Mercurial > libervia-web
view libervia/pages/blog/view/page_meta.py @ 1203:251eba911d4d
server (websockets): fixed websocket handling on HTTPS connections:
Original request used to retrieve a page was stored on dynamic pages, but after the end of
it, the channel was deleted, resulting in a isSecure() always returning False, and
troubles in chain leading to the the use of the wrong session object. This patch fixes
this by reworking the way original request is used, and creating a new wrapping class
allowing to keep an API similar to iweb.IRequest, with data coming from both the original
request and the websocket request.
fix 327
author | Goffi <goffi@goffi.org> |
---|---|
date | Sun, 14 Jul 2019 14:45:51 +0200 |
parents | 0f37b65fe7c2 |
children | 67ec22356457 |
line wrap: on
line source
#!/usr/bin/env python2.7 # -*- coding: utf-8 -*- import unicodedata import re import cgi from libervia.server.constants import Const as C from twisted.words.protocols.jabber import jid from twisted.internet import defer from sat.tools.common import data_objects from libervia.server import session_iface from sat.core.i18n import _ from sat.tools.common.template import safe from sat.tools.common import uri from sat.tools.common import data_format from libervia.server import utils from libervia.server.utils import SubPage from sat.core.log import getLogger log = getLogger(__name__) """generic blog (with service/node provided)""" name = u'blog_view' template = u"blog/articles.html" uri_handlers = {(u'pubsub', u'microblog'): 'microblog_uri'} RE_TEXT_URL = re.compile(ur'[^a-zA-Z,_]+') TEXT_MAX_LEN = 60 TEXT_WORD_MIN_LENGHT = 4 URL_LIMIT_MARK = 90 # if canonical URL is longer than that, text will not be appended def microblog_uri(self, uri_data): args = [uri_data[u'path'], uri_data[u'node']] if u'item' in uri_data: args.extend([u'id', uri_data[u'item']]) return self.getURL(*args) def parse_url(self, request): """URL is /[service]/[node]/[filter_keyword]/[item]|[other] if [node] is '@', default namespace is used if a value is unset, default one will be used keyword can be one of: id: next value is a item id tag: next value is a blog tag """ data = self.getRData(request) try: service = self.nextPath(request) except IndexError: data['service'] = u'' else: try: data[u"service"] = jid.JID(service) except Exception: log.warning(_(u"bad service entered: {}").format(service)) self.pageError(request, C.HTTP_BAD_REQUEST) try: node = self.nextPath(request) except IndexError: node = u'@' data['node'] = u'' if node == u'@' else node try: filter_kw = data['filter_keyword'] = self.nextPath(request) except IndexError: filter_kw = u'@' else: if filter_kw == u'@': # No filter, this is used when a subpage is needed, notably Atom feed pass elif filter_kw == u'id': try: data[u'item'] = self.nextPath(request) except IndexError: self.pageError(request, C.HTTP_BAD_REQUEST) # we get one more argument in case text has been added to have a nice URL try: self.nextPath(request) except IndexError: pass elif filter_kw == u'tag': try: data[u'tag'] = self.nextPath(request) except IndexError: self.pageError(request, C.HTTP_BAD_REQUEST) else: # invalid filter keyword log.warning(_(u"invalid filter keyword: {filter_kw}").format( filter_kw=filter_kw)) self.pageError(request, C.HTTP_BAD_REQUEST) # if URL is parsed here, we'll have atom.xml available and we need to # add the link to the page atom_url = self.getURLByPath( SubPage(u'blog_view'), service, node, filter_kw, SubPage(u'blog_feed_atom'), ) request.template_data.setdefault(u'links', []).append({ u"href": atom_url, u"type": "application/atom+xml", u"rel": "alternate", u"title": "{service}'s blog".format(service=service)}) @defer.inlineCallbacks def appendComments(self, blog_items, identities, profile): for blog_item in blog_items: if identities is not None: author = blog_item.author_jid if not author: log.warning(_(u"no author found for item {item_id}").format(item_id=blog_item.id)) else: if author not in identities: identities[author] = yield self.host.bridgeCall(u'identityGet', author, profile) for comment_data in blog_item.comments: service = comment_data[u'service'] node = comment_data[u'node'] try: comments_data = yield self.host.bridgeCall(u'mbGet', service, node, C.NO_LIMIT, [], {C.KEY_ORDER_BY: C.ORDER_BY_CREATION}, profile) except Exception as e: log.warning(_(u"Can't get comments at {node} (service: {service}): {msg}").format( service=service, node=node, msg=e)) continue comments = data_objects.BlogItems(comments_data) blog_item.appendCommentsItems(comments) yield appendComments(self, comments, identities, profile) @defer.inlineCallbacks def getBlogItems(self, request, service, node, item_id, extra, profile): try: if item_id: items_id = [item_id] else: items_id = [] blog_data = yield self.host.bridgeCall(u'mbGet', service.userhost(), node, C.NO_LIMIT, items_id, extra, profile) except Exception as e: # FIXME: need a better way to test errors in bridge errback if u"forbidden" in unicode(e): self.pageError(request, 401) else: log.warning(_(u"can't retrieve blog for [{service}]: {msg}".format( service = service.userhost(), msg=e))) blog_data = ([], {}) defer.returnValue(data_objects.BlogItems(blog_data)) @defer.inlineCallbacks def prepare_render(self, request): data = self.getRData(request) page_max = data.get(u"page_max", 10) # if the comments are not explicitly hidden, we show them service, node, item_id, show_comments = data.get(u'service', u''), data.get(u'node', u''), data.get(u'item'), data.get(u'show_comments', True) profile = self.getProfile(request) if profile is None: profile = C.SERVICE_PROFILE profile_connected = False else: profile_connected = True ## pagination/filtering parameters if item_id: extra = {} else: extra = self.getPubsubExtra(request, page_max=page_max) tag = data.get('tag') if tag: extra[u'mam_filter_{}'.format(C.MAM_FILTER_CATEGORY)] = tag ## main data ## # we get data from backend/XMPP here items = yield getBlogItems(self, request, service, node, item_id, extra, profile) ## navigation ## # no let's fill service, node and pagination URLs template_data = request.template_data if u'service' not in template_data: template_data[u'service'] = service if u'node' not in template_data: template_data[u'node'] = node target_profile = template_data.get(u'target_profile') if items: if not item_id: self.setPagination(request, items.metadata) else: if item_id: # if item id has been specified in URL and it's not found, # we must return an error self.pageError(request, C.HTTP_NOT_FOUND) ## identities ## # identities are used to show nice nickname or avatars identities = template_data[u'identities'] = self.host.getSessionData(request, session_iface.ISATSession).identities ## Comments ## # if comments are requested, we need to take them if show_comments: yield appendComments(self, items, identities, profile) ## URLs ## # We will fill items_http_uri and tags_http_uri in template_data with suitable urls # if we know the profile, we use it instead of service + blog (nicer url) if target_profile is None: blog_base_url_item = self.getPageByName(u'blog_view').getURL(service.full(), node or u'@', u'id') blog_base_url_tag = self.getPageByName(u'blog_view').getURL(service.full(), node or u'@', u'tag') else: blog_base_url_item = self.getURLByNames([(u'user', [target_profile]), (u'user_blog', [u'id'])]) blog_base_url_tag = self.getURLByNames([(u'user', [target_profile]), (u'user_blog', [u'tag'])]) # we also set the background image if specified by user bg_img = yield self.host.bridgeCall(u'asyncGetParamA', u'Background', u'Blog page', u'value', -1, template_data[u'target_profile']) if bg_img: template_data['dynamic_style'] = safe(u""" :root { --bg-img: url("%s"); } """ % cgi.escape(bg_img, True)) template_data[u'items'] = data[u'items'] = items if request.args.get('reverse') == ['1']: template_data[u'items'].items.reverse() template_data[u'items_http_uri'] = items_http_uri = {} template_data[u'tags_http_uri'] = tags_http_uri = {} for item in items: blog_canonical_url = u'/'.join([blog_base_url_item, utils.quote(item.id)]) if len(blog_canonical_url) > URL_LIMIT_MARK: blog_url = blog_canonical_url else: # we add text from title or body at the end of URL # to make it more human readable text = item.title or item.content # we change special chars to ascii one, trick found at https://stackoverflow.com/a/3194567 text = unicodedata.normalize('NFD', text).encode('ascii', 'ignore') text = RE_TEXT_URL.sub(u' ', text).lower() text = u'-'.join([t for t in text.split() if t and len(t)>=TEXT_WORD_MIN_LENGHT]) while len(text) > TEXT_MAX_LEN: if u'-' in text: text = text.rsplit(u'-', 1)[0] else: text = text[:TEXT_MAX_LEN] if text: blog_url = blog_canonical_url + u'/' + text else: blog_url = blog_canonical_url items_http_uri[item.id] = self.host.getExtBaseURL(request, blog_url) for tag in item.tags: if tag not in tags_http_uri: tag_url = u'/'.join([blog_base_url_tag, utils.quote(tag)]) tags_http_uri[tag] = self.host.getExtBaseURL(request, tag_url) # if True, page should display a comment box template_data[u'allow_commenting'] = data.get(u'allow_commenting', profile_connected) # last but not least, we add a xmpp: link to the node uri_args = {u'path': service.full()} if node: uri_args[u'node'] = node if item_id: uri_args[u'item'] = item_id template_data[u'xmpp_uri'] = uri.buildXMPPUri(u'pubsub', subtype='microblog', **uri_args) @defer.inlineCallbacks def on_data_post(self, request): profile = self.getProfile(request) if profile is None: self.pageError(request, C.HTTP_FORBIDDEN) type_ = self.getPostedData(request, u'type') if type_ == u'comment': service, node, body = self.getPostedData(request, (u'service', u'node', u'body')) if not body: self.pageError(request, C.HTTP_BAD_REQUEST) comment_data = {u"content": body} try: yield self.host.bridgeCall(u'mbSend', service, node, data_format.serialise(comment_data), profile) except Exception as e: if u"forbidden" in unicode(e): self.pageError(request, 401) else: raise e else: log.warning(_(u"Unhandled data type: {}").format(type_))