searxng/searx/utils.py

253 lines
7.2 KiB
Python
Raw Normal View History

2014-10-19 10:41:04 +00:00
# import htmlentitydefs
2015-01-11 12:26:40 +00:00
import locale
import dateutil.parser
import cStringIO
import csv
import os
import re
2014-01-20 01:31:20 +00:00
from codecs import getincrementalencoder
2014-04-24 23:46:40 +00:00
from HTMLParser import HTMLParser
from random import choice
2014-11-18 10:37:42 +00:00
from searx.version import VERSION_STRING
from searx import settings
2015-01-11 12:26:40 +00:00
from searx import logger
2014-11-18 10:37:42 +00:00
2015-01-11 12:26:40 +00:00
logger = logger.getChild('utils')
2014-01-10 22:38:08 +00:00
2015-01-27 18:58:39 +00:00
ua_versions = ('31.0',
2014-11-22 17:37:42 +00:00
'32.0',
2015-01-27 18:58:39 +00:00
'33.0',
'34.0',
'35.0')
2014-11-22 17:37:42 +00:00
2014-05-20 14:55:49 +00:00
ua_os = ('Windows NT 6.3; WOW64',
'X11; Linux x86_64',
'X11; Linux x86')
2014-11-22 17:37:42 +00:00
ua = "Mozilla/5.0 ({os}) Gecko/20100101 Firefox/{version}"
blocked_tags = ('script',
'style')
2014-03-04 18:26:09 +00:00
2014-01-18 20:53:59 +00:00
def gen_useragent():
2014-01-12 19:13:14 +00:00
# TODO
return ua.format(os=choice(ua_os), version=choice(ua_versions))
2014-01-12 19:13:14 +00:00
2014-10-17 10:34:51 +00:00
def searx_useragent():
return 'searx/{searx_version} {suffix}'.format(
searx_version=VERSION_STRING,
suffix=settings['server'].get('useragent_suffix', ''))
2014-10-19 10:41:04 +00:00
2014-01-10 22:38:08 +00:00
def highlight_content(content, query):
if not content:
return None
# ignoring html contents
# TODO better html content detection
if content.find('<') != -1:
return content
query = query.decode('utf-8')
if content.lower().find(query.lower()) > -1:
query_regex = u'({0})'.format(re.escape(query))
2014-05-16 14:51:23 +00:00
content = re.sub(query_regex, '<span class="highlight">\\1</span>',
content, flags=re.I | re.U)
2014-01-10 22:38:08 +00:00
else:
regex_parts = []
for chunk in query.split():
if len(chunk) == 1:
regex_parts.append(u'\W+{0}\W+'.format(re.escape(chunk)))
else:
regex_parts.append(u'{0}'.format(re.escape(chunk)))
query_regex = u'({0})'.format('|'.join(regex_parts))
2014-05-16 14:51:23 +00:00
content = re.sub(query_regex, '<span class="highlight">\\1</span>',
content, flags=re.I | re.U)
2014-01-10 22:38:08 +00:00
return content
2013-11-08 22:44:26 +00:00
2013-11-08 22:44:26 +00:00
class HTMLTextExtractor(HTMLParser):
def __init__(self):
HTMLParser.__init__(self)
self.result = []
self.tags = []
def handle_starttag(self, tag, attrs):
self.tags.append(tag)
def handle_endtag(self, tag):
if not self.tags:
return
if tag != self.tags[-1]:
raise Exception("invalid html")
self.tags.pop()
def is_valid_tag(self):
return not self.tags or self.tags[-1] not in blocked_tags
2013-11-08 22:44:26 +00:00
def handle_data(self, d):
if not self.is_valid_tag():
return
2013-11-08 22:44:26 +00:00
self.result.append(d)
def handle_charref(self, number):
if not self.is_valid_tag():
return
2014-01-20 01:31:20 +00:00
if number[0] in (u'x', u'X'):
codepoint = int(number[1:], 16)
else:
codepoint = int(number)
2013-11-08 22:44:26 +00:00
self.result.append(unichr(codepoint))
def handle_entityref(self, name):
if not self.is_valid_tag():
return
2014-10-19 10:41:04 +00:00
# codepoint = htmlentitydefs.name2codepoint[name]
# self.result.append(unichr(codepoint))
2013-11-18 15:47:20 +00:00
self.result.append(name)
2013-11-08 22:44:26 +00:00
def get_text(self):
2015-01-27 19:03:33 +00:00
return u''.join(self.result).strip()
2013-11-08 22:44:26 +00:00
2013-11-08 22:44:26 +00:00
def html_to_text(html):
html = html.replace('\n', ' ')
html = ' '.join(html.split())
2013-11-08 22:44:26 +00:00
s = HTMLTextExtractor()
s.feed(html)
return s.get_text()
2013-11-15 17:55:18 +00:00
class UnicodeWriter:
"""
A CSV writer which will write rows to CSV file "f",
which is encoded in the given encoding.
"""
def __init__(self, f, dialect=csv.excel, encoding="utf-8", **kwds):
# Redirect output to a queue
self.queue = cStringIO.StringIO()
self.writer = csv.writer(self.queue, dialect=dialect, **kwds)
self.stream = f
2014-01-20 01:31:20 +00:00
self.encoder = getincrementalencoder(encoding)()
2013-11-15 17:55:18 +00:00
def writerow(self, row):
2014-01-20 01:31:20 +00:00
unicode_row = []
for col in row:
if type(col) == str or type(col) == unicode:
unicode_row.append(col.encode('utf-8').strip())
else:
unicode_row.append(col)
self.writer.writerow(unicode_row)
2013-11-15 17:55:18 +00:00
# Fetch UTF-8 output from the queue ...
data = self.queue.getvalue()
data = data.decode("utf-8")
# ... and reencode it into the target encoding
data = self.encoder.encode(data)
# write to the target stream
self.stream.write(data)
# empty queue
self.queue.truncate(0)
def writerows(self, rows):
for row in rows:
self.writerow(row)
2014-04-24 23:46:40 +00:00
def get_themes(root):
"""Returns available themes list."""
static_path = os.path.join(root, 'static')
templates_path = os.path.join(root, 'templates')
2015-01-01 16:48:12 +00:00
themes = os.listdir(os.path.join(static_path, 'themes'))
2014-04-24 23:46:40 +00:00
return static_path, templates_path, themes
2015-01-01 16:48:12 +00:00
def get_static_files(base_path):
2015-01-01 17:59:53 +00:00
base_path = os.path.join(base_path, 'static')
2015-01-01 16:48:12 +00:00
static_files = set()
2015-01-01 17:59:53 +00:00
base_path_length = len(base_path) + 1
for directory, _, files in os.walk(base_path):
2015-01-01 16:48:12 +00:00
for filename in files:
f = os.path.join(directory[base_path_length:], filename)
static_files.add(f)
return static_files
2015-01-01 17:59:53 +00:00
def get_result_templates(base_path):
base_path = os.path.join(base_path, 'templates')
result_templates = set()
base_path_length = len(base_path) + 1
for directory, _, files in os.walk(base_path):
if directory.endswith('result_templates'):
for filename in files:
f = os.path.join(directory[base_path_length:], filename)
result_templates.add(f)
return result_templates
2015-01-11 12:26:40 +00:00
def format_date_by_locale(date_string, locale_string):
# strftime works only on dates after 1900
parsed_date = dateutil.parser.parse(date_string)
if parsed_date.year <= 1900:
return parsed_date.isoformat().split('T')[0]
orig_locale = locale.getlocale()[0]
try:
locale.setlocale(locale.LC_ALL, locale_string)
except:
logger.warning('cannot set locale: {0}'.format(locale_string))
formatted_date = parsed_date.strftime(locale.nl_langinfo(locale.D_FMT))
try:
locale.setlocale(locale.LC_ALL, orig_locale)
except:
logger.warning('cannot set original locale: {0}'.format(orig_locale))
return formatted_date
def dict_subset(d, properties):
result = {}
for k in properties:
if k in d:
result[k] = d[k]
return result
2015-01-29 18:44:52 +00:00
def prettify_url(url):
if len(url) > 74:
return u'{0}[...]{1}'.format(url[:35], url[-35:])
else:
return url
def get_blocked_engines(engines, cookies):
if 'blocked_engines' not in cookies:
2015-02-03 17:48:52 +00:00
return [(engine_name, category) for engine_name in engines
for category in engines[engine_name].categories if engines[engine_name].disabled]
blocked_engine_strings = cookies.get('blocked_engines', '').split(',')
blocked_engines = []
if not blocked_engine_strings:
return blocked_engines
for engine_string in blocked_engine_strings:
if engine_string.find('__') > -1:
engine, category = engine_string.split('__', 1)
if engine in engines and category in engines[engine].categories:
blocked_engines.append((engine, category))
elif engine_string in engines:
for category in engines[engine_string].categories:
blocked_engines.append((engine_string, category))
return blocked_engines