def test_character_detection(self):
subtitles = [
Subtitle(0, 0, "Hello sir"),
Subtitle(0, 0, "my name is micha"),
]
script_entities = [
ScriptEntity(**{
"character": "CHAR0",
"text": "Hallo Mr",
"type": "speech"
}),
ScriptEntity(**{
"character": "CHAR1",
"text": "my name is michi",
"type": "speech"
}),
]
alignment = self._sut(script_entities, subtitles, logger=self._logger)
nw.pretty_print_grid(alignment)
assert len(alignment.subtitles) == 2
assert alignment.subtitles[0].character == "CHAR0"
assert alignment.subtitles[1].character == "CHAR1"
def test_character_difference_same_length(self):
s1 = "millegeeee"
s2 = "mille asde"
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
nw_alignment = self._sut_nw(speeches, subtitles, logger=self._logger)
nw.pretty_print_grid(nw_alignment)
assert nw_alignment.subtitles[0].character is "CHAR0"
def test_bio_string(self):
subtitles = [
Subtitle(0, 0, "GCATGCU"),
]
script_entities = [
ScriptEntity(**{
"character": "CHAR0",
"text": "GATTACA",
"type": "speech"
}),
]
alignment = self._sut(script_entities, subtitles, logger=self._logger)
nw.pretty_print_grid(alignment)
def test_unseparated_words_2(self):
s1 = "Millage Ville, Georgia."
s2 = "Milledgeville, Georgia."
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
dtw_alignment = self._sut_dtw(speeches, subtitles, self._logger,
distance_function=dtw.binary_distance)
dtw.pretty_print_grid(dtw_alignment)
assert dtw_alignment.subtitles[0].character is None
seperator()
nw_alignment = self._sut_nw(speeches, subtitles, logger=self._logger)
nw.pretty_print_grid(nw_alignment)
assert nw_alignment.subtitles[0].character is "CHAR0"
def test_character_difference_repeat(self):
s1 = "three"
s2 = "threeeee"
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
dtw_alignment = self._sut_dtw(speeches, subtitles, self._logger, distance_function=dtw.levenstein_distance)
dtw.pretty_print_grid(dtw_alignment)
assert dtw_alignment.subtitles[0].character is None
seperator()
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
nw_alignment = self._sut_nw(speeches, subtitles, logger=self._logger)
nw.pretty_print_grid(nw_alignment)
assert nw_alignment.subtitles[0].character is "CHAR0"
def test_gap_penalty_alignment_adaptive(self):
s1 = "GAAAAAAT"
s1_index = [(s, None) for s in list(s1)]
s2 = "GAAT"
s2_index = [(s, None) for s in list(s2)]
grid, traceback = nw.nw(s1_index, s2_index,
Weighting(1, -1, AdaptiveGapPenalty(-5, -1)))
alignment = nw.calculate_backtrace(grid, traceback, s1_index, s2_index)
nw.pretty_print_grid(alignment)
s1_string = "".join([char for char, _ in alignment.vertical_index])
s2_string = "".join([char for char, _ in alignment.horizontal_index])
print(s1_string)
print(s2_string)
assert s2_string == "GAA____T"
def test_bio_string_2(self):
subtitles = [
Subtitle(0, 0, "CNJRQCLU"),
]
script_entities = [
ScriptEntity(**{
"character": "CHAR0",
"text": "CJRQDLN",
"type": "speech"
}),
]
alignment = self._sut(script_entities,
subtitles,
Weighting(1, -1, AdaptiveGapPenalty(-5, -1)),
logger=self._logger)
nw.pretty_print_grid(alignment)
def test_unseperated_words_short(self):
s1 = "All right"
s2 = "Alright"
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
dtw_alignment = self._sut_dtw(speeches, subtitles, self._logger,
distance_function=dtw.levenstein_distance)
dtw.pretty_print_grid(dtw_alignment)
assert dtw_alignment.subtitles[0].character is "CHAR0"
seperator()
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
nw_alignment = self._sut_nw(speeches, subtitles, logger=self._logger)
nw.pretty_print_grid(nw_alignment)
assert nw_alignment.subtitles[0].character is "CHAR0"
def test_unseperated_words(self):
s1 = "Hey, All right. All reight."
s2 = "Alright...alright"
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
dtw_alignment = self._sut_dtw(speeches, subtitles, self._logger,
distance_function=dtw.binary_distance)
dtw.pretty_print_grid(dtw_alignment)
assert dtw_alignment.subtitles[0].character is None
seperator()
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
nw_alignment = self._sut_nw(speeches, subtitles, logger=self._logger)
nw.pretty_print_grid(nw_alignment)
assert nw_alignment.subtitles[0].character is "CHAR0"
def test_unseperated_words_long(self):
s1 = "Developed by Master Wuxi in the Third Dynasty"
s2 = "DevelopedbyMasterWuxiInTheThirdDynasty"
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
dtw_alignment = self._sut_dtw(speeches, subtitles, self._logger,
distance_function=dtw.binary_distance)
dtw.pretty_print_grid(dtw_alignment)
assert dtw_alignment.subtitles[0].character is None
seperator()
subtitles = mock_subtitles([s1])
speeches = mock_speeches([("CHAR0", s2)])
nw_alignment = self._sut_nw(speeches, subtitles, logger=self._logger)
nw.pretty_print_grid(nw_alignment)
assert nw_alignment.subtitles[0].character is "CHAR0"