Data Transformer

gbelsalvador/data-transformer is a PHP library for reading, transforming, validating, and exporting structured data across multiple formats.

It is designed for practical data workflows:

• import from CSV, JSON, XML, SQL, or XLSX
• normalize data into a consistent PHP array structure
• filter, map, and validate rows through a fluent pipeline
• export the result to another format

The project aims to offer a clean developer experience, predictable behavior, and a solid base for production-oriented data conversion workflows.

Documentation

• French README: README.fr.md
• HTML documentation: docs/index.html

Why This Library

Most conversion utilities stop at file-to-file export. Data Transformer goes further by adding a transformation pipeline between the reader and writer.

That means you can:

• convert formats
• rename and reshape fields
• filter rows
• validate data before export
• get an execution report with counts and validation errors

Features

• Multi-format support: CSV, JSON, XML, SQL, XLSX
• Fluent transformation pipeline with read(), filter(), map(), validate(), write()
• Structured SQL filters for safer queries
• Execution reporting via ExecutionResult
• Spreadsheet export hardening for CSV/XLSX formula injection
• XML parsing hardened with LIBXML_NONET
• PHPUnit test suite for pipeline behavior and file integrations
• PSR-4 autoloading and PSR-12-friendly code style

Installation

composer require gbelsalvador/data-transformer

Requirements

• PHP >= 8.0
• phpoffice/phpspreadsheet ^5.4

Supported Formats

Format	Read	Write
CSV	Yes	Yes
JSON	Yes	Yes
XML	Yes	Yes
SQL	Yes	Yes
XLSX	Yes	Yes

Quick Start

use Gbelsalvador\DataTransformer\Core\Transformer;
use Gbelsalvador\DataTransformer\Readers\CsvReader;
use Gbelsalvador\DataTransformer\Writers\JsonWriter;

$transformer = new Transformer();

$result = $transformer->transform(
    new CsvReader('input.csv'),
    new JsonWriter('output.json')
);

echo $result->rowsRead();
echo $result->rowsWritten();

Fluent Pipeline

The fluent API is the recommended approach for non-trivial workflows.

use Gbelsalvador\DataTransformer\Core\Transformer;
use Gbelsalvador\DataTransformer\Readers\CsvReader;
use Gbelsalvador\DataTransformer\Writers\JsonWriter;

$result = (new Transformer())
    ->read(new CsvReader('users.csv'))
    ->filter(fn (array $row) => ($row['active'] ?? null) === '1')
    ->map([
        'id' => 'user_id',
        'full_name' => fn (array $row) => trim(($row['first_name'] ?? '') . ' ' . ($row['last_name'] ?? '')),
        'email' => 'email',
        'country' => 'country',
    ])
    ->validate([
        'id' => 'required|integer',
        'full_name' => 'required|max:120',
        'email' => 'required|email',
        'country' => 'in:FR,BE,CH,CA',
    ])
    ->write(new JsonWriter('clean-users.json'));

echo $result->rowsRead();
echo $result->rowsWritten();
echo $result->rowsInvalid();
print_r($result->validationErrors());

Core Concepts

Readers

Readers load data from a source and return a normalized PHP array:

[
    ['column1' => 'value1', 'column2' => 'value2'],
    ['column1' => 'value3', 'column2' => 'value4'],
]

Available readers:

• CsvReader
• JsonReader
• XmlReader
• SqlReader
• XlsxReader

Writers

Writers receive normalized rows and persist them to a target format.

Available writers:

• CsvWriter
• JsonWriter
• XmlWriter
• SqlWriter
• XlsxWriter

ExecutionResult

Every completed transformation returns an ExecutionResult object with runtime information:

• rowsRead()
• rowsWritten()
• rowsFilteredOut()
• rowsInvalid()
• errorCount()
• duration()
• validationErrors()

Usage Examples

CSV to JSON

use Gbelsalvador\DataTransformer\Core\Transformer;
use Gbelsalvador\DataTransformer\Readers\CsvReader;
use Gbelsalvador\DataTransformer\Writers\JsonWriter;

$result = (new Transformer())->transform(
    new CsvReader('products.csv', delimiter: ';'),
    new JsonWriter('products.json')
);

SQL to XLSX

use Gbelsalvador\DataTransformer\Core\Transformer;
use Gbelsalvador\DataTransformer\Readers\SqlReader;
use Gbelsalvador\DataTransformer\Writers\XlsxWriter;

$pdo = new PDO('mysql:host=localhost;dbname=my_database', 'user', 'password');
$pdo->setAttribute(PDO::ATTR_ERRMODE, PDO::ERRMODE_EXCEPTION);

$result = (new Transformer())->transform(
    new SqlReader(
        pdo: $pdo,
        tableName: 'users',
        columns: ['id', 'name', 'email', 'created_at'],
        filters: [
            'active' => 1,
            'created_at' => [
                'operator' => '>=',
                'value' => '2026-01-01',
            ],
        ]
    ),
    new XlsxWriter('active-users.xlsx', sheetName: 'Users')
);

JSON to XML

use Gbelsalvador\DataTransformer\Core\Transformer;
use Gbelsalvador\DataTransformer\Readers\JsonReader;
use Gbelsalvador\DataTransformer\Writers\XmlWriter;

$result = (new Transformer())->transform(
    new JsonReader('catalog.json'),
    new XmlWriter('catalog.xml', rootElement: 'catalog', rowElement: 'item')
);

Validation Rules

The current validation layer supports:

• required
• email
• numeric
• integer
• boolean
• date
• max:<length>
• in:value1,value2,value3
• same:other_field

Example:

$transformer->validate([
    'email' => 'required|email',
    'status' => 'required|in:active,inactive,pending',
    'name' => 'max:120',
]);

SQL Safety

SqlReader supports two approaches:

1. filters
2. whereClause + whereParams

The recommended approach is filters, because it builds parameterized conditions from a structured array.

new SqlReader(
    pdo: $pdo,
    tableName: 'users',
    filters: [
        'active' => 1,
        'role' => [
            'operator' => 'in',
            'value' => ['admin', 'editor'],
        ],
    ]
);

Supported structured operators:

• =
• !=
• >
• >=
• <
• <=
• like
• in
• is_null
• is_not_null

Security Notes

The library includes several basic hardening measures:

• SQL identifiers are validated before query construction
• CSV and XLSX exports neutralize formula-like values
• XML reading uses LIBXML_NONET to reduce external entity/network risks

These protections help reduce common issues, but you should still treat file paths, SQL configuration, and user-provided inputs as untrusted data in your application layer.

API Overview

Transformer

$transformer = new Transformer();

$transformer->read(ReaderInterface $reader);
$transformer->filter(callable $callback);
$transformer->map(callable|array $mapping);
$transformer->validate(array $rules);
$result = $transformer->write(WriterInterface $writer);

Reader Signatures

new CsvReader(
    string $filePath,
    string $delimiter = ',',
    bool $hasHeader = true,
    string $enclosure = '"',
    string $escape = '\\'
);

new JsonReader(
    string $filePath,
    bool $assoc = true
);

new XmlReader(
    string $filePath,
    string $rootElement = 'root'
);

new SqlReader(
    PDO $pdo,
    string $tableName,
    array $columns = ['*'],
    ?string $whereClause = null,
    array $whereParams = [],
    array $filters = []
);

new XlsxReader(
    string $filePath,
    ?string $sheetName = null,
    bool $hasHeader = true
);

Writer Signatures

new CsvWriter(
    string $filePath,
    string $delimiter = ',',
    string $enclosure = '"',
    bool $includeHeader = true
);

new JsonWriter(
    string $filePath,
    int $flags = JSON_PRETTY_PRINT | JSON_UNESCAPED_UNICODE
);

new XmlWriter(
    string $filePath,
    string $rootElement = 'root',
    string $rowElement = 'row'
);

new SqlWriter(
    PDO $pdo,
    string $tableName,
    bool $truncateFirst = false
);

new XlsxWriter(
    string $filePath,
    ?string $sheetName = null
);

Testing

Install development dependencies:

composer install

Run the test suite:

composer test

Current coverage includes:

• pipeline behavior
• validation behavior
• CSV/JSON/XML integration flows
• SQL reader/writer tests when pdo_sqlite is available

Note: SQL integration tests are skipped automatically if the pdo_sqlite driver is not enabled in your PHP environment.

Project Structure

src/
  Contracts/
  Core/
  Exceptions/
  Readers/
  Writers/

Tests/
  FileIntegrationTest.php
  SqlReadWriteTest.php
  TransformerPipelineTest.php

Roadmap

Planned improvements for upcoming versions:

• streaming for large datasets
• richer schema validation
• more transformation primitives
• append/merge output strategies
• additional input sources such as HTTP or YAML

Contributing

Contributions, bug reports, and improvement ideas are welcome.

If you want to contribute:

1. Fork the repository
2. Create a feature branch
3. Add or update tests
4. Open a pull request with a clear description

License

This project is released under the MIT License.