He said, "'Quoted' words in a larger quote."
double_quote_sets_re = re.compile(r""""'(?=\w)""") single_quote_sets_re = re.compile(r"""'"(?=\w)""") # Special case for decade abbreviations (the '80s): decade_abbr_re = re.compile(r"""\b'(?=\d{2}s)""") # 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) # Double closing quotes: closing_double_quotes_regex = re.compile(r""" #(%s)? # character that indicates the quote should be closing " (?=\s) """ % (close_class,), re.VERBOSE) closing_double_quotes_regex_2 = re.compile(r""" (%s) # character that indicates the quote should be closing " """ % (close_class,), re.VERBOSE) # 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) closing_single_quotes_regex = re.compile(r""" (%s) ' (?!\s | s\b | \d) """ % (close_class,), re.VERBOSE) closing_single_quotes_regex_2 = re.compile(r""" (%s) ' (\s | s\b) """ % (close_class,), re.VERBOSE) def educateQuotes(str): """ Parameter: String. Returns: The string, with "educated" curly quote HTML entities. Example input: "Isn't this fun?" Example output: “Isn’t this fun?” """ # Special case if the very first character is a quote # followed by punctuation at a non-word-break. Close the quotes by brute force: str = single_quote_start_re.sub("’", str) str = double_quote_start_re.sub("”", str) # Special case for double sets of quotes, e.g.: #He said, "'Quoted' words in a larger quote."
str = double_quote_sets_re.sub("“‘", str) str = single_quote_sets_re.sub("‘“", str) # Special case for decade abbreviations (the '80s): str = decade_abbr_re.sub("’", str) str = opening_single_quotes_regex.sub(r"\1‘", str) str = closing_single_quotes_regex.sub(r"\1’", str) str = closing_single_quotes_regex_2.sub(r"\1’\2", str) # Any remaining single quotes should be opening ones: str = str.replace("'", "‘") str = opening_double_quotes_regex.sub(r"\1“", str) str = closing_double_quotes_regex.sub(r"”", str) str = closing_double_quotes_regex_2.sub(r"\1”", str) # Any remaining quotes should be opening ones. str = str.replace('"', "“") return str def educateBackticks(str): """ Parameter: String. Returns: The string, with ``backticks'' -style double quotes translated into HTML curly quote entities. Example input: ``Isn't this fun?'' Example output: “Isn't this fun?” """ return str.replace("``", "“").replace("''", "”") def educateSingleBackticks(str): """ Parameter: String. Returns: The string, with `backticks' -style single quotes translated into HTML curly quote entities. Example input: `Isn't this fun?' Example output: ‘Isn’t this fun?’ """ return str.replace('`', "‘").replace("'", "’") def educateDashesOldSchool(str): """ Parameter: String. Returns: The string, with each instance of "--" translated to an en-dash HTML entity, and each "---" translated to an em-dash HTML entity. """ return str.replace('---', "—").replace('--', "–") def educateDashesOldSchoolInverted(str): """ Parameter: String. Returns: The string, with each instance of "--" translated to an em-dash HTML entity, and each "---" translated to an en-dash HTML entity. Two reasons why: First, unlike the en- and em-dash syntax supported by EducateDashesOldSchool(), it's compatible with existing entries written before SmartyPants 1.1, back when "--" was only used for em-dashes. Second, em-dashes are more common than en-dashes, and so it sort of makes sense that the shortcut should be shorter to type. (Thanks to Aaron Swartz for the idea.) """ return str.replace('---', "–").replace('--', "—") def educateEllipses(str): """ Parameter: String. Returns: The string, with each instance of "..." translated to an ellipsis HTML entity. Example input: Huh...? Example output: Huh…? """ return str.replace('...', "…").replace('. . .', "…") __author__ = "Chad Miller