Mercurial > libervia-backend
view src/plugins/plugin_xep_0277.py @ 1446:e8c8e467964b
plugins xep-0060, xep-0277: code simplification/cleaning/fix:
- plugin xep-0060: moved rsm data to a more general metadata dict, which will contain all data relative to the node/items set. RSM metadata are prefixed with "rsm_"
- plugin xep-0060: minor docstring fixes
- plugin xep-0060: removed cache to simplify code base
- fixed broken getLastMicroblogs
- added _getLastMicroblogs as wrapper to getLastMicroblogs, for bridge
- removed lxml dependecy for this plugin, use native twisted instead
- several improvments/fixes in item2mbdata
author | Goffi <goffi@goffi.org> |
---|---|
date | Sat, 15 Aug 2015 22:13:27 +0200 |
parents | 16b1ba7ccaaa |
children | 7797dda847ae |
line wrap: on
line source
#!/usr/bin/python # -*- coding: utf-8 -*- # SAT plugin for microblogging over XMPP (xep-0277) # Copyright (C) 2009, 2010, 2011, 2012, 2013, 2014, 2015 Jérôme Poisson (goffi@goffi.org) # This program 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. # This program 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 this program. If not, see <http://www.gnu.org/licenses/>. from sat.core.i18n import _ from sat.core.constants import Const as C from sat.core.log import getLogger log = getLogger(__name__) from twisted.words.protocols.jabber import jid from twisted.internet import defer from twisted.python import failure from sat.core import exceptions from sat.tools.xml_tools import ElementParser from wokkel import pubsub from feed import atom, date import uuid from time import time import urlparse from cgi import escape NS_MICROBLOG = 'urn:xmpp:microblog:0' NS_ATOM = 'http://www.w3.org/2005/Atom' NS_XHTML = 'http://www.w3.org/1999/xhtml' NS_PUBSUB_EVENT = "{}{}".format(pubsub.NS_PUBSUB, "#event") PLUGIN_INFO = { "name": "Microblogging over XMPP Plugin", "import_name": "XEP-0277", "type": "XEP", "protocols": ["XEP-0277"], "dependencies": ["XEP-0163", "XEP-0060", "TEXT-SYNTAXES"], "recommendations": ["XEP-0059"], "main": "XEP_0277", "handler": "no", "description": _("""Implementation of microblogging Protocol""") } class NodeAccessChangeException(Exception): pass class XEP_0277(object): def __init__(self, host): log.info(_("Microblogging plugin initialization")) self.host = host self.host.plugins["XEP-0163"].addPEPEvent("MICROBLOG", NS_MICROBLOG, self.microblogCB, self.sendMicroblog, notify=False) host.bridge.addMethod("getLastMicroblogs", ".plugin", in_sign='sis', out_sign='(aa{ss}a{ss})', method=self._getLastMicroblogs, async=True, doc={'summary': 'retrieve items', 'param_0': 'jid: publisher of wanted microblog', 'param_1': 'max_items: see XEP-0060 #6.5.7', 'param_2': '%(doc_profile)s', 'return': 'list of microblog data (dict)'}) host.bridge.addMethod("setMicroblogAccess", ".plugin", in_sign='ss', out_sign='', method=self.setMicroblogAccess, async=True, doc={}) ## plugin management methods ## def microblogCB(self, itemsEvent, profile): """Callback to "MICROBLOG" PEP event.""" def manageItem(microblog_data): self.host.bridge.personalEvent(itemsEvent.sender.full(), "MICROBLOG", microblog_data, profile) for item in itemsEvent.items: self.item2mbdata(item).addCallbacks(manageItem, lambda failure: None) ## data/item transformation ## def _removeXHTMLMarkups(self, xhtml): """Remove XHTML markups from the given string. @param xhtml: the XHTML string to be cleaned @return: a Deferred instance for the cleaned string """ return self.host.plugins["TEXT-SYNTAXES"].convert(xhtml, self.host.plugins["TEXT-SYNTAXES"].SYNTAX_XHTML, self.host.plugins["TEXT-SYNTAXES"].SYNTAX_TEXT, False) @defer.inlineCallbacks def item2mbdata(self, item_elt): """Convert an XML Item to microblog data used in bridge API @param item_elt: domish.Element of microblog item @return: microblog data (dictionary) """ microblog_data = {} def check_conflict(key): if key in microblog_data: raise failure.Failure(exceptions.DataError("key {} is already present for item {}").format(key, item_elt['id'])) return key @defer.inlineCallbacks def parseElement(elem): """Parse title/content elements and fill microblog_data accordingly""" type_ = elem.getAttribute('type') if type_ == 'xhtml': data_elt = elem.firstChildElement() if data_elt.uri != NS_XHTML: raise failure.Failure(exceptions.DataError(_('Content of type XHTML must declare its namespace!'))) key = check_conflict(u'{}_xhtml'.format(elem.name)) data = unicode(data_elt) microblog_data[key] = yield self.host.plugins["TEXT-SYNTAXES"].clean_xhtml(data) else: key = check_conflict(elem.name) microblog_data[key] = unicode(elem) id_ = item_elt.getAttribute('id', '') # there can be no id for transient nodes microblog_data['id'] = id_ if item_elt.uri not in (pubsub.NS_PUBSUB, NS_PUBSUB_EVENT): msg = u"Unsupported namespace {ns} in pubsub item {id_}".format(ns=item_elt.uri, id_=id_) log.warning(msg) raise failure.Failure(exceptions.DataError(msg)) try: entry_elt = item_elt.elements(NS_ATOM, 'entry').next() except StopIteration: msg = u'No atom entry found in the pubsub item {}'.format(id_) raise failure.Failure(exceptions.DataError(msg)) try: title_elt = entry_elt.elements(NS_ATOM, 'title').next() except StopIteration: msg = u'No atom title found in the pubsub item {}'.format(id_) raise failure.Failure(exceptions.DataError(msg)) yield parseElement(title_elt) for content_elt in entry_elt.elements(NS_ATOM, 'content'): yield parseElement(content_elt) # we check that text content is present for key in ('title', 'content'): if key not in microblog_data and ('{}_xhtml'.format(key)) in microblog_data: log.warning(u"item {id_} provide a {key}_xhtml data but not a text one".format(id_, key)) # ... and do the conversion if it's not microblog_data[key] = yield self.host.plugins["TEXT-SYNTAXES"].\ convert(microblog_data['{}_xhtml'.format(key)], self.host.plugins["TEXT-SYNTAXES"].SYNTAX_XHTML, self.host.plugins["TEXT-SYNTAXES"].SYNTAX_TEXT, False) try: id_elt = entry_elt.elements(NS_ATOM, 'id').next() except StopIteration: msg = u'No atom id found in the pubsub item {}, this is not standard !'.format(id_) log.warning(msg) microblog_data['atom_id'] = "" else: microblog_data['atom_id'] = unicode(id_elt) try: updated_elt = entry_elt.elements(NS_ATOM, 'updated').next() except StopIteration: msg = u'No atom updated element found in the pubsub item {}'.format(id_) raise failure.Failure(exceptions.DataError(msg)) microblog_data['updated'] = unicode(date.rfc3339.tf_from_timestamp(unicode(updated_elt))) if 'content' not in microblog_data: # use the atom title data as the microblog body content microblog_data['content'] = microblog_data['title'] del microblog_data['title'] if 'title_xhtml' in microblog_data: microblog_data['content_xhtml'] = microblog_data['title_xhtml'] del microblog_data['title_xhtml'] try: published_elt = entry_elt.elements(NS_ATOM, 'published').next() except StopIteration: microblog_data['published'] = microblog_data['updated'] else: microblog_data['published'] = unicode(date.rfc3339.tf_from_timestamp(unicode(published_elt))) for link_elt in entry_elt.elements(NS_ATOM, 'link'): if link_elt.getAttribute('rel') == 'replies' and link_elt.getAttribute('title') == 'comments': key = check_conflict('comments') microblog_data[key] = link_elt['href'] try: service, node = self.parseCommentUrl(microblog_data[key]) except: log.warning(u"Can't parse url {}".format(microblog_data[key])) del microblog_data[key] else: microblog_data['comments_service'] = service.full() microblog_data['comments_node'] = node else: rel = link_elt.getAttribute('rel','') title = link_elt.getAttribute('title','') href = link_elt.getAttribute('href','') log.warning(u"Unmanaged link element: rel={rel} title={title} href={href}".format(rel=rel, title=title, href=href)) try: author_elt = entry_elt.elements(NS_ATOM, 'author').next() except StopIteration: log.warning("Can't find author element in item {}".format(id_)) else: try: name_elt = author_elt.elements(NS_ATOM, 'name').next() except StopIteration: log.warning("No name element found in author element of item {}".format(id_)) else: microblog_data['author'] = unicode(name_elt) defer.returnValue(microblog_data) @defer.inlineCallbacks def data2entry(self, data, profile): """Convert a data dict to en entry usable to create an item @param data: data dict as given by bridge method. @return: deferred which fire domish.Element """ #TODO: rewrite this directly with twisted (i.e. without atom / reparsing) _uuid = unicode(uuid.uuid1()) _entry = atom.Entry() _entry.title = '' # reset the default value which is not empty elems = {'title': atom.Title, 'content': atom.Content} synt = self.host.plugins["TEXT-SYNTAXES"] # loop on ('title', 'title_rich', 'title_xhtml', 'content', 'content_rich', 'content_xhtml') for key in elems.keys(): for type_ in ['', 'rich', 'xhtml']: attr = "%s_%s" % (key, type_) if type_ else key if attr in data: if type_: if type_ == 'rich': # convert input from current syntax to XHTML converted = yield synt.convert(data[attr], synt.getCurrentSyntax(profile), "XHTML") else: # clean the XHTML input converted = yield synt.clean_xhtml(data[attr]) elem = elems[key]((u'<div xmlns="%s">%s</div>' % (NS_XHTML, converted)).encode('utf-8')) elem.attrs['type'] = 'xhtml' if hasattr(_entry, '%s_xhtml' % key): raise failure.Failure(exceptions.DataError(_("Can't have xhtml and rich content at the same time"))) setattr(_entry, '%s_xhtml' % key, elem) else: # raw text only needs to be escaped to get HTML-safe sequence elem = elems[key](escape(data[attr]).encode('utf-8')) elem.attrs['type'] = 'text' setattr(_entry, key, elem) if not getattr(_entry, key).text: if hasattr(_entry, '%s_xhtml' % key): text = yield self._removeXHTMLMarkups(getattr(_entry, '%s_xhtml' % key).text) setattr(_entry, key, text) if not _entry.title.text: # eventually move the data from content to title _entry.title = _entry.content.text _entry.title.attrs['type'] = _entry.content.attrs['type'] _entry.content.text = '' _entry.content.attrs['type'] = '' if hasattr(_entry, 'content_xhtml'): _entry.title_xhtml = atom.Title(_entry.content_xhtml.text) _entry.title_xhtml.attrs['type'] = _entry.content_xhtml.attrs['type'] _entry.content_xhtml.text = '' _entry.content_xhtml.attrs['type'] = '' _entry.author = atom.Author() _entry.author.name = data.get('author', self.host.getJidNStream(profile)[0].userhost()).encode('utf-8') _entry.updated = float(data.get('updated', time())) _entry.published = float(data.get('published', time())) entry_id = data.get('id', unicode(_uuid)) _entry.id = entry_id.encode('utf-8') if 'comments' in data: link = atom.Link() link.attrs['href'] = data['comments'] link.attrs['rel'] = 'replies' link.attrs['title'] = 'comments' _entry.links.append(link) _entry_elt = ElementParser()(str(_entry).decode('utf-8')) item = pubsub.Item(id=entry_id, payload=_entry_elt) defer.returnValue(item) ## publish ## @defer.inlineCallbacks def sendMicroblog(self, data, profile): """Send XEP-0277's microblog data @param data: must include content @param profile: profile which send the mood""" if 'content' not in data: log.error("Microblog data must contain at least 'content' key") raise failure.Failure(exceptions.DataError('no "content" key found')) content = data['content'] if not content: log.error("Microblog data's content value must not be empty") raise failure.Failure(exceptions.DataError('empty content')) item = yield self.data2entry(data, profile) ret = yield self.host.plugins["XEP-0060"].publish(None, NS_MICROBLOG, [item], profile_key=profile) defer.returnValue(ret) ## get ## def _getLastMicroblogs(self, pub_jid_s, max_items=10, profile_key=C.PROF_KEY_NONE): return self.getLastMicroblogs(jid.JID(pub_jid_s), max_items, profile_key) @defer.inlineCallbacks def getLastMicroblogs(self, pub_jid, max_items=10, profile_key=C.PROF_KEY_NONE): """Get the last published microblogs @param pub_jid(jid.JID): jid of the publisher @param max_items: how many microblogs we want to get @param profile_key: profile key @return: a deferred couple with the list of items and metadatas. """ items, metadata = yield self.host.plugins["XEP-0060"].getItems(pub_jid, NS_MICROBLOG, max_items=max_items, profile_key=profile_key) dlist_result = yield defer.DeferredList(map(self.item2mbdata, items), consumeErrors=True) items_data = [] for success, value in dlist_result: if success: items_data.append(value) else: log.warning(u"Error while parsing microblog data: {}".format(value.value)) defer.returnValue((items_data, metadata)) def parseCommentUrl(self, node_url): """Parse a XMPP URI Determine the fields comments_service and comments_node of a microblog data from the href attribute of an entry's link element. For example this input: xmpp:sat-pubsub.libervia.org?node=urn%3Axmpp%3Acomments%3A_c5c4a142-2279-4b2a-ba4c-1bc33aa87634__urn%3Axmpp%3Agroupblog%3Asouliane%40libervia.org will return (JID(u'sat-pubsub.libervia.org'), 'urn:xmpp:comments:_c5c4a142-2279-4b2a-ba4c-1bc33aa87634__urn:xmpp:groupblog:souliane@libervia.org') @return: a tuple (JID, str) """ parsed_url = urlparse.urlparse(node_url, 'xmpp') service = jid.JID(parsed_url.path) queries = parsed_url.query.split(';') parsed_queries = dict() for query in queries: parsed_queries.update(urlparse.parse_qs(query)) node = parsed_queries.get('node', [''])[0] if not node: raise failure.Failure(exceptions.DataError('Invalid comments link')) return (service, node) ## configure ## def setMicroblogAccess(self, access="presence", profile_key=C.PROF_KEY_NONE): """Create a microblog node on PEP with given access If the node already exists, it change options @param access: Node access model, according to xep-0060 #4.5 @param profile_key: profile key""" _jid, xmlstream = self.host.getJidNStream(profile_key) if not _jid: log.error(_("Can't find profile's jid")) return C = self.host.plugins["XEP-0060"] _options = {C.OPT_ACCESS_MODEL: access, C.OPT_PERSIST_ITEMS: 1, C.OPT_MAX_ITEMS: -1, C.OPT_DELIVER_PAYLOADS: 1, C.OPT_SEND_ITEM_SUBSCRIBE: 1} def cb(result): #Node is created with right permission log.debug(_(u"Microblog node has now access %s") % access) def fatal_err(s_error): #Something went wrong log.error(_("Can't set microblog access")) raise NodeAccessChangeException() def err_cb(s_error): #If the node already exists, the condition is "conflict", #else we have an unmanaged error if s_error.value.condition == 'conflict': #d = self.host.plugins["XEP-0060"].deleteNode(_jid.userhostJID(), NS_MICROBLOG, profile_key=profile_key) #d.addCallback(lambda x: create_node().addCallback(cb).addErrback(fatal_err)) change_node_options().addCallback(cb).addErrback(fatal_err) else: fatal_err(s_error) def create_node(): return self.host.plugins["XEP-0060"].createNode(_jid.userhostJID(), NS_MICROBLOG, _options, profile_key=profile_key) def change_node_options(): return self.host.plugins["XEP-0060"].setOptions(_jid.userhostJID(), NS_MICROBLOG, _jid.userhostJID(), _options, profile_key=profile_key) create_node().addCallback(cb).addErrback(err_cb)