"""Check the running local site without sending forms or creating contacts."""
import json
import concurrent.futures
import urllib.request
import urllib.error
import subprocess
import argparse
from pathlib import Path
from html.parser import HTMLParser

ROOT = Path(__file__).resolve().parents[1]
parser = argparse.ArgumentParser(description=__doc__)
parser.add_argument('--base', default='http://localhost:3000')
parser.add_argument('--report', default='http-checks.json')
args = parser.parse_args()
BASE = args.base.rstrip('/')
modules = json.loads(subprocess.check_output([
    'node', '--input-type=module', '-e',
    "import {modules} from './data/cvim/course.mjs'; console.log(JSON.stringify(modules.map(m=>({path:m.path,title:m.title,topics:m.topics.map(t=>({path:t.path,title:t.title}))}))));"
], cwd=ROOT, encoding='utf-8'))

class Page(HTMLParser):
    def __init__(self):
        super().__init__()
        self.h1 = False
        self.headings = []
        self.canonical = None
        self.schemas = []
        self.script = False
        self.script_text = ''
        self.meta = {}
        self.article = False
        self.article_text = []
    def handle_starttag(self, tag, attrs):
        attrs = dict(attrs)
        if tag == 'h1': self.h1 = True
        if tag == 'link' and attrs.get('rel') == 'canonical': self.canonical = attrs.get('href')
        if tag == 'meta': self.meta[attrs.get('name') or attrs.get('property')] = attrs.get('content', '')
        if tag == 'article': self.article = True
        if tag == 'script' and attrs.get('type') == 'application/ld+json': self.script = True; self.script_text = ''
    def handle_endtag(self, tag):
        if tag == 'h1': self.h1 = False
        if tag == 'article': self.article = False
        if tag == 'script' and self.script:
            self.schemas.append(json.loads(self.script_text))
            self.script = False
    def handle_data(self, text):
        if self.h1: self.headings.append(text)
        if self.script: self.script_text += text
        if self.article and not self.script: self.article_text.append(text)

targets = [{'path': '/cvim/resources', 'title': 'Understand the task.'}]
for module in modules:
    targets.append({'path': module['path'], 'title': module['title']})
    targets.extend(module['topics'])

def check(target):
    path = target['path']
    with urllib.request.urlopen(BASE + path, timeout=60) as response:
        html = response.read().decode()
        assert response.status == 200
    page = Page()
    page.feed(html)
    assert target['title'] in ''.join(page.headings), (path, 'wrong or missing heading')
    assert page.canonical and page.canonical.endswith(path), (path, 'missing canonical')
    assert any(isinstance(schema, (list, dict)) for schema in page.schemas), (path, 'missing schema')
    assert page.meta.get('description'), (path, 'missing description')
    assert page.meta.get('twitter:card') == 'summary_large_image', (path, 'missing social preview')
    assert '/cvim/resources/share-image' in page.meta.get('og:image', ''), (path, 'missing topic share image')
    assert 'IELTS Masterclass guidance' in html, (path, 'missing advisor card')
    assert 'https://www.linkedin.com/in/aisha-vats-4b969121b/' in html
    if path.count('/') == 4:
        assert 'A worked example' in html
        assert 'A practical sequence to follow' in html
        assert 'Sources and authorship' in html
        assert len(' '.join(page.article_text).split()) >= 500, (path, 'insufficient rendered content')
    assert 'NEXT_HTTP_ERROR_FALLBACK;500' not in html
    return path

results = []
errors = []
with concurrent.futures.ThreadPoolExecutor(max_workers=3) as executor:
    futures = {executor.submit(check, target): target for target in targets}
    for future in concurrent.futures.as_completed(futures):
        try:
            results.append(future.result())
            if len(results) % 25 == 0: print(f'Checked {len(results)}/{len(targets)} pages', flush=True)
        except Exception as error: errors.append({'path': futures[future]['path'], 'error': str(error)})

for path in ['/cvim/resources/missing-module', '/cvim/resources/ielts-foundations/missing-topic']:
    try:
        urllib.request.urlopen(BASE + path, timeout=60)
        errors.append({'path': path, 'error': 'Expected a 404'})
    except urllib.error.HTTPError as error:
        if error.code != 404: errors.append({'path': path, 'error': str(error)})

source_count = 0
from urllib.parse import quote
for path in (ROOT / 'uploads/pdf/cvim').glob('*.pdf'):
    try:
        with urllib.request.urlopen(BASE + '/uploads/pdf/cvim/' + quote(path.name), timeout=60) as response:
            assert response.headers.get_content_type() == 'application/pdf'
            assert response.read(5) == b'%PDF-'
            source_count += 1
    except Exception as error: errors.append({'path': str(path), 'error': str(error)})

report = {'pagesPassed': len(results), 'pagesExpected': len(targets), 'sourcePdfsPassed': source_count, 'errors': errors}
(ROOT / 'reports/cvim' / args.report).write_text(json.dumps(report, indent=2) + '\n', encoding='utf-8')
print(json.dumps(report, indent=2), flush=True)
raise SystemExit(1 if errors else 0)
