios = Folios((Folio("WGL", "3"), Folio("XXX", "3")))
record = Record((RecordEntry("test", RecordType.TRANSLITERATION), ))
line_to_vec = ((
LineToVecEncoding.TEXT_LINE,
LineToVecEncoding.TEXT_LINE,
LineToVecEncoding.TEXT_LINE,
LineToVecEncoding.TEXT_LINE,
LineToVecEncoding.TEXT_LINE,
LineToVecEncoding.SINGLE_RULING,
), )
@pytest.mark.parametrize(
"line,expected_line",
[
(
"// cf. F K.1 &d tablet* o! iii? 1",
ParallelFragment(
True,
MuseumNumber.of("K.1"),
True,
Labels(
ObjectLabel.from_object(atf.Object.TABLET,
[atf.Status.COLLATION]),
SurfaceLabel.from_label(atf.Surface.OBVERSE,
[atf.Status.CORRECTION]),
ColumnLabel.from_int(3, [atf.Status.UNCERTAIN]),
),
LineNumber(1),
),
),
(
"// F K.1 1",
ParallelFragment(False, MuseumNumber.of("K.1"), False, Labels(),
LineNumber(1)),
),
(
'// cf. L I.1 OB "my name" 1',
ParallelText(
True,
TextId(Genre.LITERATURE, 1, 1),
ChapterName(Stage.OLD_BABYLONIAN, "", "my name"),
),
{
"prefix": "@",
"content": [OneOfTokenSchema().dump(ValueToken.of("surface stone wig!*"))],
"type": "SurfaceAtLine",
"surface_label": {
"status": ["CORRECTION", "COLLATION"],
"surface": "SURFACE",
"text": "stone wig",
"abbreviation": "stone wig",
},
"displayValue": "surface stone wig!*",
},
),
(
ColumnAtLine(ColumnLabel([atf.Status.CORRECTION, atf.Status.COLLATION], 1)),
{
"prefix": "@",
"content": [OneOfTokenSchema().dump(ValueToken.of("column 1!*"))],
"type": "ColumnAtLine",
"column_label": {
"status": ["CORRECTION", "COLLATION"],
"column": 1,
"abbreviation": "i",
},
"displayValue": "column 1!*",
},
),
(
SealAtLine(1),
{
"@prism!",
[
ObjectAtLine(
ObjectLabel([atf.Status.CORRECTION], atf.Object.PRISM))
],
),
("@prism", [ObjectAtLine(ObjectLabel([], atf.Object.PRISM))]),
("@object stone",
[ObjectAtLine(ObjectLabel([], atf.Object.OBJECT, "stone"))]),
("@fragment 1",
[ObjectAtLine(ObjectLabel([], atf.Object.FRAGMENT, "1"))]),
("@edge a", [SurfaceAtLine(SurfaceLabel([], atf.Surface.EDGE, "a"))]),
("@face a", [SurfaceAtLine(SurfaceLabel([], atf.Surface.FACE, "a"))]),
("@h1", [HeadingAtLine(1, tuple())]),
("@h1 foo", [HeadingAtLine(1, (StringPart("foo"), ))]),
("@column 1", [ColumnAtLine(ColumnLabel.from_int(1))]),
(
"@column 1!",
[ColumnAtLine(ColumnLabel.from_int(1, (atf.Status.CORRECTION, )))],
),
(
"@column 1!*",
[
ColumnAtLine(
ColumnLabel.from_int(
1, (atf.Status.CORRECTION, atf.Status.COLLATION)))
],
),
("@date", [DiscourseAtLine(atf.Discourse.DATE)]),
],
)
def make_label(self, data, **kwargs) -> ColumnLabel:
return ColumnLabel(data["status"], data["column"])
AnnotationFactory,
AnnotationDataFactory,
)
from ebl.tests.factories.fragment import TransliteratedFragmentFactory
from ebl.transliteration.domain import atf
from ebl.transliteration.domain.labels import ColumnLabel, SurfaceLabel, ObjectLabel
from ebl.transliteration.domain.line_label import LineLabel
from ebl.transliteration.domain.line_number import LineNumber, LineNumberRange
@pytest.mark.parametrize(
"line_label, expected",
[
(
LineLabel(
ColumnLabel.from_int(1),
SurfaceLabel([], atf.Surface.SURFACE, "Stone wig"),
ObjectLabel([], atf.Object.OBJECT, "Stone wig"),
LineNumber(2),
),
"i Stone wig Stone wig 2",
),
(
LineLabel(None, None, None,
LineNumberRange(LineNumber(1, True), LineNumber(3))),
"1'-3",
),
],
)
def test_format_line_label(line_label, expected, annotations_repository,
photo_repository, fragment_repository):
"#tr.en: translation",
TranslationLine((StringPart("translation"),), "en", None),
),
(
"#tr.ar.(2): translation",
TranslationLine(
(StringPart("translation"),), "ar", Extent(LineNumber(2), tuple())
),
),
(
"#tr.(2): translation",
TranslationLine(
(StringPart("translation"),), "en", Extent(LineNumber(2), tuple())
),
),
(
"#tr.de.(o iii 1): translation",
TranslationLine(
(StringPart("translation"),),
"de",
Extent(
LineNumber(1),
(SurfaceLabel(tuple(), Surface.OBVERSE), ColumnLabel(tuple(), 3)),
),
),
),
],
)
def test_parse_translation_line(atf, expected_line) -> None:
assert parse_translation_line(atf) == expected_line
from ebl.transliteration.domain.museum_number import MuseumNumber
from ebl.transliteration.domain.atf import Surface
from ebl.transliteration.domain.enclosure_tokens import BrokenAway
from ebl.transliteration.domain.labels import ColumnLabel, SurfaceLabel
from ebl.transliteration.domain.line_number import LineNumber
from ebl.transliteration.domain.markup import StringPart
from ebl.transliteration.domain.normalized_akkadian import AkkadianWord, Caesura
from ebl.transliteration.domain.note_line import NoteLine
from ebl.transliteration.domain.sign_tokens import Reading
from ebl.transliteration.domain.text_line import TextLine
from ebl.transliteration.domain.tokens import Joiner, ValueToken
from ebl.transliteration.domain.word_tokens import Word
from ebl.transliteration.domain.genre import Genre
MANUSCRIPT_ID = 1
LABELS = (ColumnLabel.from_int(1), )
TEXT_LINE = TextLine(
LineNumber(1),
(
Word.of([Reading.of_name("kur")],
unique_lemma=(WordId("word1"), ),
alignment=0),
Word.of([Reading.of_name("ra")],
unique_lemma=(WordId("word2"), ),
alignment=None),
),
)
NEW_MANUSCRIPT_ID = 2
NEW_LABELS = (SurfaceLabel.from_label(Surface.REVERSE), )
NEW_TEXT_LINE = TextLine(
def test_parse_labels_multiple() -> None:
labels = (SurfaceLabel.from_label(Surface.OBVERSE),
ColumnLabel.from_int(3))
assert parse_labels(" ".join(label.to_value()
for label in labels)) == labels
def ebl_atf_text_line__column(self, number, statuses):
return ColumnAtLine(ColumnLabel.from_int(number, statuses))
def test_load_and_dump_column_label_schema():
column_label = ColumnLabel([], 1)
dump = ColumnLabelSchema().dump(column_label)
assert dump == {"column": 1, "status": [], "abbreviation": "i"}
assert ColumnLabelSchema().load(dump) == column_label