def index():
"""Render index page.
Supported outputs: html, json, csv, rss.
"""
if not request.args and not request.form:
return render('index.html')
try:
search = Search(request)
except:
return render('index.html')
# TODO moar refactor - do_search integration into Search class
search.results, search.suggestions = do_search(search.query,
request,
search.engines,
search.pageno,
search.lang)
for result in search.results:
if not search.paging and engines[result['engine']].paging:
search.paging = True
if search.request_data.get('format', 'html') == 'html':
if 'content' in result:
result['content'] = highlight_content(result['content'],
search.query.encode('utf-8')) # noqa
result['title'] = highlight_content(result['title'],
search.query.encode('utf-8'))
else:
if 'content' in result:
result['content'] = html_to_text(result['content']).strip()
# removing html content and whitespace duplications
result['title'] = ' '.join(html_to_text(result['title'])
.strip().split())
if len(result['url']) > 74:
url_parts = result['url'][:35], result['url'][-35:]
result['pretty_url'] = u'{0}[...]{1}'.format(*url_parts)
else:
result['pretty_url'] = result['url']
for engine in result['engines']:
if engine in favicons:
result['favicon'] = engine
if search.request_data.get('format') == 'json':
return Response(json.dumps({'query': search.query,
'results': search.results}),
mimetype='application/json')
elif search.request_data.get('format') == 'csv':
csv = UnicodeWriter(cStringIO.StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score')
if search.results:
csv.writerow(keys)
for row in search.results:
row['host'] = row['parsed_url'].netloc
csv.writerow([row.get(key, '') for key in keys])
csv.stream.seek(0)
response = Response(csv.stream.read(), mimetype='application/csv')
cont_disp = 'attachment;Filename=searx_-_{0}.csv'.format(search.query)
response.headers.add('Content-Disposition', cont_disp)
return response
elif search.request_data.get('format') == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=search.results,
q=search.request_data['q'],
number_of_results=len(search.results),
base_url=get_base_url()
)
return Response(response_rss, mimetype='text/xml')
return render(
'results.html',
results=search.results,
q=search.request_data['q'],
selected_categories=search.categories,
paging=search.paging,
pageno=search.pageno,
base_url=get_base_url(),
suggestions=search.suggestions
)
def index():
"""Render index page.
Supported outputs: html, json, csv, rss.
"""
if not request.args and not request.form:
return render('index.html')
try:
search = Search(request)
except:
return render('index.html')
# TODO moar refactor - do_search integration into Search class
search.results, search.suggestions = do_search(search.query,
request,
search.engines,
search.pageno,
search.lang)
for result in search.results:
if not search.paging and engines[result['engine']].paging:
search.paging = True
if search.request_data.get('format', 'html') == 'html':
if 'content' in result:
result['content'] = highlight_content(result['content'],
search.query.encode('utf-8')) # noqa
result['title'] = highlight_content(result['title'],
search.query.encode('utf-8'))
else:
if 'content' in result:
result['content'] = html_to_text(result['content']).strip()
# removing html content and whitespace duplications
result['title'] = ' '.join(html_to_text(result['title'])
.strip().split())
if len(result['url']) > 74:
url_parts = result['url'][:35], result['url'][-35:]
result['pretty_url'] = u'{0}[...]{1}'.format(*url_parts)
else:
result['pretty_url'] = result['url']
for engine in result['engines']:
if engine in favicons:
result['favicon'] = engine
# TODO, check if timezone is calculated right
if 'publishedDate' in result:
if result['publishedDate'].replace(tzinfo=None)\
>= datetime.now() - timedelta(days=1):
timedifference = datetime.now() - result['publishedDate']\
.replace(tzinfo=None)
minutes = int((timedifference.seconds / 60) % 60)
hours = int(timedifference.seconds / 60 / 60)
if hours == 0:
result['publishedDate'] = gettext(u'{minutes} minute(s) ago').format(minutes=minutes) # noqa
else:
result['publishedDate'] = gettext(u'{hours} hour(s), {minutes} minute(s) ago').format(hours=hours, minutes=minutes) # noqa
else:
result['pubdate'] = result['publishedDate']\
.strftime('%a, %d %b %Y %H:%M:%S %z')
result['publishedDate'] = format_date(result['publishedDate'])
if search.request_data.get('format') == 'json':
return Response(json.dumps({'query': search.query,
'results': search.results}),
mimetype='application/json')
elif search.request_data.get('format') == 'csv':
csv = UnicodeWriter(cStringIO.StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score')
if search.results:
csv.writerow(keys)
for row in search.results:
row['host'] = row['parsed_url'].netloc
csv.writerow([row.get(key, '') for key in keys])
csv.stream.seek(0)
response = Response(csv.stream.read(), mimetype='application/csv')
cont_disp = 'attachment;Filename=searx_-_{0}.csv'.format(search.query)
response.headers.add('Content-Disposition', cont_disp)
return response
elif search.request_data.get('format') == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=search.results,
q=search.request_data['q'],
number_of_results=len(search.results),
base_url=get_base_url()
)
return Response(response_rss, mimetype='text/xml')
return render(
'results.html',
results=search.results,
q=search.request_data['q'],
selected_categories=search.categories,
paging=search.paging,
pageno=search.pageno,
base_url=get_base_url(),
suggestions=search.suggestions
)
def index():
"""Render index page.
Supported outputs: html, json, csv, rss.
"""
if not request.args and not request.form:
return render(
'index.html',
)
try:
search = Search(request)
except:
return render(
'index.html',
)
search.results, search.suggestions,\
search.answers, search.infoboxes = search.search(request)
for result in search.results:
if not search.paging and engines[result['engine']].paging:
search.paging = True
# check if HTTPS rewrite is required
if settings['server']['https_rewrite']\
and result['parsed_url'].scheme == 'http':
result = https_url_rewrite(result)
if search.request_data.get('format', 'html') == 'html':
if 'content' in result:
result['content'] = highlight_content(result['content'],
search.query.encode('utf-8')) # noqa
result['title'] = highlight_content(result['title'],
search.query.encode('utf-8'))
else:
if 'content' in result:
result['content'] = html_to_text(result['content']).strip()
# removing html content and whitespace duplications
result['title'] = ' '.join(html_to_text(result['title'])
.strip().split())
result['pretty_url'] = prettify_url(result['url'])
# TODO, check if timezone is calculated right
if 'publishedDate' in result:
result['pubdate'] = result['publishedDate'].strftime('%Y-%m-%d %H:%M:%S%z')
if result['publishedDate'].replace(tzinfo=None) >= datetime.now() - timedelta(days=1):
timedifference = datetime.now() - result['publishedDate'].replace(tzinfo=None)
minutes = int((timedifference.seconds / 60) % 60)
hours = int(timedifference.seconds / 60 / 60)
if hours == 0:
result['publishedDate'] = gettext(u'{minutes} minute(s) ago').format(minutes=minutes) # noqa
else:
result['publishedDate'] = gettext(u'{hours} hour(s), {minutes} minute(s) ago').format(hours=hours, minutes=minutes) # noqa
else:
result['publishedDate'] = format_date(result['publishedDate'])
if search.request_data.get('format') == 'json':
return Response(json.dumps({'query': search.query,
'results': search.results}),
mimetype='application/json')
elif search.request_data.get('format') == 'csv':
csv = UnicodeWriter(cStringIO.StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score')
if search.results:
csv.writerow(keys)
for row in search.results:
row['host'] = row['parsed_url'].netloc
csv.writerow([row.get(key, '') for key in keys])
csv.stream.seek(0)
response = Response(csv.stream.read(), mimetype='application/csv')
cont_disp = 'attachment;Filename=searx_-_{0}.csv'.format(search.query)
response.headers.add('Content-Disposition', cont_disp)
return response
elif search.request_data.get('format') == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=search.results,
q=search.request_data['q'],
number_of_results=len(search.results),
base_url=get_base_url()
)
return Response(response_rss, mimetype='text/xml')
return render(
'results.html',
results=search.results,
q=search.request_data['q'],
selected_categories=search.categories,
paging=search.paging,
pageno=search.pageno,
base_url=get_base_url(),
suggestions=search.suggestions,
answers=search.answers,
infoboxes=search.infoboxes,
theme=get_current_theme_name(),
favicons=global_favicons[themes.index(get_current_theme_name())]
)
def index():
"""Render index page.
Supported outputs: html, json, csv, rss.
"""
if not request.args and not request.form:
return render('index.html', )
try:
search = Search(request)
except:
return render('index.html', )
# TODO moar refactor - do_search integration into Search class
search.results, search.suggestions = do_search(search.query, request,
search.engines,
search.pageno, search.lang)
for result in search.results:
if not search.paging and engines[result['engine']].paging:
search.paging = True
if search.request_data.get('format', 'html') == 'html':
if 'content' in result:
result['content'] = highlight_content(
result['content'], search.query.encode('utf-8')) # noqa
result['title'] = highlight_content(result['title'],
search.query.encode('utf-8'))
else:
if 'content' in result:
result['content'] = html_to_text(result['content']).strip()
# removing html content and whitespace duplications
result['title'] = ' '.join(
html_to_text(result['title']).strip().split())
if len(result['url']) > 74:
url_parts = result['url'][:35], result['url'][-35:]
result['pretty_url'] = u'{0}[...]{1}'.format(*url_parts)
else:
result['pretty_url'] = result['url']
for engine in result['engines']:
if engine in favicons:
result['favicon'] = engine
# TODO, check if timezone is calculated right
if 'publishedDate' in result:
if result['publishedDate'].replace(tzinfo=None)\
>= datetime.now() - timedelta(days=1):
timedifference = datetime.now() - result['publishedDate']\
.replace(tzinfo=None)
minutes = int((timedifference.seconds / 60) % 60)
hours = int(timedifference.seconds / 60 / 60)
if hours == 0:
result['publishedDate'] = gettext(
u'{minutes} minute(s) ago').format(
minutes=minutes) # noqa
else:
result['publishedDate'] = gettext(
u'{hours} hour(s), {minutes} minute(s) ago').format(
hours=hours, minutes=minutes) # noqa
else:
result['pubdate'] = result['publishedDate']\
.strftime('%a, %d %b %Y %H:%M:%S %z')
result['publishedDate'] = format_date(result['publishedDate'])
if search.request_data.get('format') == 'json':
return Response(json.dumps({
'query': search.query,
'results': search.results
}),
mimetype='application/json')
elif search.request_data.get('format') == 'csv':
csv = UnicodeWriter(cStringIO.StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score')
if search.results:
csv.writerow(keys)
for row in search.results:
row['host'] = row['parsed_url'].netloc
csv.writerow([row.get(key, '') for key in keys])
csv.stream.seek(0)
response = Response(csv.stream.read(), mimetype='application/csv')
cont_disp = 'attachment;Filename=searx_-_{0}.csv'.format(search.query)
response.headers.add('Content-Disposition', cont_disp)
return response
elif search.request_data.get('format') == 'rss':
response_rss = render('opensearch_response_rss.xml',
results=search.results,
q=search.request_data['q'],
number_of_results=len(search.results),
base_url=get_base_url())
return Response(response_rss, mimetype='text/xml')
return render('results.html',
results=search.results,
q=search.request_data['q'],
selected_categories=search.categories,
paging=search.paging,
pageno=search.pageno,
base_url=get_base_url(),
suggestions=search.suggestions)
def index():
"""Render index page.
Supported outputs: html, json, csv, rss.
"""
if not request.args and not request.form:
return render(
'index.html',
)
try:
search = Search(request)
except:
return render(
'index.html',
)
search.results, search.suggestions,\
search.answers, search.infoboxes = search.search(request)
for result in search.results:
if not search.paging and engines[result['engine']].paging:
search.paging = True
# check if HTTPS rewrite is required
if settings['server']['https_rewrite']\
and result['parsed_url'].scheme == 'http':
skip_https_rewrite = False
# check if HTTPS rewrite is possible
for target, rules, exclusions in https_rules:
# check if target regex match with url
if target.match(result['url']):
# process exclusions
for exclusion in exclusions:
# check if exclusion match with url
if exclusion.match(result['url']):
skip_https_rewrite = True
break
# skip https rewrite if required
if skip_https_rewrite:
break
# process rules
for rule in rules:
try:
# TODO, precompile rule
p = re.compile(rule[0])
# rewrite url if possible
new_result_url = p.sub(rule[1], result['url'])
except:
break
# parse new url
new_parsed_url = urlparse(new_result_url)
# continiue if nothing was rewritten
if result['url'] == new_result_url:
continue
# get domainname from result
# TODO, does only work correct with TLD's like
# asdf.com, not for asdf.com.de
# TODO, using publicsuffix instead of this rewrite rule
old_result_domainname = '.'.join(
result['parsed_url'].hostname.split('.')[-2:])
new_result_domainname = '.'.join(
new_parsed_url.hostname.split('.')[-2:])
# check if rewritten hostname is the same,
# to protect against wrong or malicious rewrite rules
if old_result_domainname == new_result_domainname:
# set new url
result['url'] = new_result_url
# target has matched, do not search over the other rules
break
if search.request_data.get('format', 'html') == 'html':
if 'content' in result:
result['content'] = highlight_content(result['content'],
search.query.encode('utf-8')) # noqa
result['title'] = highlight_content(result['title'],
search.query.encode('utf-8'))
else:
if 'content' in result:
result['content'] = html_to_text(result['content']).strip()
# removing html content and whitespace duplications
result['title'] = ' '.join(html_to_text(result['title'])
.strip().split())
if len(result['url']) > 74:
url_parts = result['url'][:35], result['url'][-35:]
result['pretty_url'] = u'{0}[...]{1}'.format(*url_parts)
else:
result['pretty_url'] = result['url']
for engine in result['engines']:
if engine in favicons:
result['favicon'] = engine
# TODO, check if timezone is calculated right
if 'publishedDate' in result:
if result['publishedDate'].replace(tzinfo=None)\
>= datetime.now() - timedelta(days=1):
timedifference = datetime.now() - result['publishedDate']\
.replace(tzinfo=None)
minutes = int((timedifference.seconds / 60) % 60)
hours = int(timedifference.seconds / 60 / 60)
if hours == 0:
result['publishedDate'] = gettext(u'{minutes} minute(s) ago').format(minutes=minutes) # noqa
else:
result['publishedDate'] = gettext(u'{hours} hour(s), {minutes} minute(s) ago').format(hours=hours, minutes=minutes) # noqa
else:
result['pubdate'] = result['publishedDate']\
.strftime('%a, %d %b %Y %H:%M:%S %z')
result['publishedDate'] = format_date(result['publishedDate'])
if search.request_data.get('format') == 'json':
return Response(json.dumps({'query': search.query,
'results': search.results}),
mimetype='application/json')
elif search.request_data.get('format') == 'csv':
csv = UnicodeWriter(cStringIO.StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score')
if search.results:
csv.writerow(keys)
for row in search.results:
row['host'] = row['parsed_url'].netloc
csv.writerow([row.get(key, '') for key in keys])
csv.stream.seek(0)
response = Response(csv.stream.read(), mimetype='application/csv')
cont_disp = 'attachment;Filename=searx_-_{0}.csv'.format(search.query)
response.headers.add('Content-Disposition', cont_disp)
return response
elif search.request_data.get('format') == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=search.results,
q=search.request_data['q'],
number_of_results=len(search.results),
base_url=get_base_url()
)
return Response(response_rss, mimetype='text/xml')
return render(
'results.html',
results=search.results,
q=search.request_data['q'],
selected_categories=search.categories,
paging=search.paging,
pageno=search.pageno,
base_url=get_base_url(),
suggestions=search.suggestions,
answers=search.answers,
infoboxes=search.infoboxes,
theme=get_current_theme_name()
)
