def __init__(
self,
constants=ScansionConstants(),
syllabifier=Syllabifier(),
optional_tranform=False,
*args,
**kwargs
):
super().__init__(*args, **kwargs)
self.constants = constants
self.remove_punct_map = string_utils.remove_punctuation_dict()
self.punctuation_substitutions = string_utils.punctuation_for_spaces_dict()
self.metrical_validator = MetricalValidator(constants)
self.formatter = ScansionFormatter(constants)
self.syllabifier = syllabifier
self.inverted_amphibrach_re = re.compile(
r"{}\s*{}\s*{}".format(
self.constants.STRESSED,
self.constants.UNSTRESSED,
self.constants.STRESSED,
)
)
self.syllable_matcher = re.compile(
r"[{}]".format(
self.constants.VOWELS
+ self.constants.ACCENTED_VOWELS
+ self.constants.LIQUIDS
+ self.constants.MUTES
)
)
self.optional_transform = optional_tranform
class PentameterScanner(VerseScanner):
"""The scansion symbols used can be configured by passing a suitable constants class to
the constructor."""
def __init__(self,
constants=ScansionConstants(),
syllabifier=Syllabifier(),
optional_transform: bool = False,
*args,
**kwargs):
super().__init__(*args, **kwargs)
self.constants = constants
self.remove_punct_map = string_utils.remove_punctuation_dict()
self.punctuation_substitutions = string_utils.punctuation_for_spaces_dict(
)
self.metrical_validator = MetricalValidator(constants)
self.formatter = ScansionFormatter(constants)
self.syllabifier = syllabifier
self.optional_transform = optional_transform
self.inverted_amphibrach_re = re.compile(r"{}\s*{}\s*{}".format(
self.constants.STRESSED,
self.constants.UNSTRESSED,
self.constants.STRESSED,
))
self.syllable_matcher = re.compile(
r"[{}]".format(self.constants.VOWELS +
self.constants.ACCENTED_VOWELS +
self.constants.LIQUIDS + self.constants.MUTES))
self.SPONDAIC_PENTAMETER = (self.constants.SPONDEE +
self.constants.SPONDEE +
self.constants.STRESSED +
self.constants.DACTYL +
self.constants.DACTYL +
self.constants.OPTIONAL_ENDING)
self.DACTYLIC_PENTAMETER = (self.constants.DACTYL +
self.constants.DACTYL +
self.constants.STRESSED +
self.constants.DACTYL +
self.constants.DACTYL +
self.constants.OPTIONAL_ENDING)
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)
# conservative i to j
line = self.transform_i_to_j(line)
working_line = self.elide_all(line)
working_line = self.accent_by_position(working_line)
syllables = self.syllabifier.syllabify(working_line)
if optional_transform:
working_line = self.transform_i_to_j_optional(line)
working_line = self.elide_all(working_line)
working_line = self.accent_by_position(working_line)
syllables = self.syllabifier.syllabify(working_line)
verse.scansion_notes += [
self.constants.NOTE_MAP["optional i to j"]
]
verse.working_line = working_line
verse.syllable_count = self.syllabifier.get_syllable_count(syllables)
verse.syllables = syllables
if verse.syllable_count < 12:
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["< 12p"]]
return verse
stresses = self.flag_dipthongs(syllables)
syllables_wspaces = string_utils.to_syllables_with_trailing_spaces(
working_line, syllables)
offset_map = self.calc_offset(syllables_wspaces)
for idx, syl in enumerate(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 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 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 verse.valid:
return self.scan(original_line, optional_transform=True)
verse.accented = self.formatter.merge_line_scansion(
verse.original, verse.scansion)
return verse
def make_spondaic(self, scansion: str) -> str:
"""
If a pentameter line has 12 syllables, then it must start with double spondees.
:param scansion: a string of scansion patterns
:return: a scansion pattern string starting with two spondees
>>> print(PentameterScanner().make_spondaic("U U U U U U U U U U U U"))
- - - - - - U U - U U U
"""
mark_list = string_utils.mark_list(scansion)
vals = list(scansion.replace(" ", ""))
new_vals = self.SPONDAIC_PENTAMETER[:-1] + vals[-1]
corrected = "".join(new_vals)
new_line = list(" " * len(scansion))
for idx, car in enumerate(corrected):
new_line[mark_list[idx]] = car
return "".join(new_line)
def make_dactyls(self, scansion: str) -> str:
"""
If a pentameter line has 14 syllables, it starts and ends with double dactyls.
:param scansion: a string of scansion patterns
:return: a scansion pattern string starting and ending with double dactyls
>>> print(PentameterScanner().make_dactyls("U U U U U U U U U U U U U U"))
- U U - U U - - U U - U U U
"""
mark_list = string_utils.mark_list(scansion)
vals = list(scansion.replace(" ", ""))
new_vals = self.DACTYLIC_PENTAMETER[:-1] + vals[-1]
corrected = "".join(new_vals)
new_line = list(" " * len(scansion))
for idx, car in enumerate(corrected):
new_line[mark_list[idx]] = car
return "".join(new_line)
def correct_penultimate_dactyl_chain(self, scansion: str) -> str:
"""
For pentameter the last two feet of the verse are predictable dactyls,
and do not regularly allow substitutions.
:param scansion: scansion line thus far
:return: corrected line of scansion
>>> print(PentameterScanner().correct_penultimate_dactyl_chain(
... "U U U U U U U U U U U U U U"))
U U U U U U U - U U - U U U
"""
mark_list = string_utils.mark_list(scansion)
vals = list(scansion.replace(" ", ""))
n_vals = (vals[:-7] + [self.constants.DACTYL + self.constants.DACTYL] +
[vals[-1]])
corrected = "".join(n_vals)
new_line = list(" " * len(scansion))
for idx, car in enumerate(corrected):
new_line[mark_list[idx]] = car
return "".join(new_line)
class HendecasyllableScanner(VerseScanner):
"""The scansion symbols used can be configured by passing a suitable constants class to
the constructor."""
def __init__(
self,
constants=ScansionConstants(),
syllabifier=Syllabifier(),
optional_tranform=False,
*args,
**kwargs
):
super().__init__(*args, **kwargs)
self.constants = constants
self.remove_punct_map = string_utils.remove_punctuation_dict()
self.punctuation_substitutions = string_utils.punctuation_for_spaces_dict()
self.metrical_validator = MetricalValidator(constants)
self.formatter = ScansionFormatter(constants)
self.syllabifier = syllabifier
self.inverted_amphibrach_re = re.compile(
r"{}\s*{}\s*{}".format(
self.constants.STRESSED,
self.constants.UNSTRESSED,
self.constants.STRESSED,
)
)
self.syllable_matcher = re.compile(
r"[{}]".format(
self.constants.VOWELS
+ self.constants.ACCENTED_VOWELS
+ self.constants.LIQUIDS
+ self.constants.MUTES
)
)
self.optional_transform = optional_tranform
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)
# conservative i to j
line = self.transform_i_to_j(line)
working_line = self.elide_all(line)
working_line = self.accent_by_position(working_line)
syllables = self.syllabifier.syllabify(working_line)
if optional_transform:
working_line = self.transform_i_to_j_optional(line)
working_line = self.elide_all(working_line)
working_line = self.accent_by_position(working_line)
syllables = self.syllabifier.syllabify(working_line)
verse.scansion_notes += [self.constants.NOTE_MAP["optional i to j"]]
verse.working_line = working_line
verse.syllable_count = self.syllabifier.get_syllable_count(syllables)
verse.syllables = 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(syllables)
syllables_wspaces = string_utils.to_syllables_with_trailing_spaces(
working_line, syllables
)
offset_map = self.calc_offset(syllables_wspaces)
for idx, syl in enumerate(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 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 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 verse.valid:
return self.scan(original_line, optional_transform=True)
verse.accented = self.formatter.merge_line_scansion(
verse.original, verse.scansion
)
return verse
def correct_invalid_start(self, scansion: str) -> str:
"""
The third syllable of a hendecasyllabic line is long, so we will convert it.
:param scansion: scansion string
:return: scansion string with corrected start
>>> print(HendecasyllableScanner().correct_invalid_start(
... "- U U U U - U - U - U").strip())
- U - U U - U - U - U
"""
mark_list = string_utils.mark_list(scansion)
vals = list(scansion.replace(" ", ""))
corrected = vals[:2] + [self.constants.STRESSED] + vals[3:]
new_line = list(" " * len(scansion))
for idx, car in enumerate(corrected):
new_line[mark_list[idx]] = car
return "".join(new_line)
def correct_antepenult_chain(self, scansion: str) -> str:
"""
For hendecasyllables the last three feet of the verse are predictable
and do not regularly allow substitutions.
:param scansion: scansion line thus far
:return: corrected line of scansion
>>> print(HendecasyllableScanner().correct_antepenult_chain(
... "-U -UU UU UU UX").strip())
-U -UU -U -U -X
"""
mark_list = string_utils.mark_list(scansion)
vals = list(scansion.replace(" ", ""))
new_vals = (
vals[: len(vals) - 6]
+ [
self.constants.TROCHEE
+ self.constants.TROCHEE
+ self.constants.STRESSED
]
+ vals[-1:]
)
corrected = "".join(new_vals)
new_line = list(" " * len(scansion))
for idx, car in enumerate(corrected):
new_line[mark_list[idx]] = car
return "".join(new_line)
class HexameterScanner(VerseScanner):
"""The scansion symbols used can be configured by passing a suitable constants class to
the constructor."""
def __init__(self,
constants=ScansionConstants(),
syllabifier=Syllabifier(),
optional_transform=False,
*args,
**kwargs):
super().__init__(*args, **kwargs)
self.constants = constants
self.remove_punct_map = string_utils.remove_punctuation_dict()
self.punctuation_substitutions = string_utils.punctuation_for_spaces_dict(
)
self.metrical_validator = MetricalValidator(constants)
self.formatter = ScansionFormatter(constants)
self.syllabifier = syllabifier
self.inverted_amphibrach_re = re.compile(r"{}\s*{}\s*{}".format(
self.constants.STRESSED,
self.constants.UNSTRESSED,
self.constants.STRESSED,
))
self.syllable_matcher = re.compile(
r"[{}]".format(self.constants.VOWELS +
self.constants.ACCENTED_VOWELS +
self.constants.LIQUIDS + self.constants.MUTES))
self.optional_transform = optional_transform
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 - -
"""
verse = Verse(original_line, meter="hexameter")
# replace punctuation with spaces
line = original_line.translate(self.punctuation_substitutions)
# conservative i to j
line = self.transform_i_to_j(line)
working_line = self.elide_all(line)
working_line = self.accent_by_position(working_line)
syllables = self.syllabifier.syllabify(working_line)
if optional_transform:
working_line = self.transform_i_to_j_optional(line)
working_line = self.elide_all(working_line)
working_line = self.accent_by_position(working_line)
syllables = self.syllabifier.syllabify(working_line)
verse.scansion_notes += [
self.constants.NOTE_MAP["optional i to j"]
]
verse.working_line = working_line
verse.syllable_count = self.syllabifier.get_syllable_count(syllables)
verse.syllables = syllables
if verse.syllable_count < 12:
verse.valid = False
verse.scansion_notes += [self.constants.NOTE_MAP["< 12"]]
return verse
stresses = self.flag_dipthongs(syllables)
syllables_wspaces = string_utils.to_syllables_with_trailing_spaces(
working_line, syllables)
offset_map = self.calc_offset(syllables_wspaces)
for idx, syl in enumerate(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 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 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 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 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 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 verse.valid:
return self.scan(original_line,
optional_transform=True,
dactyl_smoothing=True)
return verse
def correct_invalid_fifth_foot(self, scansion: str) -> str:
"""
The 'inverted amphibrach': stressed_unstressed_stressed syllable pattern is invalid
in hexameters, so here we coerce it to stressed when it occurs at the end of a line
:param scansion: the scansion pattern
:return corrected scansion: the corrected scansion pattern
>>> print(HexameterScanner().correct_invalid_fifth_foot(
... " - - - U U - U U U - - U U U - x")) # doctest: +NORMALIZE_WHITESPACE
- - - U U - U U U - - - U U - x
"""
scansion_wo_spaces = (scansion.replace(" ", "")[:-1] +
self.constants.OPTIONAL_ENDING)
if scansion_wo_spaces.endswith(self.constants.DACTYL +
self.constants.IAMB +
self.constants.OPTIONAL_ENDING):
matches = list(
re.compile(r"{}\s*{}\s*{}\s*{}\s*{}".format(
self.constants.STRESSED,
self.constants.UNSTRESSED,
self.constants.UNSTRESSED,
self.constants.UNSTRESSED,
self.constants.STRESSED,
)).finditer(scansion))
(start, end) = matches[len(matches) - 1].span()
unstressed_idx = scansion.index(self.constants.UNSTRESSED, start)
new_line = (scansion[:unstressed_idx] + self.constants.STRESSED +
scansion[unstressed_idx + 1:])
return new_line
return scansion
def invalid_foot_to_spondee(self, feet: list, foot: str, idx: int) -> str:
"""
In hexameters, a single foot that is a unstressed_stressed syllable pattern is often
just a double spondee, so here we coerce it to stressed.
:param feet: list of string representations of meterical feet
:param foot: the bad foot to correct
:param idx: the index of the foot to correct
:return: corrected scansion
>>> print(HexameterScanner().invalid_foot_to_spondee(
... ['-UU', '--', '-U', 'U-', '--', '-UU'],'-U', 2)) # doctest: +NORMALIZE_WHITESPACE
-UU----U----UU
"""
new_foot = foot.replace(self.constants.UNSTRESSED,
self.constants.STRESSED)
feet[idx] = new_foot
return "".join(feet)
def correct_dactyl_chain(self, scansion: str) -> str:
"""
Three or more unstressed accents in a row is a broken dactyl chain, best detected and
processed backwards.
Since this method takes a Procrustean approach to modifying the scansion pattern,
it is not used by default in the scan method; however, it is available as an optional
keyword parameter, and users looking to further automate the generation of scansion
candidates should consider using this as a fall back.
:param scansion: scansion with broken dactyl chain; inverted amphibrachs not allowed
:return: corrected line of scansion
>>> print(HexameterScanner().correct_dactyl_chain(
... "- U U - - U U - - - U U - x"))
- - - - - U U - - - U U - x
>>> print(HexameterScanner().correct_dactyl_chain(
... "- U U U U - - - - - U U - U")) # doctest: +NORMALIZE_WHITESPACE
- - - U U - - - - - U U - U
"""
mark_list = string_utils.mark_list(scansion)
vals = list(scansion.replace(" ", ""))
# ignore last two positions, save them
feet = [vals.pop(), vals.pop()]
length = len(vals)
idx = length - 1
while idx > 0:
one = vals[idx]
two = vals[idx - 1]
if idx > 1:
three = vals[idx - 2]
else:
three = ""
# Dactyl foot is okay, no corrections
if (one == self.constants.UNSTRESSED
and two == self.constants.UNSTRESSED
and three == self.constants.STRESSED):
feet += [one]
feet += [two]
feet += [three]
idx -= 3
continue
# Spondee foot is okay, no corrections
if one == self.constants.STRESSED and two == self.constants.STRESSED:
feet += [one]
feet += [two]
idx -= 2
continue
# handle "U U U" foot as "- U U"
if (one == self.constants.UNSTRESSED
and two == self.constants.UNSTRESSED
and three == self.constants.UNSTRESSED):
feet += [one]
feet += [two]
feet += [self.constants.STRESSED]
idx -= 3
continue
# handle "U U -" foot as "- -"
if (one == self.constants.STRESSED
and two == self.constants.UNSTRESSED
and three == self.constants.UNSTRESSED):
feet += [self.constants.STRESSED]
feet += [self.constants.STRESSED]
idx -= 2
continue
# handle "- U" foot as "- -"
if one == self.constants.UNSTRESSED and two == self.constants.STRESSED:
feet += [self.constants.STRESSED]
feet += [two]
idx -= 2
continue
corrected = "".join(feet[::-1])
new_line = list(" " * len(scansion))
for idx, car in enumerate(corrected):
new_line[mark_list[idx]] = car
return "".join(new_line)
def correct_inverted_amphibrachs(self, scansion: str) -> str:
"""
The 'inverted amphibrach': stressed_unstressed_stressed syllable pattern is invalid
in hexameters, so here we coerce it to stressed: - U - -> - - -
:param scansion: the scansion stress pattern
:return: a string with the corrected scansion pattern
>>> print(HexameterScanner().correct_inverted_amphibrachs(
... " - U - - U - U U U U - U - x")) # doctest: +NORMALIZE_WHITESPACE
- - - - - - U U U U - - - x
>>> print(HexameterScanner().correct_inverted_amphibrachs(
... " - - - U - - U U U U U- - U - x")) # doctest: +NORMALIZE_WHITESPACE
- - - - - - U U U U U- - - - x
>>> print(HexameterScanner().correct_inverted_amphibrachs(
... "- - - - - U - U U - U U - -")) # doctest: +NORMALIZE_WHITESPACE
- - - - - - - U U - U U - -
>>> print(HexameterScanner().correct_inverted_amphibrachs(
... "- UU- U - U - - U U U U- U")) # doctest: +NORMALIZE_WHITESPACE
- UU- - - - - - U U U U- U
"""
new_line = scansion
while list(self.inverted_amphibrach_re.finditer(new_line)):
matches = list(self.inverted_amphibrach_re.finditer(new_line))
for match in matches:
(start, end) = match.span() # pylint: disable=unused-variable
unstressed_idx = new_line.index(self.constants.UNSTRESSED,
start)
new_line = (new_line[:unstressed_idx] +
self.constants.STRESSED +
new_line[unstressed_idx + 1:])
return new_line