Skip to main content
The FQL\Query\Debugger class provides CLI output utilities for inspecting queries, measuring execution time, and benchmarking iteration performance.
Debugger requires the tracy/tracy package. Install it as a dev dependency:
composer require --dev tracy/tracy

​
Starting and stopping

Call Debugger::start() at the beginning of your script. It records the start timestamp and prints an initial memory snapshot. Call Debugger::end() when you are done to print final timing and memory totals. 
use FQL\Query\Debugger;

Debugger::start();

// ... your queries ...

Debugger::end();

​
Inspecting a query

Debugger::inspectQuery() executes a query and prints the SQL representation, result count, memory size, and optionally all rows. 
use FQL\Query\Debugger;
use FQL\Stream\Xml;

$ordersFile = Xml::open(__DIR__ . '/data/orders.xml');
$query = $ordersFile->query()
    ->select('id')->as('orderId')
    ->select('user_id')->as('userId')
    ->select('total_price')->as('totalPrice')
    ->from('orders.order');

// Show SQL, result count, and first row
Debugger::inspectQuery($query);

// Show SQL, result count, and all rows
Debugger::inspectQuery($query, listResults: true);

​
Inspecting a raw SQL string

Debugger::inspectStreamSql() parses a raw FQL string against an existing stream and shows what the applied query looks like compared to the original input. 
use FQL\Query\Debugger;
use FQL\Stream\Xml;

$ordersFile = Xml::open(__DIR__ . '/data/orders.xml');

Debugger::inspectStreamSql(
    $ordersFile,
    'SELECT id, user_id, total_price FROM orders.order'
);
Use Debugger::inspectSql() when you want to parse a self-contained FQL string (with format and path embedded) and have the Query object returned: 
$query = Debugger::inspectSql(
    'SELECT id, name FROM json(./data/products.json).data.products WHERE price > 100'
);

Debugger::inspectQuery($query, listResults: true);

​
Benchmarking a query

Debugger::benchmarkQuery() runs a query in both stream and in-memory modes for the specified number of iterations, then prints throughput stats. 
use FQL\Query\Debugger;
use FQL\Stream\Xml;

$query = Xml::open(__DIR__ . '/data/orders.xml')->query()
    ->select('id')->as('orderId')
    ->select('user_id')->as('userId')
    ->select('total_price')->as('totalPrice')
    ->from('orders.order');

Debugger::benchmarkQuery($query, iterations: 1000);
The default number of iterations is 2500.

​
Dumping a value

Debugger::dump() delegates to Tracy’s dump() function for pretty-printing any PHP value: 
Debugger::dump($results->fetch());

​
Example output

Running composer example:csv produces output similar to this: 
=========================
### Debugger started: ###
=========================
> Memory usage (MB): 1.3191 (emalloc)
> Memory peak usage (MB): 1.7326 (emalloc)
------------------------------
> Execution time (s): 8.5E-5
> Execution time (ms): 0.085
> Execution time (µs): 85
> Execution memory peak usage (MB): 0
=========================
### Inspecting query: ###
=========================
==================
### SQL query: ###
==================
> SELECT
>   ean ,
>   defaultCategory ,
>   EXPLODE(defaultCategory, " > ") AS categoryArray ,
>   price ,
>   ROUND(price, 2) AS price_rounded ,
>   MOD(price, 100) AS modulo_100 ,
>   MOD(price, 54) AS modulo_54
> FROM csv(products-w-1250.csv, "windows-1250", ";").*
> GROUP BY defaultCategory
> ORDER BY defaultCategory DESC
================
### Results: ###
================
> Result class: FQL\Results\InMemory
> Results size memory (KB): 3.55
> Result exists: true
> Result count: 15
========================
### Fetch first row: ###
========================
array (7)
   'ean' => 5010232964877
   'defaultCategory' => 'Testování > Drogerie'
   'categoryArray' => array (2)
   |  0 => 'Testování'
   |  1 => 'Drogerie'
   'price' => 121.0
   'price_rounded' => 121.0
   'modulo_100' => 21.0
   'modulo_54' => 13.0

>>> SPLIT TIME <<<
> Memory usage (MB): 3.1451 (emalloc)
> Memory peak usage (MB): 3.2262 (emalloc)
------------------------------
> Execution time (s): 0.040016
> Execution time (ms): 40.016
> Execution time (µs): 40016
> Execution memory peak usage (MB): 1.4936
========================
### Benchmark Query: ###
========================
> 2 500 iterations
=========================
### STREAM BENCHMARK: ###
=========================
> Size (KB): 2.78
> Count: 15
> Iterated results: 37 500
>>> SPLIT TIME <<<
> Execution time (ms): 36402.098
> Execution memory peak usage (MB): 0
============================
### IN_MEMORY BENCHMARK: ###
============================
> Size (KB): 3.55
> Count: 15
> Iterated results: 37 500
>>> SPLIT TIME <<<
> Execution time (ms): 17.43
> Execution memory peak usage (MB): 0
=======================
### Debugger ended: ###
=======================
> Final execution time (ms): 36459.756
Notice the large difference in iteration time between STREAM BENCHMARK (re-reads the file each time) and IN_MEMORY BENCHMARK (iterates a cached result set). Use in-memory results when you need to iterate over the same data multiple times.

​
Method reference

MethodDescription
Debugger::start()Record start time and print initial memory stats
Debugger::end()Print final elapsed time and memory totals
Debugger::inspectQuery($query, $listResults)Execute query and print SQL, count, memory, and rows
Debugger::inspectStreamSql($stream, $sql)Parse FQL against a stream and show applied SQL diff
Debugger::inspectSql($sql)Parse a self-contained FQL string and return the Query
Debugger::benchmarkQuery($query, $iterations)Run N iterations in both stream and in-memory modes
Debugger::dump($value)Pretty-print a PHP value via Tracy
Debugger::split()Print elapsed time since the last split point