def transform_i_to_j_optional(self, line: str) -> str:
"""
Sometimes for the demands of meter a more permissive i to j transformation is warranted.
:param line:
:return:
>>> print(VerseScanner().transform_i_to_j_optional("Italiam"))
Italjam
>>> print(VerseScanner().transform_i_to_j_optional("Lāvīniaque"))
Lāvīnjaque
>>> print(VerseScanner().transform_i_to_j_optional("omnium"))
omnjum
"""
words = line.split(" ")
space_list = string_utils.space_list(line)
corrected_words = []
for word in words:
found = False
for prefix in self.constants.PREFIXES:
if word.startswith(prefix) and word != prefix:
corrected_words.append(
self.syllabifier.convert_consonantal_i(prefix))
corrected_words.append(
self.syllabifier.convert_consonantal_i(
word[len(prefix):]))
found = True
break
if not found:
corrected_words.append(
self.syllabifier.convert_consonantal_i(word))
new_line = string_utils.join_syllables_spaces(corrected_words,
space_list)
# the following two may be tunable and subject to improvement
char_list = string_utils.overwrite(
list(new_line),
"[bcdfgjkmpqrstvwxzBCDFGHJKMPQRSTVWXZ][i][{}]".format(
self.constants.VOWELS_WO_I), "j", 1)
char_list = string_utils.overwrite(
char_list, "[{}][iI][{}]".format(self.constants.LIQUIDS,
self.constants.VOWELS_WO_I), "j",
1)
return "".join(char_list)
def transform_i_to_j(self, line: str) -> str:
"""
Transform instances of consonantal i to j
:param line:
:return:
>>> print(VerseScanner().transform_i_to_j("iactātus"))
jactātus
>>> print(VerseScanner().transform_i_to_j("bracchia"))
bracchia
"""
words = line.split(" ")
space_list = string_utils.space_list(line)
corrected_words = []
for word in words:
found = False
for prefix in self.constants.PREFIXES:
if word.startswith(prefix) and word != prefix:
corrected_words.append(
self.syllabifier.convert_consonantal_i(prefix))
corrected_words.append(
self.syllabifier.convert_consonantal_i(
word[len(prefix):]))
found = True
break
if not found:
corrected_words.append(
self.syllabifier.convert_consonantal_i(word))
new_line = string_utils.join_syllables_spaces(corrected_words,
space_list)
char_list = string_utils.overwrite(
list(new_line),
r"\b[iī][{}]".format(self.constants.VOWELS +
self.constants.ACCENTED_VOWELS), "j")
char_list = string_utils.overwrite(
char_list, r"\b[I][{}]".format(self.constants.VOWELS_WO_I), "J")
char_list = string_utils.overwrite(
char_list, r"[{}][i][{}]".format(self.constants.VOWELS_WO_I,
self.constants.VOWELS), "j", 1)
return "".join(char_list)
def transform_i_to_j_optional(self, line: str) -> str:
"""
Sometimes for the demands of meter a more permissive i to j transformation is warranted.
:param line:
:return:
>>> print(VerseScanner().transform_i_to_j_optional("Italiam"))
Italjam
>>> print(VerseScanner().transform_i_to_j_optional("Lāvīniaque"))
Lāvīnjaque
>>> print(VerseScanner().transform_i_to_j_optional("omnium"))
omnjum
"""
words = line.split(" ")
space_list = string_utils.space_list(line)
corrected_words = []
for word in words:
found = False
for prefix in self.constants.PREFIXES:
if word.startswith(prefix) and word != prefix:
corrected_words.append(self.syllabifier.convert_consonantal_i(prefix))
corrected_words.append(
self.syllabifier.convert_consonantal_i(word[len(prefix):]))
found = True
break
if not found:
corrected_words.append(self.syllabifier.convert_consonantal_i(word))
new_line = string_utils.join_syllables_spaces(corrected_words, space_list)
# the following two may be tunable and subject to improvement
char_list = string_utils.overwrite(list(new_line),
"[bcdfgjkmpqrstvwxzBCDFGHJKMPQRSTVWXZ][i][{}]".format(
self.constants.VOWELS_WO_I),
"j", 1)
char_list = string_utils.overwrite(char_list,
"[{}][iI][{}]".format(self.constants.LIQUIDS,
self.constants.VOWELS_WO_I),
"j", 1)
return "".join(char_list)
def transform_i_to_j(self, line: str) -> str:
"""
Transform instances of consonantal i to j
:param line:
:return:
>>> print(VerseScanner().transform_i_to_j("iactātus"))
jactātus
>>> print(VerseScanner().transform_i_to_j("bracchia"))
bracchia
"""
words = line.split(" ")
space_list = string_utils.space_list(line)
corrected_words = []
for word in words:
found = False
for prefix in self.constants.PREFIXES:
if word.startswith(prefix) and word != prefix:
corrected_words.append(self.syllabifier.convert_consonantal_i(prefix))
corrected_words.append(
self.syllabifier.convert_consonantal_i(word[len(prefix):]))
found = True
break
if not found:
corrected_words.append(self.syllabifier.convert_consonantal_i(word))
new_line = string_utils.join_syllables_spaces(corrected_words, space_list)
char_list = string_utils.overwrite(list(new_line),
r"\b[iī][{}]".format(
self.constants.VOWELS + self.constants.ACCENTED_VOWELS),
"j")
char_list = string_utils.overwrite(char_list,
r"\b[I][{}]".format(self.constants.VOWELS_WO_I),
"J")
char_list = string_utils.overwrite(char_list, r"[{}][i][{}]".format(
self.constants.VOWELS_WO_I, self.constants.VOWELS),
"j", 1)
return "".join(char_list)
def accent_by_position(self, verse_line: str) -> str:
"""
Accent vowels according to the rules of scansion.
:param verse_line: a line of unaccented verse
:return: the same line with vowels accented by position
>>> print(VerseScanner().accent_by_position(
... "Arma virumque cano, Troiae qui primus ab oris").lstrip())
Ārma virūmque canō Trojae qui primus ab oris
"""
line = verse_line.translate(self.punctuation_substitutions)
line = self.transform_i_to_j(line)
marks = list(line)
# locate and save dipthong positions since we don't want them being accented
dipthong_positions = []
for dipth in self.constants.DIPTHONGS:
if dipth in line:
dipthong_positions.append(line.find(dipth))
# Vowels followed by 2 consonants
# The digraphs ch, ph, th, qu and sometimes gu and su count as single consonants.
# see http://people.virginia.edu/~jdk3t/epicintrog/scansion.htm
marks = string_utils.overwrite(
marks, "[{}][{}][{}]".format(self.constants.VOWELS,
self.constants.CONSONANTS,
self.constants.CONSONANTS_WO_H),
self.constants.STRESSED)
# one space (or more for 'dropped' punctuation may intervene)
marks = string_utils.overwrite(
marks, r"[{}][{}]\s*[{}]".format(self.constants.VOWELS,
self.constants.CONSONANTS,
self.constants.CONSONANTS_WO_H),
self.constants.STRESSED)
# ... if both consonants are in the next word, the vowel may be long
# .... but it could be short if the vowel is not on the thesis/emphatic part of the foot
# ... see Gildersleeve and Lodge p.446
marks = string_utils.overwrite(
marks, r"[{}]\s*[{}][{}]".format(self.constants.VOWELS,
self.constants.CONSONANTS,
self.constants.CONSONANTS_WO_H),
self.constants.STRESSED)
# x is considered as two letters
marks = string_utils.overwrite(
marks, "[{}][xX]".format(self.constants.VOWELS),
self.constants.STRESSED)
# z is considered as two letters
marks = string_utils.overwrite(
marks, r"[{}][zZ]".format(self.constants.VOWELS),
self.constants.STRESSED)
original_verse = list(line)
for idx, word in enumerate(original_verse):
if marks[idx] == self.constants.STRESSED:
original_verse[idx] = self.constants.VOWELS_TO_ACCENTS[
original_verse[idx]]
# make sure dipthongs aren't accented
for idx in dipthong_positions:
if original_verse[idx + 1] in self.constants.ACCENTS_TO_VOWELS:
original_verse[idx + 1] = self.constants.ACCENTS_TO_VOWELS[
original_verse[idx + 1]]
return "".join(original_verse)
def accent_by_position(self, verse_line: str) -> str:
"""
Accent vowels according to the rules of scansion.
:param verse_line: a line of unaccented verse
:return: the same line with vowels accented by position
>>> print(VerseScanner().accent_by_position(
... "Arma virumque cano, Troiae qui primus ab oris").lstrip())
Ārma virūmque canō Trojae qui primus ab oris
"""
line = verse_line.translate(self.punctuation_substitutions)
line = self.transform_i_to_j(line)
marks = list(line)
# locate and save dipthong positions since we don't want them being accented
dipthong_positions = []
for dipth in self.constants.DIPTHONGS:
if dipth in line:
dipthong_positions.append(line.find(dipth))
# Vowels followed by 2 consonants
# The digraphs ch, ph, th, qu and sometimes gu and su count as single consonants.
# see http://people.virginia.edu/~jdk3t/epicintrog/scansion.htm
marks = string_utils.overwrite(marks, "[{}][{}][{}]".format(
self.constants.VOWELS,
self.constants.CONSONANTS,
self.constants.CONSONANTS_WO_H),
self.constants.STRESSED)
# one space (or more for 'dropped' punctuation may intervene)
marks = string_utils.overwrite(marks,
r"[{}][{}]\s*[{}]".format(
self.constants.VOWELS,
self.constants.CONSONANTS,
self.constants.CONSONANTS_WO_H),
self.constants.STRESSED)
# ... if both consonants are in the next word, the vowel may be long
# .... but it could be short if the vowel is not on the thesis/emphatic part of the foot
# ... see Gildersleeve and Lodge p.446
marks = string_utils.overwrite(marks,
r"[{}]\s*[{}][{}]".format(
self.constants.VOWELS,
self.constants.CONSONANTS,
self.constants.CONSONANTS_WO_H),
self.constants.STRESSED)
# x is considered as two letters
marks = string_utils.overwrite(marks,
"[{}][xX]".format(self.constants.VOWELS),
self.constants.STRESSED)
# z is considered as two letters
marks = string_utils.overwrite(marks,
r"[{}][zZ]".format(self.constants.VOWELS),
self.constants.STRESSED)
original_verse = list(line)
for idx, word in enumerate(original_verse):
if marks[idx] == self.constants.STRESSED:
original_verse[idx] = self.constants.VOWELS_TO_ACCENTS[original_verse[idx]]
# make sure dipthongs aren't accented
for idx in dipthong_positions:
if original_verse[idx + 1] in self.constants.ACCENTS_TO_VOWELS:
original_verse[idx + 1] = self.constants.ACCENTS_TO_VOWELS[original_verse[idx + 1]]
return "".join(original_verse)