function process($hdrList, $curMsg, $endMsg)
{
$this->displayStatus("progress", $curMsg . " till " . $endMsg);
$this->_db->beginTransaction();
$signedCount = 0;
$hdrsRetrieved = 0;
$fullsRetrieved = 0;
$modCount = 0;
$skipCount = 0;
# pak onze lijst met messageid's, en kijk welke er al in de database zitten
$t = microtime(true);
$dbIdList = $this->_db->matchMessageIds($hdrList);
# en loop door elke header heen
foreach ($hdrList as $msgid => $msgheader) {
# Reset timelimit
set_time_limit(120);
# messageid to check
$msgId = substr($msgheader['Message-ID'], 1, -1);
# als we de spot overview nog niet in de database hebben, haal hem dan op
if (!in_array($msgId, $dbIdList['spot'])) {
$hdrsRetrieved++;
$spotParser = new SpotParser();
$spot = $spotParser->parseXover($msgheader['Subject'], $msgheader['From'], $msgheader['Message-ID'], $this->_rsakeys);
# als er een parse error was, negeren we de spot volledig, ook niet
# verified spots gooien we weg.
if ($spot === false || !$spot['Verified']) {
continue;
}
# if
if ($spot['KeyID'] == 2) {
$commandAr = explode(' ', strtolower($spot['Title']));
$validCommands = array('delete', 'dispose', 'remove');
# is dit een geldig commando?
if (array_search($commandAr[0], $validCommands) !== false) {
switch ($this->_settings['spot_moderation']) {
case 'disable':
break;
case 'markspot':
$this->_db->markSpotModerated($commandAr[1]);
break;
default:
$this->_db->deleteSpot($commandAr[1]);
break;
}
# switch
$modCount++;
}
# if
} else {
// Oudere spots niet toevoegen, hoeven we het later ook niet te verwijderen
if ($this->_settings['retention'] > 0 && $spot['Stamp'] < time() - $this->_settings['retention'] * 24 * 60 * 60) {
$skipCount++;
} else {
$this->_db->addSpot($spot);
$dbIdList['spot'][] = $msgId;
if ($spot['WasSigned']) {
$signedCount++;
}
# if
}
# if
}
# else
} else {
# anders halen we hem uit de database want we hebben die nodig
$spot = $this->_db->getFullSpot($msgId);
}
# else
# We willen enkel de volledige spot ophalen als de header in de database zit, omdat
# we dat hierboven eventueel doen, is het enkel daarop checken voldoende
if (in_array($msgId, $dbIdList['spot']) && !in_array($msgId, $dbIdList['fullspot'])) {
#
# We gebruiken altijd XOVER, dit is namelijk handig omdat eventueel ontbrekende
# artikel nummers (en soms zijn dat er duizenden) niet hoeven op te vragen, nu
# vragen we enkel de de headers op van de artikelen die er daadwerkelijk zijn
#
# KeyID 2 is een 'moderator' post en kan dus niet getrieved worden
#
if ($this->_retrieveFull && $spot['KeyID'] != 2) {
$fullSpot = array();
try {
$fullsRetrieved++;
$fullSpot = $this->_spotnntp->getFullSpot(substr($msgheader['Message-ID'], 1, -1));
# en voeg hem aan de database toe
$this->_db->addFullSpot($fullSpot);
} catch (ParseSpotXmlException $x) {
# swallow error
} catch (Exception $x) {
# messed up index aan de kant van de server ofzo? iig, dit gebeurt. soms, if so,
# swallow the error
if ($x->getMessage() == 'No such article found') {
} elseif ($x->getMessage() == 'String could not be parsed as XML') {
} else {
throw $x;
}
# else
}
# catch
}
# if retrievefull
}
# if fullspot is not in db yet
}
# foreach
if (count($hdrList) > 0) {
$this->displayStatus("hdrparsed", $hdrsRetrieved);
$this->displayStatus("fullretrieved", $fullsRetrieved);
$this->displayStatus("verified", $signedCount);
$this->displayStatus("modcount", $modCount);
$this->displayStatus("skipcount", $skipCount);
$this->displayStatus("loopcount", count($hdrList));
} else {
$this->displayStatus("hdrparsed", 0);
$this->displayStatus("fullretrieved", 0);
$this->displayStatus("verified", 0);
$this->displayStatus("modcount", 0);
$this->displayStatus("skipcount", 0);
$this->displayStatus("loopcount", 0);
}
# else
$this->_db->setMaxArticleid($this->_server['host'], $curMsg);
$this->_db->commitTransaction();
return count($hdrList);
}
function process($hdrList, $curMsg, $increment)
{
$this->displayStatus("progress", $curMsg . " till " . ($curMsg + $increment));
$this->_db->beginTransaction();
$signedCount = 0;
foreach ($hdrList as $msgid => $msgheader) {
# Reset timelimit
set_time_limit(120);
$spotParser = new SpotParser();
$spot = $spotParser->parseXover($msgheader['Subject'], $msgheader['From'], $msgheader['Message-ID'], $this->_rsakeys);
if ($spot != null && $spot['Verified']) {
$this->_db->addSpot($spot);
}
# if
if ($spot['Verified']) {
if ($spot['WasSigned']) {
$signedCount++;
}
# if
}
# if
}
# foreach
if (count($hdrList) > 0) {
$this->displayStatus("verified", $signedCount);
$this->displayStatus("loopcount", count($hdrList));
} else {
$this->displayStatus("verified", 0);
$this->displayStatus("loopcount", 0);
}
# else
$this->_db->setMaxArticleid($this->_server['host'], $curMsg);
$this->_db->commitTransaction();
return count($hdrList);
}
function process($hdrList, $curMsg, $endMsg, $timer)
{
$this->displayStatus("progress", $curMsg . " till " . $endMsg);
$spotParser = new SpotParser();
$signedCount = 0;
$hdrsRetrieved = 0;
$fullsRetrieved = 0;
$msgCounter = 0;
$modCount = 0;
$skipCount = 0;
$lastProcessedId = '';
$fullSpotDbList = array();
$spotDbList = array();
$moderationList = array();
$processingStartTime = time();
/*
* Determine the cutoff date (unixtimestamp) from whereon we do not want to
* load the spots
*/
if ($this->_settings->get('retention') > 0) {
$retentionStamp = time() - $this->_settings->get('retention') * 24 * 60 * 60;
} else {
$retentionStamp = 0;
}
# else
$this->debug('retentionStamp=' . $retentionStamp);
$this->debug('hdrList=' . serialize($hdrList));
/**
* We ask the database to match our messageid's we just retrieved with
* the list of id's we have just retrieved from the server
*/
$dbIdList = $this->_db->matchSpotMessageIds($hdrList);
$this->debug('dbIdList=' . serialize($dbIdList));
# if we need to fetch images or nzb files, we need an spotsoverview instance
if ($this->_retrieveFull && $this->_prefetch_image || $this->_prefetch_nzb) {
$spotsOverview = new SpotsOverview($this->_db, $this->_settings);
$spotsOverview->setActiveRetriever(true);
/*
* Only create a new NZB instance if the server differs from the
* header host, else re-use the connection
*/
$settings_nntp_nzb = $this->_settings->get('nntp_nzb');
if ($this->_server['host'] == $settings_nntp_nzb['host']) {
$nntp_nzb = $this->_spotnntp;
} else {
$nntp_nzb = new SpotNntp($settings_nntp_nzb);
$nntp_nzb->selectGroup($this->_settings->get('nzb_group'));
}
# else
}
# if
foreach ($hdrList as $msgheader) {
$msgCounter++;
$this->debug('foreach-loop, start. msgId= ' . $msgCounter);
/*
* Keep te usenet server alive when processing is slow.
*/
if ($processingStartTime - time() > 30) {
$this->_spotnntp->sendNoop();
if (isset($nntp_nzb) && $nntp_nzb != $this->_spotnntp) {
$nntp_nzb->sendNoop();
}
# if
$processingStartTime = time();
}
# if
/*
* We keep track whether we actually fetched this header and fullspot
* to add it to the database, because only then we can update the
* titel from the spots title or rely on our database to fetch
* the fullspot
*/
$didFetchHeader = false;
$didFetchFullSpot = false;
# Reset timelimit
set_time_limit(120);
# messageid to check
$msgId = substr($msgheader['Message-ID'], 1, -1);
/*
* We prepare some variables to we don't have to perform an array
* lookup for each check and the code is easier to read.
*/
$header_isInDb = isset($dbIdList['spot'][$msgId]);
$fullspot_isInDb = isset($dbIdList['fullspot'][$msgId]);
/*
* If the spotheader is not yet added to the database, parse the header
* information.
*
* If the header is present, but we don't have the fullspot yet or we are
* running in 'retro' mode, parse the header as well because some fields
* are only in the header and not in the full.
*
* We need some of those fields (for example KeyID)
*/
if (!$header_isInDb || (!$fullspot_isInDb || $this->_retro) && $this->_retrieveFull) {
$hdrsRetrieved++;
$this->debug('foreach-loop, parsingXover, start. msgId= ' . $msgCounter);
$spot = $spotParser->parseXover($msgheader['Subject'], $msgheader['From'], $msgheader['Date'], $msgheader['Message-ID'], $this->_rsakeys);
$this->debug('foreach-loop, parsingXover, done. msgId= ' . $msgCounter);
/*
* When a parse error occured, we ignore the spot, also unverified
* spots are ignored
*/
if ($spot === false || !$spot['verified']) {
$this->debug('foreach-loop, spot is either false or not verified');
continue;
}
# if
/*
* Special moderator commands always have keyid 2
*/
if ($spot['keyid'] == 2) {
$this->debug('foreach-loop, spot is a moderation spot');
$commandAr = explode(' ', strtolower($spot['title']));
$validCommands = array('delete', 'dispose', 'remove');
# is this one of the defined valid commands?
if (in_array($commandAr[0], $validCommands) !== false) {
$moderationList[$commandAr[1]] = 1;
$modCount++;
}
# if
} else {
/*
* Don't add spots older than specified for the retention stamp
*/
if ($retentionStamp > 0 && $spot['stamp'] < $retentionStamp && $this->_settings->get('retentiontype') == 'everything') {
$this->debug('foreach-loop, spot is expired: ' . $spot['stamp']);
continue;
} elseif ($spot['stamp'] < $this->_settings->get('retrieve_newer_than')) {
$this->debug('foreach-loop, spot is too old: ' . $spot['stamp']);
$skipCount++;
} else {
/*
* Do we have the header in the database? If not, lets add it
*/
if (!$header_isInDb) {
$spotDbList[] = $spot;
/*
* Some buggy NNTP servers give us the same messageid
* in one XOVER statement, hence we update the list of
* messageid's we already have retrieved and are ready
* to be added to the database
*/
$dbIdList['spot'][$msgId] = 1;
$header_isInDb = true;
$lastProcessedId = $msgId;
$didFetchHeader = true;
if ($spot['wassigned']) {
$signedCount++;
}
# if
}
# if
}
# if
}
# else
} else {
$lastProcessedId = $msgId;
}
# else
/*
* We don't want to retrieve the fullspot if we don't have the header
* in the database. Because we try to add headers in the above code we just have
* to check if the header is in the database.
*
* We cannot collapse this code with the header fetching code because we want to
* be able to add the fullspot to a system after all the headers are retrieved
*/
if ($header_isInDb && !$fullspot_isInDb) {
/*
* Don't add older fullspots than specified for the retention stamp
*/
if ($retentionStamp > 0 && strtotime($msgheader['Date']) < $retentionStamp) {
continue;
}
# if
if ($this->_retrieveFull) {
$fullSpot = array();
try {
$fullsRetrieved++;
$this->debug('foreach-loop, getFullSpot, start. msgId= ' . $msgId);
$fullSpot = $this->_spotnntp->getFullSpot($msgId);
$this->debug('foreach-loop, getFullSpot, done. msgId= ' . $msgId);
# add this spot to the database
$fullSpotDbList[] = $fullSpot;
$fullspot_isInDb = true;
$didFetchFullSpot = true;
/*
* Some buggy NNTP servers give us the same messageid
* in once XOVER statement, hence we update the list of
* messageid's we already have retrieved and are ready
* to be added to the database
*/
$dbIdList['fullspot'][$msgId] = 1;
/*
* Overwrite the spots' title because the fullspot contains the title in
* UTF-8 format.
* We also overwrite the spotterid from the spotsfull because the spotterid
* is only in the header in more recent spots.
*/
if ($didFetchHeader) {
$spotDbList[count($spotDbList) - 1]['title'] = $fullSpot['title'];
$spotDbList[count($spotDbList) - 1]['spotterid'] = $fullSpot['spotterid'];
}
# if
} catch (ParseSpotXmlException $x) {
# swallow error
} catch (Exception $x) {
/**
* Sometimes we get an 'No such article' error for a header we just retrieved,
* if we want to retrieve the full article. This is messed up, but let's just
* swallow the error
*/
if ($x->getCode() == 430) {
} elseif ($x->getMessage() == 'String could not be parsed as XML') {
} else {
throw $x;
}
# else
}
# catch
}
# if retrievefull
}
# if fullspot is not in db yet
if ($this->_retrieveFull && $header_isInDb && ($this->_prefetch_image || $this->_prefetch_nzb)) {
try {
/*
* If we are running in 'retro' mode, it is possible both the header and spot are in the
* database already, however -- we need the information from the fullspot so we retrieve it
* again
*/
if (!$didFetchFullSpot) {
$fullSpot = $this->_db->getFullSpot($msgId, SPOTWEB_ANONYMOUS_USERID);
$fullSpot = array_merge($spotParser->parseFull($fullSpot['fullxml']), $fullSpot);
}
# if
/*
* Prefetch (cache) the spots' image
*/
if ($this->_prefetch_image) {
/*
* If the spot is older than 30 days, and the image is on the web, we do not
* prefetch the image.
*/
if (is_array($fullSpot['image']) || $fullSpot['stamp'] > (int) time() - 30 * 24 * 60 * 60) {
$this->debug('foreach-loop, getImage(), start. msgId= ' . $msgId);
$spotsOverview->getImage($fullSpot, $nntp_nzb);
$this->debug('foreach-loop, getImage(), done. msgId= ' . $msgId);
}
# if
}
# if
/*
* Prefetch (cache) the spots' NZB file
*/
if ($this->_prefetch_nzb) {
/*
* Only do so if we can expect an NZB file
*/
if (!empty($fullSpot['nzb']) && $fullSpot['stamp'] > 1290578400) {
$this->debug('foreach-loop, getNzb(), start. msgId= ' . $msgId);
$spotsOverview->getNzb($fullSpot, $nntp_nzb);
$this->debug('foreach-loop, getNzb(), done. msgId= ' . $msgId);
}
# if
}
# if
} catch (ParseSpotXmlException $x) {
# swallow error
} catch (Exception $x) {
/**
* Sometimes we get an 'No such article' error for a header we just retrieved,
* if we want to retrieve the full article. This is messed up, but let's just
* swallow the error
*/
if ($x->getCode() == 430) {
} elseif ($x->getMessage() == 'String could not be parsed as XML') {
} else {
throw $x;
}
# else
}
# catch
}
# if prefetch image and/or nzb
$this->debug('foreach-loop, done. msgId= ' . $msgCounter);
}
# foreach
if (count($hdrList) > 0) {
$this->displayStatus("hdrparsed", $hdrsRetrieved);
$this->displayStatus("fullretrieved", $fullsRetrieved);
$this->displayStatus("verified", $signedCount);
$this->displayStatus("modcount", $modCount);
$this->displayStatus("skipcount", $skipCount);
$this->displayStatus("loopcount", count($hdrList));
} else {
$this->displayStatus("hdrparsed", 0);
$this->displayStatus("fullretrieved", 0);
$this->displayStatus("verified", 0);
$this->displayStatus("modcount", 0);
$this->displayStatus("skipcount", 0);
$this->displayStatus("loopcount", 0);
}
# else
/*
* Add the spots to the database and update the last article
* number found
*/
$this->_db->addSpots($spotDbList, $fullSpotDbList);
$this->debug('added Spots, spotDbList=' . serialize($spotDbList));
$this->debug('added Spots, fullSpotDbList=' . serialize($fullSpotDbList));
/*
* Actually act on the moderation settings. We cannot process this inline
* because a spot can be added and moderated within the same iteration
*/
switch ($this->_settings->get('spot_moderation')) {
case 'disable':
break;
case 'markspot':
$this->_db->markCommentsModerated($moderationList);
$this->_db->markSpotsModerated($moderationList);
break;
# case 'markspot'
# case 'markspot'
default:
$this->_db->removeSpots($moderationList);
$this->_db->removeComments($moderationList);
break;
# default
}
# switch
# update the maximum article id
if ($this->_retro) {
$this->_db->setMaxArticleid('spots_retro', $endMsg);
} else {
$this->_db->setMaxArticleid($this->_server['host'], $endMsg);
}
# if
$this->debug('loop finished, setMaxArticleId=' . serialize($endMsg));
$this->displayStatus("timer", round(microtime(true) - $timer, 2));
return array('count' => count($hdrList), 'headercount' => $hdrsRetrieved, 'lastmsgid' => $lastProcessedId);
}
require_once "lib/SpotSecurity.php";
require_once "lib/SpotTiming.php";
require_once "settings.php";
require_once "lib/SpotDb.php";
define('USERID', 30);
/* -------------------------------------------------------------------- */
echo "Included PHP classes... " . PHP_EOL;
$db = new SpotDb($settings['db']);
$db->connect();
echo "Connected to the database.." . PHP_EOL;
$spotSettings = SpotSettings::singleton($db, $settings);
$spotSigning = new SpotSigning();
$spotPosting = new SpotPosting($db, $spotSettings);
$spotUserSystem = new SpotUserSystem($db, $spotSettings);
echo "Initialized classes.." . PHP_EOL;
$rsaKeys = $spotSettings->get('rsa_keys');
$retriever = new SpotRetriever_Spots($spotSettings->get('nntp_hdr'), $db, $spotSettings, $rsaKeys, '', $spotSettings->get('retrieve_full'));
$msgdata = $retriever->connect($spotSettings->get('hdr_group'));
var_dump($msgdata);
# Connct thru our own NNTP session to the server so we have an XOVER list
$headerServer = $spotSettings->get('nntp_hdr');
$spotnntp = new SpotNntp($spotSettings->get('nntp_hdr'));
$spotnntp->selectGroup($spotSettings->get('hdr_group'));
$hdrList = $spotnntp->getOverview(3244937, 3244938);
foreach ($hdrList as $msgid => $msgheader) {
$spotParser = new SpotParser();
$spot = $spotParser->parseXover($msgheader['Subject'], $msgheader['From'], $msgheader['Date'], $msgheader['Message-ID'], $rsaKeys);
var_dump($spot);
}
# and signal quit
$retriever->quit();
function process($hdrList, $curMsg, $endMsg)
{
$this->displayStatus("progress", $curMsg . " till " . $endMsg);
$this->_db->beginTransaction();
$signedCount = 0;
$hdrsRetrieved = 0;
$fullsRetrieved = 0;
$modCount = 0;
$skipCount = 0;
$lastProcessedId = '';
# pak onze lijst met messageid's, en kijk welke er al in de database zitten
$dbIdList = $this->_db->matchSpotMessageIds($hdrList);
#var_dump($hdrList);
# en loop door elke header heen
$spotParser = new SpotParser();
foreach ($hdrList as $msgid => $msgheader) {
# Reset timelimit
set_time_limit(120);
# messageid to check
$msgId = substr($msgheader['Message-ID'], 1, -1);
# definieer een paar booleans zodat we niet steeds een array lookup moeten doen
# en de code wat duidelijker is
$header_isInDb = isset($dbIdList['spot'][$msgId]);
$fullspot_isInDb = isset($dbIdList['fullspot'][$msgId]);
# als we de spot overview nog niet in de database hebben, haal hem dan op,
# ook als de fullspot er nog niet is, moeten we dit doen want een aantal velden
# die wel in de header zitten, zitten niet in de database (denk aan 'keyid')
if (!$header_isInDb || !$fullspot_isInDb) {
$hdrsRetrieved++;
$spot = $spotParser->parseXover($msgheader['Subject'], $msgheader['From'], $msgheader['Date'], $msgheader['Message-ID'], $this->_rsakeys);
# als er een parse error was, negeren we de spot volledig, ook niet-
# verified spots gooien we weg.
if ($spot === false || !$spot['verified']) {
continue;
}
# if
if ($spot['keyid'] == 2) {
$commandAr = explode(' ', strtolower($spot['title']));
$validCommands = array('delete', 'dispose', 'remove');
# FIXME: Message-ID kan ook van een comment zijn,
# onderstaande code gaat uit van een spot.
# is dit een geldig commando?
if (in_array($commandAr[0], $validCommands) !== false) {
switch ($this->_settings->get('spot_moderation')) {
case 'disable':
break;
case 'markspot':
$this->_db->markSpotModerated($commandAr[1]);
break;
default:
$this->_db->deleteSpot($commandAr[1]);
break;
}
# switch
$modCount++;
}
# if
} else {
# Oudere spots niet toevoegen, hoeven we het later ook niet te verwijderen
if ($this->_settings->get('retention') > 0 && $spot['stamp'] < time() - $this->_settings->get('retention') * 24 * 60 * 60) {
$skipCount++;
} elseif ($spot['stamp'] < $this->_settings->get('retrieve_newer_than')) {
$skipCount++;
} else {
# Hier kijken we alleen of de spotheader niet bestaat
if (!$header_isInDb) {
$this->_db->addSpot($spot);
# definieer de header als al ontvangen, we moeten ook de
# msgid lijst updaten omdat soms een messageid meerdere
# keren per xover mee komt ...
$dbIdList['spot'][$msgId] = 1;
$header_isInDb = true;
$lastProcessedId = $msgId;
if ($spot['wassigned']) {
$signedCount++;
}
# if
}
# if
}
# if
}
# else
} else {
$lastProcessedId = $msgId;
}
# else
# We willen enkel de volledige spot ophalen als de header in de database zit, omdat
# we dat hierboven eventueel doen, is het enkel daarop checken voldoende
if ($header_isInDb && !$fullspot_isInDb) {
#
# We gebruiken altijd XOVER, dit is namelijk handig omdat eventueel ontbrekende
# artikel nummers (en soms zijn dat er duizenden) niet hoeven op te vragen, nu
# vragen we enkel de de headers op van de artikelen die er daadwerkelijk zijn
#
# KeyID 2 is een 'moderator' post en kan dus niet getrieved worden
#
if ($this->_retrieveFull && $spot['keyid'] != 2) {
$fullSpot = array();
try {
$fullsRetrieved++;
$fullSpot = $this->_spotnntp->getFullSpot($msgId);
# en voeg hem aan de database toe
$this->_db->addFullSpot($fullSpot);
$fullspot_isInDb = true;
# we moeten ook de msgid lijst updaten omdat soms een messageid meerdere
# keren per xover mee komt ...
$dbIdList['fullspot'][$msgId] = 1;
} catch (ParseSpotXmlException $x) {
# swallow error
} catch (Exception $x) {
# messed up index aan de kant van de server ofzo? iig, dit gebeurt. soms, if so,
# swallow the error
if ($x->getMessage() == 'No such article found') {
} elseif ($x->getMessage() == 'String could not be parsed as XML') {
} else {
throw $x;
}
# else
}
# catch
}
# if retrievefull
}
# if fullspot is not in db yet
}
# foreach
if (count($hdrList) > 0) {
$this->displayStatus("hdrparsed", $hdrsRetrieved);
$this->displayStatus("fullretrieved", $fullsRetrieved);
$this->displayStatus("verified", $signedCount);
$this->displayStatus("modcount", $modCount);
$this->displayStatus("skipcount", $skipCount);
$this->displayStatus("loopcount", count($hdrList));
} else {
$this->displayStatus("hdrparsed", 0);
$this->displayStatus("fullretrieved", 0);
$this->displayStatus("verified", 0);
$this->displayStatus("modcount", 0);
$this->displayStatus("skipcount", 0);
$this->displayStatus("loopcount", 0);
}
# else
$this->_db->setMaxArticleid($this->_server['host'], $curMsg);
$this->_db->commitTransaction();
return array('count' => count($hdrList), 'headercount' => $hdrsRetrieved, 'lastmsgid' => $lastProcessedId);
}