Files
oleibman 38a8330748 More Extendability for CSV Reader
A follow-up to PR #4837. PR #4827, to which I am not necessarily committed, shows that there could be a use case for letting a user perform customized logic to convert a non-UTF8-encoded CSV. In addition, several other CSV Reader properties require the use of static or locale properties, which is somewhat problematic; this PR allows them to be set as instance variables, falling back to static/locale only when unset.
2026-03-27 16:16:45 -07:00

256 lines
8.3 KiB
PHP

<?php
declare(strict_types=1);
namespace PhpOffice\PhpSpreadsheetTests\Reader\Csv;
use PhpOffice\PhpSpreadsheet\Reader\Csv;
use PhpOffice\PhpSpreadsheet\Reader\Exception as ReaderException;
use PHPUnit\Framework\Attributes\DataProvider;
use PHPUnit\Framework\TestCase;
class CsvTest extends TestCase
{
#[DataProvider('providerDelimiterDetection')]
public function testDelimiterDetection(string $filename, string $expectedDelimiter, string $cell, string|float|int|null $expectedValue): void
{
$reader = new Csv();
$delim1 = $reader->getDelimiter();
self::assertNull($delim1);
$spreadsheet = $reader->load($filename);
self::assertSame($expectedDelimiter, $reader->getDelimiter(), 'should be able to infer the delimiter');
$actual = $spreadsheet->getActiveSheet()->getCell($cell)->getValue();
self::assertSame($expectedValue, $actual, 'should be able to retrieve correct value');
$spreadsheet->disconnectWorksheets();
}
public static function providerDelimiterDetection(): array
{
return [
[
'tests/data/Reader/CSV/enclosure.csv',
',',
'C4',
'username2',
],
[
'tests/data/Reader/CSV/semicolon_separated.csv',
';',
'C2',
'25,5',
],
[
'tests/data/Reader/CSV/line_break_in_enclosure.csv',
',',
'A3',
'Test',
],
[
'tests/data/Reader/CSV/line_break_in_enclosure_with_escaped_quotes.csv',
',',
'A3',
'Test',
],
[
'tests/data/Reader/HTML/csv_with_angle_bracket.csv',
',',
'B1',
'Number of items with weight <= 50kg',
],
[
'samples/Reader2/sampleData/example1.csv',
',',
'I4',
'100%',
],
[
'samples/Reader2/sampleData/example2.csv',
',',
'D8',
-58.373161,
],
[
'tests/data/Reader/CSV/empty.csv',
',',
'A1',
null,
],
[
'tests/data/Reader/CSV/no_delimiter.csv',
',',
'A1',
'SingleLine',
],
];
}
#[DataProvider('providerCanLoad')]
public function testCanLoad(bool $expected, string $filename): void
{
$reader = new Csv();
self::assertSame($expected, $reader->canRead($filename));
}
public static function providerCanLoad(): array
{
return [
[false, 'tests/data/Reader/Ods/data.ods'],
[false, 'samples/templates/excel2003.xml'],
[true, 'tests/data/Reader/CSV/enclosure.csv'],
[true, 'tests/data/Reader/CSV/semicolon_separated.csv'],
[true, 'tests/data/Reader/CSV/contains_html.csv'],
[true, 'tests/data/Reader/CSV/csv_without_extension'],
[true, 'tests/data/Reader/HTML/csv_with_angle_bracket.csv'],
[true, 'tests/data/Reader/CSV/empty.csv'],
[true, 'samples/Reader2/sampleData/example1.csv'],
[true, 'samples/Reader2/sampleData/example2.csv'],
];
}
#[DataProvider('providerVersion')]
public function testEscapeCharacters(int $version): void
{
if ($version >= 90000) {
$this->expectException(ReaderException::class);
$this->expectExceptionMessage('Escape character must be null string');
}
$reader = new Csv();
if ($version === PHP_VERSION_ID) {
$reader->setEscapeCharacter('"');
} else {
$reader->setEscapeCharacter('"', $version);
}
$spreadsheet = $reader->load('tests/data/Reader/CSV/backslash.csv');
$worksheet = $spreadsheet->getActiveSheet();
$expected = [
['field 1', 'field 2\\'],
['field 3\\', 'field 4'],
];
self::assertSame('"', $reader->getEscapeCharacter());
self::assertSame($expected, $worksheet->toArray());
$spreadsheet->disconnectWorksheets();
}
public static function providerVersion(): array
{
return [
[PHP_VERSION_ID],
[90000],
];
}
public function testInvalidWorkSheetInfo(): void
{
$this->expectException(ReaderException::class);
$this->expectExceptionMessage('is an Invalid Spreadsheet file');
$reader = new Csv();
$reader->listWorksheetInfo('');
}
public function testLineBreakEscape(): void
{
$reader = new Csv();
$spreadsheet = $reader->load('tests/data/Reader/CSV/line_break_in_enclosure_with_escaped_quotes.csv');
$sheet = $spreadsheet->getActiveSheet();
$expected = <<<EOF
This is a "test csv file"
with both "line breaks"
and "escaped
quotes" that breaks
the delimiters
EOF;
self::assertSame($expected, $sheet->getCell('B3')->getValue());
$spreadsheet->disconnectWorksheets();
}
public function testSeparatorLine(): void
{
$reader = new Csv();
$reader->setSheetIndex(3);
$spreadsheet = $reader->load('tests/data/Reader/CSV/sep.csv');
self::assertSame(';', $reader->getDelimiter());
$sheet = $spreadsheet->getActiveSheet();
self::assertSame(3, $reader->getSheetIndex());
self::assertSame(3, $spreadsheet->getActiveSheetIndex());
self::assertSame('A', $sheet->getCell('A1')->getValue());
self::assertSame(1, $sheet->getCell('B1')->getValue());
self::assertSame(2, $sheet->getCell('A2')->getValue());
self::assertSame(3, $sheet->getCell('B2')->getValue());
$spreadsheet->disconnectWorksheets();
}
public function testDefaultSettings(): void
{
$reader = new Csv();
self::assertSame('UTF-8', $reader->getInputEncoding());
self::assertSame('"', $reader->getEnclosure());
$reader->setEnclosure('\'');
self::assertSame('\'', $reader->getEnclosure());
$reader->setEnclosure('');
self::assertSame('"', $reader->getEnclosure());
// following tests from BaseReader
self::assertTrue($reader->getReadEmptyCells());
self::assertFalse($reader->getIncludeCharts());
self::assertNull($reader->getLoadSheetsOnly());
}
public function testReadEmptyFileName(): void
{
$this->expectException(ReaderException::class);
$this->expectExceptionMessage('is an Invalid Spreadsheet file');
$reader = new Csv();
$filename = '';
$reader->load($filename);
}
public function testReadNonexistentFileName(): void
{
$this->expectException(ReaderException::class);
$this->expectExceptionMessage('is an Invalid Spreadsheet file');
$reader = new Csv();
$reader->load('tests/data/Reader/CSV/encoding.utf8.csvxxx');
}
#[DataProvider('providerEscapes')]
public function testInferSeparator(string $escape, string $delimiter, int $version = PHP_VERSION_ID): void
{
if ($version >= 90000 && $escape !== '') {
$this->expectException(ReaderException::class);
$this->expectExceptionMessage('Escape character must be null string');
}
$reader = new Csv();
if ($version === PHP_VERSION_ID) {
$reader->setEscapeCharacter($escape);
} else {
$reader->setEscapeCharacter($escape, $version);
}
$filename = 'tests/data/Reader/CSV/escape.csv';
$reader->listWorksheetInfo($filename);
self::assertSame($delimiter, $reader->getDelimiter());
}
public static function providerEscapes(): array
{
return [
['\\', ';'],
["\x0", ','],
['', ','],
['\\', ';', 90000],
];
}
public function testSetDelimiterNull(): void
{
$reader = new Csv();
$reader->setDelimiter(',');
self::assertSame(',', $reader->getDelimiter());
$reader->setDelimiter(null);
self::assertNull($reader->getDelimiter());
}
}