Refactor regex validation checks to a dedicated method (#64)

The code has been updated to move regex validation checks to a Utility method, improving reusability. Multiple classes including IsDomain, IsFloat, IsGeohash, IsInt, IsUsaMarketName, RegEx, CsvValidator, all now utilize this utility method. This enhancement provides a centralized place for handling regex validations ensuring uniform error handling across the application. Minor modifications were also made to related comments and tests.
JBZoo · Mar 19, 2024 · 14bd208 · 14bd208
1 parent fc08765
commit 14bd208
Show file tree

Hide file tree

Showing 17 changed files with 182 additions and 92 deletions.
diff --git a/README.md b/README.md
@@ -159,6 +159,7 @@ columns:
       # Of course it's an ultimatum to verify any sort of string data.
       # Please, be careful. Regex is a powerful tool, but it can be very dangerous if used incorrectly.
       # Remember that if you want to solve a problem with regex, you now have two problems.
+      # But have it your way, then happy debugging! https://regex101.com.
       regex: /^[\d]{2}$/
 
       # Checks length of a string including spaces (multibyte safe).
@@ -525,22 +526,22 @@ Schema is invalid: ./tests/schemas/demo_invalid.yml
 +-------+------------------+--------------+----- demo_invalid.yml -----------------------------------------------+
 
 (1/1) Invalid file: ./tests/fixtures/demo.csv
-+------+------------------+------------------+----------------------- demo.csv ---------------------------------------------------------------------+
-| Line | id:Column        | Rule             | Message                                                                                              |
-+------+------------------+------------------+------------------------------------------------------------------------------------------------------+
-| 1    |                  | filename_pattern | Filename "./tests/fixtures/demo.csv" does not match pattern: "/demo-[12].csv$/i"                     |
-| 6    | 0:Name           | length_min       | The length of the value "Carl" is 4, which is less than the expected "5"                             |
-| 11   | 0:Name           | length_min       | The length of the value "Lois" is 4, which is less than the expected "5"                             |
-| 1    | 1:City           | ag:is_unique     | Column has non-unique values. Unique: 9, total: 10                                                   |
-| 2    | 2:Float          | num_max          | The number of the value "4825.185", which is greater than the expected "4825.184"                    |
-| 6    | 3:Birthday       | date_min         | The date of the value "1955-05-14" is parsed as "1955-05-14 00:00:00 +00:00", which is less than the |
-|      |                  |                  | expected "1955-05-15 00:00:00 +00:00 (1955-05-15)"                                                   |
-| 8    | 3:Birthday       | date_min         | The date of the value "1955-05-14" is parsed as "1955-05-14 00:00:00 +00:00", which is less than the |
-|      |                  |                  | expected "1955-05-15 00:00:00 +00:00 (1955-05-15)"                                                   |
-| 9    | 3:Birthday       | date_max         | The date of the value "2010-07-20" is parsed as "2010-07-20 00:00:00 +00:00", which is greater than  |
-|      |                  |                  | the expected "2009-01-01 00:00:00 +00:00 (2009-01-01)"                                               |
-| 5    | 4:Favorite color | allow_values     | Value "blue" is not allowed. Allowed values: ["red", "green", "Blue"]                                |
-+------+------------------+------------------+----------------------- demo.csv ---------------------------------------------------------------------+
++-------+------------------+------------------+----------------------- demo.csv ---------------------------------------------------------------------+
+| Line  | id:Column        | Rule             | Message                                                                                              |
++-------+------------------+------------------+------------------------------------------------------------------------------------------------------+
+| undef |                  | filename_pattern | Filename "./tests/fixtures/demo.csv" does not match pattern: "/demo-[12].csv$/i"                     |
+| 6     | 0:Name           | length_min       | The length of the value "Carl" is 4, which is less than the expected "5"                             |
+| 11    | 0:Name           | length_min       | The length of the value "Lois" is 4, which is less than the expected "5"                             |
+| 1     | 1:City           | ag:is_unique     | Column has non-unique values. Unique: 9, total: 10                                                   |
+| 2     | 2:Float          | num_max          | The number of the value "4825.185", which is greater than the expected "4825.184"                    |
+| 6     | 3:Birthday       | date_min         | The date of the value "1955-05-14" is parsed as "1955-05-14 00:00:00 +00:00", which is less than the |
+|       |                  |                  | expected "1955-05-15 00:00:00 +00:00 (1955-05-15)"                                                   |
+| 8     | 3:Birthday       | date_min         | The date of the value "1955-05-14" is parsed as "1955-05-14 00:00:00 +00:00", which is less than the |
+|       |                  |                  | expected "1955-05-15 00:00:00 +00:00 (1955-05-15)"                                                   |
+| 9     | 3:Birthday       | date_max         | The date of the value "2010-07-20" is parsed as "2010-07-20 00:00:00 +00:00", which is greater than  |
+|       |                  |                  | the expected "2009-01-01 00:00:00 +00:00 (2009-01-01)"                                               |
+| 5     | 4:Favorite color | allow_values     | Value "blue" is not allowed. Allowed values: ["red", "green", "Blue"]                                |
++-------+------------------+------------------+----------------------- demo.csv ---------------------------------------------------------------------+
 
 
 Found 9 issues in CSV file.

diff --git a/csv-blueprint.php b/csv-blueprint.php
@@ -37,6 +37,10 @@
 
 \date_default_timezone_set('UTC');
 
+\set_error_handler(static function ($severity, $message, $file, $line): void {
+    throw new \ErrorException($message, 0, $severity, $file, $line);
+});
+
 (new CliApplication('CSV Blueprint', '@git-version@'))
     ->registerCommandsByPath(PATH_ROOT . '/src/Commands', __NAMESPACE__)
     ->setLogo(

diff --git a/schema-examples/full.yml b/schema-examples/full.yml
@@ -80,6 +80,7 @@ columns:
       # Of course it's an ultimatum to verify any sort of string data.
       # Please, be careful. Regex is a powerful tool, but it can be very dangerous if used incorrectly.
       # Remember that if you want to solve a problem with regex, you now have two problems.
+      # But have it your way, then happy debugging! https://regex101.com.
       regex: /^[\d]{2}$/
 
       # Checks length of a string including spaces (multibyte safe).

diff --git a/src/Rules/Cell/IsDomain.php b/src/Rules/Cell/IsDomain.php
@@ -16,6 +16,8 @@
 
 namespace JBZoo\CsvBlueprint\Rules\Cell;
 
+use JBZoo\CsvBlueprint\Utils;
+
 final class IsDomain extends AbstractCellRule
 {
     protected const HELP_OPTIONS = [
@@ -26,7 +28,7 @@ public function validateRule(string $cellValue): ?string
     {
         $domainPattern = '/^(?!-)[A-Za-z0-9-]+(\.[A-Za-z0-9-]+)*(\.[A-Za-z]{2,})$/';
 
-        if (\preg_match($domainPattern, $cellValue) === 0) {
+        if (Utils::testRegex($domainPattern, $cellValue)) {
             return "Value \"<c>{$cellValue}</c>\" is not a valid domain";
         }
 

diff --git a/src/Rules/Cell/IsFloat.php b/src/Rules/Cell/IsFloat.php
@@ -16,6 +16,8 @@
 
 namespace JBZoo\CsvBlueprint\Rules\Cell;
 
+use JBZoo\CsvBlueprint\Utils;
+
 class IsFloat extends AbstractCellRule
 {
     protected const HELP_OPTIONS = [
@@ -24,7 +26,7 @@ class IsFloat extends AbstractCellRule
 
     public function validateRule(string $cellValue): ?string
     {
-        if (\preg_match('/^-?\d+(\.\d+)?$/', $cellValue) === 0) {
+        if (Utils::testRegex('/^-?\d+(\.\d+)?$/', $cellValue)) {
             return "Value \"<c>{$cellValue}</c>\" is not a float number";
         }
 

diff --git a/src/Rules/Cell/IsGeohash.php b/src/Rules/Cell/IsGeohash.php
@@ -16,6 +16,8 @@
 
 namespace JBZoo\CsvBlueprint\Rules\Cell;
 
+use JBZoo\CsvBlueprint\Utils;
+
 class IsGeohash extends AbstractCellRule
 {
     protected const HELP_OPTIONS = [
@@ -24,7 +26,7 @@ class IsGeohash extends AbstractCellRule
 
     public function validateRule(string $cellValue): ?string
     {
-        if (\preg_match('/^[0-9b-hj-km-np-z]{1,}$/', $cellValue) === 0) {
+        if (Utils::testRegex('/^[0-9b-hj-km-np-z]{1,}$/', $cellValue)) {
             return "Value \"<c>{$cellValue}</c>\" is not a valid Geohash";
         }
 

diff --git a/src/Rules/Cell/IsInt.php b/src/Rules/Cell/IsInt.php
@@ -16,6 +16,8 @@
 
 namespace JBZoo\CsvBlueprint\Rules\Cell;
 
+use JBZoo\CsvBlueprint\Utils;
+
 final class IsInt extends AbstractCellRule
 {
     protected const HELP_OPTIONS = [
@@ -24,7 +26,7 @@ final class IsInt extends AbstractCellRule
 
     public function validateRule(string $cellValue): ?string
     {
-        if (\preg_match('/^-?\d+$/', $cellValue) === 0) {
+        if (Utils::testRegex('/^-?\d+$/', $cellValue)) {
             return "Value \"<c>{$cellValue}</c>\" is not an integer";
         }
 

diff --git a/src/Rules/Cell/IsUsaMarketName.php b/src/Rules/Cell/IsUsaMarketName.php
@@ -16,6 +16,8 @@
 
 namespace JBZoo\CsvBlueprint\Rules\Cell;
 
+use JBZoo\CsvBlueprint\Utils;
+
 final class IsUsaMarketName extends AllowValues
 {
     protected const HELP_OPTIONS = [
@@ -24,7 +26,7 @@ final class IsUsaMarketName extends AllowValues
 
     public function validateRule(string $cellValue): ?string
     {
-        if (\preg_match('/^[A-Za-z\s\'\-\.,]+, [A-Z]{2}$/u', $cellValue) === 0) {
+        if (Utils::testRegex('/^[A-Za-z\s\'\-\.,\(\)]+, [A-Z-]{2,6}$/u', $cellValue)) {
             return "Invalid market name format for value \"<c>{$cellValue}</c>\". " .
                 'Market name must have format "<green>New York, NY</green>"';
         }

diff --git a/src/Rules/Cell/Regex.php b/src/Rules/Cell/Regex.php
@@ -25,6 +25,7 @@ final class Regex extends AbstractCellRule
         "Of course it's an ultimatum to verify any sort of string data.",
         'Please, be careful. Regex is a powerful tool, but it can be very dangerous if used incorrectly.',
         'Remember that if you want to solve a problem with regex, you now have two problems.',
+        'But have it your way, then happy debugging! https://regex101.com',
     ];
 
     protected const HELP_OPTIONS = [
@@ -42,7 +43,7 @@ public function validateRule(string $cellValue): ?string
             return 'Regex pattern is not defined';
         }
 
-        if (\preg_match($regex, $cellValue) === 0) {
+        if (Utils::testRegex($regex, $cellValue)) {
             return "Value \"<c>{$cellValue}</c>\" does not match the pattern \"<green>{$regex}</green>\"";
         }
 

diff --git a/src/Utils.php b/src/Utils.php
@@ -201,4 +201,21 @@ public static function matchTypes(
         return isset($mapOfValidConvertions[$expectedType])
             && \in_array($actualType, $mapOfValidConvertions[$expectedType], true);
     }
+
+    public static function testRegex(string $regex, string $cellValue): bool
+    {
+        if ($regex === '' || $cellValue === '') {
+            return false;
+        }
+
+        try {
+            if (\preg_match($regex, $cellValue) === 0) {
+                return true;
+            }
+        } catch (\Throwable) {
+            return false;
+        }
+
+        return false;
+    }
 }
diff --git a/src/Validators/CsvValidator.php b/src/Validators/CsvValidator.php
@@ -103,14 +103,14 @@ private function validateFile(bool $quickStop = false): ErrorSuite
         if (
             $filenamePattern !== null
             && $filenamePattern !== ''
-            && \preg_match($filenamePattern, $this->csv->getCsvFilename()) === 0
+            && Utils::testRegex($filenamePattern, $this->csv->getCsvFilename())
         ) {
             $error = new Error(
                 'filename_pattern',
                 'Filename "<c>' . Utils::cutPath($this->csv->getCsvFilename()) .
                 "</c>\" does not match pattern: \"<c>{$filenamePattern}</c>\"",
                 '',
-                ColumnValidator::FALLBACK_LINE,
+                Error::UNDEFINED_LINE,
             );
 
             $errors->addError($error);

diff --git a/src/Validators/Error.php b/src/Validators/Error.php
@@ -30,12 +30,13 @@ public function __construct(
 
     public function __toString(): string
     {
+        $columnStr = $this->getColumnName() === '' ? '' : ", column \"{$this->getColumnName()}\"";
+
         if ($this->line === self::UNDEFINED_LINE) {
-            return "\"{$this->getRuleCode()}\", column \"{$this->getColumnName()}\". {$this->getMessage()}.";
+            return "\"{$this->getRuleCode()}\"{$columnStr}. {$this->getMessage()}.";
         }
 
-        return "\"{$this->getRuleCode()}\" at line <red>{$this->getLine()}</red>, " .
-            "column \"{$this->getColumnName()}\". {$this->getMessage()}.";
+        return "\"{$this->getRuleCode()}\" at line <red>{$this->getLine()}</red>{$columnStr}. {$this->getMessage()}.";
     }
 
     public function getRuleCode(): string