import os, re, json
from playwright.sync_api import sync_playwright

TYPO = "\u2019"
ASCII_APOS_PATTERNS = [r"\bl'(?=\w)", r"\bL'(?=\w)", r"\bd'(?=\w)",
                       r"\bD'(?=\w)", r"\bqu'(?=\w)", r"\bQu'(?=\w)",
                       r"\bn'(?=\w)", r"\bjusqu'(?=\w)"]

def count_ascii(content):
    total = 0
    for p in ASCII_APOS_PATTERNS:
        total += len(re.findall(p, content))
    return total

with sync_playwright() as p:
    browser = p.chromium.launch(headless=True, args=["--no-sandbox"])
    context = browser.new_context(viewport={"width": 1440, "height": 900})
    page = context.new_page()

    stages = [
        ("Homepage", "https://weval-consulting.com/"),
        ("Marketplace", "https://weval-consulting.com/marketplace"),
        ("Products catalogue", "https://weval-consulting.com/products/"),
        ("WEVIA EM", "https://weval-consulting.com/products/wevia-em.html"),
        ("Consulting Pro", "https://weval-consulting.com/products/consulting-pro.html"),
        ("Suite Enterprise", "https://weval-consulting.com/solutions/wevia-enterprise.html"),
        ("Suite Advisory", "https://weval-consulting.com/solutions/advisory.html"),
    ]

    print(f"{'Page':<25} {'ASCII':>7} {'Typo':>7} {'Status'}")
    print("-"*55)
    clean = 0
    for name, url in stages:
        page.goto(url + "?cb=v109-final", wait_until="networkidle", timeout=25000)
        page.wait_for_timeout(2500)
        c = page.content()
        a = count_ascii(c)
        t = c.count(TYPO)
        status = "✓ CLEAN" if a == 0 else f"✗ {a}"
        if a == 0: clean += 1
        print(f"{name:<25} {a:>7} {t:>7}  {status}")
        # Screenshot
        sdir = "/opt/weval-l99/pw-screenshots-v109"
        os.makedirs(sdir, exist_ok=True)
        fn = f"{sdir}/{name.replace(' ','-').lower()}.png"
        page.screenshot(path=fn, full_page=False)
    
    # Homepage client logos check
    page.goto("https://weval-consulting.com/?cb=logos", wait_until="networkidle", timeout=25000)
    page.wait_for_timeout(2500)
    logos_ok = []
    for n in ["Abbott", "AbbVie", "Michelin", "LOreal", "Carrefour", "Nestle", "Servier"]:
        imgs = page.locator(f'img[src*="{n}"]').all()
        for img in imgs[:2]:
            try:
                if img.is_visible():
                    logos_ok.append(n); break
            except: pass
    
    print(f"\n{clean}/{len(stages)} pages CLEAN")
    print(f"Client logos visible: {logos_ok}")
    
    browser.close()