def track_pa11y_stats(pa11y_results, spider):
"""
Keep track of the number of pa11y errors, warnings, and notices that
we've seen so far, using the Scrapy stats collector:
http://doc.scrapy.org/en/1.1/topics/stats.html
"""
num_err, num_warn, num_notice = pa11y_counts(pa11y_results)
stats = spider.crawler.stats
stats.inc_value("pa11y/error", count=num_err, spider=spider)
stats.inc_value("pa11y/warning", count=num_warn, spider=spider)
stats.inc_value("pa11y/notice", count=num_notice, spider=spider)
def render_html(data_dir, output_dir):
"""
The main workhorse of this script. Finds all the JSON data files
from pa11ycrawler, and transforms them into HTML files via Jinja2 templating.
"""
env = Environment(loader=PackageLoader('pa11ycrawler', 'templates'))
env.globals["wcag_refs"] = wcag_refs
pages = []
counter = collections.Counter()
copy_assets(output_dir)
# render detail templates
template = env.get_template("detail.html")
for data_file in data_dir.files('*.json'):
data = json.load(data_file.open())
num_error, num_warning, num_notice = pa11y_counts(data['pa11y'])
data["num_error"] = num_error
data["num_warning"] = num_warning
data["num_notice"] = num_notice
rendered_html = template.render(**data) # pylint: disable=no-member
# replace `.json` with `.html`
fname = data_file.namebase + ".html"
html_path = output_dir / fname
html_path.write_text(rendered_html, encoding='utf-8')
data["filename"] = fname
pages.append(data)
counter["error"] += num_error
counter["warning"] += num_warning
counter["notice"] += num_notice
def extract_nums(page):
"Used for sorting"
return (
page["num_error"],
page["num_warning"],
page["num_notice"],
)
# render index template
index_template = env.get_template("index.html")
context = {
"pages": sorted(pages, key=extract_nums, reverse=True),
"num_error": counter["error"],
"num_warning": counter["warning"],
"num_notice": counter["notice"],
}
rendered_html = index_template.render(**context) # pylint: disable=no-member
html_path = output_dir / "index.html"
html_path.write_text(rendered_html, encoding='utf-8')
def render_html(data_dir, output_dir):
"""
The main workhorse of this script. Finds all the JSON data files
from pa11ycrawler, and transforms them into HTML files via Jinja2 templating.
"""
env = Environment(loader=PackageLoader('pa11ycrawler', 'templates'))
env.globals["wcag_refs"] = wcag_refs
pages = []
counter = collections.Counter()
grouped_violations = collections.defaultdict(dict)
# render detail templates
for data_file in data_dir.files('*.json'):
data = json.load(data_file.open())
num_error, num_warning, num_notice = pa11y_counts(data['pa11y'])
data["num_error"] = num_error
data["num_warning"] = num_warning
data["num_notice"] = num_notice
fname = data_file.namebase + ".html"
html_path = output_dir / fname
render_template(env, html_path, 'detail.html', data)
data["filename"] = fname
pages.append(data)
for violation in data['pa11y']:
violation_id = hashlib.md5(
(violation['selector'] +
violation['code']).encode('utf-8')).hexdigest()
if violation_id not in grouped_violations[violation['type']]:
violation['pages'] = []
grouped_violations[violation['type']][violation_id] = violation
counter[violation['type']] += 1
grouped_violations[
violation['type']][violation_id]['pages'].append({
'url':
data['url'],
'page_title':
data['page_title']
})
def extract_nums(page):
"Used to sort pages by violation counts"
return (
page["num_error"],
page["num_warning"],
page["num_notice"],
)
index_path = output_dir / INDEX_TEMPLATE
render_template(
env, index_path, INDEX_TEMPLATE, {
"pages": sorted(pages, key=extract_nums, reverse=True),
"num_error": counter["error"],
"num_warning": counter["warning"],
"num_notice": counter["notice"]
})
for violation_type in grouped_violations:
unique_path = output_dir / u'{}s.html'.format(violation_type)
render_template(
env, unique_path, UNIQUE_TEMPLATE, {
"grouped_violations":
sorted(grouped_violations[violation_type].values(),
key=lambda item: len(item['pages']),
reverse=True),
"current_type":
violation_type,
"violation_counts":
counter
})
def render_html(data_dir, output_dir):
"""
The main workhorse of this script. Finds all the JSON data files
from pa11ycrawler, and transforms them into HTML files via Jinja2 templating.
"""
env = Environment(loader=PackageLoader('pa11ycrawler', 'templates'))
env.globals["wcag_refs"] = wcag_refs
pages = []
counter = collections.Counter()
grouped_violations = collections.defaultdict(dict)
# render detail templates
for data_file in data_dir.files('*.json'):
data = json.load(data_file.open())
num_error, num_warning, num_notice = pa11y_counts(data['pa11y'])
data["num_error"] = num_error
data["num_warning"] = num_warning
data["num_notice"] = num_notice
fname = data_file.namebase + ".html"
html_path = output_dir / fname
render_template(env, html_path, 'detail.html', data)
data["filename"] = fname
pages.append(data)
for violation in data['pa11y']:
violation_id = hashlib.md5(
(violation['selector'] + violation['code']).encode('utf-8')
).hexdigest()
if violation_id not in grouped_violations[violation['type']]:
violation['pages'] = []
grouped_violations[violation['type']][violation_id] = violation
counter[violation['type']] += 1
grouped_violations[violation['type']][violation_id]['pages'].append({
'url': data['url'],
'page_title': data['page_title']
})
def extract_nums(page):
"Used to sort pages by violation counts"
return (
page["num_error"],
page["num_warning"],
page["num_notice"],
)
index_path = output_dir / INDEX_TEMPLATE
render_template(env, index_path, INDEX_TEMPLATE, {
"pages": sorted(pages, key=extract_nums, reverse=True),
"num_error": counter["error"],
"num_warning": counter["warning"],
"num_notice": counter["notice"]
})
for violation_type in grouped_violations:
unique_path = output_dir / u'{}s.html'.format(violation_type)
render_template(env, unique_path, UNIQUE_TEMPLATE, {
"grouped_violations": sorted(
grouped_violations[violation_type].values(),
key=lambda item: len(item['pages']),
reverse=True
),
"current_type": violation_type,
"violation_counts": counter
})