[enh] Add command line engines: git grep, find, etc. (#2128)

A new "base" engine called command is introduced. It is the foundation for all command line engines for now.
You can use this engine to create your own command line engine.

Add some engines (commented out to make sure no one enables anything accidentally):
* git grep: This engine lets you grep in the searx repo.
* locate: If locate is installed and initialized, you can search on the FS.
* find: You can find files with a specific name from where you started searx.
* pattern search in files: This engine utilizes the command fgrep.
* regex search in files: This engine runs `grep` to find a file based on its contents.
This commit is contained in:
Noémi Ványi 2020-09-08 09:51:53 +02:00 committed by GitHub
parent 3397382754
commit f0ca1c3483
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
4 changed files with 497 additions and 1 deletions

184
searx/engines/command.py Normal file
View file

@ -0,0 +1,184 @@
'''
searx is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
searx is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with searx. If not, see < http://www.gnu.org/licenses/ >.
'''
from os.path import expanduser, isabs, realpath, commonprefix
from re import MULTILINE, search as re_search
from shlex import split as shlex_split
from subprocess import Popen, PIPE
from time import time
from threading import Thread
from searx import logger
offline = True
paging = True
command = []
delimiter = {}
parse_regex = {}
query_type = ''
query_enum = []
environment_variables = {}
working_dir = realpath('.')
result_separator = '\n'
result_template = 'key-value.html'
timeout = 4.0
_command_logger = logger.getChild('command')
_compiled_parse_regex = {}
def init(engine_settings):
check_parsing_options(engine_settings)
if 'command' not in engine_settings:
raise ValueError('engine command : missing configuration key: command')
global command, working_dir, result_template, delimiter, parse_regex, timeout, environment_variables
command = engine_settings['command']
if 'working_dir' in engine_settings:
working_dir = engine_settings['working_dir']
if not isabs(engine_settings['working_dir']):
working_dir = realpath(working_dir)
if 'parse_regex' in engine_settings:
parse_regex = engine_settings['parse_regex']
for result_key, regex in parse_regex.items():
_compiled_parse_regex[result_key] = re.compile(regex, flags=MULTILINE)
if 'delimiter' in engine_settings:
delimiter = engine_settings['delimiter']
if 'environment_variables' in engine_settings:
environment_variables = engine_settings['environment_variables']
def search(query, params):
cmd = _get_command_to_run(query)
if not cmd:
return []
results = []
reader_thread = Thread(target=_get_results_from_process, args=(results, cmd, params['pageno']))
reader_thread.start()
reader_thread.join(timeout=timeout)
return results
def _get_command_to_run(query):
params = shlex_split(query.decode('utf-8'))
__check_query_params(params)
cmd = []
for c in command:
if c == '{{QUERY}}':
cmd.extend(params)
else:
cmd.append(c)
return cmd
def _get_results_from_process(results, cmd, pageno):
leftover = ''
count = 0
start, end = __get_results_limits(pageno)
with Popen(cmd, stdout=PIPE, stderr=PIPE, env=environment_variables) as process:
line = process.stdout.readline()
while line:
buf = leftover + line.decode('utf-8')
raw_results = buf.split(result_separator)
if raw_results[-1]:
leftover = raw_results[-1]
raw_results = raw_results[:-1]
for raw_result in raw_results:
result = __parse_single_result(raw_result)
if result is None:
_command_logger.debug('skipped result:', raw_result)
continue
if start <= count and count <= end:
result['template'] = result_template
results.append(result)
count += 1
if end < count:
return results
line = process.stdout.readline()
return_code = process.wait(timeout=timeout)
if return_code != 0:
raise RuntimeError('non-zero return code when running command', cmd, return_code)
def __get_results_limits(pageno):
start = (pageno - 1) * 10
end = start + 9
return start, end
def __check_query_params(params):
if not query_type:
return
if query_type == 'path':
query_path = params[-1]
query_path = expanduser(query_path)
if commonprefix([realpath(query_path), working_dir]) != working_dir:
raise ValueError('requested path is outside of configured working directory')
elif query_type == 'enum' and len(query_enum) > 0:
for param in params:
if param not in query_enum:
raise ValueError('submitted query params is not allowed', param, 'allowed params:', query_enum)
def check_parsing_options(engine_settings):
""" Checks if delimiter based parsing or regex parsing is configured correctly """
if 'delimiter' not in engine_settings and 'parse_regex' not in engine_settings:
raise ValueError('failed to init settings for parsing lines: missing delimiter or parse_regex')
if 'delimiter' in engine_settings and 'parse_regex' in engine_settings:
raise ValueError('failed to init settings for parsing lines: too many settings')
if 'delimiter' in engine_settings:
if 'chars' not in engine_settings['delimiter'] or 'keys' not in engine_settings['delimiter']:
raise ValueError
def __parse_single_result(raw_result):
""" Parses command line output based on configuration """
result = {}
if delimiter:
elements = raw_result.split(delimiter['chars'], maxsplit=len(delimiter['keys']) - 1)
if len(elements) != len(delimiter['keys']):
return {}
for i in range(len(elements)):
result[delimiter['keys'][i]] = elements[i]
if parse_regex:
for result_key, regex in _compiled_parse_regex.items():
found = regex.search(raw_result)
if not found:
return {}
result[result_key] = raw_result[found.start():found.end()]
return result

View file

@ -920,6 +920,77 @@ engines:
# shortcut : uw # shortcut : uw
# base_url : 'http://doc.ubuntu-fr.org' # base_url : 'http://doc.ubuntu-fr.org'
# Be careful when enabling this engine if you are
# running a public instance. Do not expose any sensitive
# information. You can restrict access by configuring a list
# of access tokens under tokens.
# - name: git grep
# engine: command
# command: ['git', 'grep', '{{QUERY}}']
# shortcut: gg
# tokens: []
# disabled: True
# delimiter:
# chars: ':'
# keys: ['filepath', 'code']
# Be careful when enabling this engine if you are
# running a public instance. Do not expose any sensitive
# information. You can restrict access by configuring a list
# of access tokens under tokens.
# - name: locate
# engine: command
# command: ['locate', '{{QUERY}}']
# shortcut: loc
# tokens: []
# disabled: True
# delimiter:
# chars: ' '
# keys: ['line']
# Be careful when enabling this engine if you are
# running a public instance. Do not expose any sensitive
# information. You can restrict access by configuring a list
# of access tokens under tokens.
# - name: find
# engine: command
# command: ['find', '.', '-name', '{{QUERY}}']
# query_type: path
# shortcut: fnd
# tokens: []
# disabled: True
# delimiter:
# chars: ' '
# keys: ['line']
# Be careful when enabling this engine if you are
# running a public instance. Do not expose any sensitive
# information. You can restrict access by configuring a list
# of access tokens under tokens.
# - name: pattern search in files
# engine: command
# command: ['fgrep', '{{QUERY}}']
# shortcut: fgr
# tokens: []
# disabled: True
# delimiter:
# chars: ' '
# keys: ['line']
# Be careful when enabling this engine if you are
# running a public instance. Do not expose any sensitive
# information. You can restrict access by configuring a list
# of access tokens under tokens.
# - name: regex search in files
# engine: command
# command: ['grep', '{{QUERY}}']
# shortcut: gr
# tokens: []
# disabled: True
# delimiter:
# chars: ' '
# keys: ['line']
locales: locales:
en : English en : English
ar : العَرَبِيَّة (Arabic) ar : العَرَبِيَّة (Arabic)

View file

@ -6,7 +6,7 @@
{% continue %} {% continue %}
{% endif %} {% endif %}
<tr> <tr>
<td><b>{{ key|upper }}</b>: {{ value }}</td> <td><b>{{ key|upper }}</b>: {{ value|truncate }}</td>
</tr> </tr>
{% endfor %} {% endfor %}
</table> </table>

View file

@ -0,0 +1,241 @@
'''
searx is free software: you can redistribute it and/or modify
it under the terms of the GNU Affero General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
searx is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU Affero General Public License for more details.
You should have received a copy of the GNU Affero General Public License
along with searx. If not, see < http://www.gnu.org/licenses/ >.
'''
from sys import version_info
from searx.engines import command as command_engine
from searx.testing import SearxTestCase
class TestCommandEngine(SearxTestCase):
def test_basic_seq_command_engine(self):
ls_engine = command_engine
ls_engine.command = ['seq', '{{QUERY}}']
ls_engine.delimiter = {'chars': ' ', 'keys': ['number']}
expected_results = [
{'number': '1', 'template': 'key-value.html'},
{'number': '2', 'template': 'key-value.html'},
{'number': '3', 'template': 'key-value.html'},
{'number': '4', 'template': 'key-value.html'},
{'number': '5', 'template': 'key-value.html'},
]
results = ls_engine.search('5'.encode('utf-8'), {'pageno': 1})
self.assertEqual(results, expected_results)
def test_delimiter_parsing_command_engine(self):
searx_logs = '''DEBUG:searx.webapp:static directory is /home/n/p/searx/searx/static
DEBUG:searx.webapp:templates directory is /home/n/p/searx/searx/templates
DEBUG:searx.engines:soundcloud engine: Starting background initialization
DEBUG:searx.engines:wolframalpha engine: Starting background initialization
DEBUG:searx.engines:locate engine: Starting background initialization
DEBUG:searx.engines:regex search in files engine: Starting background initialization
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): www.wolframalpha.com
DEBUG:urllib3.connectionpool:Starting new HTTPS connection (1): soundcloud.com
DEBUG:searx.engines:find engine: Starting background initialization
DEBUG:searx.engines:pattern search in files engine: Starting background initialization
DEBUG:searx.webapp:starting webserver on 127.0.0.1:8888
WARNING:werkzeug: * Debugger is active!
INFO:werkzeug: * Debugger PIN: 299-578-362'''
echo_engine = command_engine
echo_engine.command = ['echo', searx_logs]
echo_engine.delimiter = {'chars': ':', 'keys': ['level', 'component', 'message']}
expected_results_by_page = [
[
{
'component': 'searx.webapp',
'message': 'static directory is /home/n/p/searx/searx/static',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'searx.webapp',
'message': 'templates directory is /home/n/p/searx/searx/templates',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'searx.engines',
'message': 'soundcloud engine: Starting background initialization',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'searx.engines',
'message': 'wolframalpha engine: Starting background initialization',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'searx.engines',
'message': 'locate engine: Starting background initialization',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'searx.engines',
'message': 'regex search in files engine: Starting background initialization',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'urllib3.connectionpool',
'message': 'Starting new HTTPS connection (1): www.wolframalpha.com',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'urllib3.connectionpool',
'message': 'Starting new HTTPS connection (1): soundcloud.com',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'searx.engines',
'message': 'find engine: Starting background initialization',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'searx.engines',
'message': 'pattern search in files engine: Starting background initialization',
'template': 'key-value.html',
'level': 'DEBUG',
},
],
[
{
'component': 'searx.webapp',
'message': 'starting webserver on 127.0.0.1:8888',
'template': 'key-value.html',
'level': 'DEBUG',
},
{
'component': 'werkzeug',
'message': ' * Debugger is active!',
'template': 'key-value.html',
'level': 'WARNING',
},
{
'component': 'werkzeug',
'message': ' * Debugger PIN: 299-578-362',
'template': 'key-value.html',
'level': 'INFO',
},
],
]
for i in [0, 1]:
results = echo_engine.search(''.encode('utf-8'), {'pageno': i + 1})
self.assertEqual(results, expected_results_by_page[i])
def test_regex_parsing_command_engine(self):
txt = '''commit 35f9a8c81d162a361b826bbcd4a1081a4fbe76a7
Author: Noémi Ványi <sitbackandwait@gmail.com>
Date: Tue Oct 15 11:31:33 2019 +0200
first interesting message
commit 6c3c206316153ccc422755512bceaa9ab0b14faa
Author: Noémi Ványi <sitbackandwait@gmail.com>
Date: Mon Oct 14 17:10:08 2019 +0200
second interesting message
commit d8594d2689b4d5e0d2f80250223886c3a1805ef5
Author: Noémi Ványi <sitbackandwait@gmail.com>
Date: Mon Oct 14 14:45:05 2019 +0200
third interesting message
commit '''
git_log_engine = command_engine
git_log_engine.command = ['echo', txt]
git_log_engine.result_separator = '\n\ncommit '
git_log_engine.delimiter = {}
git_log_engine.parse_regex = {
'commit': '\w{40}',
'author': '[\w* ]* <\w*@?\w*\.?\w*>',
'date': 'Date: .*',
'message': '\n\n.*$'
}
expected_results = [
{
'commit': '35f9a8c81d162a361b826bbcd4a1081a4fbe76a7',
'author': ' Noémi Ványi <sitbackandwait@gmail.com>',
'date': 'Date: Tue Oct 15 11:31:33 2019 +0200',
'message': '\n\nfirst interesting message',
'template': 'key-value.html',
},
{
'commit': '6c3c206316153ccc422755512bceaa9ab0b14faa',
'author': ' Noémi Ványi <sitbackandwait@gmail.com>',
'date': 'Date: Mon Oct 14 17:10:08 2019 +0200',
'message': '\n\nsecond interesting message',
'template': 'key-value.html',
},
{
'commit': 'd8594d2689b4d5e0d2f80250223886c3a1805ef5',
'author': ' Noémi Ványi <sitbackandwait@gmail.com>',
'date': 'Date: Mon Oct 14 14:45:05 2019 +0200',
'message': '\n\nthird interesting message',
'template': 'key-value.html',
},
]
results = git_log_engine.search(''.encode('utf-8'), {'pageno': 1})
self.assertEqual(results, expected_results)
def test_working_dir_path_query(self):
ls_engine = command_engine
ls_engine.command = ['ls', '{{QUERY}}']
ls_engine.result_separator = '\n'
ls_engine.delimiter = {'chars': ' ', 'keys': ['file']}
ls_engine.query_type = 'path'
results = ls_engine.search('.'.encode(), {'pageno': 1})
self.assertTrue(len(results) != 0)
forbidden_paths = [
'..',
'../..',
'./..',
'~',
'/var',
]
for forbidden_path in forbidden_paths:
self.assertRaises(ValueError, ls_engine.search, '..'.encode(), {'pageno': 1})
def test_enum_queries(self):
echo_engine = command_engine
echo_engine.command = ['echo', '{{QUERY}}']
echo_engine.query_type = 'enum'
echo_engine.query_enum = ['i-am-allowed-to-say-this', 'and-that']
for allowed in echo_engine.query_enum:
results = echo_engine.search(allowed.encode(), {'pageno': 1})
self.assertTrue(len(results) != 0)
forbidden_queries = [
'forbidden',
'banned',
'prohibited',
]
for forbidden in forbidden_queries:
self.assertRaises(ValueError, echo_engine.search, forbidden.encode(), {'pageno': 1})