Which poly teaches the course you want?

import collections, re

POLYS = [('NANYANG POLYTECHNIC', 'NYP'), ('NGEE ANN POLYTECHNIC', 'NP'), ('REPUBLIC POLYTECHNIC', 'RP'),
         ('SINGAPORE POLYTECHNIC', 'SP'), ('TEMASEK POLYTECHNIC', 'TP')]
# a course line: the title, the course code, the score type and the 2026 range ("-" for a new course)
COURSE = re.compile(r'^(?P<left>.*?\S)\s{2,}(?P<code>[A-Z]\d{2,3})\*?\s+(?P<type>ELR2B2-[A-D]|ELB4-[A-D]|ELR1B3-[A-D])\s+(?P<rng>\d+\s+to\s+\d+\*?|-)(?:\s{2,}|$)')
CODE_ONLY = re.compile(r'^\s+(?P<code>[A-Z]\d{2,3})\*?\s+(?P<type>ELR2B2-[A-D]|ELB4-[A-D]|ELR1B3-[A-D])\s+(?P<rng>\d+\s+to\s+\d+\*?|-)(?:\s{2,}|$)')
MARKS = '①-⑳-Ⓐ-ⓩ'          # footnote symbols the PDF draws as characters
HEADINGS = ('COURSES', 'NANYANG POLYTECHNIC', 'NGEE ANN POLYTECHNIC', 'REPUBLIC POLYTECHNIC', 'SINGAPORE POLYTECHNIC', 'TEMASEK POLYTECHNIC')


def parse(text):
    lines = text.split('\n')
    starts = {}
    for i, l in enumerate(lines):                      # each polytechnic section starts under "ENTRY REQUIREMENTS OF COURSES TO"
        if 'ENTRY REQUIREMENTS OF COURSES TO' in l:
            for j in range(i + 1, i + 4):
                for name, ab in POLYS:
                    if lines[j].strip() == name:
                        starts[ab] = j
    assert len(starts) == 5, starts
    ite = [i for i, l in enumerate(lines) if l.strip() == 'INSTITUTE OF TECHNICAL EDUCATION']
    order = sorted(starts.items(), key=lambda kv: kv[1])
    bounds = {}
    for k, (ab, s) in enumerate(order):                # a section ends where the next begins; the last one where the ITE section starts
        after = [i for i in ite if i > s]
        e = order[k + 1][1] if k + 1 < len(order) else (min(after) if after else len(lines))
        bounds[ab] = (s, e)
    rows = []
    for ab, (s, e) in bounds.items():
        field, cur = '', None
        for i in range(s, e):
            l = lines[i]
            if '\f' in l:
                cur = None
            l = l.replace('\f', '')
            st = l.strip()
            mo = CODE_ONLY.match(l)
            if mo and not COURSE.match(l):             # the title sits on the line above the code
                prev = lines[i - 1]
                title = re.split(r'\s{3,}', prev.strip())[0]
                cur = dict(poly=ab, field=field, code=mo.group('code'), title=title, type=mo.group('type'), rng=mo.group('rng'),
                           notes=[], indent=len(prev) - len(prev.lstrip()))
                rows.append(cur)
                continue
            m = COURSE.match(l)
            if m:
                indent = len(l) - len(l.lstrip())
                cur = dict(poly=ab, field=field, code=m.group('code'), title=m.group('left').strip(), type=m.group('type'), rng=m.group('rng'),
                           notes=[], indent=indent)
                rows.append(cur)
                j = i + 1                              # a long title wraps onto the next line
                while j < e and lines[j].strip() and not COURSE.match(lines[j]):
                    part = lines[j][indent:indent + 42].rstrip().strip()
                    if not part or lines[j][:indent].strip():
                        break
                    if re.match(r'^(At the end|\(|•|Specialisations|This is|The first|Course Title|These|Note|Common|\*|Subject|\d)', part) or part.startswith('Notes'):
                        break
                    if len(part) <= 34 and len(lines[j]) - len(lines[j].lstrip()) == indent and not part.endswith(':'):
                        cur['title'] += ' ' + part
                        j += 1
                    else:
                        break
                continue
            if st and re.fullmatch(r"[A-Z&,\-' /\(\)]{6,}", st) and st not in HEADINGS and not st.startswith('ELR') \
                    and 'ENTRY REQUIREMENTS' not in st and 'COURSES TO' not in st and len(l) - len(l.lstrip()) < 12:
                field, cur = st, None                  # a capitalised heading is a field
                continue
            if cur is not None:                        # bullets under a course are the options inside it
                left = l[cur['indent']:cur['indent'] + 42].strip() if len(l) > cur['indent'] else ''
                lind = len(l) - len(l.lstrip())
                if left.startswith('•'):
                    cur['notes'].append(left.lstrip('• ').strip())
                    cur['open'] = True
                elif left and cur.get('open') and cur['notes'] and cur['indent'] < lind < cur['indent'] + 40 \
                        and not left.startswith(('Specialisations', 'Dip', 'The first', 'At the end')):
                    cur['notes'][-1] += ' ' + left
                else:
                    cur['open'] = False
    out = []
    for r in rows:
        low = high = None
        if r['rng'] != '-':
            low, high = [int(x) for x in re.findall(r'\d+', r['rng'])[:2]]
        title = re.sub(r'\s+', ' ', re.sub('[' + MARKS + ']', '', r['title'])).strip()
        out.append(dict(poly=r['poly'], code=r['code'], title=title, field=r['field'].title(), aggregate=r['type'], low=low, high=high,
                        common=title.lower().startswith('common '), new=r['rng'] == '-',
                        options=[re.sub(r'\s+', ' ', re.sub('[' + MARKS + ']', '', n)).strip() for n in r['notes']]))
    return out


def same_title(t):
    """Two courses have the same title when the names match, ignoring capitals and treating "&" as "and"."""
    return re.sub(r'\s+', ' ', t.lower().replace('&', 'and')).strip()


def count_titles(rows):
    polys = collections.defaultdict(set)
    for r in rows:
        polys[same_title(r['title'])].add(r['poly'])
    return len(polys), sorted(collections.Counter(len(v) for v in polys.values()).items())

# usage:  text = open("jae2027.txt", encoding="utf-8").read()   (from: pdftotext -layout jae2027.pdf jae2027.txt)
#         rows = parse(text); print(len(rows), "courses"); print(count_titles(rows))   # 195 courses; (136, [(1, 105), (2, 16), (3, 7), (4, 3), (5, 5)])