I’m rocking along and all is working well for initial tests of the algorithm (header + first row of actual data) but when I turn-on processing for the other 9,999 rows, all I get stored into mongo is the first row of data.

I add an echo statement after the insert and I see 10,000 names scroll across my terminal.  So the problem isn’t that I’m not getting the data, it’s that the data isn’t being stored into Mongo.  I try turning on safe writes on my $mongo->insert() function and *bam*, error message:

PHP Fatal error:  Uncaught exception ‘MongoCursorException’ with message ‘E11000 duplicate key error index: insite.testdata.$_id_  dup key: { : ObjectId(’4d25fd9a7e03972618000000′) }’ in /htdocs/framework/parseCSV.php:449
Stack trace:
#0 /htdocs/framework/parseCSV.php(449): MongoCollection->insert(Object(mongorriffic), Array)
#1 {main}
thrown in /htdocs/framework/parseCSV.php on line 449

My code looks like this:

if (($handle = fopen($argv[1], "r")) !== FALSE) {
while (($data = fgetcsv($handle, 0, ",")) !== FALSE) {
if (!$row) {
// use the csv headers to instantiate the new mongo object
$objMongo = new mongorriffic($data);
} else {
// parse the data into the existing structure
$objMongo->storeData($data);
$mongoCollection->insert($objMongo);
}
$row++;
}
fclose($handle);
}

I’m calling my internal method (storeDate()) but I’m not creating, populating or resetting an “_id” field value because mongodb is supposed to automagically handle that for me.  What’s actually happening is that mongodb  is creating a valid “_id” value for the first record, but since my method does nothing to manipulate the field, the value persists through iterations of the data. I fix this by adding the following line of code after I invoke the insert method:

unset($objMongo->_id);

I re-run my script and *success*!   10,000 records are stored in less than 6-seconds using 4 collections.  BTW,  am *loving* Mongodb….