Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Rewrite regex and add extra check to variable names #74

Open
wants to merge 6 commits into
base: master
Choose a base branch
from
Open

Rewrite regex and add extra check to variable names #74

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
668efe8
rewrite regex and add extra check to ensure it follows current spss n…
sinnbeck Feb 13, 2024
c2ebe1d
add tests
sinnbeck Feb 13, 2024
e9d395e
dont throw exception on reserved names
sinnbeck Feb 14, 2024
41af280
add tests for illegal characters in variable names
sinnbeck Feb 14, 2024
bc259f4
ensure unique names for reserved variable names
sinnbeck Feb 16, 2024
bfe88d7
ensure variable names in tests ends with a letter
sinnbeck Feb 16, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 7 additions & 4 deletions src/Sav/Writer.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -59,7 +59,7 @@ public function __construct($data = [], $buffer = null)
$this->write($data);
}
}

/**
* @param array $data
* @param string $file
Expand Down Expand Up @@ -101,7 +101,7 @@ public function write($data)

$this->data = new Record\Data();

$nominalIdx = 0;
$nominalIdx = 0;

/** @var Variable $var */
// for ($idx = 0; $idx <= $variablesCount; $idx++) {
Expand All @@ -110,12 +110,15 @@ public function write($data)
$var = new Variable($var);
}

//if (! preg_match('/^[A-Za-z0-9_]+$/', $var->name)) {
// UTF-8 and '.' characters could pass here
if (!preg_match('/^[A-Za-z0-9_\.\x{4e00}-\x{9fa5}]+$/u', $var->name)) {
if (!preg_match('/^(?!#|\$|\.)[\w0-9_.#@$\x{4e00}-\x{9fa5}]+(?<!\.|_)$/u', $var->name)) {
throw new \InvalidArgumentException(sprintf('Variable name `%s` contains an illegal character.', $var->name));
}

if (in_array($var->name, ['ALL', 'AND', 'BY', 'EQ', 'GE', 'GT', 'LE', 'LT', 'NE', 'NOT', 'OR', 'TO', 'WITH'])) {
$var->name = \uniqid($var->name);
}

if (empty($var->width)) {
throw new \InvalidArgumentException(sprintf('Invalid field width. Should be an integer number greater than zero.'));
}
Expand Down
80 changes: 80 additions & 0 deletions tests/NamingTest.php
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,80 @@
<?php

use SPSS\Sav\Reader;
use SPSS\Sav\Record\Info\LongVariableNames;
use SPSS\Sav\Writer;
use SPSS\Tests\TestCase;

class NamingTest extends TestCase
{
public function illegalNameProvider()
{
return [
['#FOO', ''],
['$FOO', ''],
['.FOO', ''],
['FOO.', ''],
['FOO_', ''],
];
}

public function testReservedNames()
{
$data = [
'header' => [
'prodName' => '@(#) IBM SPSS STATISTICS',
'layoutCode' => 2,
'creationDate' => date('d M y'),
'creationTime' => date('H:i:s'),
],
'variables' => [
[
'name' => 'WITH',
'width' => 16,
'format' => 1,
],
[
'name' => 'OR',
'format' => 5,
],
],
];
$writer = new Writer($data);

$buffer = $writer->getBuffer();
$buffer->rewind();

$reader = Reader::fromString($buffer->getStream())->read();

$this->assertRegExp('/^' . $data['variables'][0]['name'] . '[\w]{13}$/', $reader->info[LongVariableNames::SUBTYPE]['V00001']);
$this->assertRegExp('/^' . $data['variables'][1]['name'] . '[\w]{13}$/', $reader->info[LongVariableNames::SUBTYPE]['V00002']);
}


/**
* @dataProvider illegalNameProvider
*/
public function testIllegalNames($name)
{
$this->expectException(\InvalidArgumentException::class);
$this->expectExceptionMessage(sprintf('Variable name `%s` contains an illegal character.', $name));

$data = [
'header' => [
'prodName' => '@(#) IBM SPSS STATISTICS',
'layoutCode' => 2,
'creationDate' => date('d M y'),
'creationTime' => date('H:i:s'),
],
'variables' => [
[
'name' => $name,
'width' => 16,
'format' => 1,
],
],
];

new Writer($data);
}
}
4 changes: 2 additions & 2 deletions tests/SavRandomReadWriteTest.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -37,7 +37,7 @@ public function provider()
$count = 1; // mt_rand(1, 20);
for ($i = 0; $i < $count; $i++) {
$var = $this->generateVariable([
'id' => $this->generateRandomString(mt_rand(2, 100)),
'id' => $this->generateRandomString(mt_rand(2, 100)) . 'a',
'casesCount' => $header['casesCount'],
]
);
Expand All @@ -50,7 +50,7 @@ public function provider()
$header['casesCount'] = 5;
for ($i = 0; $i < 100; $i++) {
$variable = $this->generateVariable([
'id' => $this->generateRandomString(mt_rand(2, 100)),
'id' => $this->generateRandomString(mt_rand(2, 100)) . 'a',
'casesCount' => $header['casesCount'],
]);
$header['nominalCaseSize'] = Utils::widthToOcts($variable['width']);
Expand Down
42 changes: 39 additions & 3 deletions tests/WriteMultibyteTest.php
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
namespace SPSS\Tests;

use SPSS\Sav\Reader;
use SPSS\Sav\Record\Info\LongVariableNames;
use SPSS\Sav\Variable;
use SPSS\Sav\Writer;

Expand All @@ -11,7 +12,7 @@ class WriteMultibyteTest extends TestCase
public function testMultiByteLabel()
{
$data = [
'header' => [
'header' => [
'prodName' => '@(#) IBM SPSS STATISTICS',
'layoutCode' => 2,
'creationDate' => date('d M y'),
Expand Down Expand Up @@ -55,13 +56,12 @@ public function testMultiByteLabel()

/**
* ISSUE #20.
*
* Chinese value labels seem to work fine, but free text does not work
*/
public function testChinese()
{
$input = [
'header' => [
'header' => [
'prodName' => '@(#) IBM SPSS STATISTICS 64-bit Macintosh 23.0.0.0',
'creationDate' => '05 Oct 18',
'creationTime' => '01:36:53',
Expand Down Expand Up @@ -124,4 +124,40 @@ public function testChinese()
$expected[2][1] = $input['variables'][1]['data'][2];
$this->assertEquals($expected, $reader->data);
}

public function testMultiByteVariableName()
{
$data = [
'header' => [
'prodName' => '@(#) IBM SPSS STATISTICS',
'layoutCode' => 2,
'creationDate' => date('d M y'),
'creationTime' => date('H:i:s'),
],
'variables' => [
[
'name' => 'Å',
'width' => 16,
'format' => 1,
],
[
'name' => 'DSADÆØØÅÅÅÅÅSAAA',
'format' => 5,
],
],
];
$writer = new Writer($data);

$buffer = $writer->getBuffer();
$buffer->rewind();

$reader = Reader::fromString($buffer->getStream())->read();

// Short variable name
$this->assertEquals($data['variables'][0]['name'], $reader->info[LongVariableNames::SUBTYPE]['V00001']);
// Long variable name
$this->assertEquals($data['variables'][1]['name'], $reader->info[LongVariableNames::SUBTYPE]['V00002']);

}

}