-
Notifications
You must be signed in to change notification settings - Fork 1
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Add reader for txt files with fixed columns size (#22)
* Add reader for txt files with fixed columns size * Updated fixed size file reader with latest changed from 0.x * Fixed static analyis and checkstyle * Fixed static analysis
- Loading branch information
1 parent
d9df3a9
commit f546a33
Showing
3 changed files
with
203 additions
and
0 deletions.
There are no files selected for viewing
106 changes: 106 additions & 0 deletions
106
src/batch/src/Job/Item/Reader/Filesystem/FixedColumnSizeFileReader.php
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,106 @@ | ||
<?php | ||
|
||
declare(strict_types=1); | ||
|
||
namespace Yokai\Batch\Job\Item\Reader\Filesystem; | ||
|
||
use Generator; | ||
use Yokai\Batch\Exception\RuntimeException; | ||
use Yokai\Batch\Exception\UnexpectedValueException; | ||
use Yokai\Batch\Job\Item\ItemReaderInterface; | ||
use Yokai\Batch\Job\JobExecutionAwareInterface; | ||
use Yokai\Batch\Job\JobExecutionAwareTrait; | ||
use Yokai\Batch\Job\Parameters\JobParameterAccessorInterface; | ||
|
||
final class FixedColumnSizeFileReader implements | ||
ItemReaderInterface, | ||
JobExecutionAwareInterface | ||
{ | ||
use JobExecutionAwareTrait; | ||
|
||
public const HEADERS_MODE_SKIP = 'skip'; | ||
public const HEADERS_MODE_COMBINE = 'combine'; | ||
public const HEADERS_MODE_NONE = 'none'; | ||
private const AVAILABLE_HEADERS_MODES = [ | ||
self::HEADERS_MODE_SKIP, | ||
self::HEADERS_MODE_COMBINE, | ||
self::HEADERS_MODE_NONE, | ||
]; | ||
|
||
/** | ||
* @var int[] | ||
*/ | ||
private array $columns; | ||
|
||
/** | ||
* @var string | ||
*/ | ||
private string $headersMode; | ||
|
||
/** | ||
* @var JobParameterAccessorInterface | ||
*/ | ||
private JobParameterAccessorInterface $filePath; | ||
|
||
/** | ||
* @param int[] $columns | ||
*/ | ||
public function __construct( | ||
array $columns, | ||
JobParameterAccessorInterface $filePath, | ||
string $headersMode = self::HEADERS_MODE_NONE | ||
) { | ||
if (!\in_array($headersMode, self::AVAILABLE_HEADERS_MODES, true)) { | ||
throw UnexpectedValueException::enum(self::AVAILABLE_HEADERS_MODES, $headersMode, 'Invalid header mode.'); | ||
} | ||
|
||
$this->columns = $columns; | ||
$this->headersMode = $headersMode; | ||
$this->filePath = $filePath; | ||
} | ||
|
||
/** | ||
* @inheritdoc | ||
* @phpstan-return Generator<array<mixed>> | ||
*/ | ||
public function read(): Generator | ||
{ | ||
$path = (string)$this->filePath->get($this->jobExecution); | ||
$handle = @\fopen($path, 'r'); | ||
if ($handle === false) { | ||
throw new RuntimeException(\sprintf('Cannot read %s.', $path)); | ||
} | ||
|
||
$headers = \array_keys($this->columns); | ||
|
||
$index = -1; | ||
|
||
while (($line = \fgets($handle)) !== false) { | ||
$index++; | ||
|
||
$start = 0; | ||
$row = []; | ||
foreach ($this->columns as $size) { | ||
$row[] = \trim(\mb_substr($line, $start, $size)); | ||
$start += $size; | ||
} | ||
|
||
if ($index === 0) { | ||
if ($this->headersMode === self::HEADERS_MODE_COMBINE) { | ||
$headers = $row; | ||
} | ||
if (\in_array($this->headersMode, [self::HEADERS_MODE_COMBINE, self::HEADERS_MODE_SKIP], true)) { | ||
continue; | ||
} | ||
} | ||
|
||
if (\is_array($headers)) { | ||
$row = \array_combine($headers, $row); | ||
} | ||
|
||
yield $row; | ||
} | ||
|
||
\fclose($handle); | ||
} | ||
} |
93 changes: 93 additions & 0 deletions
93
src/batch/tests/Job/Item/Reader/Filesystem/FixedColumnSizeFileReaderTest.php
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,93 @@ | ||
<?php | ||
|
||
declare(strict_types=1); | ||
|
||
namespace Yokai\Batch\Tests\Job\Item\Reader\Filesystem; | ||
|
||
use Generator; | ||
use PHPUnit\Framework\TestCase; | ||
use Yokai\Batch\Exception\RuntimeException; | ||
use Yokai\Batch\Exception\UnexpectedValueException; | ||
use Yokai\Batch\Job\Item\Reader\Filesystem\FixedColumnSizeFileReader; | ||
use Yokai\Batch\Job\Parameters\StaticValueParameterAccessor; | ||
use Yokai\Batch\JobExecution; | ||
|
||
class FixedColumnSizeFileReaderTest extends TestCase | ||
{ | ||
/** | ||
* @dataProvider config | ||
*/ | ||
public function test(array $columns, string $headersMode, array $expected): void | ||
{ | ||
$execution = JobExecution::createRoot('123456', 'testing'); | ||
$reader = new FixedColumnSizeFileReader( | ||
$columns, | ||
new StaticValueParameterAccessor(__DIR__ . '/fixtures/fixed-column-size.txt'), | ||
$headersMode | ||
); | ||
$reader->setJobExecution($execution); | ||
self::assertSame($expected, \iterator_to_array($reader->read())); | ||
} | ||
|
||
public function testInvalidHeaderMode(): void | ||
{ | ||
$this->expectException(UnexpectedValueException::class); | ||
new FixedColumnSizeFileReader([10, 10], new StaticValueParameterAccessor(null), 'wrong header mode'); | ||
} | ||
|
||
public function testFileNotFound(): void | ||
{ | ||
$this->expectException(RuntimeException::class); | ||
$execution = JobExecution::createRoot('123456', 'testing'); | ||
$reader = new FixedColumnSizeFileReader( | ||
[10, 10], | ||
new StaticValueParameterAccessor(__DIR__ . '/fixtures/unknown-file.ext') | ||
); | ||
$reader->setJobExecution($execution); | ||
\iterator_to_array($reader->read()); | ||
} | ||
|
||
public function config(): Generator | ||
{ | ||
$columnsWithoutNames = [10, 9, 8, -1]; | ||
$columnsWithNames = ['firstName' => 10, 'lastName' => 9, 'country' => 8, 'city' => -1]; | ||
|
||
yield [ | ||
$columnsWithoutNames, | ||
FixedColumnSizeFileReader::HEADERS_MODE_COMBINE, | ||
[ | ||
['firstName' => 'John', 'lastName' => 'Doe', 'country' => 'USA', 'city' => 'Washington'], | ||
['firstName' => 'Jane', 'lastName' => 'Doe', 'country' => 'USA', 'city' => 'Seattle'], | ||
['firstName' => 'Jack', 'lastName' => 'Doe', 'country' => 'USA', 'city' => 'San Francisco'], | ||
], | ||
]; | ||
yield [ | ||
$columnsWithNames, | ||
FixedColumnSizeFileReader::HEADERS_MODE_SKIP, | ||
[ | ||
['firstName' => 'John', 'lastName' => 'Doe', 'country' => 'USA', 'city' => 'Washington'], | ||
['firstName' => 'Jane', 'lastName' => 'Doe', 'country' => 'USA', 'city' => 'Seattle'], | ||
['firstName' => 'Jack', 'lastName' => 'Doe', 'country' => 'USA', 'city' => 'San Francisco'], | ||
], | ||
]; | ||
yield [ | ||
$columnsWithoutNames, | ||
FixedColumnSizeFileReader::HEADERS_MODE_NONE, | ||
[ | ||
['firstName', 'lastName', 'country', 'city'], | ||
['John', 'Doe', 'USA', 'Washington'], | ||
['Jane', 'Doe', 'USA', 'Seattle'], | ||
['Jack', 'Doe', 'USA', 'San Francisco'], | ||
], | ||
]; | ||
yield [ | ||
$columnsWithoutNames, | ||
FixedColumnSizeFileReader::HEADERS_MODE_SKIP, | ||
[ | ||
['John', 'Doe', 'USA', 'Washington'], | ||
['Jane', 'Doe', 'USA', 'Seattle'], | ||
['Jack', 'Doe', 'USA', 'San Francisco'], | ||
], | ||
]; | ||
} | ||
} |
4 changes: 4 additions & 0 deletions
4
src/batch/tests/Job/Item/Reader/Filesystem/fixtures/fixed-column-size.txt
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,4 @@ | ||
firstName lastName country city | ||
John Doe USA Washington | ||
Jane Doe USA Seattle | ||
Jack Doe USA San Francisco |