def educateQuotes(text, language='en'):
# type: (str, str) -> str
"""
Parameter: - text string (unicode or bytes).
- language (`BCP 47` language tag.)
Returns: The `text`, with "educated" curly quote characters.
Example input: "Isn't this fun?"
Example output: “Isn’t this fun?“;
"""
smart = smartquotes.smartchars(language)
try:
apostrophe = smart.apostrophe
except Exception:
apostrophe = u'’'
# oldtext = text
punct_class = r"""[!"#\$\%'()*+,-.\/:;<=>?\@\[\\\]\^_`{|}~]"""
# Special case if the very first character is a quote
# followed by punctuation at a non-word-break.
# Close the quotes by brute force:
text = re.sub(r"""^'(?=%s\\B)""" % (punct_class,), smart.csquote, text)
text = re.sub(r"""^"(?=%s\\B)""" % (punct_class,), smart.cpquote, text)
# Special case for double sets of quotes, e.g.:
# <p>He said, "'Quoted' words in a larger quote."</p>
text = re.sub(r""""'(?=\w)""", smart.opquote + smart.osquote, text)
text = re.sub(r"""'"(?=\w)""", smart.osquote + smart.opquote, text)
# Special case for decade abbreviations (the '80s):
if language.startswith('en'): # TODO similar cases in other languages?
text = re.sub(r"""'(?=\d{2}s)""", apostrophe, text, re.UNICODE)
close_class = r"""[^\ \t\r\n\[\{\(\-]"""
dec_dashes = r"""–|—"""
# Get most opening single quotes:
opening_single_quotes_regex = re.compile(r"""
(
\s | # a whitespace char, or
| # a non-breaking space entity, or
-- | # dashes, or
&[mn]dash; | # named dash entities
%s | # or decimal entities
&\#x201[34]; # or hex
)
' # the quote
(?=\w) # followed by a word character
""" % (dec_dashes,), re.VERBOSE | re.UNICODE)
text = opening_single_quotes_regex.sub(r'\1' + smart.osquote, text)
# In many locales, single closing quotes are different from apostrophe:
if smart.csquote != apostrophe:
apostrophe_regex = re.compile(r"(?<=(\w|\d))'(?=\w)", re.UNICODE)
text = apostrophe_regex.sub(apostrophe, text)
# TODO: keep track of quoting level to recognize apostrophe in, e.g.,
# "Ich fass' es nicht."
closing_single_quotes_regex = re.compile(r"""
(%s)
'
(?!\s | # whitespace
s\b |
\d # digits ('80s)
)
""" % (close_class,), re.VERBOSE | re.UNICODE)
text = closing_single_quotes_regex.sub(r'\1' + smart.csquote, text)
closing_single_quotes_regex = re.compile(r"""
(%s)
'
(\s | s\b)
""" % (close_class,), re.VERBOSE | re.UNICODE)
text = closing_single_quotes_regex.sub(r'\1%s\2' % smart.csquote, text)
# Any remaining single quotes should be opening ones:
text = re.sub(r"""'""", smart.osquote, text)
# Get most opening double quotes:
opening_double_quotes_regex = re.compile(r"""
(
\s | # a whitespace char, or
| # a non-breaking space entity, or
-- | # dashes, or
&[mn]dash; | # named dash entities
%s | # or decimal entities
&\#x201[34]; # or hex
)
" # the quote
(?=\w) # followed by a word character
""" % (dec_dashes,), re.VERBOSE)
text = opening_double_quotes_regex.sub(r'\1' + smart.opquote, text)
# Double closing quotes:
closing_double_quotes_regex = re.compile(r"""
#(%s)? # character that indicates the quote should be closing
"
(?=\s)
""" % (close_class,), re.VERBOSE)
text = closing_double_quotes_regex.sub(smart.cpquote, text)
closing_double_quotes_regex = re.compile(r"""
(%s) # character that indicates the quote should be closing
"
""" % (close_class,), re.VERBOSE)
text = closing_double_quotes_regex.sub(r'\1' + smart.cpquote, text)
# Any remaining quotes should be opening ones.
text = re.sub(r'"', smart.opquote, text)
return text
def educateQuotes(text: str, language: str = 'en') -> str:
"""
Parameter: - text string (unicode or bytes).
- language (`BCP 47` language tag.)
Returns: The `text`, with "educated" curly quote characters.
Example input: "Isn't this fun?"
Example output: “Isn’t this fun?“;
"""
smart = smartquotes.smartchars(language)
try:
apostrophe = smart.apostrophe
except Exception:
apostrophe = '’'
# oldtext = text
punct_class = r"""[!"#\$\%'()*+,-.\/:;<=>?\@\[\\\]\^_`{|}~]"""
# Special case if the very first character is a quote
# followed by punctuation at a non-word-break.
# Close the quotes by brute force:
text = re.sub(r"""^'(?=%s\\B)""" % (punct_class, ), smart.csquote, text)
text = re.sub(r"""^"(?=%s\\B)""" % (punct_class, ), smart.cpquote, text)
# Special case for double sets of quotes, e.g.:
# <p>He said, "'Quoted' words in a larger quote."</p>
text = re.sub(r""""'(?=\w)""", smart.opquote + smart.osquote, text)
text = re.sub(r"""'"(?=\w)""", smart.osquote + smart.opquote, text)
# Special case for decade abbreviations (the '80s):
if language.startswith('en'): # TODO similar cases in other languages?
text = re.sub(r"""'(?=\d{2}s)""", apostrophe, text, flags=re.UNICODE)
close_class = r"""[^\ \t\r\n\[\{\(\-]"""
dec_dashes = r"""–|—"""
# Get most opening single quotes:
opening_single_quotes_regex = re.compile(
r"""
(
\s | # a whitespace char, or
| # a non-breaking space entity, or
-- | # dashes, or
&[mn]dash; | # named dash entities
%s | # or decimal entities
&\#x201[34]; # or hex
)
' # the quote
(?=\w) # followed by a word character
""" % (dec_dashes, ), re.VERBOSE | re.UNICODE)
text = opening_single_quotes_regex.sub(r'\1' + smart.osquote, text)
# In many locales, single closing quotes are different from apostrophe:
if smart.csquote != apostrophe:
apostrophe_regex = re.compile(r"(?<=(\w|\d))'(?=\w)", re.UNICODE)
text = apostrophe_regex.sub(apostrophe, text)
# TODO: keep track of quoting level to recognize apostrophe in, e.g.,
# "Ich fass' es nicht."
closing_single_quotes_regex = re.compile(
r"""
(%s)
'
(?!\s | # whitespace
s\b |
\d # digits ('80s)
)
""" % (close_class, ), re.VERBOSE | re.UNICODE)
text = closing_single_quotes_regex.sub(r'\1' + smart.csquote, text)
closing_single_quotes_regex = re.compile(
r"""
(%s)
'
(\s | s\b)
""" % (close_class, ), re.VERBOSE | re.UNICODE)
text = closing_single_quotes_regex.sub(r'\1%s\2' % smart.csquote, text)
# Any remaining single quotes should be opening ones:
text = re.sub(r"""'""", smart.osquote, text)
# Get most opening double quotes:
opening_double_quotes_regex = re.compile(
r"""
(
\s | # a whitespace char, or
| # a non-breaking space entity, or
-- | # dashes, or
&[mn]dash; | # named dash entities
%s | # or decimal entities
&\#x201[34]; # or hex
)
" # the quote
(?=\w) # followed by a word character
""" % (dec_dashes, ), re.VERBOSE)
text = opening_double_quotes_regex.sub(r'\1' + smart.opquote, text)
# Double closing quotes:
closing_double_quotes_regex = re.compile(
r"""
#(%s)? # character that indicates the quote should be closing
"
(?=\s)
""" % (close_class, ), re.VERBOSE)
text = closing_double_quotes_regex.sub(smart.cpquote, text)
closing_double_quotes_regex = re.compile(
r"""
(%s) # character that indicates the quote should be closing
"
""" % (close_class, ), re.VERBOSE)
text = closing_double_quotes_regex.sub(r'\1' + smart.cpquote, text)
# Any remaining quotes should be opening ones.
text = re.sub(r'"', smart.opquote, text)
return text