def attempt_single_item(self, item): """Attempts to preview or publish a single mf2 item. Args: item: mf2 item dict from mf2py Returns: CreationResult """ self.maybe_inject_silo_content(item) obj = microformats2.json_to_object(item) ignore_formatting = self.ignore_formatting(item) if ignore_formatting: prop = microformats2.first_props(item.get('properties', {})) content = microformats2.get_text(prop.get('content')) if content: obj['content'] = content.strip() # which original post URL to include? in order of preference: # 1. rel-shortlink (background: https://github.com/snarfed/bridgy/issues/173) # 2. original user-provided URL if it redirected # 3. u-url if available # 4. actual final fetched URL if self.shortlink: obj['url'] = self.shortlink elif self.source_url() != self.fetched.url: obj['url'] = self.source_url() elif 'url' not in obj: obj['url'] = self.fetched.url logging.debug('Converted to ActivityStreams object: %s', json.dumps(obj, indent=2)) # posts and comments need content obj_type = obj.get('objectType') if obj_type in ('note', 'article', 'comment'): if (not obj.get('content') and not obj.get('summary') and not obj.get('displayName')): return gr_source.creation_result( abort=False, error_plain='Could not find content in %s' % self.fetched.url, error_html= 'Could not find <a href="http://microformats.org/">content</a> in %s' % self.fetched.url) self.preprocess(obj) include_link = self.include_link(item) if not self.authorize(): return gr_source.creation_result(abort=True) # RIP Facebook comments/likes. https://github.com/snarfed/bridgy/issues/350 if (isinstance(self.source, FacebookPage) and (obj_type == 'comment' or obj.get('verb') == 'like')): return gr_source.creation_result( abort=True, error_plain= 'Facebook comments and likes are no longer supported. :(', error_html= '<a href="https://github.com/snarfed/bridgy/issues/350">' 'Facebook comments and likes are no longer supported.</a> :(') if self.PREVIEW: result = self.source.gr_source.preview_create( obj, include_link=include_link, ignore_formatting=ignore_formatting) self.entity.published = result.content or result.description if not self.entity.published: return result # there was an error state = { 'source_key': self.source.key.urlsafe(), 'source_url': self.source_url(), 'target_url': self.target_url(), 'include_link': include_link, } vars = { 'source': self.preprocess_source(self.source), 'preview': result.content, 'description': result.description, 'webmention_endpoint': self.request.host_url + '/publish/webmention', 'state': self.encode_state_parameter(state), } vars.update(state) logging.info('Rendering preview with template vars %s', pprint.pformat(vars)) return gr_source.creation_result( template.render('templates/preview.html', vars)) else: result = self.source.gr_source.create( obj, include_link=include_link, ignore_formatting=ignore_formatting) self.entity.published = result.content if not result.content: return result # there was an error if 'url' not in self.entity.published: self.entity.published['url'] = obj.get('url') self.entity.type = self.entity.published.get( 'type') or models.get_type(obj) self.entity.type_label = self.source.TYPE_LABELS.get( self.entity.type) self.response.headers['Content-Type'] = 'application/json' logging.info('Returning %s', json.dumps(self.entity.published, indent=2)) self.response.headers['Location'] = self.entity.published[ 'url'].encode('utf-8') self.response.status = 201 return gr_source.creation_result( json.dumps(self.entity.published, indent=2))
def attempt_single_item(self, item): """Attempts to preview or publish a single mf2 item. Args: item: mf2 item dict from mf2py Returns: CreationResult """ self.maybe_inject_silo_content(item) obj = microformats2.json_to_object(item) ignore_formatting = self.ignore_formatting(item) if ignore_formatting: prop = microformats2.first_props(item.get('properties', {})) content = microformats2.get_text(prop.get('content')) if content: obj['content'] = content.strip() # which original post URL to include? in order of preference: # 1. rel-shortlink (background: https://github.com/snarfed/bridgy/issues/173) # 2. original user-provided URL if it redirected # 3. u-url if available # 4. actual final fetched URL if self.shortlink: obj['url'] = self.shortlink elif self.source_url() != self.fetched.url: obj['url'] = self.source_url() elif 'url' not in obj: obj['url'] = self.fetched.url logging.debug('Converted to ActivityStreams object: %s', json_dumps(obj, indent=2)) # posts and comments need content obj_type = obj.get('objectType') if obj_type in ('note', 'article', 'comment'): if (not obj.get('content') and not obj.get('summary') and not obj.get('displayName')): return gr_source.creation_result( abort=False, error_plain='Could not find content in %s' % self.fetched.url, error_html='Could not find <a href="http://microformats.org/">content</a> in %s' % self.fetched.url) self.preprocess(obj) include_link = self.include_link(item) if not self.authorize(): return gr_source.creation_result(abort=True) if self.PREVIEW: result = self.source.gr_source.preview_create( obj, include_link=include_link, ignore_formatting=ignore_formatting) previewed = result.content or result.description if self.entity.type == 'preview': self.entity.published = previewed if not previewed: return result # there was an error return self._render_preview(result, include_link=include_link) else: result = self.source.gr_source.create( obj, include_link=include_link, ignore_formatting=ignore_formatting) self.entity.published = result.content if not result.content: return result # there was an error if 'url' not in self.entity.published: self.entity.published['url'] = obj.get('url') self.entity.type = self.entity.published.get('type') or models.get_type(obj) self.response.headers['Content-Type'] = 'application/json' logging.info('Returning %s', json_dumps(self.entity.published, indent=2)) self.response.headers['Location'] = self.entity.published['url'] self.response.status = 201 return gr_source.creation_result( json_dumps(self.entity.published, indent=2))
def attempt_single_item(self, item): """Attempts to preview or publish a single mf2 item. Args: item: mf2 item dict from mf2py Returns: CreationResult """ self.maybe_inject_silo_content(item) obj = microformats2.json_to_object(item) ignore_formatting = self.ignore_formatting(item) if ignore_formatting: prop = microformats2.first_props(item.get('properties', {})) content = microformats2.get_text(prop.get('content')) if content: obj['content'] = content.strip() # which original post URL to include? in order of preference: # 1. rel-shortlink (background: https://github.com/snarfed/bridgy/issues/173) # 2. original user-provided URL if it redirected # 3. u-url if available # 4. actual final fetched URL if self.shortlink: obj['url'] = self.shortlink elif self.source_url() != self.fetched.url: obj['url'] = self.source_url() elif 'url' not in obj: obj['url'] = self.fetched.url logging.debug('Converted to ActivityStreams object: %s', json.dumps(obj, indent=2)) # posts and comments need content obj_type = obj.get('objectType') if obj_type in ('note', 'article', 'comment'): if (not obj.get('content') and not obj.get('summary') and not obj.get('displayName')): return gr_source.creation_result( abort=False, error_plain='Could not find content in %s' % self.fetched.url, error_html='Could not find <a href="http://microformats.org/">content</a> in %s' % self.fetched.url) self.preprocess(obj) include_link = self.include_link(item) if not self.authorize(): return gr_source.creation_result(abort=True) # RIP Facebook. # https://github.com/snarfed/bridgy/issues/817 # https://github.com/snarfed/bridgy/issues/350 verb = obj.get('verb') if isinstance(self.source, FacebookPage): return gr_source.creation_result( abort=True, error_plain='Facebook is no longer supported. So long, and thanks for all the fish!', error_html='<a href="https://brid.gy/about#rip-facebook">Facebook is no longer supported. So long, and thanks for all the fish!</a>') if self.PREVIEW: result = self.source.gr_source.preview_create( obj, include_link=include_link, ignore_formatting=ignore_formatting) self.entity.published = result.content or result.description if not self.entity.published: return result # there was an error return self._render_preview(result, include_link=include_link) else: result = self.source.gr_source.create( obj, include_link=include_link, ignore_formatting=ignore_formatting) self.entity.published = result.content if not result.content: return result # there was an error if 'url' not in self.entity.published: self.entity.published['url'] = obj.get('url') self.entity.type = self.entity.published.get('type') or models.get_type(obj) self.response.headers['Content-Type'] = 'application/json' logging.info('Returning %s', json.dumps(self.entity.published, indent=2)) self.response.headers['Location'] = self.entity.published['url'].encode('utf-8') self.response.status = 201 return gr_source.creation_result( json.dumps(self.entity.published, indent=2))
def attempt_single_item(self, item): """Attempts to preview or publish a single mf2 item. Args: item: mf2 item dict from mf2py Returns: CreationResult """ self.maybe_inject_silo_content(item) obj = microformats2.json_to_object(item) ignore_formatting = self.ignore_formatting(item) if ignore_formatting: prop = microformats2.first_props(item.get('properties', {})) content = microformats2.get_text(prop.get('content')) if content: obj['content'] = content.strip() # which original post URL to include? in order of preference: # 1. rel-shortlink (background: https://github.com/snarfed/bridgy/issues/173) # 2. original user-provided URL if it redirected # 3. u-url if available # 4. actual final fetched URL if self.shortlink: obj['url'] = self.shortlink elif self.source_url() != self.fetched.url: obj['url'] = self.source_url() elif 'url' not in obj: obj['url'] = self.fetched.url logging.debug('Converted to ActivityStreams object: %s', json.dumps(obj, indent=2)) # posts and comments need content obj_type = obj.get('objectType') if obj_type in ('note', 'article', 'comment'): if (not obj.get('content') and not obj.get('summary') and not obj.get('displayName')): return gr_source.creation_result( abort=False, error_plain='Could not find content in %s' % self.fetched.url, error_html='Could not find <a href="http://microformats.org/">content</a> in %s' % self.fetched.url) self.preprocess(obj) omit_link = self.omit_link(item) if not self.authorize(): return gr_source.creation_result(abort=True) # RIP Facebook comments/likes. https://github.com/snarfed/bridgy/issues/350 if (isinstance(self.source, FacebookPage) and (obj_type == 'comment' or obj.get('verb') == 'like')): return gr_source.creation_result( abort=True, error_plain='Facebook comments and likes are no longer supported. :(', error_html='<a href="https://github.com/snarfed/bridgy/issues/350">' 'Facebook comments and likes are no longer supported.</a> :(') if self.PREVIEW: result = self.source.gr_source.preview_create( obj, include_link=not omit_link, ignore_formatting=ignore_formatting) self.entity.published = result.content or result.description if not self.entity.published: return result # there was an error state = { 'source_key': self.source.key.urlsafe(), 'source_url': self.source_url(), 'target_url': self.target_url(), 'bridgy_omit_link': omit_link, } vars = {'source': self.preprocess_source(self.source), 'preview': result.content, 'description': result.description, 'webmention_endpoint': self.request.host_url + '/publish/webmention', 'state': self.encode_state_parameter(state), } vars.update(state) logging.info('Rendering preview with template vars %s', pprint.pformat(vars)) return gr_source.creation_result( template.render('templates/preview.html', vars)) else: result = self.source.gr_source.create(obj, include_link=not omit_link, ignore_formatting=ignore_formatting) self.entity.published = result.content if not result.content: return result # there was an error if 'url' not in self.entity.published: self.entity.published['url'] = obj.get('url') self.entity.type = self.entity.published.get('type') or models.get_type(obj) self.entity.type_label = self.source.TYPE_LABELS.get(self.entity.type) self.response.headers['Content-Type'] = 'application/json' logging.info('Returning %s', json.dumps(self.entity.published, indent=2)) self.response.headers['Location'] = self.entity.published['url'].encode('utf-8') self.response.status = 201 return gr_source.creation_result( json.dumps(self.entity.published, indent=2))
def template_vars(self, domain=None, url=None): logging.debug(f'Headers: {list(request.headers.items())}') if domain.split('.')[-1] in NON_TLDS: error(f"{domain} doesn't look like a domain", status=404) # find representative h-card. try url, then url's home page, then domain urls = [f'http://{domain}/'] if url: urls = [url, urllib.parse.urljoin(url, '/')] + urls for candidate in urls: resp = common.requests_get(candidate) parsed = util.parse_html(resp) mf2 = util.parse_mf2(parsed, url=resp.url) # logging.debug(f'Parsed mf2 for {resp.url}: {json_dumps(mf2, indent=2)}') hcard = mf2util.representative_hcard(mf2, resp.url) if hcard: logging.info( f'Representative h-card: {json_dumps(hcard, indent=2)}') break else: error( f"didn't find a representative h-card (http://microformats.org/wiki/representative-hcard-parsing) on {resp.url}" ) logging.info(f'Generating WebFinger data for {domain}') key = models.MagicKey.get_or_create(domain) props = hcard.get('properties', {}) urls = util.dedupe_urls(props.get('url', []) + [resp.url]) canonical_url = urls[0] acct = f'{domain}@{domain}' for url in urls: if url.startswith('acct:'): urluser, urldomain = util.parse_acct_uri(url) if urldomain == domain: acct = f'{urluser}@{domain}' logging.info(f'Found custom username: acct:{acct}') break # discover atom feed, if any atom = parsed.find('link', rel='alternate', type=common.CONTENT_TYPE_ATOM) if atom and atom['href']: atom = urllib.parse.urljoin(resp.url, atom['href']) else: atom = 'https://granary.io/url?' + urllib.parse.urlencode( { 'input': 'html', 'output': 'atom', 'url': resp.url, 'hub': resp.url, }) # discover PuSH, if any for link in resp.headers.get('Link', '').split(','): match = common.LINK_HEADER_RE.match(link) if match and match.group(2) == 'hub': hub = match.group(1) else: hub = 'https://bridgy-fed.superfeedr.com/' # generate webfinger content data = util.trim_nulls({ 'subject': 'acct:' + acct, 'aliases': urls, 'magic_keys': [{ 'value': key.href() }], 'links': sum(([{ 'rel': 'http://webfinger.net/rel/profile-page', 'type': 'text/html', 'href': url, }] for url in urls if url.startswith("http")), []) + [{ 'rel': 'http://webfinger.net/rel/avatar', 'href': get_text(url), } for url in props.get('photo', [])] + [ { 'rel': 'canonical_uri', 'type': 'text/html', 'href': canonical_url, }, # ActivityPub { 'rel': 'self', 'type': common.CONTENT_TYPE_AS2, # WARNING: in python 2 sometimes request.host_url lost port, # http://localhost:8080 would become just http://localhost. no # clue how or why. pay attention here if that happens again. 'href': f'{request.host_url}{domain}', }, { 'rel': 'inbox', 'type': common.CONTENT_TYPE_AS2, 'href': f'{request.host_url}{domain}/inbox', }, # OStatus { 'rel': 'http://schemas.google.com/g/2010#updates-from', 'type': common.CONTENT_TYPE_ATOM, 'href': atom, }, { 'rel': 'hub', 'href': hub, }, { 'rel': 'magic-public-key', 'href': key.href(), }, { 'rel': 'salmon', 'href': f'{request.host_url}{domain}/salmon', } ] }) logging.info(f'Returning WebFinger data: {json_dumps(data, indent=2)}') return data