diff options
| -rw-r--r-- | actions/apimediaupload.php | 141 | ||||
| -rw-r--r-- | actions/apistatusnetconfig.php | 10 | ||||
| -rw-r--r-- | actions/deleteuser.php | 10 | ||||
| -rw-r--r-- | classes/Profile.php | 3 | ||||
| -rw-r--r-- | classes/Profile_role.php | 1 | ||||
| -rw-r--r-- | classes/Safe_DataObject.php | 19 | ||||
| -rw-r--r-- | classes/Subscription.php | 4 | ||||
| -rw-r--r-- | lib/activity.php | 10 | ||||
| -rw-r--r-- | lib/deluserqueuehandler.php | 95 | ||||
| -rw-r--r-- | lib/queuemanager.php | 3 | ||||
| -rw-r--r-- | lib/router.php | 6 | ||||
| -rw-r--r-- | lib/userprofile.php | 11 | ||||
| -rw-r--r-- | plugins/OStatus/actions/ostatussub.php | 5 | ||||
| -rw-r--r-- | plugins/OStatus/classes/Ostatus_profile.php | 233 | ||||
| -rw-r--r-- | plugins/OStatus/lib/discovery.php | 78 | ||||
| -rw-r--r-- | plugins/OStatus/lib/discoveryhints.php | 182 | ||||
| -rw-r--r-- | plugins/OStatus/lib/feeddiscovery.php | 4 | ||||
| -rw-r--r-- | plugins/OStatus/lib/linkheader.php | 63 | ||||
| -rwxr-xr-x | scripts/fixup_files.php | 77 |
19 files changed, 779 insertions, 176 deletions
diff --git a/actions/apimediaupload.php b/actions/apimediaupload.php new file mode 100644 index 000000000..ec316edc8 --- /dev/null +++ b/actions/apimediaupload.php @@ -0,0 +1,141 @@ +<?php +/** + * StatusNet, the distributed open-source microblogging tool + * + * Upload an image via the API + * + * PHP version 5 + * + * LICENCE: 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/>. + * + * @category API + * @author Zach Copley <zach@status.net> + * @copyright 2010 StatusNet, Inc. + * @license http://www.fsf.org/licensing/licenses/agpl-3.0.html GNU Affero General Public License version 3.0 + * @link http://status.net/ + */ + +if (!defined('STATUSNET')) { + exit(1); +} + +require_once INSTALLDIR . '/lib/apiauth.php'; +require_once INSTALLDIR . '/lib/mediafile.php'; + +/** + * Upload an image via the API. Returns a shortened URL for the image + * to the user. + * + * @category API + * @package StatusNet + * @author Zach Copley <zach@status.net> + * @license http://www.fsf.org/licensing/licenses/agpl-3.0.html GNU Affero General Public License version 3.0 + * @link http://status.net/ + */ + +class ApiMediaUploadAction extends ApiAuthAction +{ + /** + * Handle the request + * + * Grab the file from the 'media' param, then store, and shorten + * + * @todo Upload throttle! + * + * @param array $args $_REQUEST data (unused) + * + * @return void + */ + + function handle($args) + { + parent::handle($args); + + if ($_SERVER['REQUEST_METHOD'] != 'POST') { + $this->clientError( + _('This method requires a POST.'), + 400, $this->format + ); + return; + } + + // Workaround for PHP returning empty $_POST and $_FILES when POST + // length > post_max_size in php.ini + + if (empty($_FILES) + && empty($_POST) + && ($_SERVER['CONTENT_LENGTH'] > 0) + ) { + $msg = _('The server was unable to handle that much POST ' . + 'data (%s bytes) due to its current configuration.'); + + $this->clientError(sprintf($msg, $_SERVER['CONTENT_LENGTH'])); + return; + } + + $upload = null; + + try { + $upload = MediaFile::fromUpload('media', $this->auth_user); + } catch (ClientException $ce) { + $this->clientError($ce->getMessage()); + return; + } + + if (isset($upload)) { + $this->showResponse($upload); + } else { + $this->clientError('Upload failed.'); + return; + } + } + + /** + * Show a Twitpic-like response with the ID of the media file + * and a (hopefully) shortened URL for it. + * + * @param File $upload the uploaded file + * + * @return void + */ + function showResponse($upload) + { + $this->initDocument(); + $this->elementStart('rsp', array('stat' => 'ok')); + $this->element('mediaid', null, $upload->fileRecord->id); + $this->element('mediaurl', null, $upload->shortUrl()); + $this->elementEnd('rsp'); + $this->endDocument(); + } + + /** + * Overrided clientError to show a more Twitpic-like error + * + * @param String $msg an error message + * + */ + function clientError($msg) + { + $this->initDocument(); + $this->elementStart('rsp', array('stat' => 'fail')); + + // @todo add in error code + $errAttr = array('msg' => $msg); + + $this->element('err', $errAttr, null); + $this->elementEnd('rsp'); + $this->endDocument(); + } + +} diff --git a/actions/apistatusnetconfig.php b/actions/apistatusnetconfig.php index bff8313b5..66b23c02d 100644 --- a/actions/apistatusnetconfig.php +++ b/actions/apistatusnetconfig.php @@ -97,8 +97,6 @@ class ApiStatusnetConfigAction extends ApiAction // XXX: check that all sections and settings are legal XML elements - common_debug(var_export($this->keys, true)); - foreach ($this->keys as $section => $settings) { $this->elementStart($section); foreach ($settings as $setting) { @@ -110,6 +108,14 @@ class ApiStatusnetConfigAction extends ApiAction } else if ($value === true) { $value = 'true'; } + + // return theme logo if there's no site specific one + if (empty($value)) { + if ($section == 'site' && $setting == 'logo') { + $value = Theme::path('logo.png'); + } + } + $this->element($setting, null, $value); } $this->elementEnd($section); diff --git a/actions/deleteuser.php b/actions/deleteuser.php index c4f84fad2..4e6b27395 100644 --- a/actions/deleteuser.php +++ b/actions/deleteuser.php @@ -162,7 +162,15 @@ class DeleteuserAction extends ProfileFormAction function handlePost() { if (Event::handle('StartDeleteUser', array($this, $this->user))) { - $this->user->delete(); + // Mark the account as deleted and shove low-level deletion tasks + // to background queues. Removing a lot of posts can take a while... + if (!$this->user->hasRole(Profile_role::DELETED)) { + $this->user->grantRole(Profile_role::DELETED); + } + + $qm = QueueManager::get(); + $qm->enqueue($this->user, 'deluser'); + Event::handle('EndDeleteUser', array($this, $this->user)); } } diff --git a/classes/Profile.php b/classes/Profile.php index 91f6e4692..eded1ff71 100644 --- a/classes/Profile.php +++ b/classes/Profile.php @@ -732,6 +732,9 @@ class Profile extends Memcached_DataObject function hasRight($right) { $result = false; + if ($this->hasRole(Profile_role::DELETED)) { + return false; + } if (Event::handle('UserRightsCheck', array($this, $right, &$result))) { switch ($right) { diff --git a/classes/Profile_role.php b/classes/Profile_role.php index d0a0b31f0..e7aa1f0f0 100644 --- a/classes/Profile_role.php +++ b/classes/Profile_role.php @@ -53,6 +53,7 @@ class Profile_role extends Memcached_DataObject const ADMINISTRATOR = 'administrator'; const SANDBOXED = 'sandboxed'; const SILENCED = 'silenced'; + const DELETED = 'deleted'; // Pending final deletion of notices... public static function isValid($role) { diff --git a/classes/Safe_DataObject.php b/classes/Safe_DataObject.php index 021f7b506..08bc6846f 100644 --- a/classes/Safe_DataObject.php +++ b/classes/Safe_DataObject.php @@ -43,6 +43,25 @@ class Safe_DataObject extends DB_DataObject } /** + * Magic function called at clone() time. + * + * We use this to drop connection with some global resources. + * This supports the fairly common pattern where individual + * items being read in a loop via a single object are cloned + * for individual processing, then fall out of scope when the + * loop comes around again. + * + * As that triggers the destructor, we want to make sure that + * the original object doesn't have its database result killed. + * It will still be freed properly when the original object + * gets destroyed. + */ + function __clone() + { + $this->_DB_resultid = false; + } + + /** * Magic function called at serialize() time. * * We use this to drop a couple process-specific references diff --git a/classes/Subscription.php b/classes/Subscription.php index 9cef2df1a..5ac95f922 100644 --- a/classes/Subscription.php +++ b/classes/Subscription.php @@ -105,8 +105,8 @@ class Subscription extends Memcached_DataObject $auto = new Subscription(); - $auto->subscriber = $subscriber->id; - $auto->subscribed = $other->id; + $auto->subscriber = $other->id; + $auto->subscribed = $subscriber->id; $auto->created = common_sql_now(); $result = $auto->insert(); diff --git a/lib/activity.php b/lib/activity.php index ae65fe36f..d84eabf7c 100644 --- a/lib/activity.php +++ b/lib/activity.php @@ -1083,15 +1083,11 @@ class Activity $this->entry = $entry; - // @fixme Don't send in a DOMDocument + // Insist on a feed's root DOMElement; don't allow a DOMDocument if ($feed instanceof DOMDocument) { - common_log( - LOG_WARNING, - 'Activity::__construct() - ' - . 'DOMDocument passed in for feed by mistake. ' - . "Expecting a 'feed' DOMElement." + throw new ClientException( + _("Expecting a root feed element but got a whole XML document.") ); - $feed = $feed->getElementsByTagName('feed')->item(0); } $this->feed = $feed; diff --git a/lib/deluserqueuehandler.php b/lib/deluserqueuehandler.php new file mode 100644 index 000000000..4a1233a5e --- /dev/null +++ b/lib/deluserqueuehandler.php @@ -0,0 +1,95 @@ +<?php +/* + * StatusNet - the distributed open-source microblogging tool + * Copyright (C) 2010, StatusNet, Inc. + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU Affero General Public License as published by + * the Free Software Foundation, either version 3 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Affero General Public License for more details. + * + * You should have received a copy of the GNU Affero General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +/** + * Background job to delete prolific users without disrupting front-end too much. + * + * Up to 50 messages are deleted on each run through; when all messages are gone, + * the actual account is deleted. + * + * @package QueueHandler + * @maintainer Brion Vibber <brion@status.net> + */ + +class DelUserQueueHandler extends QueueHandler +{ + const DELETION_WINDOW = 50; + + public function transport() + { + return 'deluser'; + } + + public function handle($user) + { + if (!($user instanceof User)) { + common_log(LOG_ERR, "Got a bogus user, not deleting"); + return true; + } + + $user = User::staticGet('id', $user->id); + if (!$user) { + common_log(LOG_INFO, "User {$user->nickname} was deleted before we got here."); + return true; + } + + if (!$user->hasRole(Profile_role::DELETED)) { + common_log(LOG_INFO, "User {$user->nickname} is not pending deletion; aborting."); + return true; + } + + $notice = $this->getNextBatch($user); + if ($notice->N) { + common_log(LOG_INFO, "Deleting next {$notice->N} notices by {$user->nickname}"); + while ($notice->fetch()) { + $del = clone($notice); + $del->delete(); + } + + // @todo improve reliability in case we died during the above deletions + // with a fatal error. If the job is lost, we should perform some kind + // of garbage collection later. + + // Queue up the next batch. + $qm = QueueManager::get(); + $qm->enqueue($user, 'deluser'); + } else { + // Out of notices? Let's finish deleting this guy! + $user->delete(); + common_log(LOG_INFO, "User $user->id $user->nickname deleted."); + return true; + } + + return true; + } + + /** + * Fetch the next self::DELETION_WINDOW messages for this user. + * @return Notice + */ + protected function getNextBatch(User $user) + { + $notice = new Notice(); + $notice->profile_id = $user->id; + $notice->limit(self::DELETION_WINDOW); + $notice->find(); + return $notice; + } + +} diff --git a/lib/queuemanager.php b/lib/queuemanager.php index 87bd356aa..0829c8a8b 100644 --- a/lib/queuemanager.php +++ b/lib/queuemanager.php @@ -264,6 +264,9 @@ abstract class QueueManager extends IoManager $this->connect('sms', 'SmsQueueHandler'); } + // Background user management tasks... + $this->connect('deluser', 'DelUserQueueHandler'); + // Broadcasting profile updates to OMB remote subscribers $this->connect('profile', 'ProfileQueueHandler'); diff --git a/lib/router.php b/lib/router.php index 706120e0b..a48ee875e 100644 --- a/lib/router.php +++ b/lib/router.php @@ -628,6 +628,12 @@ class Router array('action' => 'ApiTimelineTag', 'format' => '(xmljson|rss|atom)')); + // media related + $m->connect( + 'api/statusnet/media/upload', + array('action' => 'ApiMediaUpload') + ); + // search $m->connect('api/search.atom', array('action' => 'twitapisearchatom')); $m->connect('api/search.json', array('action' => 'twitapisearchjson')); diff --git a/lib/userprofile.php b/lib/userprofile.php index 1e4543a5a..ca060842b 100644 --- a/lib/userprofile.php +++ b/lib/userprofile.php @@ -229,6 +229,17 @@ class UserProfile extends Widget function showEntityActions() { + if ($this->profile->hasRole(Profile_role::DELETED)) { + $this->out->elementStart('div', 'entity_actions'); + $this->out->element('h2', null, _('User actions')); + $this->out->elementStart('ul'); + $this->out->elementStart('p', array('class' => 'profile_deleted')); + $this->out->text(_('User deletion in progress...')); + $this->out->elementEnd('p'); + $this->out->elementEnd('ul'); + $this->out->elementEnd('div'); + return; + } if (Event::handle('StartProfilePageActionsSection', array(&$this->out, $this->profile))) { $cur = common_current_user(); diff --git a/plugins/OStatus/actions/ostatussub.php b/plugins/OStatus/actions/ostatussub.php index 65dee2392..07081c2c6 100644 --- a/plugins/OStatus/actions/ostatussub.php +++ b/plugins/OStatus/actions/ostatussub.php @@ -149,7 +149,7 @@ class OStatusSubAction extends Action $fullname = $entity->fullname; $homepage = $entity->homepage; $location = $entity->location; - + if (!$avatar) { $avatar = Avatar::defaultImage(AVATAR_PROFILE_SIZE); } @@ -242,7 +242,7 @@ class OStatusSubAction extends Action if (Validate::email($this->profile_uri)) { $this->oprofile = Ostatus_profile::ensureWebfinger($this->profile_uri); } else if (Validate::uri($this->profile_uri)) { - $this->oprofile = Ostatus_profile::ensureProfile($this->profile_uri); + $this->oprofile = Ostatus_profile::ensureProfileURL($this->profile_uri); } else { $this->error = _m("Sorry, we could not reach that address. Please make sure that the OStatus address is like nickname@example.com or http://example.net/nickname"); common_debug('Invalid address format.', __FILE__); @@ -339,7 +339,6 @@ class OStatusSubAction extends Action } } - /** * Handle posts to this form * diff --git a/plugins/OStatus/classes/Ostatus_profile.php b/plugins/OStatus/classes/Ostatus_profile.php index 6ae8e4fd5..73f5d2322 100644 --- a/plugins/OStatus/classes/Ostatus_profile.php +++ b/plugins/OStatus/classes/Ostatus_profile.php @@ -708,18 +708,122 @@ class Ostatus_profile extends Memcached_DataObject * @return Ostatus_profile * @throws FeedSubException */ - public static function ensureProfile($profile_uri, $hints=array()) + + public static function ensureProfileURL($profile_url, $hints=array()) { - // Get the canonical feed URI and check it + $oprofile = self::getFromProfileURL($profile_url); + + if (!empty($oprofile)) { + return $oprofile; + } + + $hints['profileurl'] = $profile_url; + + // Fetch the URL + // XXX: HTTP caching + + $client = new HTTPClient(); + $client->setHeader('Accept', 'text/html,application/xhtml+xml'); + $response = $client->get($profile_url); + + if (!$response->isOk()) { + return null; + } + + // Check if we have a non-canonical URL + + $finalUrl = $response->getUrl(); + + if ($finalUrl != $profile_url) { + + $hints['profileurl'] = $finalUrl; + + $oprofile = self::getFromProfileURL($finalUrl); + + if (!empty($oprofile)) { + return $oprofile; + } + } + + // Try to get some hCard data + + $body = $response->getBody(); + + $hcardHints = DiscoveryHints::hcardHints($body, $finalUrl); + + if (!empty($hcardHints)) { + $hints = array_merge($hints, $hcardHints); + } + + // Check if they've got an LRDD header + + $lrdd = LinkHeader::getLink($response, 'lrdd', 'application/xrd+xml'); + + if (!empty($lrdd)) { + + $xrd = Discovery::fetchXrd($lrdd); + $xrdHints = DiscoveryHints::fromXRD($xrd); + + $hints = array_merge($hints, $xrdHints); + } + + // If discovery found a feedurl (probably from LRDD), use it. + + if (array_key_exists('feedurl', $hints)) { + return self::ensureFeedURL($hints['feedurl'], $hints); + } + + // Get the feed URL from HTML + $discover = new FeedDiscovery(); - if (isset($hints['feedurl'])) { - $feeduri = $hints['feedurl']; - $feeduri = $discover->discoverFromFeedURL($feeduri); - } else { - $feeduri = $discover->discoverFromURL($profile_uri); - $hints['feedurl'] = $feeduri; + + $feedurl = $discover->discoverFromHTML($finalUrl, $body); + + if (!empty($feedurl)) { + $hints['feedurl'] = $feedurl; + + return self::ensureFeedURL($feedurl, $hints); + } + } + + static function getFromProfileURL($profile_url) + { + $profile = Profile::staticGet('profileurl', $profile_url); + + if (empty($profile)) { + return null; + } + + // Is it a known Ostatus profile? + + $oprofile = Ostatus_profile::staticGet('profile_id', $profile->id); + + if (!empty($oprofile)) { + return $oprofile; } + // Is it a local user? + + $user = User::staticGet('id', $profile->id); + + if (!empty($user)) { + throw new Exception("'$profile_url' is the profile for local user '{$user->nickname}'."); + } + + // Continue discovery; it's a remote profile + // for OMB or some other protocol, may also + // support OStatus + + return null; + } + + public static function ensureFeedURL($feed_url, $hints=array()) + { + $discover = new FeedDiscovery(); + + $feeduri = $discover->discoverFromFeedURL($feed_url); + $hints['feedurl'] = $feeduri; + $huburi = $discover->getAtomLink('hub'); $hints['hub'] = $huburi; $salmonuri = $discover->getAtomLink(Salmon::NS_REPLIES); @@ -1303,7 +1407,7 @@ class Ostatus_profile extends Memcached_DataObject } } - // First, look it up + // Try looking it up $oprofile = Ostatus_profile::staticGet('uri', 'acct:'.$addr); @@ -1317,7 +1421,7 @@ class Ostatus_profile extends Memcached_DataObject $disco = new Discovery(); try { - $result = $disco->lookup($addr); + $xrd = $disco->lookup($addr); } catch (Exception $e) { // Save negative cache entry so we don't waste time looking it up again. // @fixme distinguish temporary failures? @@ -1327,38 +1431,26 @@ class Ostatus_profile extends Memcached_DataObject $hints = array('webfinger' => $addr); - foreach ($result->links as $link) { - switch ($link['rel']) { - case Discovery::PROFILEPAGE: - $hints['profileurl'] = $profileUrl = $link['href']; - break; - case Salmon::NS_REPLIES: - $hints['salmon'] = $salmonEndpoint = $link['href']; - break; - case Discovery::UPDATESFROM: - $hints['feedurl'] = $feedUrl = $link['href']; - break; - case Discovery::HCARD: - $hcardUrl = $link['href']; - break; - default: - common_log(LOG_NOTICE, "Don't know what to do with rel = '{$link['rel']}'"); - break; - } - } + $dhints = DiscoveryHints::fromXRD($xrd); + + $hints = array_merge($hints, $dhints); + + // If there's an Hcard, let's grab its info - if (isset($hcardUrl)) { - $hcardHints = self::slurpHcard($hcardUrl); - // Note: Webfinger > hcard - $hints = array_merge($hcardHints, $hints); + if (array_key_exists('hcard', $hints)) { + if (!array_key_exists('profileurl', $hints) || + $hints['hcard'] != $hints['profileurl']) { + $hcardHints = DiscoveryHints::fromHcardUrl($hints['hcard']); + $hints = array_merge($hcardHints, $hints); + } } // If we got a feed URL, try that - if (isset($feedUrl)) { + if (array_key_exists('feedurl', $hints)) { try { common_log(LOG_INFO, "Discovery on acct:$addr with feed URL $feedUrl"); - $oprofile = self::ensureProfile($feedUrl, $hints); + $oprofile = self::ensureFeedURL($hints['feedurl'], $hints); self::cacheSet(sprintf('ostatus_profile:webfinger:%s', $addr), $oprofile->uri); return $oprofile; } catch (Exception $e) { @@ -1369,10 +1461,10 @@ class Ostatus_profile extends Memcached_DataObject // If we got a profile page, try that! - if (isset($profileUrl)) { + if (array_key_exists('profileurl', $hints)) { try { common_log(LOG_INFO, "Discovery on acct:$addr with profile URL $profileUrl"); - $oprofile = self::ensureProfile($profileUrl, $hints); + $oprofile = self::ensureProfile($hints['profileurl'], $hints); self::cacheSet(sprintf('ostatus_profile:webfinger:%s', $addr), $oprofile->uri); return $oprofile; } catch (Exception $e) { @@ -1384,7 +1476,9 @@ class Ostatus_profile extends Memcached_DataObject // XXX: try hcard // XXX: try FOAF - if (isset($salmonEndpoint)) { + if (array_key_exists('salmon', $hints)) { + + $salmonEndpoint = $hints['salmon']; // An account URL, a salmon endpoint, and a dream? Not much to go // on, but let's give it a try @@ -1464,67 +1558,4 @@ class Ostatus_profile extends Memcached_DataObject return $file; } - - protected static function slurpHcard($url) - { - set_include_path(get_include_path() . PATH_SEPARATOR . INSTALLDIR . '/plugins/OStatus/extlib/hkit/'); - require_once('hkit.class.php'); - - $h = new hKit; - - // Google Buzz hcards need to be tidied. Probably others too. - - $h->tidy_mode = 'proxy'; // 'proxy', 'exec', 'php' or 'none' - - // Get by URL - $hcards = $h->getByURL('hcard', $url); - - if (empty($hcards)) { - return array(); - } - - // @fixme more intelligent guess on multi-hcard pages - $hcard = $hcards[0]; - - $hints = array(); - - $hints['profileurl'] = $url; - - if (array_key_exists('nickname', $hcard)) { - $hints['nickname'] = $hcard['nickname']; - } - - if (array_key_exists('fn', $hcard)) { - $hints['fullname'] = $hcard['fn']; - } else if (array_key_exists('n', $hcard)) { - $hints['fullname'] = implode(' ', $hcard['n']); - } - - if (array_key_exists('photo', $hcard)) { - $hints['avatar'] = $hcard['photo']; - } - - if (array_key_exists('note', $hcard)) { - $hints['bio'] = $hcard['note']; - } - - if (array_key_exists('adr', $hcard)) { - if (is_string($hcard['adr'])) { - $hints['location'] = $hcard['adr']; - } else if (is_array($hcard['adr'])) { - $hints['location'] = implode(' ', $hcard['adr']); - } - } - - if (array_key_exists('url', $hcard)) { - if (is_string($hcard['url'])) { - $hints['homepage'] = $hcard['url']; - } else if (is_array($hcard['url'])) { - // HACK get the last one; that's how our hcards look - $hints['homepage'] = $hcard['url'][count($hcard['url'])-1]; - } - } - - return $hints; - } } diff --git a/plugins/OStatus/lib/discovery.php b/plugins/OStatus/lib/discovery.php index df2fea64f..44fad62fb 100644 --- a/plugins/OStatus/lib/discovery.php +++ b/plugins/OStatus/lib/discovery.php @@ -40,7 +40,7 @@ class Discovery const PROFILEPAGE = 'http://webfinger.net/rel/profile-page'; const UPDATESFROM = 'http://schemas.google.com/g/2010#updates-from'; const HCARD = 'http://microformats.org/profile/hcard'; - + public $methods = array(); public function __construct() @@ -50,12 +50,11 @@ class Discovery $this->registerMethod('Discovery_LRDD_Link_HTML'); } - public function registerMethod($class) { $this->methods[] = $class; } - + /** * Given a "user id" make sure it's normalized to either a webfinger * acct: uri or a profile HTTP URL. @@ -78,7 +77,7 @@ class Discovery public static function isWebfinger($user_id) { $uri = Discovery::normalize($user_id); - + return (substr($uri, 0, 5) == 'acct:'); } @@ -99,7 +98,7 @@ class Discovery } else { $xrd_uri = $link['href']; } - + $xrd = $this->fetchXrd($xrd_uri); if ($xrd) { return $xrd; @@ -114,14 +113,13 @@ class Discovery if (!is_array($links)) { return false; } - + foreach ($links as $link) { if ($link['rel'] == $service) { return $link; } } } - public static function applyTemplate($template, $id) { @@ -130,7 +128,6 @@ class Discovery return $template; } - public static function fetchXrd($url) { try { @@ -172,7 +169,7 @@ class Discovery_LRDD_Host_Meta implements Discovery_LRDD if ($xrd->host != $domain) { return false; } - + return $xrd->links; } } @@ -188,7 +185,7 @@ class Discovery_LRDD_Link_Header implements Discovery_LRDD } catch (HTTP_Request2_Exception $e) { return false; } - + if ($response->getStatus() != 200) { return false; } @@ -197,51 +194,17 @@ class Discovery_LRDD_Link_Header implements Discovery_LRDD if (!$link_header) { // return false; } - + return Discovery_LRDD_Link_Header::parseHeader($link_header); } protected static function parseHeader($header) { - preg_match('/^<[^>]+>/', $header, $uri_reference); - //if (empty($uri_reference)) return; + $lh = new LinkHeader($header); - $links = array(); - - $link_uri = trim($uri_reference[0], '<>'); - $link_rel = array(); - $link_type = null; - - // remove uri-reference from header - $header = substr($header, strlen($uri_reference[0])); - - // parse link-params - $params = explode(';', $header); - - foreach ($params as $param) { - if (empty($param)) continue; - list($param_name, $param_value) = explode('=', $param, 2); - $param_name = trim($param_name); - $param_value = preg_replace('(^"|"$)', '', trim($param_value)); - - // for now we only care about 'rel' and 'type' link params - // TODO do something with the other links-params - switch ($param_name) { - case 'rel': - $link_rel = trim($param_value); - break; - - case 'type': - $link_type = trim($param_value); - } - } - - $links[] = array( - 'href' => $link_uri, - 'rel' => $link_rel, - 'type' => $link_type); - - return $links; + return array('href' => $lh->href, + 'rel' => $lh->rel, + 'type' => $lh->type); } } @@ -263,49 +226,48 @@ class Discovery_LRDD_Link_HTML implements Discovery_LRDD return Discovery_LRDD_Link_HTML::parse($response->getBody()); } - public function parse($html) { $links = array(); - + preg_match('/<head(\s[^>]*)?>(.*?)<\/head>/is', $html, $head_matches); $head_html = $head_matches[2]; - + preg_match_all('/<link\s[^>]*>/i', $head_html, $link_matches); - + foreach ($link_matches[0] as $link_html) { $link_url = null; $link_rel = null; $link_type = null; - + preg_match('/\srel=(("|\')([^\\2]*?)\\2|[^"\'\s]+)/i', $link_html, $rel_matches); if ( isset($rel_matches[3]) ) { $link_rel = $rel_matches[3]; } else if ( isset($rel_matches[1]) ) { $link_rel = $rel_matches[1]; } - + preg_match('/\shref=(("|\')([^\\2]*?)\\2|[^"\'\s]+)/i', $link_html, $href_matches); if ( isset($href_matches[3]) ) { $link_uri = $href_matches[3]; } else if ( isset($href_matches[1]) ) { $link_uri = $href_matches[1]; } - + preg_match('/\stype=(("|\')([^\\2]*?)\\2|[^"\'\s]+)/i', $link_html, $type_matches); if ( isset($type_matches[3]) ) { $link_type = $type_matches[3]; } else if ( isset($type_matches[1]) ) { $link_type = $type_matches[1]; } - + $links[] = array( 'href' => $link_url, 'rel' => $link_rel, 'type' => $link_type, ); } - + return $links; } } diff --git a/plugins/OStatus/lib/discoveryhints.php b/plugins/OStatus/lib/discoveryhints.php new file mode 100644 index 000000000..db13793dd --- /dev/null +++ b/plugins/OStatus/lib/discoveryhints.php @@ -0,0 +1,182 @@ +<?php +/* + * StatusNet - the distributed open-source microblogging tool + * Copyright (C) 2010, StatusNet, Inc. + * + * Some utilities for generating hint data + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU Affero General Public License as published by + * the Free Software Foundation, either version 3 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Affero General Public License for more details. + * + * You should have received a copy of the GNU Affero General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +class DiscoveryHints { + + static function fromXRD($xrd) + { + $hints = array(); + + foreach ($xrd->links as $link) { + switch ($link['rel']) { + case Discovery::PROFILEPAGE: + $hints['profileurl'] = $link['href']; + break; + case Salmon::NS_REPLIES: + $hints['salmon'] = $link['href']; + break; + case Discovery::UPDATESFROM: + $hints['feedurl'] = $link['href']; + break; + case Discovery::HCARD: + $hints['hcardurl'] = $link['href']; + break; + default: + break; + } + } + + return $hints; + } + + static function fromHcardUrl($url) + { + $client = new HTTPClient(); + $client->setHeader('Accept', 'text/html,application/xhtml+xml'); + $response = $client->get($url); + + if (!$response->isOk()) { + return null; + } + + return self::hcardHints($response->getBody(), + $response->getUrl()); + } + + static function hcardHints($body, $url) + { + common_debug("starting tidy"); + + $body = self::_tidy($body); + + common_debug("done with tidy"); + + set_include_path(get_include_path() . PATH_SEPARATOR . INSTALLDIR . '/plugins/OStatus/extlib/hkit/'); + require_once('hkit.class.php'); + + $h = new hKit; + + $hcards = $h->getByString('hcard', $body); + + if (empty($hcards)) { + return array(); + } + + if (count($hcards) == 1) { + $hcard = $hcards[0]; + } else { + foreach ($hcards as $try) { + if (array_key_exists('url', $try)) { + if (is_string($try['url']) && $try['url'] == $url) { + $hcard = $try; + break; + } else if (is_array($try['url'])) { + foreach ($try['url'] as $tryurl) { + if ($tryurl == $url) { + $hcard = $try; + break 2; + } + } + } + } + } + // last chance; grab the first one + if (empty($hcard)) { + $hcard = $hcards[0]; + } + } + + $hints = array(); + + if (array_key_exists('nickname', $hcard)) { + $hints['nickname'] = $hcard['nickname']; + } + + if (array_key_exists('fn', $hcard)) { + $hints['fullname'] = $hcard['fn']; + } else if (array_key_exists('n', $hcard)) { + $hints['fullname'] = implode(' ', $hcard['n']); + } + + if (array_key_exists('photo', $hcard)) { + $hints['avatar'] = $hcard['photo']; + } + + if (array_key_exists('note', $hcard)) { + $hints['bio'] = $hcard['note']; + } + + if (array_key_exists('adr', $hcard)) { + if (is_string($hcard['adr'])) { + $hints['location'] = $hcard['adr']; + } else if (is_array($hcard['adr'])) { + $hints['location'] = implode(' ', $hcard['adr']); + } + } + + if (array_key_exists('url', $hcard)) { + if (is_string($hcard['url'])) { + $hints['homepage'] = $hcard['url']; + } else if (is_array($hcard['url'])) { + // HACK get the last one; that's how our hcards look + $hints['homepage'] = $hcard['url'][count($hcard['url'])-1]; + } + } + + return $hints; + } + + private static function _tidy($body) + { + if (function_exists('tidy_parse_string')) { + common_debug("Tidying with extension"); + $text = tidy_parse_string($body); + $text = tidy_clean_repair($text); + return $body; + } else if ($fullpath = self::_findProgram('tidy')) { + common_debug("Tidying with program $fullpath"); + $tempfile = tempnam('/tmp', 'snht'); // statusnet hcard tidy + file_put_contents($tempfile, $source); + exec("$fullpath -utf8 -indent -asxhtml -numeric -bare -quiet $tempfile", $tidy); + unlink($tempfile); + return implode("\n", $tidy); + } else { + common_debug("Not tidying."); + return $body; + } + } + + private static function _findProgram($name) + { + $path = $_ENV['PATH']; + + $parts = explode(':', $path); + + foreach ($parts as $part) { + $fullpath = $part . '/' . $name; + if (is_executable($fullpath)) { + return $fullpath; + } + } + + return null; + } +} diff --git a/plugins/OStatus/lib/feeddiscovery.php b/plugins/OStatus/lib/feeddiscovery.php index 7761ea583..7de80b335 100644 --- a/plugins/OStatus/lib/feeddiscovery.php +++ b/plugins/OStatus/lib/feeddiscovery.php @@ -117,7 +117,7 @@ class FeedDiscovery return $this->discoverFromURL($target, false); } } - + return $this->initFromResponse($response); } @@ -202,7 +202,7 @@ class FeedDiscovery 'application/atom+xml' => false, 'application/rss+xml' => false, ); - + $nodes = $dom->getElementsByTagName('link'); for ($i = 0; $i < $nodes->length; $i++) { $node = $nodes->item($i); diff --git a/plugins/OStatus/lib/linkheader.php b/plugins/OStatus/lib/linkheader.php new file mode 100644 index 000000000..2f6c66dc9 --- /dev/null +++ b/plugins/OStatus/lib/linkheader.php @@ -0,0 +1,63 @@ +<?php + +class LinkHeader +{ + var $href; + var $rel; + var $type; + + function __construct($str) + { + preg_match('/^<[^>]+>/', $str, $uri_reference); + //if (empty($uri_reference)) return; + + $this->uri = trim($uri_reference[0], '<>'); + $this->rel = array(); + $this->type = null; + + // remove uri-reference from header + $str = substr($str, strlen($uri_reference[0])); + + // parse link-params + $params = explode(';', $str); + + foreach ($params as $param) { + if (empty($param)) continue; + list($param_name, $param_value) = explode('=', $param, 2); + $param_name = trim($param_name); + $param_value = preg_replace('(^"|"$)', '', trim($param_value)); + + // for now we only care about 'rel' and 'type' link params + // TODO do something with the other links-params + switch ($param_name) { + case 'rel': + $this->rel = trim($param_value); + break; + + case 'type': + $this->type = trim($param_value); + } + } + } + + static function getLink($response, $rel=null, $type=null) + { + $headers = $response->getHeader('Link'); + + // Can get an array or string, so try to simplify the path + if (!is_array($headers)) { + $headers = array($headers); + } + + foreach ($headers as $header) { + $lh = new LinkHeader($header); + + if ((is_null($rel) || $lh->rel == $rel) && + (is_null($type) || $lh->type == $type)) { + return $lh->href; + } + } + + return null; + } +}
\ No newline at end of file diff --git a/scripts/fixup_files.php b/scripts/fixup_files.php new file mode 100755 index 000000000..18feaf221 --- /dev/null +++ b/scripts/fixup_files.php @@ -0,0 +1,77 @@ +#!/usr/bin/env php +<?php +/* + * StatusNet - a distributed open-source microblogging tool + * Copyright (C) 2010 StatusNet, Inc. + * + * This program is free software: you can redistribute it and/or modify + * it under the terms of the GNU Affero General Public License as published by + * the Free Software Foundation, either version 3 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU Affero General Public License for more details. + * + * You should have received a copy of the GNU Affero General Public License + * along with this program. If not, see <http://www.gnu.org/licenses/>. + */ + +define('INSTALLDIR', realpath(dirname(__FILE__) . '/..')); + +$longoptions = array('dry-run'); + +$helptext = <<<END_OF_USERROLE_HELP +fixup_files.php [options] +Patches up file entries with corrupted types and titles (the "h bug"). + + --dry-run look but don't touch + +END_OF_USERROLE_HELP; + +require_once INSTALLDIR.'/scripts/commandline.inc'; + +$dry = have_option('dry-run'); + +$f = new File(); +$f->title = 'h'; +$f->mimetype = 'h'; +$f->size = 0; +$f->protected = 0; +$f->find(); +echo "Found $f->N bad items:\n"; + +while ($f->fetch()) { + echo "$f->id $f->url"; + + $data = File_redirection::lookupWhere($f->url); + if ($dry) { + if (is_array($data)) { + echo " (unchanged)\n"; + } else { + echo " (unchanged, but embedding lookup failed)\n"; + } + } else { + // NULL out the mime/title/size/protected fields + $sql = sprintf("UPDATE file " . + "SET mimetype=null,title=null,size=null,protected=null " . + "WHERE id=%d", + $f->id); + $f->query($sql); + $f->decache(); + + if (is_array($data)) { + if ($f->saveOembed($data, $f->url)) { + echo " (ok)\n"; + } else { + echo " (ok, no embedding data)\n"; + } + } else { + echo " (ok, but embedding lookup failed)\n"; + } + } +} + +echo "done.\n"; + |