def test_at_line_surface_no_status() -> None:
at_line = SurfaceAtLine(SurfaceLabel([], atf.Surface.SURFACE, "Stone wig"))
assert at_line.lemmatization == (LemmatizationToken("surface Stone wig"), )
assert at_line.surface_label == SurfaceLabel([], atf.Surface.SURFACE,
"Stone wig")
assert at_line.display_value == "surface Stone wig"
def test_at_line_surface_instantiate_text_with_wrong_surface() -> None:
with pytest.raises(ValueError):
at_line = SurfaceAtLine(
SurfaceLabel((atf.Status.CORRECTION, ), atf.Surface.OBVERSE,
"Stone wig"))
assert at_line.lemmatization == (
LemmatizationToken("obverse Stone wig!"), )
assert at_line.surface_label == SurfaceLabel(
(atf.Status.CORRECTION, ), atf.Surface.OBVERSE, "Stone wig")
assert at_line.display_value == "obverse Stone wig!"
def test_at_line_surface() -> None:
at_line = SurfaceAtLine(
SurfaceLabel((atf.Status.CORRECTION, ), atf.Surface.SURFACE,
"Stone wig"))
assert at_line.lemmatization == (
LemmatizationToken("surface Stone wig!"), )
assert at_line.surface_label == SurfaceLabel(
(atf.Status.CORRECTION, ), atf.Surface.SURFACE, "Stone wig")
assert at_line.display_value == "surface Stone wig!"
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")])]),
])
def test_load_and_dump_surface_label_schema():
surface_label = SurfaceLabel([atf.Status.CORRECTION], atf.Surface.SURFACE,
"stone wig")
dump = SurfaceLabelSchema().dump(surface_label)
assert dump == {
"status": ["CORRECTION"],
"surface": "SURFACE",
"text": "stone wig",
"abbreviation": "stone wig",
}
assert SurfaceLabelSchema().load(dump) == surface_label
ChapterName,
Labels,
ParallelComposition,
ParallelFragment,
ParallelText,
)
@pytest.mark.parametrize(
"cf,duplicates,labels,display_value",
[
(
True,
True,
Labels(
surface=SurfaceLabel((Status.CORRECTION, ), Surface.OBVERSE)),
"cf. F K.1 &d o! 1",
),
(False, False, Labels(), "F K.1 1"),
],
)
def test_parallel_fragment(cf, duplicates, labels, display_value) -> None:
museum_number = MuseumNumber.of("K.1")
line_number = LineNumber(1)
line = ParallelFragment(cf, museum_number, duplicates, labels, line_number)
assert line.has_cf is cf
assert line.museum_number == museum_number
assert line.has_duplicates is duplicates
assert line.labels == labels
assert line.line_number == line_number
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),
))
"#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
"content": [OneOfTokenSchema().dump(ValueToken.of("object stone wig!*"))],
"type": "ObjectAtLine",
"label": {
"status": ["CORRECTION", "COLLATION"],
"object": "OBJECT",
"text": "stone wig",
"abbreviation": "stone wig",
},
"displayValue": "object stone wig!*",
},
),
(
SurfaceAtLine(
SurfaceLabel(
[atf.Status.CORRECTION, atf.Status.COLLATION],
atf.Surface.SURFACE,
"stone wig",
)
),
{
"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!*",
},
"line,expected_tokens",
[
("@composite", [CompositeAtLine(atf.Composite.COMPOSITE, "")]),
("@m=locator o 1", [CompositeAtLine(atf.Composite.MILESTONE, "o", 1)]),
("@m=locator o", [CompositeAtLine(atf.Composite.MILESTONE, "o")]),
("@div part 1", [CompositeAtLine(atf.Composite.DIV, "part", 1)]),
("@end part", [CompositeAtLine(atf.Composite.END, "part")]),
("@m=division paragraph ", [DivisionAtLine("paragraph")]),
("@m=division paragraph 1", [DivisionAtLine("paragraph", 1)]),
("@date", [DiscourseAtLine(atf.Discourse.DATE)]),
(
"@reverse!*",
[
SurfaceAtLine(
SurfaceLabel(
[atf.Status.CORRECTION, atf.Status.COLLATION],
atf.Surface.REVERSE,
))
],
),
(
"@reverse!",
[
SurfaceAtLine(
SurfaceLabel([atf.Status.CORRECTION], atf.Surface.REVERSE))
],
),
("@reverse", [SurfaceAtLine(SurfaceLabel([], atf.Surface.REVERSE))]),
("@edge", [SurfaceAtLine(SurfaceLabel([], atf.Surface.EDGE, ""))]),
("@edge a", [SurfaceAtLine(SurfaceLabel([], atf.Surface.EDGE, "a"))]),
(
"@surface stone",
[
ControlLine("#", "first"),
EmptyLine(),
ControlLine("#", "second")
],
),
(
"#first\n \n#second",
[
ControlLine("#", "first"),
EmptyLine(),
ControlLine("#", "second")
],
),
("&K11111", [ControlLine("&", "K11111")]),
("@reverse", [SurfaceAtLine(SurfaceLabel([], atf.Surface.REVERSE))]),
(
"$ (end of side)",
[
StateDollarLine(
None,
atf.Extent.END_OF,
ScopeContainer(atf.Scope.SIDE, ""),
None,
None,
)
],
),
("#some notes", [ControlLine("#", "some notes")]),
("=: continuation", [ControlLine("=:", " continuation")]),
],
def make_label(self, data, **kwargs) -> SurfaceLabel:
return SurfaceLabel(data["status"], data["surface"], data["text"])
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):
image_extractor = AnnotationImageExtractor(fragment_repository,
import pytest
from ebl.transliteration.domain.atf import Status, Surface, Object
from ebl.transliteration.domain.labels import (
parse_labels,
ColumnLabel,
Label,
LabelVisitor,
SurfaceLabel,
ObjectLabel,
)
from ebl.transliteration.domain.lark_parser import PARSE_ERRORS
LABELS: List[Tuple[str, str, str, Label]] = [
("o", "", "@obverse", SurfaceLabel(tuple(), Surface.OBVERSE)),
("r", "", "@reverse", SurfaceLabel(tuple(), Surface.REVERSE)),
("b.e.", "", "@bottom", SurfaceLabel(tuple(), Surface.BOTTOM)),
("e.", "", "@edge", SurfaceLabel(tuple(), Surface.EDGE)),
("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.",
"*!",
def ebl_atf_text_line__surface_with_status(self, surface: SurfaceLabel,
statuses):
return SurfaceAtLine(
SurfaceLabel(statuses, surface.surface, surface.text))
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,
), )