nemunaire/nemubot
1
0
Fork 0
Code Releases Activity

New module mediawiki

Browse source
This commit is contained in:
nemunaire 2014-09-02 21:19:08 +02:00
commit 7387fabee1
3 changed files with 92 additions and 107 deletions
Download patch file Download diff file Expand all files Collapse all files

56
modules/ddg/Wikipedia.py
View file

@ -1,56 +0,0 @@
# coding=utf-8
import re
from urllib.parse import quote
import urllib.request
import xmlparser
class Wikipedia:
def __init__(self, terms, lang="fr", site="wikipedia.org", section=0):
self.terms = terms
self.lang = lang
self.curRT = section
raw = urllib.request.urlopen(urllib.request.Request("http://" + self.lang + "." + site + "/w/api.php?format=xml&redirects&action=query&prop=revisions&rvprop=content&titles=%s" % (quote(terms)), headers={"User-agent": "Nemubot v3"}))
self.wres = xmlparser.parse_string(raw.read())
if self.wres is None or not (self.wres.hasNode("query") and self.wres.getFirstNode("query").hasNode("pages") and self.wres.getFirstNode("query").getFirstNode("pages").hasNode("page") and self.wres.getFirstNode("query").getFirstNode("pages").getFirstNode("page").hasNode("revisions")):
self.wres = None
else:
self.wres = self.wres.getFirstNode("query").getFirstNode("pages").getFirstNode("page").getFirstNode("revisions").getFirstNode("rev").getContent()
self.wres = striplink(self.wres)
@property
def nextRes(self):
if self.wres is not None:
for cnt in self.wres.split("\n"):
if self.curRT > 0:
self.curRT -= 1
continue
(c, u) = RGXP_s.subn(' ', cnt)
c = c.strip()
if c != "":
yield c
RGXP_p = re.compile(r"(<!--.*-->|<ref[^>]*/>|<ref[^>]*>[^>]*</ref>|<dfn[^>]*>[^>]*</dfn>|\{\{[^{}]*\}\}|\[\[([^\[\]]*\[\[[^\]\[]*\]\])+[^\[\]]*\]\]|\{\{([^{}]*\{\{[^{}]*\}\}[^{}]*)+\}\}|\{\{([^{}]*\{\{([^{}]*\{\{[^{}]*\}\}[^{}]*)+\}\}[^{}]*)+\}\}|\[\[[^\]|]+(\|[^\]\|]+)*\]\])|#\* ''" + "\n", re.I)
RGXP_l = re.compile(r'\{\{(nobr|lang\|[^|}]+)\|([^}]+)\}\}', re.I)
RGXP_m = re.compile(r'\{\{pron\|([^|}]+)\|[^}]+\}\}', re.I)
RGXP_t = re.compile("==+ *([^=]+) *=+=\n+([^\n])", re.I)
RGXP_q = re.compile(r'\[\[([^\[\]|]+)\|([^\]|]+)]]', re.I)
RGXP_r = re.compile(r'\[\[([^\[\]|]+)\]\]', re.I)
RGXP_s = re.compile(r'\s+')
def striplink(s):
s.replace("{{m}}", "masculin").replace("{{f}}", "feminin").replace("{{n}}", "neutre")
(s, n) = RGXP_m.subn(r"[\1]", s)
(s, n) = RGXP_l.subn(r"\2", s)
(s, n) = RGXP_q.subn(r"\1", s)
(s, n) = RGXP_r.subn(r"\1", s)
(s, n) = RGXP_p.subn('', s)
if s == "": return s
(s, n) = RGXP_t.subn("\x03\x16" + r"\1" + " :\x03\x16 " + r"\2", s)
return s.replace("'''", "\x03\x02").replace("''", "\x03\x1f")

51
modules/ddg/__init__.py
View file

@ -11,7 +11,6 @@ nemubotversion = 3.4
from . import DDGSearch
from . import UrbanDictionnary
from . import WFASearch
from . import Wikipedia
def load(context):
global CONF
@ -98,53 +97,3 @@ def calculate(msg):
return res
else:
return Response(msg.sender, s.error, msg.channel)
@hook("cmd_hook", "wikipedia")
def wikipedia(msg):
return wiki("wikipedia.org", 0, msg)
@hook("cmd_hook", "wiktionary")
def wiktionary(msg):
return wiki("wiktionary.org", 1, msg)
@hook("cmd_hook", "etymology")
def wiktionary(msg):
return wiki("wiktionary.org", 0, msg)
def wiki(site, section, msg):
if len(msg.cmds) <= 1:
return Response(msg.sender,
"Indicate a term to search",
msg.channel, nick=msg.nick)
if len(msg.cmds) > 2 and len(msg.cmds[1]) < 4:
lang = msg.cmds[1]
extract = 2
else:
lang = "fr"
extract = 1
s = Wikipedia.Wikipedia(' '.join(msg.cmds[extract:]), lang, site, section)
res = Response(msg.sender, channel=msg.channel, nomore="No more results")
if site == "wiktionary.org":
tout = [result for result in s.nextRes if result.find("\x03\x16 :\x03\x16 ") != 0]
if len(tout) > 0:
defI=1
for t in tout:
if t.find("# ") == 0:
t = t.replace("# ", "%d. " % defI)
defI += 1
elif t.find("#* ") == 0:
t = t.replace("#* ", " * ")
res.append_message(t)
else:
for result in s.nextRes:
res.append_message(result)
if len(res.messages) > 0:
return res
else:
return Response(msg.sender,
"No information about " + " ".join(msg.cmds[extract:]),
msg.channel)