### # Copyright (c) 2002-2005, Jeremiah Fincher # Copyright (c) 2008, 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 gc import os import csv import time import codecs import string import fnmatch import os.path import threading import collections.abc from .. import callbacks, conf, dbi, ircdb, ircutils, i18n, log, utils, world from ..commands import * _ = i18n.PluginInternationalization() class NoSuitableDatabase(Exception): def __init__(self, suitable): self.suitable = list(suitable) self.suitable.sort() def __str__(self): return format('No suitable databases were found. Suitable databases ' 'include %L. If you have one of these databases ' 'installed, make sure it is listed in the ' 'supybot.databases configuration variable.', self.suitable) def DB(filename, types): # We don't care if any of the DBs are actually available when # documenting, so just fake that we found something suitable if world.documenting: def junk(*args, **kwargs): pass return junk def MakeDB(*args, **kwargs): for type in conf.supybot.databases(): # Can't do this because Python sucks. Go ahead, try it! # filename = '.'.join([filename, type, 'db']) fn = '.'.join([filename, type, 'db']) fn = utils.file.sanitizeName(fn) path = conf.supybot.directories.data.dirize(fn) try: return types[type](path, *args, **kwargs) except KeyError: continue raise NoSuitableDatabase(types.keys()) return MakeDB def makeChannelFilename(filename, channel=None, dirname=None): assert channel is not None, 'Channel should not be None' filename = os.path.basename(filename) channelSpecific = conf.supybot.databases.plugins.channelSpecific channel = channelSpecific.getChannelLink(channel) channel = utils.file.sanitizeName(ircutils.toLower(channel)) if dirname is None: dirname = conf.supybot.directories.data.dirize(channel) if not os.path.exists(dirname): os.makedirs(dirname) return os.path.join(dirname, filename) def getChannel(channel): assert channel is not None, 'Channel should not be None' channelSpecific = conf.supybot.databases.plugins.channelSpecific return channelSpecific.getChannelLink(channel) # XXX This shouldn't be a mixin. This should be contained by classes that # want such behavior. But at this point, it wouldn't gain much for us # to refactor it. # XXX We need to get rid of this, it's ugly and opposed to # database-independence. class ChannelDBHandler(object): """A class to handle database stuff for individual channels transparently. """ suffix = '.db' def __init__(self, suffix='.db'): self.dbCache = ircutils.IrcDict() suffix = self.suffix if self.suffix and self.suffix[0] != '.': suffix = '.' + suffix self.suffix = suffix def makeFilename(self, channel): """Override this to specialize the filenames of your databases.""" channel = ircutils.toLower(channel) className = self.__class__.__name__ return makeChannelFilename(className + self.suffix, channel) def makeDb(self, filename): """Override this to create your databases.""" raise NotImplementedError def getDb(self, channel): """Use this to get a database for a specific channel.""" currentThread = threading.currentThread() if channel not in self.dbCache and currentThread == world.mainThread: self.dbCache[channel] = self.makeDb(self.makeFilename(channel)) if currentThread != world.mainThread: db = self.makeDb(self.makeFilename(channel)) else: db = self.dbCache[channel] db.isolation_level = None return db def die(self): for db in self.dbCache.values(): try: db.commit() except AttributeError: # In case it's not an SQLite database. pass try: db.close() except AttributeError: # In case it doesn't have a close method. pass del db gc.collect() class DbiChannelDB(object): """This just handles some of the general stuff for Channel DBI databases. Check out ChannelIdDatabasePlugin for an example of how to use this.""" def __init__(self, filename): self.filename = filename self.dbs = ircutils.IrcDict() def _getDb(self, channel): filename = makeChannelFilename(self.filename, channel) try: db = self.dbs[channel] except KeyError: db = self.DB(filename) self.dbs[channel] = db return db def close(self): for db in self.dbs.values(): db.close() def flush(self): for db in self.dbs.values(): db.flush() def __getattr__(self, attr): def _getDbAndDispatcher(channel, *args, **kwargs): db = self._getDb(channel) return getattr(db, attr)(*args, **kwargs) return _getDbAndDispatcher class ChannelUserDictionary(collections.abc.MutableMapping): IdDict = dict def __init__(self): self.channels = ircutils.IrcDict() def __getitem__(self, key): (channel, id) = key return self.channels[channel][id] def __setitem__(self, key, v): (channel, id) = key if channel not in self.channels: self.channels[channel] = self.IdDict() self.channels[channel][id] = v def __delitem__(self, key): (channel, id) = key del self.channels[channel][id] def __iter__(self): for channel, ids in self.channels.items(): for id_, value in ids.items(): yield (channel, id_) def __len__(self): return sum([len(x) for x in self.channels]) def items(self): for (channel, ids) in self.channels.items(): for (id, v) in ids.items(): yield ((channel, id), v) def keys(self): L = [] for (k, _) in self.items(): L.append(k) return L # XXX The interface to this needs to be made *much* more like the dbi.DB # interface. This is just too odd and not extensible; any extension # would very much feel like an extension, rather than part of the db # itself. class ChannelUserDB(ChannelUserDictionary): def __init__(self, filename): ChannelUserDictionary.__init__(self) self.filename = filename try: fd = codecs.open(self.filename, encoding='utf8') except EnvironmentError as e: log.warning('Couldn\'t open %s: %s.', self.filename, e) return reader = csv.reader(fd) try: lineno = 0 for t in reader: lineno += 1 try: channel = t.pop(0) id = t.pop(0) try: id = int(id) except ValueError: # We'll skip over this so, say, nicks can be kept here. pass v = self.deserialize(channel, id, t) self[channel, id] = v except Exception as e: log.warning('Invalid line #%s in %s.', lineno, self.__class__.__name__) log.debug('Exception: %s', utils.exnToString(e)) except Exception as e: # This catches exceptions from csv.reader. log.warning('Invalid line #%s in %s.', lineno, self.__class__.__name__) log.debug('Exception: %s', utils.exnToString(e)) def flush(self): mode = 'wb' if utils.minisix.PY2 else 'w' fd = utils.file.AtomicFile(self.filename, mode, makeBackupIfSmaller=False) writer = csv.writer(fd) items = list(self.items()) if not items: log.debug('%s: Refusing to write blank file.', self.__class__.__name__) fd.rollback() return try: items.sort() except TypeError: # FIXME: Implement an algorithm that can order dictionnaries # with both strings and integers as keys. pass for ((channel, id), v) in items: L = self.serialize(v) L.insert(0, id) L.insert(0, channel) writer.writerow(L) fd.close() def close(self): self.flush() self.clear() def deserialize(self, channel, id, L): """Should take a list of strings and return an object to be accessed via self.get(channel, id).""" raise NotImplementedError def serialize(self, x): """Should take an object (as returned by self.get(channel, id)) and return a list (of any type serializable to csv).""" raise NotImplementedError def getUserName(id): if isinstance(id, int): try: return ircdb.users.getUser(id).name except KeyError: return 'a user that is no longer registered' else: return id class ChannelIdDatabasePlugin(callbacks.Plugin): class DB(DbiChannelDB): class DB(dbi.DB): class Record(dbi.Record): __fields__ = [ 'at', 'by', 'text' ] def add(self, at, by, text, **kwargs): record = self.Record(at=at, by=by, text=text, **kwargs) return super(self.__class__, self).add(record) def __init__(self, irc): self.__parent = super(ChannelIdDatabasePlugin, self) self.__parent.__init__(irc) self.db = DB(self.name(), {'flat': self.DB})() def die(self): self.db.close() self.__parent.die() def typeSubstitutions(self): """Returns a dict with keys Types/Type/types/type, whose values are the plugin name with matching capitalization and plural.""" return { 'Types': format('%p', self.name()), 'Type': self.name(), 'types': format('%p', self.name().lower()), 'type': self.name().lower(), } def getCommandHelp(self, name, simpleSyntax=None): helpTemplate = self.__parent.getCommandHelp(name, simpleSyntax) return helpTemplate.substitute(self._typeSubstitutions()) def noSuchRecord(self, irc, channel, id): irc.error(_('There is no %s with id #%s in my database for %s.') % (self.name(), id, channel)) def checkChangeAllowed(self, irc, msg, channel, user, record): # Checks and returns True if either the user ID (integer) # or the hostmask of the caller match. if (hasattr(user, 'id') and user.id == record.by) or user == record.by: return True cap = ircdb.makeChannelCapability(channel, 'op') if ircdb.checkCapability(msg.prefix, cap): return True irc.errorNoCapability(cap) def addValidator(self, irc, text): """This should irc.error or raise an exception if text is invalid.""" pass def getUserId(self, irc, prefix, channel=None): try: user = ircdb.users.getUser(prefix) return user.id except KeyError: if conf.get(conf.supybot.databases.plugins.requireRegistration, channel=channel, network=irc.network): irc.errorNotRegistered(Raise=True) return def add(self, irc, msg, args, channel, text): """[] Adds to the $type database for . is only necessary if the message isn't sent in the channel itself. """ user = self.getUserId(irc, msg.prefix, channel) or msg.prefix at = time.time() self.addValidator(irc, text) if text is not None: id = self.db.add(channel, at, user, text) irc.replySuccess(_('%s #%s added.') % (self.name(), id)) add = wrap(add, ['channeldb', 'text']) def remove(self, irc, msg, args, channel, id): """[] Removes the $type with id from the $type database for . is only necessary if the message isn't sent in the channel itself. """ user = self.getUserId(irc, msg.prefix, channel) or msg.prefix try: record = self.db.get(channel, id) self.checkChangeAllowed(irc, msg, channel, user, record) self.db.remove(channel, id) irc.replySuccess() except KeyError: self.noSuchRecord(irc, channel, id) remove = wrap(remove, ['channeldb', 'id']) def searchSerializeRecord(self, record): text = utils.str.ellipsisify(record.text, 50) return format(_('#%s: %q'), record.id, text) def search(self, irc, msg, args, channel, optlist, glob): """[] [--{regexp,by} ] [] Searches for $types matching the criteria given. """ predicates = [] def p(record): for predicate in predicates: if not predicate(record): return False return True for (opt, arg) in optlist: if opt == 'by': predicates.append(lambda r, arg=arg: r.by == arg.id) elif opt == 'regexp': if not ircdb.checkCapability(msg.prefix, 'trusted'): # Limited --regexp to trusted users, because specially # crafted regexps can freeze the bot. See # https://github.com/ProgVal/Limnoria/issues/855 for details irc.errorNoCapability('trusted') predicates.append(lambda r: regexp_wrapper(r.text, reobj=arg, timeout=0.1, plugin_name=self.name(), fcn_name='search')) if glob: def globP(r, glob=glob.lower()): return fnmatch.fnmatch(r.text.lower(), glob) predicates.append(globP) L = [] for record in self.db.select(channel, p): L.append(self.searchSerializeRecord(record)) if L: L.sort() irc.reply(format(_('%s found: %L'), len(L), L)) else: what = self.name().lower() irc.reply(format(_('No matching %p were found.'), what)) search = wrap(search, ['channeldb', getopts({'by': 'otherUser', 'regexp': 'regexpMatcher'}), additional(rest('glob'))]) def showRecord(self, record): template = string.Template(conf.supybot.replies.databaseRecord()) username = getUserName(record.by) nick = username.split('!')[0] # nick==username iff this is a registered user return template.substitute( id=record.id, text=utils.str.quoted(record.text), userid=record.by, username=username, nick=nick, at=utils.str.timestamp(record.at), **self.typeSubstitutions() ) def get(self, irc, msg, args, channel, id): """[] Gets the $type with id from the $type database for . is only necessary if the message isn't sent in the channel itself. """ try: record = self.db.get(channel, id) irc.reply(self.showRecord(record)) except KeyError: self.noSuchRecord(irc, channel, id) get = wrap(get, ['channeldb', 'id']) def change(self, irc, msg, args, channel, id, replacer): """[] Changes the $type with id according to the regular expression . is only necessary if the message isn't sent in the channel itself. """ user = self.getUserId(irc, msg.prefix, channel) or msg.prefix try: record = self.db.get(channel, id) self.checkChangeAllowed(irc, msg, channel, user, record) record.text = replacer(record.text) self.db.set(channel, id, record) irc.replySuccess() except KeyError: self.noSuchRecord(irc, channel, id) change = wrap(change, ['channeldb', 'id', 'regexpReplacer']) def stats(self, irc, msg, args, channel): """[] Returns the number of $types in the database for . is only necessary if the message isn't sent in the channel itself. """ n = self.db.size(channel) whats = self.name().lower() irc.reply(format(_('There %b %n in my database.'), n, (n, whats))) stats = wrap(stats, ['channeldb']) class PeriodicFileDownloader(object): """A class to periodically download a file/files. A class-level dictionary 'periodicFiles' maps names of files to three-tuples of (url, seconds between downloads, function to run with downloaded file). 'url' should be in some form that urllib2.urlopen can handle (do note that urllib2.urlopen handles file:// links perfectly well.) 'seconds between downloads' is the number of seconds between downloads, obviously. An important point to remember, however, is that it is only engaged when a command is run. I.e., if you say you want the file downloaded every day, but no commands that use it are run in a week, the next time such a command is run, it'll be using a week-old file. If you don't want such behavior, you'll have to give an error mess age to the user and tell them to call you back in the morning. 'function to run with downloaded file' is a function that will be passed a string *filename* of the downloaded file. This will be some random filename probably generated via some mktemp-type-thing. You can do what you want with this; you may want to build a database, take some stats, or simply rename the file. You can pass None as your function and the file with automatically be renamed to match the filename you have it listed under. It'll be in conf.supybot.directories.data, of course. Aside from that dictionary, simply use self.getFile(filename) in any method that makes use of a periodically downloaded file, and you'll be set. """ periodicFiles = None def __init__(self): if self.periodicFiles is None: raise ValueError('You must provide files to download') self.lastDownloaded = {} self.downloadedCounter = {} for filename in self.periodicFiles: if self.periodicFiles[filename][-1] is None: fullname = os.path.join(conf.supybot.directories.data(), filename) if os.path.exists(fullname): self.lastDownloaded[filename] = os.stat(fullname).st_ctime else: self.lastDownloaded[filename] = 0 else: self.lastDownloaded[filename] = 0 self.currentlyDownloading = set() self.downloadedCounter[filename] = 0 self.getFile(filename) def _downloadFile(self, filename, url, f): self.currentlyDownloading.add(filename) try: try: infd = utils.web.getUrlFd(url) except IOError as e: self.log.warning('Error downloading %s: %s', url, e) return except utils.web.Error as e: self.log.warning('Error downloading %s: %s', url, e) return confDir = conf.supybot.directories.data() newFilename = os.path.join(confDir, utils.file.mktemp()) outfd = open(newFilename, 'wb') start = time.time() s = infd.read(4096) while s: outfd.write(s) s = infd.read(4096) infd.close() outfd.close() self.log.info('Downloaded %s in %s seconds', filename, time.time()-start) self.downloadedCounter[filename] += 1 self.lastDownloaded[filename] = time.time() if f is None: toFilename = os.path.join(confDir, filename) if os.name == 'nt': # Windows, grrr... if os.path.exists(toFilename): os.remove(toFilename) os.rename(newFilename, toFilename) else: start = time.time() f(newFilename) total = time.time() - start self.log.info('Function ran on %s in %s seconds', filename, total) finally: self.currentlyDownloading.remove(filename) def getFile(self, filename): if world.documenting: return (url, timeLimit, f) = self.periodicFiles[filename] if time.time() - self.lastDownloaded[filename] > timeLimit and \ filename not in self.currentlyDownloading: self.log.info('Beginning download of %s', url) args = (filename, url, f) name = '%s #%s' % (filename, self.downloadedCounter[filename]) t = threading.Thread(target=self._downloadFile, name=name, args=(filename, url, f)) t.setDaemon(True) t.start() world.threadsSpawned += 1 # vim:set shiftwidth=4 softtabstop=4 expandtab textwidth=79: