Refactor cell value validation technique and docs (#102)

Switched from using Utils regex method to Respect\Validation\Validator
in IsFloat and IsInt classes for better and more accurate cell value
validation. This update also includes correct handling for empty string
values and returns null instead. Corresponding tests have been modified
to adapt to these changes for consistency.

.github/workflows/main.yml

@@ -100,6 +100,7 @@ jobs:
         uses: shivammathur/setup-php@v2
         with:
           php-version: 8.1
+          coverage: none
           tools: composer
           extensions: ast
 
@@ -138,6 +139,7 @@ jobs:
         uses: shivammathur/setup-php@v2
         with:
           php-version: 8.3
+          coverage: none
           tools: composer
           extensions: ast
 

README.md

@@ -4,7 +4,7 @@
 [![GitHub Release](https://img.shields.io/github/v/release/jbzoo/csv-blueprint?label=Latest)](https://github.com/jbzoo/csv-blueprint/releases)    [![Total Downloads](https://poser.pugx.org/jbzoo/csv-blueprint/downloads)](https://packagist.org/packages/jbzoo/csv-blueprint/stats)    [![Docker Pulls](https://img.shields.io/docker/pulls/jbzoo/csv-blueprint.svg)](https://hub.docker.com/r/jbzoo/csv-blueprint/tags)    [![Docker Image Size](https://img.shields.io/docker/image-size/jbzoo/csv-blueprint)](https://hub.docker.com/r/jbzoo/csv-blueprint/tags)
 
 <!-- rules-counter -->
-[![Static Badge](https://img.shields.io/badge/Rules-292-green?label=Total%20number%20of%20rules&labelColor=darkgreen&color=gray)](schema-examples/full.yml)    [![Static Badge](https://img.shields.io/badge/Rules-81-green?label=Cell%20rules&labelColor=blue&color=gray)](src/Rules/Cell)    [![Static Badge](https://img.shields.io/badge/Rules-206-green?label=Aggregate%20rules&labelColor=blue&color=gray)](src/Rules/Aggregate)    [![Static Badge](https://img.shields.io/badge/Rules-5-green?label=Extra%20checks&labelColor=blue&color=gray)](#extra-checks)    [![Static Badge](https://img.shields.io/badge/Rules-199-green?label=Plan%20to%20add&labelColor=gray&color=gray)](tests/schemas/todo.yml)
+[![Static Badge](https://img.shields.io/badge/Rules-292-green?label=Total%20number%20of%20rules&labelColor=darkgreen&color=gray)](schema-examples/full.yml)    [![Static Badge](https://img.shields.io/badge/Rules-81-green?label=Cell%20rules&labelColor=blue&color=gray)](src/Rules/Cell)    [![Static Badge](https://img.shields.io/badge/Rules-206-green?label=Aggregate%20rules&labelColor=blue&color=gray)](src/Rules/Aggregate)    [![Static Badge](https://img.shields.io/badge/Rules-5-green?label=Extra%20checks&labelColor=blue&color=gray)](#extra-checks)    [![Static Badge](https://img.shields.io/badge/Rules-142/54/8-green?label=Plan%20to%20add&labelColor=gray&color=gray)](tests/schemas/todo.yml)
 <!-- /rules-counter -->
 
 ## Introduction
@@ -198,8 +198,8 @@ columns:
       contains_one: [ a, b ]            # Only one of the strings must be part of the CSV value.
       contains_any: [ a, b ]            # At least one of the string must be part of the CSV value.
       contains_all: [ a, b ]            # All the strings must be part of a CSV value.
-      starts_with: "prefix "            # Example: "prefix Hello World".
-      ends_with: " suffix"              # Example: "Hello World suffix".
+      starts_with: 'prefix '            # Example: "prefix Hello World".
+      ends_with: ' suffix'              # Example: "Hello World suffix".
 
       # Under the hood it converts and compares as float values.
       # Comparison accuracy is 10 digits after a dot.
@@ -519,13 +519,13 @@ columns:
       # See: https://en.wikipedia.org/wiki/Quartile
       # There are multiple methods for computing quartiles: ["exclusive", "inclusive"]. Exclusive is ussually classic.
       # Available types: ["0%", "Q1", "Q2", "Q3", "100%", "IQR"] ("IQR" is Interquartile Range)
-      # Example: `[ inclusive, 'Q3', 42.0 ]` - the Q3 inclusive quartile is 50.0
-      quartiles_min: [ 'exclusive', '0%', 1.0 ]             # x >= 1.0
-      quartiles_greater: [ 'inclusive', 'Q1', 2.0 ]         # x >  2.0
-      quartiles_not: [ 'exclusive', 'Q2', 5.0 ]             # x != 5.0
-      quartiles: [ 'inclusive', 'Q3', 7.0 ]                 # x == 7.0
-      quartiles_less: [ 'exclusive', '100%', 8.0 ]          # x <  8.0
-      quartiles_max: [ 'inclusive', 'IQR', 9.0 ]            # x <= 9.0
+      # Example: `[ inclusive, 'Q3', 42.0 ]` - the Q3 inclusive quartile is 42.0
+      quartiles_min: [ exclusive, '0%', 1.0 ]               # x >= 1.0
+      quartiles_greater: [ inclusive, 'Q1', 2.0 ]           # x >  2.0
+      quartiles_not: [ exclusive, 'Q2', 5.0 ]               # x != 5.0
+      quartiles: [ inclusive, 'Q3', 7.0 ]                   # x == 7.0
+      quartiles_less: [ exclusive, '100%', 8.0 ]            # x <  8.0
+      quartiles_max: [ inclusive, 'IQR', 9.0 ]              # x <= 9.0
 
       # Midhinge. The average of the first and third quartiles and is thus a measure of location.
       # Equivalently, it is the 25% trimmed mid-range or 25% midsummary; it is an L-estimator.
@@ -938,7 +938,7 @@ But... it's not a problem for most cases. And it solves the problem of validatin
 
 The utility is made to just pick up and use and not think about how it works internally.
 Moreover, everything is covered as strictly as possible by tests, strict typing of variables + `~7` linters and static analyzers (max level of rules). 
-Also, if you look, you'll see that any PR goes through about `~30` different checks on GitHub Actions (matrix of PHP versions and mods).
+Also, if you look, you'll see that any PR goes through about `~10` different checks on GitHub Actions (matrix of PHP versions and mods).
 Since I don't know under what conditions the code will be used, everything I can think of is covered. The wonderful world of Open Source.
 
 So... as strictly as possible in today's PHP world. I think it works as expected.

schema-examples/full.yml

@@ -112,8 +112,8 @@ columns:
       contains_one: [ a, b ]            # Only one of the strings must be part of the CSV value.
       contains_any: [ a, b ]            # At least one of the string must be part of the CSV value.
       contains_all: [ a, b ]            # All the strings must be part of a CSV value.
-      starts_with: "prefix "            # Example: "prefix Hello World".
-      ends_with: " suffix"              # Example: "Hello World suffix".
+      starts_with: 'prefix '            # Example: "prefix Hello World".
+      ends_with: ' suffix'              # Example: "Hello World suffix".
 
       # Under the hood it converts and compares as float values.
       # Comparison accuracy is 10 digits after a dot.
@@ -433,13 +433,13 @@ columns:
       # See: https://en.wikipedia.org/wiki/Quartile
       # There are multiple methods for computing quartiles: ["exclusive", "inclusive"]. Exclusive is ussually classic.
       # Available types: ["0%", "Q1", "Q2", "Q3", "100%", "IQR"] ("IQR" is Interquartile Range)
-      # Example: `[ inclusive, 'Q3', 42.0 ]` - the Q3 inclusive quartile is 50.0
-      quartiles_min: [ 'exclusive', '0%', 1.0 ]             # x >= 1.0
-      quartiles_greater: [ 'inclusive', 'Q1', 2.0 ]         # x >  2.0
-      quartiles_not: [ 'exclusive', 'Q2', 5.0 ]             # x != 5.0
-      quartiles: [ 'inclusive', 'Q3', 7.0 ]                 # x == 7.0
-      quartiles_less: [ 'exclusive', '100%', 8.0 ]          # x <  8.0
-      quartiles_max: [ 'inclusive', 'IQR', 9.0 ]            # x <= 9.0
+      # Example: `[ inclusive, 'Q3', 42.0 ]` - the Q3 inclusive quartile is 42.0
+      quartiles_min: [ exclusive, '0%', 1.0 ]               # x >= 1.0
+      quartiles_greater: [ inclusive, 'Q1', 2.0 ]           # x >  2.0
+      quartiles_not: [ exclusive, 'Q2', 5.0 ]               # x != 5.0
+      quartiles: [ inclusive, 'Q3', 7.0 ]                   # x == 7.0
+      quartiles_less: [ exclusive, '100%', 8.0 ]            # x <  8.0
+      quartiles_max: [ inclusive, 'IQR', 9.0 ]              # x <= 9.0
 
       # Midhinge. The average of the first and third quartiles and is thus a measure of location.
       # Equivalently, it is the 25% trimmed mid-range or 25% midsummary; it is an L-estimator.

src/Rules/Aggregate/ComboQuartiles.php

@@ -49,15 +49,15 @@ public function getHelpMeta(): array
                 'Available types: ' . Utils::printList(self::TYPES) . ' ("IQR" is Interquartile Range)',
                 // Example
                 'Example: `[ ' . self::METHODS[1] . ", '" . self::TYPES[3] . "', 42.0 ]`" .
-                ' - the ' . self::TYPES[3] . ' ' . self::METHODS[1] . ' quartile is 50.0',
+                ' - the ' . self::TYPES[3] . ' ' . self::METHODS[1] . ' quartile is 42.0',
             ],
             [
-                self::MIN     => ["[ '" . self::METHODS[0] . "', '" . self::TYPES[0] . "', 1.0 ]", 'x >= 1.0'],
-                self::GREATER => ["[ '" . self::METHODS[1] . "', '" . self::TYPES[1] . "', 2.0 ]", 'x >  2.0'],
-                self::NOT     => ["[ '" . self::METHODS[0] . "', '" . self::TYPES[2] . "', 5.0 ]", 'x != 5.0'],
-                self::EQ      => ["[ '" . self::METHODS[1] . "', '" . self::TYPES[3] . "', 7.0 ]", 'x == 7.0'],
-                self::LESS    => ["[ '" . self::METHODS[0] . "', '" . self::TYPES[4] . "', 8.0 ]", 'x <  8.0'],
-                self::MAX     => ["[ '" . self::METHODS[1] . "', '" . self::TYPES[5] . "', 9.0 ]", 'x <= 9.0'],
+                self::MIN     => ['[ ' . self::METHODS[0] . ", '" . self::TYPES[0] . "', 1.0 ]", 'x >= 1.0'],
+                self::GREATER => ['[ ' . self::METHODS[1] . ", '" . self::TYPES[1] . "', 2.0 ]", 'x >  2.0'],
+                self::NOT     => ['[ ' . self::METHODS[0] . ", '" . self::TYPES[2] . "', 5.0 ]", 'x != 5.0'],
+                self::EQ      => ['[ ' . self::METHODS[1] . ", '" . self::TYPES[3] . "', 7.0 ]", 'x == 7.0'],
+                self::LESS    => ['[ ' . self::METHODS[0] . ", '" . self::TYPES[4] . "', 8.0 ]", 'x <  8.0'],
+                self::MAX     => ['[ ' . self::METHODS[1] . ", '" . self::TYPES[5] . "', 9.0 ]", 'x <= 9.0'],
             ],
         ];
     }

src/Rules/Cell/EndsWith.php

@@ -23,7 +23,7 @@ public function getHelpMeta(): array
         return [
             [],
             [
-                self::DEFAULT => ['" suffix"', 'Example: "Hello World suffix".'],
+                self::DEFAULT => ["' suffix'", 'Example: "Hello World suffix".'],
             ],
         ];
     }

src/Rules/Cell/IsFloat.php

@@ -16,7 +16,7 @@
 
 namespace JBZoo\CsvBlueprint\Rules\Cell;
 
-use JBZoo\CsvBlueprint\Utils;
+use Respect\Validation\Validator;
 
 class IsFloat extends AbstractCellRule
 {
@@ -32,7 +32,11 @@ public function getHelpMeta(): array
 
     public function validateRule(string $cellValue): ?string
     {
-        if (Utils::testRegex('/^-?\d+(\.\d+)?$/', $cellValue)) {
+        if ($cellValue === '') {
+            return null;
+        }
+
+        if (!Validator::floatVal()->validate($cellValue)) {
             return "Value \"<c>{$cellValue}</c>\" is not a float number";
         }
 

src/Rules/Cell/IsInt.php

@@ -16,7 +16,7 @@
 
 namespace JBZoo\CsvBlueprint\Rules\Cell;
 
-use JBZoo\CsvBlueprint\Utils;
+use Respect\Validation\Validator;
 
 final class IsInt extends AbstractCellRule
 {
@@ -32,7 +32,11 @@ public function getHelpMeta(): array
 
     public function validateRule(string $cellValue): ?string
     {
-        if (Utils::testRegex('/^-?\d+$/', $cellValue)) {
+        if ($cellValue === '') {
+            return null;
+        }
+
+        if (!Validator::intVal()->validate($cellValue)) {
             return "Value \"<c>{$cellValue}</c>\" is not an integer";
         }
 

src/Rules/Cell/StartsWith.php

@@ -23,7 +23,7 @@ public function getHelpMeta(): array
         return [
             [],
             [
-                self::DEFAULT => ['"prefix "', 'Example: "prefix Hello World".'],
+                self::DEFAULT => ["'prefix '", 'Example: "prefix Hello World".'],
             ],
         ];
     }

tests/ReadmeTest.php

@@ -78,9 +78,9 @@ public function testBadgeOfRules(): void
         $totalRules = $cellRules + $aggRules + $extraRules;
 
         $todoYml = yml(Tools::SCHEMA_TODO);
-        $planToAdd = \count($todoYml->findArray('columns.0.rules')) +
-            (\count($todoYml->findArray('columns.0.aggregate_rules')) * 6)
-            + \count([
+        $planToAdd = \count($todoYml->findArray('columns.0.rules')) . '/' .
+            (\count($todoYml->findArray('columns.0.aggregate_rules')) * 6) . '/' .
+            \count([
                 'required',
                 'null_values',
                 'multiple + separator',
@@ -89,16 +89,9 @@ public function testBadgeOfRules(): void
                 'complex_rules. one example',
                 'inherit',
                 'rule not found',
-            ])
-            - \count([
-                'first_value',
-                'second_value',
-                'last_value',
-                'sorted',
-                'custom_func',
             ]);
 
-        $badge = static function (string $label, int $count, string $url, string $color): string {
+        $badge = static function (string $label, int|string $count, string $url, string $color): string {
             $label = \str_replace(' ', '%20', $label);
             $badge = "![Static Badge](https://img.shields.io/badge/Rules-{$count}-green" .
                 "?label={$label}&labelColor={$color}&color=gray)";

tests/Rules/Cell/IsFloatTest.php

@@ -28,16 +28,21 @@ final class IsFloatTest extends TestAbstractCellRule
     public function testPositive(): void
     {
         $rule = $this->create(true);
-        isSame(null, $rule->validate(''));
+        isSame('', $rule->test(''));
         isSame('', $rule->test('1'));
         isSame('', $rule->test('01'));
         isSame('', $rule->test('1.0'));
         isSame('', $rule->test('01.0'));
+        isSame('', $rule->test('.0'));
+        isSame('', $rule->test('.1'));
         isSame('', $rule->test('-1'));
         isSame('', $rule->test('-1.0'));
+        isSame('', $rule->test('1e5'));
+        isSame('', $rule->test('1E5'));
+        isSame('', $rule->test(' 1E5'));
 
         $rule = $this->create(false);
-        isSame(null, $rule->validate(' 1'));
+        isSame(null, $rule->validate(' q'));
     }
 
     public function testNegative(): void
@@ -48,8 +53,12 @@ public function testNegative(): void
             $rule->test('1.000.000'),
         );
         isSame(
-            'Value " 1" is not a float number',
-            $rule->test(' 1'),
+            'Value "1.000 000" is not a float number',
+            $rule->test('1.000 000'),
+        );
+        isSame(
+            'Value " q" is not a float number',
+            $rule->test(' q'),
         );
     }
 }

tests/Rules/Cell/IsIntTest.php

@@ -34,6 +34,7 @@ public function testPositive(): void
         isSame('', $rule->test('0'));
         isSame('', $rule->test('00'));
         isSame('', $rule->test('-1'));
+        isSame('', $rule->test('089'));
 
         $rule = $this->create(false);
         isSame(null, $rule->validate(' 1'));
@@ -42,6 +43,14 @@ public function testPositive(): void
     public function testNegative(): void
     {
         $rule = $this->create(true);
+        isSame(
+            'Value "1_000_000" is not an integer',
+            $rule->test('1_000_000'),
+        );
+        isSame(
+            'Value "1000 000" is not an integer',
+            $rule->test('1000 000'),
+        );
         isSame(
             'Value "1.000.000" is not an integer',
             $rule->test('1.000.000'),