(PECL mongo >=1.5.0)
A command cursor is similar to a MongoCursor except that you use it for iterating through the results of a database command instead of a normal query. Command cursors are useful for iterating over large result sets that might exceed the document size limit (currently 16MB) of a single MongoDB::command() response.
While you can create command cursors using MongoCommandCursor::__construct() or the MongoCommandCursor::createFromDocument() factory method, you will generally want to use command-specific helpers such as MongoCollection::aggregateCursor().
Note that the cursor does not "contain" the database command's results; it just manages iteration through them. Thus, if you print a cursor (f.e. with var_dump() or print_r()), you will see the cursor object but not the result documents.
A MongoCommandCursor has two "life stages": pre- and post- command. When a cursor is created, it has not yet contacted the database, so it is in its pre-command state. When the client first attempts to get a result (by calling MongoCommandCursor::rewind(), directly or indirectly), the cursor moves into the post-command state.
The command cursor's batch size and socket timeout may be configured in both the pre- and post- command states.
Example #1 Adding options to MongoCommandCursor
<?php
$cursor = new MongoCommandCursor(...);
$cursor = $cursor->batchSize( 4 );
foreach ($cursor as $result) {
var_dump($result);
}
?>
$connection
, string $hash
, array $document
): MongoCommandCursor