'' import re,unicodedata ORD=r'[0-9]+(?:,[0-9]+)*(?i:st|nd|rd|th)' LETTER=r'[^\W\d_]' class DomainViolation(ValueError): def __init__(self,reason,start,end):super().__init__(reason);self.reason=reason;self.span=[start,end] def scalar(atom): o=re.fullmatch(r'([0-9]+(?:,[0-9]+)*)(st|nd|rd|th)',atom,re.I) if o: digits,suffix=o.groups() if ','in digits and not re.fullmatch(r'[0-9]{1,3}(?:,[0-9]{3})+',digits):raise ValueError('ordinal grouping') n=int(digits.replace(',',''));expected='th'if n%100 in(11,12,13)else{1:'st',2:'nd',3:'rd'}.get(n%10,'th') if n>9999 or suffix.lower()!=expected:raise ValueError('ordinal bound/suffix') return if ','in atom and not re.fullmatch(r'[0-9]{1,3}(?:,[0-9]{3})+(?:\.[0-9]+)?',atom):raise ValueError('cardinal grouping') atom=atom.replace(',','') if not re.fullmatch(r'[0-9]+(?:\.[0-9]+)?',atom):raise ValueError('scalar grammar') if int(atom.partition('.')[0])>=10**15:raise ValueError('cardinal ceiling') def validate_domain(text): guard=list(text) for m in re.finditer(r'&[cC]\.(?!\w)',text):guard[m.start()]=' ' for m in re.finditer(r'\byear[ \t]+(?P—)[ \t]*[0-9]+(?![A-Za-z0-9_]|[.,][0-9])',text,re.I):guard[m.start('dash')]=' ' guard=''.join(guard) for i,c in enumerate(guard): if ((c.isnumeric()and c not in'0123456789')or(unicodedata.category(c).startswith('S')and c not in'$£€')or c in'%‰‱&@#'):raise DomainViolation('unsupported numeric/semantic symbol',i,i+1) adjacent=list(guard) for m in re.finditer(r'(?