Limnoria/plugins/Google/plugin.py
Valentin Lorentz 833af36b08 all plugins: auto-generate READMEs
Starting with this commit, there should be no "original" text
in READMEs.
Instead, the entire text should be in docstrings, and READMEs
should be automatically generated from these docstrings.

Motivation:

* The same help is available both via IRC and in the README
  (although the README can be more detailed, as only the first
  paragraph will be shown on IRC)
* This will allow auto-generating plugin help on docs.limnoria.net
  using the same content as the README, as it's sourced from the
  docstrings

Additionally, this converts all READMEs from Markdown to ReST,
because the documentation already uses ReST, and other docstrings
in the codebase are in ReST for this reason.
2021-04-05 17:42:31 +02:00

402 lines
15 KiB
Python

###
# Copyright (c) 2002-2004, Jeremiah Fincher
# Copyright (c) 2008-2010, James McCoy
# All rights reserved.
#
# Redistribution and use in source and binary forms, with or without
# modification, are permitted provided that the following conditions are met:
#
# * Redistributions of source code must retain the above copyright notice,
# this list of conditions, and the following disclaimer.
# * Redistributions in binary form must reproduce the above copyright notice,
# this list of conditions, and the following disclaimer in the
# documentation and/or other materials provided with the distribution.
# * Neither the name of the author of this software nor the name of
# contributors to this software may be used to endorse or promote products
# derived from this software without specific prior written consent.
#
# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
# ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
# LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
# CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
# SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
# INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
# CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
# ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
# POSSIBILITY OF SUCH DAMAGE.
###
import re
import sys
import json
import supybot.conf as conf
import supybot.utils as utils
import supybot.world as world
from supybot.commands import *
import supybot.utils.minisix as minisix
import supybot.ircmsgs as ircmsgs
import supybot.ircutils as ircutils
import supybot.callbacks as callbacks
from supybot.i18n import PluginInternationalization, internationalizeDocstring
_ = PluginInternationalization('Google')
from .parser import GoogleHTMLParser
class Google(callbacks.PluginRegexp):
"""
This is a simple plugin to provide access to the Google services we
all know and love from our favorite IRC bot.
1. google
Searches for a string and gives you 3 results from Google search
``!google something``
2. lucky
Return the first result (Google's "I'm Feeling Lucky" search)
``!lucky something``
3. calc
Does mathematic calculations
``!calc 5+4``
4. translate
Translates a string
``!translate en ar test``
Check: `Supported language codes`_
.. _Supported language codes: <https://cloud.google.com/translate/v2/using_rest#language-params>`
"""
threaded = True
callBefore = ['Web']
regexps = ['googleSnarfer']
_colorGoogles = {}
def _getColorGoogle(self, m):
s = m.group(1)
ret = self._colorGoogles.get(s)
if not ret:
L = list(s)
L[0] = ircutils.mircColor(L[0], 'blue')[:-1]
L[1] = ircutils.mircColor(L[1], 'red')[:-1]
L[2] = ircutils.mircColor(L[2], 'yellow')[:-1]
L[3] = ircutils.mircColor(L[3], 'blue')[:-1]
L[4] = ircutils.mircColor(L[4], 'green')[:-1]
L[5] = ircutils.mircColor(L[5], 'red')
ret = ''.join(L)
self._colorGoogles[s] = ret
return ircutils.bold(ret)
_googleRe = re.compile(r'\b(google)\b', re.I)
def outFilter(self, irc, msg):
if msg.command == 'PRIVMSG' and \
self.registryValue('colorfulFilter', msg.channel, irc.network):
s = msg.args[1]
s = re.sub(self._googleRe, self._getColorGoogle, s)
msg = ircmsgs.privmsg(msg.args[0], s, msg=msg)
return msg
@classmethod
def decode(cls, text):
parser = GoogleHTMLParser()
parser.feed(text)
return parser.results
_gsearchUrl = 'https://www.google.com/search'
def search(self, query, channel, network, options={}):
"""search("search phrase", options={})
Valid options are:
smallsearch - True/False (Default: False)
filter - {active,moderate,off} (Default: "moderate")
language - Restrict search to documents in the given language
(Default: "lang_en")
"""
self.log.warning('The Google plugin search is deprecated since '
'Google closed their public API and will be removed in a '
'future release. Please consider switching to an other '
'plugin for your searches, like '
'<https://github.com/Hoaas/Supybot-plugins/tree/master/DuckDuckGo>, '
'<https://github.com/joulez/GoogleCSE>, or '
'<https://github.com/GLolol/SupyPlugins/tree/master/DDG>.')
ref = self.registryValue('referer')
if not ref:
ref = 'http://%s/%s' % (dynamic.irc.server,
dynamic.irc.nick)
headers = dict(utils.web.defaultHeaders)
headers['Referer'] = ref
headers['User-agent'] = 'Mozilla/5.0 (compatible; utils.web python module)'
opts = {'q': query, 'gbv': '2'}
for (k, v) in options.items():
if k == 'smallsearch':
if v:
opts['rsz'] = 'small'
else:
opts['rsz'] = 'large'
elif k == 'filter':
opts['safe'] = v
elif k == 'language':
opts['hl'] = v
defLang = self.registryValue('defaultLanguage', channel, network)
if 'hl' not in opts and defLang:
opts['hl'] = defLang.strip('lang_')
if 'safe' not in opts:
opts['safe'] = self.registryValue('searchFilter', channel, network)
if 'rsz' not in opts:
opts['rsz'] = 'large'
text = utils.web.getUrl('%s?%s' % (self._gsearchUrl,
utils.web.urlencode(opts)),
headers=headers).decode('utf8')
return text
def formatData(self, data, bold=True, max=0, onetoone=False):
data = self.decode(data)
results = []
if max:
data = data[:max]
for result in data:
title = utils.web.htmlToText(result.title.encode('utf-8'))
url = result.link
if minisix.PY2:
url = url.encode('utf-8')
if title:
if bold:
title = ircutils.bold(title)
results.append(format('%s: %u', title, url))
else:
results.append(url)
if minisix.PY2:
repl = lambda x:x if isinstance(x, unicode) else unicode(x, 'utf8')
results = list(map(repl, results))
if not results:
return [_('No matches found.')]
elif onetoone:
return results
else:
return [minisix.u('; ').join(results)]
@internationalizeDocstring
def lucky(self, irc, msg, args, opts, text):
"""[--snippet] <search>
Does a google search, but only returns the first result.
If option --snippet is given, returns also the page text snippet.
"""
opts = dict(opts)
data = self.search(text, msg.channel, irc.network,
{'smallsearch': True})
data = self.decode(data)
if data:
url = data[0].link
if 'snippet' in opts:
snippet = data[0].snippet
snippet = " | " + utils.web.htmlToText(snippet, tagReplace='')
else:
snippet = ""
result = url + snippet
irc.reply(result)
else:
irc.reply(_('Google found nothing.'))
lucky = wrap(lucky, [getopts({'snippet':'',}), 'text'])
@internationalizeDocstring
def google(self, irc, msg, args, optlist, text):
"""<search> [--{filter,language} <value>]
Searches google.com for the given string. As many results as can fit
are included. --language accepts a language abbreviation; --filter
accepts a filtering level ('active', 'moderate', 'off').
"""
if 'language' in optlist and optlist['language'].lower() not in \
conf.supybot.plugins.Google.safesearch.validStrings:
irc.errorInvalid('language')
data = self.search(text, msg.channel, irc.network, dict(optlist))
bold = self.registryValue('bold', msg.channel, irc.network)
max = self.registryValue('maximumResults', msg.channel, irc.network)
# We don't use supybot.reply.oneToOne here, because you generally
# do not want @google to echo ~20 lines of results, even if you
# have reply.oneToOne enabled.
onetoone = self.registryValue('oneToOne', msg.channel, irc.network)
for result in self.formatData(data,
bold=bold, max=max, onetoone=onetoone):
irc.reply(result)
google = wrap(google, [getopts({'language':'something',
'filter':''}),
'text'])
@internationalizeDocstring
def cache(self, irc, msg, args, url):
"""<url>
Returns a link to the cached version of <url> if it is available.
"""
data = self.search(url, msg.channel, irc.network, {'smallsearch': True})
if data:
m = data[0]
if m['cacheUrl']:
url = m['cacheUrl'].encode('utf-8')
irc.reply(url)
return
irc.error(_('Google seems to have no cache for that site.'))
cache = wrap(cache, ['url'])
_fight_re = re.compile(r'id="resultStats"[^>]*>(?P<stats>[^<]*)')
@internationalizeDocstring
def fight(self, irc, msg, args):
"""<search string> <search string> [<search string> ...]
Returns the results of each search, in order, from greatest number
of results to least.
"""
channel = msg.channel
network = irc.network
results = []
for arg in args:
text = self.search(arg, channel, network, {'smallsearch': True})
i = text.find('id="resultStats"')
stats = utils.web.htmlToText(self._fight_re.search(text).group('stats'))
if stats == '':
results.append((0, args))
continue
count = ''.join(filter('0123456789'.__contains__, stats))
results.append((int(count), arg))
results.sort()
results.reverse()
if self.registryValue('bold', channel, network):
bold = ircutils.bold
else:
bold = repr
s = ', '.join([format('%s: %i', bold(s), i) for (i, s) in results])
irc.reply(s)
def _translate(self, sourceLang, targetLang, text):
headers = dict(utils.web.defaultHeaders)
headers['User-agent'] = ('Mozilla/5.0 (X11; U; Linux i686) '
'Gecko/20071127 Firefox/2.0.0.11')
sourceLang = utils.web.urlquote(sourceLang)
targetLang = utils.web.urlquote(targetLang)
text = utils.web.urlquote(text)
result = utils.web.getUrlFd('http://translate.googleapis.com/translate_a/single'
'?client=gtx&dt=t&sl=%s&tl=%s&q='
'%s' % (sourceLang, targetLang, text),
headers).read().decode('utf8')
while ',,' in result:
result = result.replace(',,', ',null,')
while '[,' in result:
result = result.replace('[,', '[')
data = json.loads(result)
try:
language = data[2]
except:
language = 'unknown'
if data[0]:
return (''.join(x[0] for x in data[0]), language)
else:
return (_('No translations found.'), language)
@internationalizeDocstring
def translate(self, irc, msg, args, sourceLang, targetLang, text):
"""<source language> [to] <target language> <text>
Returns <text> translated from <source language> into <target
language>. <source language> and <target language> take language
codes (not language names), which are listed here:
https://cloud.google.com/translate/docs/languages
"""
(text, language) = self._translate(sourceLang, targetLang, text)
irc.reply(text, language)
translate = wrap(translate, ['something', 'to', 'something', 'text'])
def googleSnarfer(self, irc, msg, match):
r"^google\s+(.*)$"
if not self.registryValue('searchSnarfer', msg.channel, irc.network):
return
searchString = match.group(1)
data = self.search(searchString, msg.channel, irc.network,
{'smallsearch': True})
if data['responseData']['results']:
url = data['responseData']['results'][0]['unescapedUrl']
irc.reply(url, prefixNick=False)
googleSnarfer = urlSnarfer(googleSnarfer)
def _googleUrl(self, s, channel, network):
s = utils.web.urlquote_plus(s)
url = r'http://%s/search?q=%s' % \
(self.registryValue('baseUrl', channel, network), s)
return url
_calcRe1 = re.compile(r'<span class="cwcot".*?>(.*?)</span>', re.I)
_calcRe2 = re.compile(r'<div class="vk_ans.*?>(.*?)</div>', re.I | re.S)
_calcRe3 = re.compile(r'<div class="side_div" id="rhs_div">.*?<input class="ucw_data".*?value="(.*?)"', re.I)
@internationalizeDocstring
def calc(self, irc, msg, args, expr):
"""<expression>
Uses Google's calculator to calculate the value of <expression>.
"""
url = self._googleUrl(expr, msg.channel, irc.network)
h = {"User-Agent":"Mozilla/5.0 (Windows NT 6.2; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/32.0.1667.0 Safari/537.36"}
html = utils.web.getUrl(url, headers=h).decode('utf8')
match = self._calcRe1.search(html)
if not match:
match = self._calcRe2.search(html)
if not match:
match = self._calcRe3.search(html)
if not match:
irc.reply("I could not find an output from Google Calc for: %s" % expr)
return
else:
s = match.group(1)
else:
s = match.group(1)
else:
s = match.group(1)
# do some cleanup of text
s = re.sub(r'<sup>(.*)</sup>&#8260;<sub>(.*)</sub>', r' \1/\2', s)
s = re.sub(r'<sup>(.*)</sup>', r'^\1', s)
s = utils.web.htmlToText(s)
irc.reply("%s = %s" % (expr, s))
calc = wrap(calc, ['text'])
_phoneRe = re.compile(r'Phonebook.*?<font size=-1>(.*?)<a href')
@internationalizeDocstring
def phonebook(self, irc, msg, args, phonenumber):
"""<phone number>
Looks <phone number> up on Google.
"""
url = self._googleUrl(phonenumber, msg.channel, irc.network)
html = utils.web.getUrl(url).decode('utf8')
m = self._phoneRe.search(html)
if m is not None:
s = m.group(1)
s = s.replace('<b>', '')
s = s.replace('</b>', '')
s = utils.web.htmlToText(s)
irc.reply(s)
else:
irc.reply(_('Google\'s phonebook didn\'t come up with anything.'))
phonebook = wrap(phonebook, ['text'])
Class = Google
# vim:set shiftwidth=4 softtabstop=4 expandtab textwidth=79: