def test_unknown_number_of_signs():
unknown_number_of_signs = UnknownNumberOfSigns(
frozenset({EnclosureType.BROKEN_AWAY}), ErasureState.NONE
)
expected_value = "..."
assert unknown_number_of_signs.value == expected_value
assert unknown_number_of_signs.clean_value == expected_value
assert unknown_number_of_signs.get_key() == f"UnknownNumberOfSigns⁝{expected_value}"
assert unknown_number_of_signs.lemmatizable is False
serialized = {"type": "UnknownNumberOfSigns"}
assert_token_serialization(unknown_number_of_signs, serialized)
def test_text_line_of_iterable(code: str, language: Language) -> None:
tokens = [
Word.of([Reading.of_name("first")]),
LanguageShift.of(code),
Word.of([Reading.of_name("second")]),
LanguageShift.of("%sb"),
LoneDeterminative.of([Determinative.of([Reading.of_name("third")])]),
Word.of([BrokenAway.open(),
Reading.of_name("fourth")]),
UnknownNumberOfSigns.of(),
BrokenAway.close(),
]
expected_tokens = (
Word.of([Reading.of_name("first")], DEFAULT_LANGUAGE),
LanguageShift.of(code),
Word.of([Reading.of_name("second")], language),
LanguageShift.of("%sb"),
LoneDeterminative.of([Determinative.of([Reading.of_name("third")])],
Language.AKKADIAN),
Word.of(
[
BrokenAway.open(),
Reading.of((ValueToken(
frozenset({EnclosureType.BROKEN_AWAY}),
ErasureState.NONE,
"fourth",
), )).set_enclosure_type(frozenset({EnclosureType.BROKEN_AWAY
})),
],
DEFAULT_LANGUAGE,
),
UnknownNumberOfSigns(frozenset({EnclosureType.BROKEN_AWAY}),
ErasureState.NONE),
BrokenAway.close().set_enclosure_type(
frozenset({EnclosureType.BROKEN_AWAY})),
)
line = TextLine.of_iterable(LINE_NUMBER, tokens)
assert line.line_number == LINE_NUMBER
assert line.content == expected_tokens
assert (
line.key ==
f"TextLine⁞{line.atf}⟨{'⁚'.join(token.get_key() for token in expected_tokens)}⟩"
)
assert line.atf == f"1. first {code} second %sb {{third}} [fourth ...]"
class LineVariantFactory(factory.Factory):
class Meta:
model = LineVariant
class Params:
manuscript_id = factory.Sequence(lambda n: n)
manuscript = factory.SubFactory(
ManuscriptLineFactory,
manuscript_id=factory.SelfAttribute("..manuscript_id"),
)
reconstruction = (
LanguageShift.normalized_akkadian(),
AkkadianWord.of((ValueToken.of("buāru"),)),
MetricalFootSeparator.uncertain(),
BrokenAway.open(),
UnknownNumberOfSigns.of(),
Caesura.certain(),
AkkadianWord.of(
(
UnknownNumberOfSigns.of(),
BrokenAway.close(),
Joiner.hyphen(),
ValueToken.of("buāru"),
),
(Flag.DAMAGE,),
),
)
note = factory.fuzzy.FuzzyChoice([None, NoteLine((StringPart("a note"),))])
manuscripts = factory.List([factory.SelfAttribute("..manuscript")], TupleFactory)
intertext = factory.fuzzy.FuzzyChoice([tuple(), (StringPart("bar"),)])
parallel_lines = factory.List(
[
factory.SubFactory(ParallelCompositionFactory),
factory.SubFactory(ParallelTextFactory),
factory.SubFactory(ParallelFragmentFactory),
],
TupleFactory,
)
def test_lemmatization() -> None:
line = TextLine.of_iterable(
LINE_NUMBER,
[
Word.of([Reading.of_name("bu")], unique_lemma=(WordId("nu I"), )),
UnknownNumberOfSigns.of(),
Word.of([Reading.of_name("nu")]),
],
)
assert line.lemmatization == (
LemmatizationToken("bu", (WordId("nu I"), )),
LemmatizationToken("..."),
LemmatizationToken("nu", tuple()),
)
def expected_transliteration(language: Language) -> Sequence[Token]:
return (
Word.of([Reading.of_name("bu")], language),
LanguageShift.of("%es"),
Word.of(
[
BrokenAway.open(),
Reading.of((ValueToken(
frozenset({EnclosureType.BROKEN_AWAY}),
ErasureState.NONE,
"kur",
), )).set_enclosure_type(frozenset({EnclosureType.BROKEN_AWAY
})),
],
Language.EMESAL,
),
UnknownNumberOfSigns(frozenset({EnclosureType.BROKEN_AWAY}),
ErasureState.NONE),
BrokenAway.close().set_enclosure_type(
frozenset({EnclosureType.BROKEN_AWAY})),
)
from ebl.transliteration.domain.sign_tokens import Reading
from ebl.transliteration.domain.tokens import (
EnclosureType,
ErasureState,
LanguageShift,
Token,
UnknownNumberOfSigns,
ValueToken,
)
from ebl.transliteration.domain.word_tokens import Word
TRANSLITERATION: Sequence[Token] = (
Word.of([Reading.of_name("bu")]),
LanguageShift.of("%es"),
Word.of([BrokenAway.open(), Reading.of_name("kur")]),
UnknownNumberOfSigns.of(),
BrokenAway.close(),
)
EXPECTED_ATF = "bu %es [kur ...]"
def expected_transliteration(language: Language) -> Sequence[Token]:
return (
Word.of([Reading.of_name("bu")], language),
LanguageShift.of("%es"),
Word.of(
[
BrokenAway.open(),
Reading.of((ValueToken(
frozenset({EnclosureType.BROKEN_AWAY}),
ErasureState.NONE,
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,
), )
from ebl.transliteration.domain.language import Language
from ebl.transliteration.domain.sign_tokens import Divider, Reading
from ebl.transliteration.domain.tokens import (
Column,
CommentaryProtocol,
ErasureState,
Joiner,
LanguageShift,
Tabulation,
UnknownNumberOfSigns,
ValueToken,
Variant,
)
TOKENS = [
UnknownNumberOfSigns(frozenset({EnclosureType.BROKEN_AWAY}), ErasureState.NONE),
LanguageShift.of("%sux"),
DocumentOrientedGloss.open(),
]
def test_value_token():
value = "value"
token = ValueToken.of(value)
equal = ValueToken.of(value)
other = ValueToken.of("anothervalue")
assert token.value == value
assert token.clean_value == value
assert token.get_key() == f"ValueToken⁝{value}"
assert token.lemmatizable is False
ValueToken,
Variant,
)
from ebl.transliteration.domain.unknown_sign_tokens import UnclearSign, UnidentifiedSign
from ebl.transliteration.domain.word_tokens import (
ErasureState,
InWordNewline,
LoneDeterminative,
Word,
)
@pytest.mark.parametrize( # pyre-ignore[56]
"atf,expected",
[
("...", Word.of([UnknownNumberOfSigns.of()])),
("x", Word.of([UnclearSign.of()])),
("X", Word.of([UnidentifiedSign.of()])),
("x?", Word.of([UnclearSign.of([atf.Flag.UNCERTAIN])])),
("X#", Word.of([UnidentifiedSign.of([atf.Flag.DAMAGE])])),
("12", Word.of([Number.of_name("12")])),
(
"1]2",
Word.of([
Number.of((ValueToken.of("1"), BrokenAway.close(),
ValueToken.of("2")))
]),
),
(
"1[2",
Word.of([
def make_token(self, data, **kwargs):
return UnknownNumberOfSigns(frozenset(data["enclosure_type"]), data["erasure"])
@pytest.mark.parametrize("parser,version",
[(parse_atf_lark, f"{atf.ATF_PARSER_VERSION}")])
def test_parser_version(parser, version):
assert parser("1. kur").parser_version == version
@pytest.mark.parametrize( # pyre-ignore[56]
"line,expected_tokens",
[
(
"1′. ...",
[
TextLine.of_iterable(LineNumber(1, True), (Word.of(
(UnknownNumberOfSigns.of(), )), ))
],
),
(
"1’. ...",
[
TextLine.of_iterable(LineNumber(1, True), (Word.of(
(UnknownNumberOfSigns.of(), )), ))
],
),
(
"D+113'a. ...",
[
TextLine.of_iterable(
LineNumber(113, True, "D", "a"),
(Word.of((UnknownNumberOfSigns.of(), )), ),
Joiner,
)
from ebl.transliteration.domain.unknown_sign_tokens import UnclearSign, UnidentifiedSign
from ebl.transliteration.domain.word_tokens import Word
def map_line(atf) -> Sequence[Token]:
visitor = EnclosureUpdater()
parse_line(f"1. {atf}").accept(visitor)
return visitor.tokens
@pytest.mark.parametrize(
"atf, expected",
[
("...", (Word.of((UnknownNumberOfSigns.of(), )), )),
(
"[...]",
(Word.of((
BrokenAway.open(),
UnknownNumberOfSigns(frozenset({EnclosureType.BROKEN_AWAY}),
ErasureState.NONE),
BrokenAway.close().set_enclosure_type(
frozenset({EnclosureType.BROKEN_AWAY})),
)), ),
),
(
"(...)",
(Word.of((
PerhapsBrokenAway.open(),
UnknownNumberOfSigns(frozenset({EnclosureType.PERHAPS}),
("<ibnû>", [Emendation.open(), ValueToken.of("ibnû"), Emendation.close(), []]),
("ib<nû", [ValueToken.of("ib"), Emendation.open(), ValueToken.of("nû"), []]),
("ib>nû", [ValueToken.of("ib"), Emendation.close(), ValueToken.of("nû"), []]),
(
"i<b>nû",
[
ValueToken.of("i"),
Emendation.open(),
ValueToken.of("b"),
Emendation.close(),
ValueToken.of("nû"),
[],
],
),
("ibnû?>", [ValueToken.of("ibnû"), Emendation.close(), [Flag.UNCERTAIN]]),
("...ibnû", [UnknownNumberOfSigns.of(), ValueToken.of("ibnû"), []]),
("ibnû...", [ValueToken.of("ibnû"), UnknownNumberOfSigns.of(), []]),
(
"ib...nû",
[ValueToken.of("ib"), UnknownNumberOfSigns.of(), ValueToken.of("nû"), []],
),
(
"<...ibnû",
[Emendation.open(), UnknownNumberOfSigns.of(), ValueToken.of("ibnû"), []],
),
(
"ibnû...>",
[ValueToken.of("ibnû"), UnknownNumberOfSigns.of(), Emendation.close(), []],
),
(
"...>ibnû",
BrokenAway.close(),
)),
"[(<ib)nû>]",
True,
),
(
AkkadianWord.of(
(ValueToken.of("ibnû"), PerhapsBrokenAway.close(),
BrokenAway.close()),
(Flag.UNCERTAIN, ),
),
"ibnû?)]",
True,
),
(
AkkadianWord.of((ValueToken.of("ib"), UnknownNumberOfSigns.of(),
ValueToken.of("nû"))),
"ib...nû",
False,
),
(
AkkadianWord.of(
(ValueToken.of("ib"), Joiner.hyphen(), ValueToken.of("nû"))),
"ib-nû",
True,
),
],
)
def test_akkadian_word(word: AkkadianWord, expected: str,
lemmatizable: bool) -> None:
assert word.value == expected
def ebl_atf_text_line__unknown_number_of_signs(self, _):
return UnknownNumberOfSigns.of()
def test_extant_lines_mixed_sides() -> None:
manuscript = Manuscript(
MANUSCRIPT_ID,
siglum_disambiguator="1",
period_modifier=PeriodModifier.NONE,
period=Period.LATE_BABYLONIAN,
provenance=Provenance.BABYLON,
type=ManuscriptType.SCHOOL,
)
manuscript_line = ManuscriptLine(
MANUSCRIPT_ID,
LABELS,
TextLine(
LineNumberRange(LineNumber(1), LineNumber(3, suffix_modifier="b")),
(UnknownNumberOfSigns.of(), ),
),
PARATEXT,
OMITTED_WORDS,
)
manuscript_line2 = ManuscriptLine(
MANUSCRIPT_ID,
tuple(),
TextLine(LineNumber(2), (UnknownNumberOfSigns.of(), )),
PARATEXT,
OMITTED_WORDS,
)
manuscript_line3 = ManuscriptLine(
MANUSCRIPT_ID,
LABELS,
TextLine(LineNumber(3), (UnknownNumberOfSigns.of(), )),
PARATEXT,
OMITTED_WORDS,
)
chapter = Chapter(
TEXT_ID,
manuscripts=(manuscript, ),
lines=(
Line(
LineNumber(1),
(LineVariant(LINE_RECONSTRUCTION, NOTE,
(manuscript_line, ), PARALLEL_LINES), ),
),
Line(
LineNumber(2),
(LineVariant(LINE_RECONSTRUCTION, NOTE,
(manuscript_line2, ), PARALLEL_LINES), ),
),
Line(
LineNumber(3),
(LineVariant(LINE_RECONSTRUCTION, NOTE,
(manuscript_line3, ), PARALLEL_LINES), ),
),
),
)
assert chapter.extant_lines == {
manuscript.siglum: {
manuscript_line.labels: [
ExtantLine(manuscript_line.labels, LineNumber(1), True),
ExtantLine(manuscript_line3.labels, LineNumber(3), False),
],
manuscript_line2.labels:
[ExtantLine(manuscript_line2.labels, LineNumber(2), False)],
}
}