|
)', r'┳\1', dc) dc = re.sub(r'([^┳]*)(?=\n\n)', r'\1┳', dc) dc = dc.replace('\n', '').replace('┳', '\n') dc = re.sub(r' {2,}', r' ', dc) with open('ndc.htm', 'w') as fwn: fwn.write(dc) with open('ndc.htm', 'r') as frn: lines = [str(BeautifulSoup(line, 'html5lib').entry) for line in frn if line.strip().startswith('')] hds = set() for line in lines: line = line.replace('adj.', 'adj.') line = line.replace('adjs.', 'adjs.') line = line.replace('—Also ', '
—Also ') line = line.replace('What′-d\'ye-call (-it, -\'em),', 'What′-d\'ye-call-it, What′-d\'ye-call-\'em,') line = line.replace('Sterēo tom′ic', 'Sterēotom′ic') line = line.replace('Czar′evitch, Tsar-', 'Czar′evitch, Tsar′evitch') line = line.replace('Hereunto′ (also -un′-)', 'Hereunto′ (also Hereun′to)') line = line.replace('Domesday-, Doomsday-book', 'Domesday-book, Doomsday-book') line = re.sub(r'(
Hæm([^<> ]+), Hem)-()', r'\1\2\3', line) line = re.sub(r'(
([^<> ]+), )-([^<> ]+)', r'\1\2\3', line) line = re.sub(r'([^<> ]+-)(, [^<> ]+-([^<> ]+))', r'\1\3\2', line) line = re.sub(r'(?<= )([^<>]+)', r'\1', line) line = re.sub(r'([^<>]+)(?= \((?!)|,? (?:or |and )?(?!))', r'\1', line) line = re.sub(r'(?<=; )([^<>]+)(?! \((?!)|,? (?:or |and )?(?!))', r'\1', line) line = re.sub(r'(?<=—)([^<>]+)(?! \((?!)|,? (?:or |and )?(?!))', r'\1', line) retmp0 = re.compile(r'(([^<> ]+)-[^<> ]+, (?:or )?)(-[^<> ]+)') while retmp0.search(line): line = retmp0.sub(r'\1\2\3', line) retmp1 = re.compile(r'(([^<> -]+-?)[^<> -]*, )-([^<> -]+)') while retmp1.search(line): line = retmp1.sub(r'\1\2\3', line) line = line.replace('[', '[').replace(']', ']') soup1 = BeautifulSoup(line, 'html5lib') s1e = soup1.entry for bb in s1e.select("b + b"): bps = bb.find_previous_sibling("b") if not bb.string or not bps.string or bps.string.startswith('-'): continue listtmp = ['al', 'al,', 'al.', 'ed', 'd', 's', 'ic', 'e', 'y', 'er', 'a', 'ous', 'n',] dicttmp = { 'or': ['er'], 'ent': ['ant'], 'ence': ['ance'], 'er': ['or'], 'jingle': ['clink'], 'temporary': ['temporaneous'], 'polit′ical': ['pol′itic'], 'typ′ical': ['typ′al'], 'bowman': ['bower'], 'rhē′al': ['rhœt′ic'], 'rhē′ic': ['rhē′al'], 'rhet′ic': ['rhē′ic'], 'cy': ['ce'], 'cote': ['cot'], 'ous': ['ose'], 'cum': ['con'], 'wards': ['ward'], 'abouts': ['about'], 'maid': ['man'], 'shirt': ['dress'], 'laryn′geal': ['glos′sal'], 'nā′sal': ['laryn′geal'], 'ō′ral': ['nā′sal'], 'eable': ['able'], 'uise': ['ise'], 'tor': ['ter'], 'streaked': ['straked'], 'abouts′': ['about′'], 'te′': ['′'], 'dor': ['der'], 'tos': ['ti'], } for ktmp, vtmps in dicttmp.items(): for vtmp in vtmps: if bb.string == '-' + ktmp and bps.string.endswith(vtmp): bb.string = re.sub(fr'{vtmp}$', fr'{ktmp}', bps.string) for tmp in listtmp: if bb.string == '-' + tmp: bb.string = bps.string + tmp for pbb in s1e.select("entry > p:first-of-type > b:first-child + b"): if len(list(pbb.descendants)) == 1: pbb.string.wrap(soup1.new_tag("span", class_="nw")) for nen in s1e.select(".nw:not(b > .nw),.etym .etym,.etym .nw"): nen.unwrap() for etym in s1e.select(".etym"): if '.' not in etym.get_text(): etym.unwrap() for bn in s1e.select("b > .nw"): bnp = bn.parent if len(bnp.contents) > 1: bn.unwrap() tmp1 = str(s1e) tmp1 = re.sub(r'(?[^<>]+ )?)', r'
—', tmp1) ''' tmp1 = re.sub(r'(?[^<>]+ )?)', r'
—', tmp1) tmp1 = re.sub(r'(?[^<>]+ )?[^<>]+(?: \((?!)|,? (?:or |and )?(?!)))', r'
—', tmp1) ''' retmp2 = re.compile(r'(, (?:[^<> ]+ )?)([A-Z][^<>]+)(?! \((?!)|,? (?:or |and )?(?!))') while retmp2.search(tmp1): tmp1 = retmp2.sub(r'\1\2', tmp1) soup2 = BeautifulSoup(tmp1, 'html5lib') s2e = soup2.entry hws = list() dfl = s2e.select_one('entry > div.figleft:first-child') if dfl: hw = dfl.img['alt'].strip() if hw in string.ascii_uppercase: hws.append(hw) for bfn in s2e.select("entry > p:first-of-type > b:first-child,.nw"): hw = bfn.get_text().strip() if not hw: continue hw = hw.replace("′", "").strip("; ") hw = unidecode(hw, errors="preserve") hw = re.sub(r'\W', r'', hw) if hw and hw not in hws: hws.append(hw) for hw in hws: for bfn2 in reversed(s2e.select("entry > p:first-of-type > b:first-child,.nw")): hw2 = bfn2.get_text().strip() if not hw2: continue hw2 = hw2.replace("′", "").strip("; ") hw2 = unidecode(hw2, errors="preserve") if hw == re.sub(r'\W', r'', hw2): bfn2.wrap(soup2.new_tag("mark")) tmp2 = hw2 dfl2 = s2e.select_one('entry > div.figleft:first-child') if dfl2: hw2 = dfl2.img['alt'].strip() if hw2 in string.ascii_uppercase and hw == hw2: tmp2 = hw2 for mm in s2e.select("mark mark"): mm.unwrap() dc = str(s2e) hd = (tmp2, dc) for mark in s2e("mark"): mark.unwrap() if hd in hds: continue hds.add(hd) CSS = '' with open('dict.txt', 'a') as fad: fad.write(f'{tmp2}\n{CSS}{dc}\n\n') print(tmp2)