def test_language_shift(value, expected_language, normalized):
shift = LanguageShift.of(value)
equal = LanguageShift.of(value)
other = ValueToken.of(r"%bar")
assert shift.value == value
assert shift.clean_value == value
assert shift.get_key() == f"LanguageShift⁝{value}"
assert shift.lemmatizable is False
assert shift.normalized == normalized
assert shift.language == expected_language
serialized = {
"type": "LanguageShift",
"normalized": normalized,
"language": shift.language.name,
}
assert_token_serialization(shift, serialized)
assert shift == equal
assert hash(shift) == hash(equal)
assert shift != other
assert hash(shift) != hash(other)
assert shift != ValueToken.of(value)
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 ...]"
def test_parse_atf_language_shifts(code: str,
expected_language: Language) -> None:
word = "ha-am"
parts = [Reading.of_name("ha"), Joiner.hyphen(), Reading.of_name("am")]
line = f"1. {word} {code} {word} %sb {word}"
expected = Text((TextLine.of_iterable(
LineNumber(1),
(
Word.of(parts, DEFAULT_LANGUAGE),
LanguageShift.of(code),
Word.of(parts, expected_language),
LanguageShift.of("%sb"),
Word.of(parts, Language.AKKADIAN),
),
), ))
assert parse_atf_lark(line).lines == expected.lines
def test_parse_normalized_akkadain_shift() -> None:
word = "ha"
line = f"1. {word} %n {word} %sux {word}"
expected = Text((TextLine.of_iterable(
LineNumber(1),
(
Word.of((Reading.of_name(word), ), DEFAULT_LANGUAGE),
LanguageShift.normalized_akkadian(),
AkkadianWord.of((ValueToken.of(word), )),
LanguageShift.of("%sux"),
Word.of((Reading.of_name(word), ), Language.SUMERIAN),
),
), ))
assert parse_atf_lark(line).lines == expected.lines
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.markup import EmphasisPart, LanguagePart, StringPart
from ebl.transliteration.domain.note_line import NoteLine
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(
LineNumber(1),
[
Word.of(unique_lemma=(WordId("nu I"), ),
parts=[Reading.of_name("bu")])
],
),
),
(
TextLine.of_iterable(
LineNumber(1),
[
Word.of(unique_lemma=(WordId("nu I"), ),
parts=[Reading.of_name("bu")])
],
),
TextLine.of_iterable(LineNumber(1), [LanguageShift.of("%sux")]),
TextLine.of_iterable(LineNumber(1), [LanguageShift.of("%sux")]),
),
(
TextLine.of_iterable(
LineNumber(1),
[
Word.of(unique_lemma=(WordId("nu I"), ),
parts=[Reading.of_name("bu")])
],
),
TextLine.of_iterable(LineNumber(1),
[Word.of([Reading.of_name("mu")])]),
TextLine.of_iterable(LineNumber(1),
[Word.of([Reading.of_name("mu")])]),
),
def ebl_atf_text_line__greek_shift(self, value):
return LanguageShift.of(str(value))
def ebl_atf_text_line__normalized_akkadian_shift(self, value):
return LanguageShift.of(str(value))
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
Joiner.hyphen(),
UnknownNumberOfSigns.of(),
Joiner.hyphen(),
Reading.of_name("ad"),
)),
),
)
],
),
(
"1. %grc ΑαΒβΓγΔδΕεΖζΗηΘθΙιΚκΛλΜμΝνΞξΟοΠπΡρΣσςΤτΥυΦφΧχΨψΩω",
[
TextLine.of_iterable(
LineNumber(1),
(
LanguageShift.of("%grc"),
GreekWord.of([
GreekLetter.of(letter)
for letter in ("ΑαΒβΓγΔδΕεΖζΗηΘθΙιΚκΛλΜμ"
"ΝνΞξΟοΠπΡρΣσςΤτΥυΦφΧχΨψΩω")
]),
),
)
],
),
(
"1. %akkgrc α",
[
TextLine.of_iterable(
LineNumber(1),
(