from bs4 import BeautifulSoup
import re
s = BeautifulSoup(open('pfl.html',encoding='utf-8',errors='ignore').read(),'lxml')

print("=== STRUCTURE / A11Y SCAN: pfl_claim_process ===")
html=s.find('html')
print("lang attribute:", html.get('lang') if html else 'none')

imgs=s.find_all('img')
noalt=[i for i in imgs if i.get('alt') is None]
emptyalt=[i for i in imgs if i.get('alt')=='']
print(f"images: {len(imgs)} | missing alt attr: {len(noalt)} | decorative alt='': {len(emptyalt)}")

hs=[(int(h.name[1]), h.get_text(' ',strip=True)[:60]) for h in s.find_all(re.compile('^h[1-6]$'))]
print(f"\nheadings: {len(hs)}")
prev=0; skips=[]
for lvl,txt in hs:
    if prev and lvl>prev+1: skips.append((prev,lvl,txt))
    prev=lvl
print("h1 count:", sum(1 for l,_ in hs if l==1))
print("heading-level skips (WCAG 1.3.1):", len(skips))
for a,b,t in skips[:8]: print(f"   h{a} -> h{b} at: {t}")

# accordion triggers
exp=s.find_all(attrs={'aria-expanded':True})
print(f"\nelements with aria-expanded: {len(exp)}")
vals=set(e.get('aria-expanded') for e in exp)
print("  values present:", vals)

# links
links=s.find_all('a')
generic=[a.get_text(' ',strip=True) for a in links if a.get_text(' ',strip=True).lower() in
         ('click here','here','more','read more','learn more','this page','link')]
empty=[a for a in links if not a.get_text(strip=True) and not a.find('img') and not a.get('aria-label')]
print(f"\nlinks: {len(links)} | non-descriptive text: {len(generic)} | empty accessible name: {len(empty)}")

# the mismatched link
for a in links:
    t=a.get_text(' ',strip=True)
    if 'Stop or Continue' in t or 'Continue Your Benefits' in t:
        print(f"\n  LINK TEXT : {t}\n  HREF      : {a.get('href')}")

# tabindex / target=_blank
tb=[a for a in links if a.get('target')=='_blank']
print(f"\nlinks opening new window: {len(tb)}")

# iframes
ifr=s.find_all('iframe')
print(f"iframes: {len(ifr)} | missing title: {sum(1 for i in ifr if not i.get('title'))}")