public static function Update($keyID, $vCode)
{
$row_count = 500;
// Start and validate they key pair
BaseApi::bootstrap();
BaseApi::validateKeyPair($keyID, $vCode);
// Set key scopes and check if the call is banned
$scope = 'Corp';
$api = 'WalletTransactions';
if (BaseApi::isBannedCall($api, $scope, $keyID)) {
return;
}
// Get the characters for this key
$characters = BaseApi::findKeyCharacters($keyID);
// Check if this key has any characters associated with it
if (!$characters) {
return;
}
// Lock the call so that we are the only instance of this running now()
// If it is already locked, just return without doing anything
if (!BaseApi::isLockedCall($api, $scope, $keyID)) {
$lockhash = BaseApi::lockCall($api, $scope, $keyID);
} else {
return;
}
// So I think a corporation key will only ever have one character
// attached to it. So we will just use that characterID for auth
// things, but the corporationID for locking etc.
$corporationID = BaseApi::findCharacterCorporation($characters[0]);
// Prepare the Pheal instance
$pheal = new Pheal($keyID, $vCode);
// Prepare blank wallet_transactions to start
$wallet_transactions = null;
// Next, start our loop over the wallet divisions for this corporation
foreach (\EveCorporationAccountBalance::where('corporationID', '=', $corporationID)->get() as $walletdivision) {
// Start a infinite loop for the Journal Walking. We will break out of this once
// we have reached the end of the records that we can get
// TODO: This needs a lot more brain thingies applied in order to figure out how
// we are going to go about the database cached_untill timer. For now, we will just
// ignore the DB level one and rely entirely on pheal-ng to cache the XML's
$first_request = true;
$from_id = PHP_INT_MAX;
// Use the maximum size for this PHP arch
while (true) {
// Do the actual API call. pheal-ng actually handles some internal
// caching too.
try {
if ($first_request) {
$wallet_transactions = $pheal->corpScope->WalletTransactions(array('characterID' => $characters[0], 'rowCount' => $row_count, 'accountKey' => $walletdivision->accountKey));
// flip the first_request as those that get processed from here need to be from the `fromID`
$first_request = false;
} else {
$wallet_transactions = $pheal->corpScope->WalletTransactions(array('characterID' => $characters[0], 'rowCount' => $row_count, 'accountKey' => $walletdivision->accountKey, 'fromID' => $from_id));
}
} catch (\Pheal\Exceptions\APIException $e) {
// If we cant get account status information, prevent us from calling
// this API again
BaseApi::banCall($api, $scope, $keyID, 0, $e->getCode() . ': ' . $e->getMessage());
return;
} catch (\Pheal\Exceptions\PhealException $e) {
throw $e;
}
// Process the transactions
foreach ($wallet_transactions->transactions as $transaction) {
// Ensure that $from_id is at its lowest
$from_id = min($transaction->transactionID, $from_id);
// Generate a transaction hash. It would seem that refID's could possibly be cycled.
$transaction_hash = md5(implode(',', array($corporationID, $walletdivision->accountKey, $transaction->transactionDateTime, $transaction->clientID, $transaction->transactionID)));
// In order try try and relieve some strain on the database, we will
// cache the hashes that we find we have knowledge of. The main
// goal of this is to make the lookups faster, and leave
// MySQL to do stuff it should rather be doing
// First, find thee entry in the cache. The cache key is determined
// as: <transaction hash + corporationID + apitype>
if (!\Cache::has($transaction_hash . $corporationID . $api)) {
// If the $transaction_hash is not in the Cache, ask the database
// if it knows about it. Again, if it exists, we will continue,
// but we will also add a new Cache entry to make the next
// lookip faster
$transaction_data = \EveCorporationWalletTransactions::where('corporationID', '=', $corporationID)->where('hash', '=', $transaction_hash)->first();
// Check if the database found the entry.
if (!$transaction_data) {
$transaction_data = new \EveCorporationWalletTransactions();
} else {
// This entry exists in the database. Put a new Cache entry
// so that the next lookup may be faster. This cache
// entry will live for 1 week.
\Cache::put($transaction_hash . $corporationID . $api, true, 60 * 24 * 7);
// Continue to the next transaction
continue;
}
$transaction_data->corporationID = $corporationID;
$transaction_data->hash = $transaction_hash;
$transaction_data->accountKey = $walletdivision->accountKey;
$transaction_data->transactionID = $transaction->transactionID;
$transaction_data->transactionDateTime = $transaction->transactionDateTime;
$transaction_data->quantity = $transaction->quantity;
$transaction_data->typeName = $transaction->typeName;
$transaction_data->typeID = $transaction->typeID;
$transaction_data->price = $transaction->price;
$transaction_data->clientID = $transaction->clientID;
$transaction_data->clientName = $transaction->clientName;
$transaction_data->stationID = $transaction->stationID;
$transaction_data->stationName = $transaction->stationName;
$transaction_data->transactionType = $transaction->transactionType;
$transaction_data->transactionFor = $transaction->transactionFor;
$transaction_data->journalTransactionID = $transaction->journalTransactionID;
$transaction_data->clientTypeID = $transaction->clientTypeID;
$transaction_data->save();
} else {
// This entry already exists as the hash is cached.
continue;
}
}
// Check how many entries we got back. If it us less that $row_count, we know we have
// walked back the entire journal
if (count($wallet_transactions->transactions) < $row_count) {
break;
}
// Break the while loop
}
}
// Unlock the call
BaseApi::unlockCall($lockhash);
return $wallet_transactions;
}
