def test_damerau_levenshtein_distance(self):
"""Test for Damerau-Levenshtein Distance between two words"""
l = Levenshtein()
dist = l.Damerau_Levenshtein_Distance(
"all haile whose solempne glorious concepcioun",
"fresche floure in quhom the hevinlie dewe doun fell")
self.assertEqual(dist, 35)
def test_levenshtein_distance(self):
"""Test for Levenshtein Distance between two words"""
l = Levenshtein()
dist = l.Levenshtein_Distance(
"now grete glorious god through grace of himselven",
"and the precious prayer of his pris moder")
self.assertEqual(dist, 36)
def levenshtein_distance(self, other_word: str) -> int:
"""
Basic use of Levenshtein distance function
:param other_word:
:return:
"""
return Levenshtein.Levenshtein_Distance(self.word, other_word)
def _calculate_ratios(self, list_a, list_b):
"""
Calulate a matrix of string comparisons given two input lists
:param list_a: list [object]
:param list_b: list [object]
:return: list [[Comparison]]
"""
comparisons = []
l = Levenshtein()
# For all strings in list a
for i, str_a in enumerate(list_a):
# Add a new list to our list of lists of comparisons
comparisons.append([])
# Compare str_a to every string in list_b
for str_b in list_b:
# If the sanitize, input flag is set, make the ratio with the sanitized values
if self.sanitize_input:
new_comparison = Comparison(
str_a['text'],
str_b['text'],
l.ratio(str_a['sanitized'], str_b['sanitized'])
)
# Otherwise, make the ratio with the original, unsanitize text strings
else:
new_comparison = Comparison(
str_a['text'],
str_b['text'],
l.ratio(str_a['text'], str_b['text'])
)
# If text metadata is set on this class for text a or b, save that data with the
# comparison
if self.text_ref_a:
new_comparison.set_ref_a(self.text_ref_a)
if self.text_ref_b:
new_comparison.set_ref_b(self.text_ref_b)
# Finally, append the new comparison to the list of comparisons
comparisons[i].append(new_comparison)
return comparisons
def compute_distance_matrix(proper_nouns: List[str]):
"""
Distance matrix with Levenshtein distance.
:param proper_nouns: Items of proper_nouns must be unique
:return:
"""
levenshtein = Levenshtein()
# We try to keep regroup different forms of a lemma
distance_matrix = np.zeros((len(proper_nouns), len(proper_nouns)))
for i in range(len(proper_nouns)):
for j in range(len(proper_nouns)):
distance_matrix[i, j] = levenshtein.Levenshtein_Distance(proper_nouns[i], proper_nouns[j])
return distance_matrix
def _closest_patterns(self, patterns: List[str],
scansion: str) -> List[str]:
"""
Find the closest group of matching valid patterns.
:patterns: a list of patterns
:scansion: the scansion pattern thus far
:return: list of the closest valid patterns; only candidates with a matching
length/number of syllables are considered.
"""
pattern = scansion.replace(" ", "")
pattern = pattern.replace(self.constants.FOOT_SEPARATOR, "")
ending = pattern[-1]
candidate = pattern[:len(pattern) - 1] + self.constants.OPTIONAL_ENDING
cans = [(Levenshtein.levenshtein_distance(candidate, x), x)
for x in patterns if len(x) == len(candidate)]
if cans:
cans = sorted(cans, key=lambda tup: tup[0])
top = cans[0][0]
return [can[1][:-1] + ending for can in cans if can[0] == top]
return []
def test_distance_ratio(self):
"""Test returning simple Levenshtein distance calculation ratio between two strings"""
l = Levenshtein()
ratio = l.ratio("dique deaeque omnes, studium quibus arua tueri,",
"dique deaeque omnes, quibus est tutela per agros,")
self.assertEqual(ratio, 0.71)
def test_distance_ratio(self):
"""Test returning simple Levenshtein distance calculation ratio between two strings"""
l = Levenshtein()
ratio = l.ratio("dique deaeque omnes, studium quibus arua tueri,", "dique deaeque omnes, quibus est tutela per agros,")
self.assertEqual(ratio, 0.71)
def compare_levenshtein(self, other_text):
return Levenshtein().ratio(self.data, other_text)
def scan(self, original_line: str, optional_transform: bool = False) -> Verse:
"""
Scan a line of Latin pentameter and produce a scansion pattern, and other data.
:param original_line: the original line of Latin verse
:param optional_transform: whether or not to perform i to j transform for syllabification
:return: a Verse object
>>> scanner = PentameterScanner()
>>> print(scanner.scan('ex hoc ingrato gaudia amore tibi.'))
Verse(original='ex hoc ingrato gaudia amore tibi.', scansion='- - - - - - U U - U U U ', meter='pentameter', valid=True, syllable_count=12, accented='ēx hōc īngrātō gaudia amōre tibi.', scansion_notes=['Spondaic pentameter'], syllables = ['ēx', 'hoc', 'īn', 'gra', 'to', 'gau', 'di', 'a', 'mo', 're', 'ti', 'bi'])
>>> print(scanner.scan(
... "in vento et rapida scribere oportet aqua.").scansion) # doctest: +NORMALIZE_WHITESPACE
- - - U U - - U U - U U U
"""
verse = Verse(original_line, meter='pentameter')
# replace punctuation with spaces
line = original_line.translate(self.punctuation_substitutions)
if optional_transform:
working_line = self.transform_i_to_j_optional(line)
verse.scansion_notes += [self.constants.NOTE_MAP["optional i to j"]]
else:
working_line = self.transform_i_to_j(line) # conservative i to j
working_line = self.elide_all(working_line)
verse.working_line = self.accent_by_position(working_line)
verse.syllables = self.syllabifier.syllabify(verse.working_line)
verse.syllable_count = self.syllabifier.get_syllable_count(verse.syllables)
if verse.syllable_count < 12:
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["< 12p"]]
return verse
stresses = self.flag_dipthongs(verse.syllables)
syllables_wspaces = string_utils.to_syllables_with_trailing_spaces(verse.working_line, verse.syllables)
offset_map = self.calc_offset(syllables_wspaces)
for idx, syl in enumerate(verse.syllables):
for accented in self.constants.ACCENTED_VOWELS:
if accented in syl:
stresses.append(idx)
# first syllable is always long in Pentameter
stresses.append(0)
# second to last syllable is always long
stresses.append(verse.syllable_count - 2)
verse.scansion = self.produce_scansion(stresses,
syllables_wspaces, offset_map)
if len(string_utils.stress_positions(self.constants.STRESSED, verse.scansion)) != \
len(set(stresses)):
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["invalid syllables"]]
return verse
if self.metrical_validator.is_valid_pentameter(verse.scansion):
verse.scansion_notes += [self.constants.NOTE_MAP["positionally"]]
return self.assign_candidate(verse, verse.scansion)
# identify some obvious and probably choices based on number of syllables
if verse.syllable_count == 12: # produce spondees where possible
candidate = self.make_spondaic(verse.scansion)
verse.scansion_notes += [self.constants.NOTE_MAP["12p"]]
return self.assign_candidate(verse, candidate)
if verse.syllable_count == 14: # produce spondees where possible
candidate = self.make_dactyls(verse.scansion)
verse.scansion_notes += [self.constants.NOTE_MAP["14p"]]
return self.assign_candidate(verse, candidate)
if verse.syllable_count > 14:
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["> 14"]]
return verse
smoothed = self.correct_first_two_dactyls(verse.scansion)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [self.constants.NOTE_MAP["invalid start"]]
verse.scansion = smoothed
stresses += string_utils.differences(verse.scansion, smoothed)
if self.metrical_validator.is_valid_pentameter(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
smoothed = self.correct_penultimate_dactyl_chain(verse.scansion)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [self.constants.NOTE_MAP["penultimate dactyl chain"]]
verse.scansion = smoothed
stresses += string_utils.differences(verse.scansion, smoothed)
if self.metrical_validator.is_valid_pentameter(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
candidates = self.metrical_validator.closest_pentameter_patterns(verse.scansion)
if candidates is not None:
if len(candidates) == 1 \
and len(verse.scansion.replace(" ", "")) == len(candidates[0]) \
and len(string_utils.differences(verse.scansion, candidates[0])) == 1:
tmp_scansion = self.produce_scansion(
string_utils.differences(verse.scansion, candidates[0]),
syllables_wspaces, offset_map)
if self.metrical_validator.is_valid_pentameter(tmp_scansion):
verse.scansion_notes += [self.constants.NOTE_MAP["closest match"]]
return self.assign_candidate(verse, tmp_scansion)
# if the line doesn't scan "as is", it may scan if the optional i to j transformations
# are made, so here we set them and try again.
if self.optional_transform and not optional_transform and not verse.valid:
return self.scan(original_line, optional_transform=True)
verse.accented = self.formatter.merge_line_scansion(verse.original, verse.scansion)
return verse
def scan(self,
original_line: str,
optional_transform: bool = False) -> Verse:
"""
Scan a line of Latin hendecasyllables and produce a scansion pattern, and other data.
:param original_line: the original line of Latin verse
:param optional_transform: whether or not to perform i to j transform for syllabification
:return: a Verse object
>>> scanner = HendecasyllableScanner()
>>> print(scanner.scan("Cui dono lepidum novum libellum"))
Verse(original='Cui dono lepidum novum libellum', scansion=' - U - U U - U - U - U ', meter='hendecasyllable', valid=True, syllable_count=11, accented='Cui donō lepidūm novūm libēllum', scansion_notes=['Corrected invalid start.'], syllables = ['Cui', 'do', 'no', 'le', 'pi', 'dūm', 'no', 'vūm', 'li', 'bēl', 'lum'])
>>> print(scanner.scan(
... "ārida modo pumice expolitum?").scansion) # doctest: +NORMALIZE_WHITESPACE
- U - U U - U - U - U
"""
verse = Verse(original_line, meter='hendecasyllable')
# replace punctuation with spaces
line = original_line.translate(self.punctuation_substitutions)
if optional_transform:
working_line = self.transform_i_to_j_optional(line)
verse.scansion_notes += [
self.constants.NOTE_MAP["optional i to j"]
]
else:
working_line = self.transform_i_to_j(line) # conservative i to j
working_line = self.elide_all(working_line)
verse.working_line = self.accent_by_position(working_line)
verse.syllables = self.syllabifier.syllabify(verse.working_line)
verse.syllable_count = self.syllabifier.get_syllable_count(
verse.syllables)
# identify some obvious and probably choices based on number of syllables
if verse.syllable_count > 11:
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["> 11"]]
return verse
if verse.syllable_count < 11:
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["< 11"]]
return verse
stresses = self.flag_dipthongs(verse.syllables)
syllables_wspaces = string_utils.to_syllables_with_trailing_spaces(
verse.working_line, verse.syllables)
offset_map = self.calc_offset(syllables_wspaces)
for idx, syl in enumerate(verse.syllables):
for accented in self.constants.ACCENTED_VOWELS:
if accented in syl:
stresses.append(idx)
# second to last syllable is always long
stresses.append(verse.syllable_count - 2)
verse.scansion = self.produce_scansion(stresses, syllables_wspaces,
offset_map)
if len(string_utils.stress_positions(self.constants.STRESSED, verse.scansion)) != \
len(set(stresses)):
verse.valid = False
verse.scansion_notes += [
self.constants.NOTE_MAP["invalid syllables"]
]
return verse
if self.metrical_validator.is_valid_hendecasyllables(verse.scansion):
verse.scansion_notes += [self.constants.NOTE_MAP["positionally"]]
return self.assign_candidate(verse, verse.scansion)
smoothed = self.correct_invalid_start(verse.scansion)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [self.constants.NOTE_MAP["invalid start"]]
verse.scansion = smoothed
stresses += string_utils.differences(verse.scansion, smoothed)
if self.metrical_validator.is_valid_hendecasyllables(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
smoothed = self.correct_antepenult_chain(verse.scansion)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [
self.constants.NOTE_MAP["antepenult chain"]
]
verse.scansion = smoothed
stresses += string_utils.differences(verse.scansion, smoothed)
if self.metrical_validator.is_valid_hendecasyllables(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
candidates = self.metrical_validator.closest_hendecasyllable_patterns(
verse.scansion)
if candidates is not None:
if len(candidates) == 1 \
and len(verse.scansion.replace(" ", "")) == len(candidates[0]) \
and len(string_utils.differences(verse.scansion, candidates[0])) == 1:
tmp_scansion = self.produce_scansion(
string_utils.differences(verse.scansion, candidates[0]),
syllables_wspaces, offset_map)
if self.metrical_validator.is_valid_hendecasyllables(
tmp_scansion):
verse.scansion_notes += [
self.constants.NOTE_MAP["closest match"]
]
return self.assign_candidate(verse, tmp_scansion)
# if the line doesn't scan "as is", if may scan if the optional i to j transformations
# are made, so here we set them and try again.
if self.optional_transform and not optional_transform and not verse.valid:
return self.scan(original_line, optional_transform=True)
verse.accented = self.formatter.merge_line_scansion(
verse.original, verse.scansion)
return verse
def scan(self,
original_line: str,
optional_transform: bool = False,
dactyl_smoothing: bool = False) -> Verse:
"""
Scan a line of Latin hexameter and produce a scansion pattern, and other data.
:param original_line: the original line of Latin verse
:param optional_transform: whether or not to perform i to j transform for syllabification
:param dactyl_smoothing: whether or not to perform dactyl smoothing
:return: a Verse object
>>> scanner = HexameterScanner()
>>> print(HexameterScanner().scan(
... "ēxiguām sedēm pariturae tērra negavit").scansion) # doctest: +NORMALIZE_WHITESPACE
- - - - - U U - - - U U - U
>>> print(scanner.scan("impulerit. Tantaene animis caelestibus irae?"))
Verse(original='impulerit. Tantaene animis caelestibus irae?', scansion='- U U - - - U U - - - U U - - ', meter='hexameter', valid=True, syllable_count=15, accented='īmpulerīt. Tāntaene animīs caelēstibus īrae?', scansion_notes=['Valid by positional stresses.'], syllables = ['īm', 'pu', 'le', 'rīt', 'Tān', 'taen', 'a', 'ni', 'mīs', 'cae', 'lēs', 'ti', 'bus', 'i', 'rae'])
>>> print(scanner.scan(
... "Arma virumque cano, Troiae qui prīmus ab ōrīs").scansion) # doctest: +NORMALIZE_WHITESPACE
- U U - U U - - - - - U U - -
>>> # some hexameters need the optional transformations:
>>> optional_transform_scanner = HexameterScanner(optional_transform=True)
>>> print(optional_transform_scanner.scan(
... "Ītaliam, fāto profugus, Lāvīniaque vēnit").scansion) # doctest: +NORMALIZE_WHITESPACE
- - - - - U U - - - U U - U
>>> print(HexameterScanner().scan(
... "lītora, multum ille et terrīs iactātus et alto").scansion) # doctest: +NORMALIZE_WHITESPACE
- U U - - - - - - - U U - U
>>> print(HexameterScanner().scan(
... "vī superum saevae memorem Iūnōnis ob īram;").scansion) # doctest: +NORMALIZE_WHITESPACE
- U U - - - U U - - - U U - U
>>> # handle multiple elisions
>>> print(scanner.scan("monstrum horrendum, informe, ingens, cui lumen ademptum").scansion) # doctest: +NORMALIZE_WHITESPACE
- - - - - - - - - U U - U
>>> # if we have 17 syllables, create a chain of all dactyls
>>> print(scanner.scan("quadrupedante putrem sonitu quatit ungula campum"
... ).scansion) # doctest: +NORMALIZE_WHITESPACE
- U U - U U - U U - U U - U U - U
>>> # if we have 13 syllables exactly, we'll create a spondaic hexameter
>>> print(HexameterScanner().scan(
... "illi inter sese multa vi bracchia tollunt").scansion) # doctest: +NORMALIZE_WHITESPACE
- - - - - - - - - UU - -
>>> print(HexameterScanner().scan(
... "dat latus; insequitur cumulo praeruptus aquae mons").scansion) # doctest: +NORMALIZE_WHITESPACE
- U U - U U - U U - - - U U - -
>>> print(optional_transform_scanner.scan(
... "Non quivis videt inmodulata poëmata iudex").scansion) # doctest: +NORMALIZE_WHITESPACE
- - - U U - U U - U U- U U - -
>>> print(HexameterScanner().scan(
... "certabant urbem Romam Remoramne vocarent").scansion) # doctest: +NORMALIZE_WHITESPACE
- - - - - - - U U - U U - -
>>> # advanced smoothing is available via keyword flags: dactyl_smoothing
>>> print(HexameterScanner().scan(
... "his verbis: 'o gnata, tibi sunt ante ferendae",
... dactyl_smoothing=True).scansion) # doctest: +NORMALIZE_WHITESPACE
- - - - - U U - - - U U - -
>>> HexameterScanner().scan('Italiam non sponte sequor.')
Verse(original='Italiam non sponte sequor.', scansion='', meter='hexameter', valid=False, syllable_count=9, accented='', scansion_notes=['Incomplete hexameter; not enough syllables.'], syllables = ['I', 'ta', 'li', 'ām', 'nōn', 'spōn', 'te', 'se', 'quor'])
>>> HexameterScanner().scan('Phaselus ille, quem videtis, hospites')
Verse(original='Phaselus ille, quem videtis, hospites', scansion=' - U U - - - U U U - - U ', meter='hexameter', valid=False, syllable_count=12, accented='', scansion_notes=['Inverted amphibrachs corrected.'], syllables = ['Pha', 'se', 'lus', 'īl', 'le', 'quēm', 'vi', 'de', 'tis', 'hōs', 'pi', 'tes'])
"""
verse = Verse(original_line, meter='hexameter')
# replace punctuation with spaces
line = original_line.translate(self.punctuation_substitutions)
if optional_transform:
working_line = self.transform_i_to_j_optional(line)
verse.scansion_notes += [
self.constants.NOTE_MAP["optional i to j"]
]
else:
working_line = self.transform_i_to_j(line) # conservative i to j
working_line = self.elide_all(working_line)
verse.working_line = self.accent_by_position(working_line)
verse.syllables = self.syllabifier.syllabify(verse.working_line)
verse.syllable_count = self.syllabifier.get_syllable_count(
verse.syllables)
if verse.syllable_count < 12:
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["< 12"]]
return verse
stresses = self.flag_dipthongs(verse.syllables)
syllables_wspaces = string_utils.to_syllables_with_trailing_spaces(
verse.working_line, verse.syllables)
offset_map = self.calc_offset(syllables_wspaces)
for idx, syl in enumerate(verse.syllables):
for accented in self.constants.ACCENTED_VOWELS:
if accented in syl:
stresses.append(idx)
# first syllable is always long in hexameter
stresses.append(0)
# second to last syllable is always long
stresses.append(verse.syllable_count - 2)
verse.scansion = self.produce_scansion(stresses, syllables_wspaces,
offset_map)
if len(string_utils.stress_positions(self.constants.STRESSED, verse.scansion)) != \
len(set(stresses)):
verse.valid = False
verse.scansion_notes += [
self.constants.NOTE_MAP["invalid syllables"]
]
return verse
if self.metrical_validator.is_valid_hexameter(verse.scansion):
verse.scansion_notes += [self.constants.NOTE_MAP["positionally"]]
return self.assign_candidate(verse, verse.scansion)
# identify some obvious and probably choices based on number of syllables
if verse.syllable_count == 17: # produce all dactyls
candidate = self.produce_scansion(
self.metrical_validator.hexameter_known_stresses(),
syllables_wspaces, offset_map)
verse.scansion_notes += [self.constants.NOTE_MAP["17"]]
if self.metrical_validator.is_valid_hexameter(candidate):
return self.assign_candidate(verse, candidate)
if verse.syllable_count == 12: # create all spondee hexameter
candidate = self.produce_scansion(list(range(12)),
syllables_wspaces, offset_map)
if self.metrical_validator.is_valid_hexameter(verse.scansion):
verse.scansion_notes += [self.constants.NOTE_MAP["12"]]
return self.assign_candidate(verse, candidate)
if verse.syllable_count == 13: # create spondee hexameter with a dactyl at 5th foot
known_unaccents = [9, 10]
last_syllable_accented = False
for vowel in self.constants.ACCENTED_VOWELS:
if vowel in verse.syllables[12]:
last_syllable_accented = True
if not last_syllable_accented:
known_unaccents.append(12)
if set(known_unaccents) - set(stresses) != len(known_unaccents):
verse.scansion = self.produce_scansion(
[x for x in range(13) if x not in known_unaccents],
syllables_wspaces, offset_map)
verse.scansion_notes += [self.constants.NOTE_MAP["5th dactyl"]]
if self.metrical_validator.is_valid_hexameter(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
if verse.syllable_count > 17:
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["> 17"]]
return verse
smoothed = self.correct_inverted_amphibrachs(verse.scansion)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [self.constants.NOTE_MAP["inverted"]]
verse.scansion = smoothed
stresses += string_utils.differences(verse.scansion, smoothed)
if self.metrical_validator.is_valid_hexameter(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
smoothed = self.correct_first_two_dactyls(verse.scansion)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [self.constants.NOTE_MAP["invalid start"]]
verse.scansion = smoothed
stresses += string_utils.differences(verse.scansion, smoothed)
if self.metrical_validator.is_valid_hexameter(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
smoothed = self.correct_invalid_fifth_foot(verse.scansion)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [self.constants.NOTE_MAP["invalid 5th"]]
verse.scansion = smoothed
stresses += string_utils.differences(verse.scansion, smoothed)
if self.metrical_validator.is_valid_hexameter(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
feet = self.metrical_validator.hexameter_feet(
verse.scansion.replace(" ", ""))
if feet:
# Normal good citizens are unwelcome in the house of hexameter
invalid_feet_in_hexameter = [
self.constants.IAMB, self.constants.TROCHEE
]
current_foot = 0
ending = feet.pop(
) # don't process the ending, a possible trochee, add it back after
scanned_line = ""
for foot in feet:
if foot.replace(" ", "") in invalid_feet_in_hexameter:
scanned_line = self.invalid_foot_to_spondee(
feet, foot, current_foot)
scanned_line = scanned_line + ending
current_foot += 1
smoothed = self.produce_scansion(
stresses + string_utils.stress_positions(
self.constants.STRESSED, scanned_line), syllables_wspaces,
offset_map)
if self.metrical_validator.is_valid_hexameter(smoothed):
verse.scansion_notes += [
self.constants.NOTE_MAP["invalid foot"]
]
return self.assign_candidate(verse, smoothed)
# need to do this again, since the scansion has changed
smoothed = self.correct_inverted_amphibrachs(verse.scansion)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [self.constants.NOTE_MAP["inverted"]]
verse.scansion = smoothed
stresses += string_utils.differences(verse.scansion, smoothed)
if self.metrical_validator.is_valid_hexameter(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
candidates = self.metrical_validator.closest_hexameter_patterns(
verse.scansion)
if candidates is not None:
if len(candidates) == 1 \
and len(verse.scansion.replace(" ", "")) == len(candidates[0]) \
and len(string_utils.differences(verse.scansion, candidates[0])) == 1:
tmp_scansion = self.produce_scansion(
string_utils.differences(verse.scansion, candidates[0]),
syllables_wspaces, offset_map)
if self.metrical_validator.is_valid_hexameter(tmp_scansion):
verse.scansion_notes += [
self.constants.NOTE_MAP["closest match"]
]
return self.assign_candidate(verse, tmp_scansion)
# need to do this again, since the scansion has changed
smoothed = self.correct_inverted_amphibrachs(smoothed)
if self.metrical_validator.is_valid_hexameter(smoothed):
verse.scansion_notes += [self.constants.NOTE_MAP["inverted"]]
return self.assign_candidate(verse, smoothed)
if dactyl_smoothing:
smoothed = self.correct_dactyl_chain(smoothed)
if Levenshtein.levenshtein_distance(verse.scansion, smoothed) > 0:
verse.scansion_notes += [
self.constants.NOTE_MAP["dactyl smoothing"]
]
verse.scansion = smoothed
if self.metrical_validator.is_valid_hexameter(verse.scansion):
return self.assign_candidate(verse, verse.scansion)
# if the line doesn't scan "as is", if may scan if the optional i to j transformations
# are made, so here we set them and try again.
if self.optional_transform and not optional_transform and not verse.valid:
return self.scan(original_line,
optional_transform=True,
dactyl_smoothing=True)
return verse
