def babel_format_bytes(size, unit='B', step_size=1024):
"""
Format given numeric value to human readable string describing size in
B/KB/MB/GB/TB.
:param int size: Number to be formatted.
:param enum unit: Starting unit, possible values are ['B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB'].
:param int step_size: Size of the step between units.
:return: Formatted and localized string.
:rtype: string
"""
units = ['B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB']
idx_max = len(units) - 1
unit = unit.upper()
for idx, val in enumerate(units):
# Skip the last step, there is no next unit defined after exabyte.
if idx == idx_max:
break
if size > step_size:
if unit == val:
size = size / step_size
unit = units[idx + 1]
else:
break
return '{} {}'.format(flask_babel.format_decimal(size), unit)
def test_basics(self):
app = flask.Flask(__name__)
babel.Babel(app)
n = 1099
with app.test_request_context():
assert babel.format_number(n) == u'1,099'
assert babel.format_decimal(Decimal('1010.99')) == u'1,010.99'
assert babel.format_currency(n, 'USD') == '$1,099.00'
assert babel.format_percent(0.19) == '19%'
assert babel.format_scientific(10000) == u'1E4'
def test_basics(self):
app = flask.Flask(__name__)
b = babel.Babel(app)
n = 1099
with app.test_request_context():
assert babel.format_number(n) == u'1,099'
assert babel.format_decimal(Decimal('1010.99')) == u'1,010.99'
assert babel.format_currency(n, 'USD') == '$1,099.00'
assert babel.format_percent(0.19) == '19%'
assert babel.format_scientific(10000) == u'1E4'
def index():
anthony = gettext(u'Anthony')
num = format_decimal(12345)
d = datetime.fromtimestamp(1594460958.12345)
day = format_date(d)
results = {
'num' : num,
'date' : day,
'anthony' : anthony
}
return render_template('index.html', results = results)
def localised_number(number: float) -> str:
''' Returns a localised unicode representation of number '''
return format_decimal(number)
def index():
"""Render index page.
Supported outputs: html, json, csv, rss.
"""
# output_format
output_format = request.form.get('format', 'html')
if output_format not in ['html', 'csv', 'json', 'rss']:
output_format = 'html'
# check if there is query
if request.form.get('q') is None:
if output_format == 'html':
return render(
'index.html',
)
else:
return index_error(output_format, 'No query'), 400
# search
search_query = None
raw_text_query = None
result_container = None
try:
search_query, raw_text_query = get_search_query_from_webapp(request.preferences, request.form)
# search = Search(search_query) # without plugins
search = SearchWithPlugins(search_query, request.user_plugins, request)
result_container = search.search()
except Exception as e:
# log exception
logger.exception('search error')
# is it an invalid input parameter or something else ?
if (issubclass(e.__class__, SearxParameterException)):
return index_error(output_format, e.message), 400
else:
return index_error(output_format, gettext('search error')), 500
# results
results = result_container.get_ordered_results()
number_of_results = result_container.results_number()
if number_of_results < result_container.results_length():
number_of_results = 0
# UI
advanced_search = request.form.get('advanced_search', None)
# Server-Timing header
request.timings = result_container.get_timings()
# output
for result in results:
if output_format == 'html':
if 'content' in result and result['content']:
result['content'] = highlight_content(escape(result['content'][:1024]), search_query.query)
if 'title' in result and result['title']:
result['title'] = highlight_content(escape(result['title'] or u''), search_query.query)
else:
if result.get('content'):
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 'url' in result:
result['pretty_url'] = prettify_url(result['url'])
# TODO, check if timezone is calculated right
if 'publishedDate' in result:
try: # test if publishedDate >= 1900 (datetime module bug)
result['pubdate'] = result['publishedDate'].strftime('%Y-%m-%d %H:%M:%S%z')
except ValueError:
result['publishedDate'] = None
else:
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)
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 output_format == 'json':
return Response(json.dumps({'query': search_query.query.decode('utf-8'),
'number_of_results': number_of_results,
'results': results,
'answers': list(result_container.answers),
'corrections': list(result_container.corrections),
'infoboxes': result_container.infoboxes,
'suggestions': list(result_container.suggestions),
'unresponsive_engines': __get_translated_errors(result_container.unresponsive_engines)}, # noqa
default=lambda item: list(item) if isinstance(item, set) else item),
mimetype='application/json')
elif output_format == 'csv':
csv = UnicodeWriter(StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score', 'type')
csv.writerow(keys)
for row in results:
row['host'] = row['parsed_url'].netloc
row['type'] = 'result'
csv.writerow([row.get(key, '') for key in keys])
for a in result_container.answers:
row = {'title': a, 'type': 'answer'}
csv.writerow([row.get(key, '') for key in keys])
for a in result_container.suggestions:
row = {'title': a, 'type': 'suggestion'}
csv.writerow([row.get(key, '') for key in keys])
for a in result_container.corrections:
row = {'title': a, 'type': 'correction'}
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.query.decode('utf-8'))
response.headers.add('Content-Disposition', cont_disp)
return response
elif output_format == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=results,
answers=result_container.answers,
corrections=result_container.corrections,
suggestions=result_container.suggestions,
q=request.form['q'],
number_of_results=number_of_results,
base_url=get_base_url(),
override_theme='__common__',
)
return Response(response_rss, mimetype='text/xml')
# HTML output format
# suggestions: use RawTextQuery to get the suggestion URLs with the same bang
suggestion_urls = list(map(lambda suggestion: {
'url': raw_text_query.changeSearchQuery(suggestion).getFullQuery(),
'title': suggestion
},
result_container.suggestions))
correction_urls = list(map(lambda correction: {
'url': raw_text_query.changeSearchQuery(correction).getFullQuery(),
'title': correction
},
result_container.corrections))
#
return render(
'results.html',
results=results,
q=request.form['q'],
selected_categories=search_query.categories,
pageno=search_query.pageno,
time_range=search_query.time_range,
number_of_results=format_decimal(number_of_results),
advanced_search=advanced_search,
suggestions=suggestion_urls,
answers=result_container.answers,
corrections=correction_urls,
infoboxes=result_container.infoboxes,
paging=result_container.paging,
unresponsive_engines=__get_translated_errors(result_container.unresponsive_engines),
current_language=match_language(search_query.lang,
LANGUAGE_CODES,
fallback=request.preferences.get_value("language")),
base_url=get_base_url(),
theme=get_current_theme_name(),
favicons=global_favicons[themes.index(get_current_theme_name())],
timeout_limit=request.form.get('timeout_limit', None)
)
def index():
"""Render index page.
Supported outputs: html, json, csv, rss.
"""
# output_format
output_format = request.form.get('format', 'html')
if output_format not in ['html', 'csv', 'json', 'rss']:
output_format = 'html'
# check if there is query
if request.form.get('q') is None:
if output_format == 'html':
return render(
'index.html',
)
else:
return index_error(output_format, 'No query'), 400
# search
search_query = None
result_container = None
try:
search_query = get_search_query_from_webapp(request.preferences, request.form)
# search = Search(search_query) # without plugins
search = SearchWithPlugins(search_query, request.user_plugins, request)
result_container = search.search()
except Exception as e:
# log exception
logger.exception('search error')
# is it an invalid input parameter or something else ?
if (issubclass(e.__class__, SearxParameterException)):
return index_error(output_format, e.message), 400
else:
return index_error(output_format, gettext('search error')), 500
# results
results = result_container.get_ordered_results()
number_of_results = result_container.results_number()
if number_of_results < result_container.results_length():
number_of_results = 0
# UI
advanced_search = request.form.get('advanced_search', None)
# output
for result in results:
if output_format == 'html':
if 'content' in result and result['content']:
result['content'] = highlight_content(escape(result['content'][:1024]), search_query.query)
result['title'] = highlight_content(escape(result['title'] or u''), search_query.query)
else:
if result.get('content'):
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:
try: # test if publishedDate >= 1900 (datetime module bug)
result['pubdate'] = result['publishedDate'].strftime('%Y-%m-%d %H:%M:%S%z')
except ValueError:
result['publishedDate'] = None
else:
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)
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 output_format == 'json':
return Response(json.dumps({'query': search_query.query.decode('utf-8'),
'number_of_results': number_of_results,
'results': results,
'answers': list(result_container.answers),
'corrections': list(result_container.corrections),
'infoboxes': result_container.infoboxes,
'suggestions': list(result_container.suggestions),
'unresponsive_engines': list(result_container.unresponsive_engines)},
default=lambda item: list(item) if isinstance(item, set) else item),
mimetype='application/json')
elif output_format == 'csv':
csv = UnicodeWriter(StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score')
csv.writerow(keys)
for row in 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.query)
response.headers.add('Content-Disposition', cont_disp)
return response
elif output_format == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=results,
q=request.form['q'],
number_of_results=number_of_results,
base_url=get_base_url(),
override_theme='__common__',
)
return Response(response_rss, mimetype='text/xml')
return render(
'results.html',
results=results,
q=request.form['q'],
selected_categories=search_query.categories,
pageno=search_query.pageno,
time_range=search_query.time_range,
number_of_results=format_decimal(number_of_results),
advanced_search=advanced_search,
suggestions=result_container.suggestions,
answers=result_container.answers,
corrections=result_container.corrections,
infoboxes=result_container.infoboxes,
paging=result_container.paging,
unresponsive_engines=result_container.unresponsive_engines,
current_language=match_language(search_query.lang,
LANGUAGE_CODES,
fallback=settings['search']['language']),
base_url=get_base_url(),
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 request.form.get('q') is None:
return render(
'index.html',
)
# search
search_query = None
result_container = None
try:
search_query = get_search_query_from_webapp(request.preferences, request.form)
# search = Search(search_query) # without plugins
search = SearchWithPlugins(search_query, request)
result_container = search.search()
except:
request.errors.append(gettext('search error'))
logger.exception('search error')
return render(
'index.html',
)
results = result_container.get_ordered_results()
# UI
advanced_search = request.form.get('advanced_search', None)
output_format = request.form.get('format', 'html')
if output_format not in ['html', 'csv', 'json', 'rss']:
output_format = 'html'
# output
for result in results:
if output_format == 'html':
if 'content' in result and result['content']:
result['content'] = highlight_content(escape(result['content'][:1024]),
search_query.query.encode('utf-8'))
result['title'] = highlight_content(escape(result['title'] or u''),
search_query.query.encode('utf-8'))
else:
if result.get('content'):
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:
try: # test if publishedDate >= 1900 (datetime module bug)
result['pubdate'] = result['publishedDate'].strftime('%Y-%m-%d %H:%M:%S%z')
except ValueError:
result['publishedDate'] = None
else:
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)
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'])
number_of_results = result_container.results_number()
if number_of_results < result_container.results_length():
number_of_results = 0
if output_format == 'json':
return Response(json.dumps({'query': search_query.query,
'number_of_results': number_of_results,
'results': results,
'answers': list(result_container.answers),
'infoboxes': result_container.infoboxes,
'suggestions': list(result_container.suggestions)}),
mimetype='application/json')
elif output_format == 'csv':
csv = UnicodeWriter(cStringIO.StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score')
csv.writerow(keys)
for row in 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.query.encode('utf-8'))
response.headers.add('Content-Disposition', cont_disp)
return response
elif output_format == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=results,
q=request.form['q'],
number_of_results=number_of_results,
base_url=get_base_url()
)
return Response(response_rss, mimetype='text/xml')
return render(
'results.html',
results=results,
q=request.form['q'],
selected_categories=search_query.categories,
pageno=search_query.pageno,
time_range=search_query.time_range,
number_of_results=format_decimal(number_of_results),
advanced_search=advanced_search,
suggestions=result_container.suggestions,
answers=result_container.answers,
infoboxes=result_container.infoboxes,
paging=result_container.paging,
base_url=get_base_url(),
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',
)
if plugins.call('pre_search', request, locals()):
search.search(request)
plugins.call('post_search', request, locals())
results = search.result_container.get_ordered_results()
for result in results:
plugins.call('on_result', request, locals())
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 result.get('content'):
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:
try: # test if publishedDate >= 1900 (datetime module bug)
result['pubdate'] = result['publishedDate'].strftime('%Y-%m-%d %H:%M:%S%z')
except ValueError:
result['publishedDate'] = None
else:
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)
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'])
number_of_results = search.result_container.results_number()
if number_of_results < search.result_container.results_length():
number_of_results = 0
if search.request_data.get('format') == 'json':
return Response(json.dumps({'query': search.query,
'number_of_results': number_of_results,
'results': results}),
mimetype='application/json')
elif search.request_data.get('format') == 'csv':
csv = UnicodeWriter(cStringIO.StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score')
csv.writerow(keys)
for row in 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.encode('utf-8'))
response.headers.add('Content-Disposition', cont_disp)
return response
elif search.request_data.get('format') == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=results,
q=search.request_data['q'],
number_of_results=number_of_results,
base_url=get_base_url()
)
return Response(response_rss, mimetype='text/xml')
return render(
'results.html',
results=results,
q=search.request_data['q'],
selected_categories=search.categories,
paging=search.paging,
number_of_results=format_decimal(number_of_results),
pageno=search.pageno,
advanced_search=search.is_advanced,
time_range=search.time_range,
base_url=get_base_url(),
suggestions=search.result_container.suggestions,
answers=search.result_container.answers,
infoboxes=search.result_container.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.
"""
# output_format
output_format = request.form.get('format', 'html')
if output_format not in ['html', 'csv', 'json', 'rss']:
output_format = 'html'
# check if there is query
if request.form.get('q') is None or len(request.form.get('q')) == 0:
if output_format == 'html':
return render(
'index.html',
)
else:
return index_error(output_format, 'No query'), 400
# search
search_query = None
result_container = None
try:
search_query = get_search_query_from_webapp(request.preferences, request.form)
# search = Search(search_query) # without plugins
search = SearchWithPlugins(search_query, request.user_plugins, request)
result_container = search.search()
except Exception as e:
# log exception
logger.exception('search error')
# is it an invalid input parameter or something else ?
if (issubclass(e.__class__, SearxParameterException)):
return index_error(output_format, e.message), 400
else:
return index_error(output_format, gettext('search error')), 500
# results
results = result_container.get_ordered_results()
number_of_results = result_container.results_number()
if number_of_results < result_container.results_length():
number_of_results = 0
# UI
advanced_search = request.form.get('advanced_search', None)
# output
for result in results:
if output_format == 'html':
if 'content' in result and result['content']:
result['content'] = highlight_content(prettify_content(
escape(result['content'][:1024])), search_query.query)
result['title'] = highlight_content(escape(result['title'] or u''), search_query.query)
else:
if result.get('content'):
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 result.__contains__('showurl'):
result['pretty_url'] = prettify_url(result['showurl'])
else:
result['pretty_url'] = prettify_url(result['url'])
# TODO, check if timezone is calculated right
if 'publishedDate' in result:
try: # test if publishedDate >= 1900 (datetime module bug)
result['pubdate'] = result['publishedDate'].strftime('%Y-%m-%d %H:%M:%S%z')
except ValueError:
result['publishedDate'] = None
else:
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)
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'])
return render(
'results.html',
results=results,
q=request.form['q'],
selected_categories=search_query.categories,
pageno=search_query.pageno,
time_range=search_query.time_range if search_query.time_range else "",
number_of_results=format_decimal(round(number_of_results)),
advanced_search=advanced_search,
suggestions=result_container.suggestions,
answers=result_container.answers,
corrections=result_container.corrections,
infoboxes=result_container.infoboxes,
paging=result_container.paging,
unresponsive_engines=result_container.unresponsive_engines,
current_language=match_language(search_query.lang,
LANGUAGE_CODES,
fallback=settings['search']['language']),
base_url=get_base_url(),
theme=get_current_theme_name(),
favicons=global_favicons[themes.index(get_current_theme_name())]
)
def index():
d = date(1984, 8, 12)
t = time(22, 16)
current_locale = str(get_locale())
if current_locale in ['es', 'de', 'mt', 'se_SE']:
current_currency = 'EUR'
elif current_locale == 'en_GB':
current_currency = 'GBP'
elif current_locale in ['en_AU']:
current_currency = 'AUD'
else:
current_currency = 'USD'
translate_current = gettext('Current')
us_num = numbers.format_decimal(1.2345, locale='en_US')
us_cur = numbers.format_currency(1.2345, locale='en_US', currency='USD')
us_date = dates.format_date(d, locale='en_US')
us_time = dates.format_time(t, locale='en_US')
se_num = numbers.format_decimal(1.2345, locale='se_SE')
se_cur = numbers.format_currency(1.2345, locale='se_SE', currency='EUR')
se_date = dates.format_date(d, locale='se_SE')
se_time = dates.format_time(t, locale='se_SE')
mt_num = numbers.format_decimal(1.2345, locale='mt_MT')
mt_cur = numbers.format_currency(1.2345, locale='mt_MT', currency='EUR')
mt_date = dates.format_date(d, locale='mt_MT')
mt_time = dates.format_time(t, locale='mt_MT')
current_num = format_decimal(1.2345)
current_cur = format_currency(1.2345, currency=current_currency)
current_date = format_date(d)
current_time = format_time(t)
results = {
'us_num': us_num,
'us_cur': us_cur,
'us_date': us_date,
'us_time': us_time,
'se_num': se_num,
'se_cur': se_cur,
'se_date': se_date,
'se_time': se_time,
'mt_num': mt_num,
'mt_cur': mt_cur,
'mt_date': mt_date,
'mt_time': mt_time,
'current_num': current_num,
'current_cur': current_cur,
'current_date': current_date,
'current_time': current_time
}
return render_template('index.html',
results=results,
current_locale=current_locale,
current=translate_current)
def index():
"""Render index page.
Supported outputs: html, json, csv, rss.
"""
if request.form.get("q") is None:
return render("index.html")
# search
search_query = None
result_container = None
try:
search_query = get_search_query_from_webapp(request.preferences, request.form)
# search = Search(search_query) # without plugins
search = SearchWithPlugins(search_query, request)
result_container = search.search()
except:
request.errors.append(gettext("search error"))
logger.exception("search error")
return render("index.html")
results = result_container.get_ordered_results()
# UI
advanced_search = request.form.get("advanced_search", None)
output_format = request.form.get("format", "html")
if output_format not in ["html", "csv", "json", "rss"]:
output_format = "html"
# output
for result in results:
if output_format == "html":
if "content" in result and result["content"]:
result["content"] = highlight_content(
escape(result["content"][:1024]), search_query.query.encode("utf-8")
)
result["title"] = highlight_content(escape(result["title"] or u""), search_query.query.encode("utf-8"))
else:
if result.get("content"):
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:
try: # test if publishedDate >= 1900 (datetime module bug)
result["pubdate"] = result["publishedDate"].strftime("%Y-%m-%d %H:%M:%S%z")
except ValueError:
result["publishedDate"] = None
else:
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)
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"])
number_of_results = result_container.results_number()
if number_of_results < result_container.results_length():
number_of_results = 0
if output_format == "json":
return Response(
json.dumps(
{
"query": search_query.query,
"number_of_results": number_of_results,
"results": results,
"answers": list(result_container.answers),
"infoboxes": result_container.infoboxes,
"suggestions": list(result_container.suggestions),
}
),
mimetype="application/json",
)
elif output_format == "csv":
csv = UnicodeWriter(cStringIO.StringIO())
keys = ("title", "url", "content", "host", "engine", "score")
csv.writerow(keys)
for row in 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.query.encode("utf-8"))
response.headers.add("Content-Disposition", cont_disp)
return response
elif output_format == "rss":
response_rss = render(
"opensearch_response_rss.xml",
results=results,
q=request.form["q"],
number_of_results=number_of_results,
base_url=get_base_url(),
)
return Response(response_rss, mimetype="text/xml")
return render(
"results.html",
results=results,
q=request.form["q"],
selected_categories=search_query.categories,
pageno=search_query.pageno,
time_range=search_query.time_range,
number_of_results=format_decimal(number_of_results),
advanced_search=advanced_search,
suggestions=result_container.suggestions,
answers=result_container.answers,
infoboxes=result_container.infoboxes,
paging=result_container.paging,
base_url=get_base_url(),
theme=get_current_theme_name(),
favicons=global_favicons[themes.index(get_current_theme_name())],
)
def search():
"""Search query in q and return results.
Supported outputs: html, json, csv, rss.
"""
# pylint: disable=too-many-locals, too-many-return-statements, too-many-branches
# pylint: disable=too-many-statements
# output_format
output_format = request.form.get('format', 'html')
if output_format not in OUTPUT_FORMATS:
output_format = 'html'
if output_format not in settings['search']['formats']:
flask.abort(403)
# check if there is query (not None and not an empty string)
if not request.form.get('q'):
if output_format == 'html':
return render(
'index.html',
selected_categories=get_selected_categories(
request.preferences, request.form),
)
return index_error(output_format, 'No query'), 400
# search
search_query = None
raw_text_query = None
result_container = None
try:
search_query, raw_text_query, _, _ = get_search_query_from_webapp(
request.preferences, request.form)
# search = Search(search_query) # without plugins
search = SearchWithPlugins(search_query, request.user_plugins, request) # pylint: disable=redefined-outer-name
result_container = search.search()
except SearxParameterException as e:
logger.exception('search error: SearxParameterException')
return index_error(output_format, e.message), 400
except Exception as e: # pylint: disable=broad-except
logger.exception(e, exc_info=True)
return index_error(output_format, gettext('search error')), 500
# results
results = result_container.get_ordered_results()
number_of_results = result_container.results_number()
if number_of_results < result_container.results_length():
number_of_results = 0
# checkin for a external bang
if result_container.redirect_url:
return redirect(result_container.redirect_url)
# Server-Timing header
request.timings = result_container.get_timings() # pylint: disable=assigning-non-slot
# output
for result in results:
if output_format == 'html':
if 'content' in result and result['content']:
result['content'] = highlight_content(
escape(result['content'][:1024]), search_query.query)
if 'title' in result and result['title']:
result['title'] = highlight_content(
escape(result['title'] or ''), search_query.query)
else:
if result.get('content'):
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 'url' in result:
result['pretty_url'] = prettify_url(result['url'])
# TODO, check if timezone is calculated right # pylint: disable=fixme
if result.get(
'publishedDate'
): # do not try to get a date from an empty string or a None type
try: # test if publishedDate >= 1900 (datetime module bug)
result['pubdate'] = result['publishedDate'].strftime(
'%Y-%m-%d %H:%M:%S%z')
except ValueError:
result['publishedDate'] = None
else:
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(
'{minutes} minute(s) ago').format(minutes=minutes)
else:
result['publishedDate'] = gettext(
'{hours} hour(s), {minutes} minute(s) ago').format(
hours=hours, minutes=minutes)
else:
result['publishedDate'] = format_date(
result['publishedDate'])
if output_format == 'json':
x = {
'query':
search_query.query,
'number_of_results':
number_of_results,
'results':
results,
'answers':
list(result_container.answers),
'corrections':
list(result_container.corrections),
'infoboxes':
result_container.infoboxes,
'suggestions':
list(result_container.suggestions),
'unresponsive_engines':
__get_translated_errors(result_container.unresponsive_engines)
}
response = json.dumps(x,
default=lambda item: list(item)
if isinstance(item, set) else item)
return Response(response, mimetype='application/json')
if output_format == 'csv':
csv = UnicodeWriter(StringIO())
keys = ('title', 'url', 'content', 'host', 'engine', 'score', 'type')
csv.writerow(keys)
for row in results:
row['host'] = row['parsed_url'].netloc
row['type'] = 'result'
csv.writerow([row.get(key, '') for key in keys])
for a in result_container.answers:
row = {'title': a, 'type': 'answer'}
csv.writerow([row.get(key, '') for key in keys])
for a in result_container.suggestions:
row = {'title': a, 'type': 'suggestion'}
csv.writerow([row.get(key, '') for key in keys])
for a in result_container.corrections:
row = {'title': a, 'type': 'correction'}
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.query)
response.headers.add('Content-Disposition', cont_disp)
return response
if output_format == 'rss':
response_rss = render(
'opensearch_response_rss.xml',
results=results,
answers=result_container.answers,
corrections=result_container.corrections,
suggestions=result_container.suggestions,
q=request.form['q'],
number_of_results=number_of_results,
override_theme='__common__',
)
return Response(response_rss, mimetype='text/xml')
# HTML output format
# suggestions: use RawTextQuery to get the suggestion URLs with the same bang
suggestion_urls = list(
map(
lambda suggestion: {
'url': raw_text_query.changeQuery(suggestion).getFullQuery(),
'title': suggestion
}, result_container.suggestions))
correction_urls = list(
map(
lambda correction: {
'url': raw_text_query.changeQuery(correction).getFullQuery(),
'title': correction
}, result_container.corrections))
return render('results.html',
results=results,
q=request.form['q'],
selected_categories=search_query.categories,
pageno=search_query.pageno,
time_range=search_query.time_range,
number_of_results=format_decimal(number_of_results),
suggestions=suggestion_urls,
answers=result_container.answers,
corrections=correction_urls,
infoboxes=result_container.infoboxes,
engine_data=result_container.engine_data,
paging=result_container.paging,
unresponsive_engines=__get_translated_errors(
result_container.unresponsive_engines),
current_language=match_language(
search_query.lang,
LANGUAGE_CODES,
fallback=request.preferences.get_value("language")),
theme=get_current_theme_name(),
favicons=global_favicons[themes.index(
get_current_theme_name())],
timeout_limit=request.form.get('timeout_limit', None))
