Improves detection of numbered documents

Also removes all uses of 'array' function

Issue: documentacao-e-tarefas/scielo#912

Signed-off-by: Jhon <jhon@lepidus.com.br>

Author: JhonathanLepidus

classes/ContentParser.php

@@ -15,6 +15,7 @@ class ContentParser
 {
     private const ZERO_WIDTH_SPACE = "\x{200B}";
     private const MIN_WORD_LENGTH = 2;
+    private const NUM_DOC_LINES_SAMPLE = 5;
 
     private function cleanWord($word)
     {
@@ -30,7 +31,7 @@ private function cleanWord($word)
 
     private function parseWordsFromString($string)
     {
-        $words = array();
+        $words = [];
 
         for ($i = 0; $i < strlen($string); $i++) {
             while ($i < strlen($string) && ctype_space($string[$i])) {
@@ -56,19 +57,32 @@ private function parseWordsFromString($string)
         return $words;
     }
 
-    private function parseLine($line)
+    private function parseLine(string $line, bool $docIsNumbered)
     {
         $zeroWidthSpacePattern = '/' . self::ZERO_WIDTH_SPACE . '/u';
         $line = preg_replace($zeroWidthSpacePattern, '', $line);
         $lineWords = $this->parseWordsFromString($line);
 
-        if (!empty($lineWords) && is_numeric($lineWords[0])) {
+        if ($docIsNumbered && !empty($lineWords) && is_numeric($lineWords[0])) {
             array_shift($lineWords);
         }
 
         return $lineWords;
     }
 
+    public function checkDocumentIsNumbered(array $docLines): bool
+    {
+        for ($i = 0; $i < self::NUM_DOC_LINES_SAMPLE; $i++) {
+            $parsedLine = explode(' ', $docLines[$i]);
+            $firstWord = $parsedLine[0];
+
+            if (!is_numeric($firstWord)) {
+                return false;
+            }
+        }
+        return true;
+    }
+
     public function parseDocument($pathFile, $useRawMode = true)
     {
         $pathTxt = substr($pathFile, 0, -3) . 'txt';
@@ -77,19 +91,21 @@ public function parseDocument($pathFile, $useRawMode = true)
 
         $docText = file_get_contents($pathTxt);
         $docLines = preg_split("/\r\n|\n|\r/", $docText);
-        $docWords = array();
+        $docWords = [];
         unlink($pathTxt);
 
+        $docIsNumbered = $this->checkDocumentIsNumbered($docLines);
+
         foreach ($docLines as $line) {
-            $docWords = array_merge($docWords, $this->parseLine($line));
+            $docWords = array_merge($docWords, $this->parseLine($line, $docIsNumbered));
         }
 
         return $docWords;
     }
 
     public function createPatternFromString($string)
     {
-        $pattern = array();
+        $pattern = [];
 
         for ($i = 0; $i < strlen($string); $i++) {
             while ($i < strlen($string) && ctype_space($string[$i])) {

classes/DocumentChecker.php

@@ -212,7 +212,7 @@ public function checkTitleInEnglish($title)
         $cleanedTitle = $parser->cleanStyledText($title);
         $patternTitle = $parser->createPatternFromString($cleanedTitle);
 
-        return $this->checkForPatterns(array($patternTitle), count($patternTitle), 75, 0.75);
+        return $this->checkForPatterns([$patternTitle], count($patternTitle), 75, 0.75);
     }
 
     public function checkEthicsCommittee()

classes/DocumentChecklist.php

@@ -24,7 +24,7 @@ public function __construct($path)
 
     public function executeChecklist($submission)
     {
-        $dataChecklist = array();
+        $dataChecklist = [];
         $submissionIsArticle = !$submission->getData('nonArticle');
 
         if ($submissionIsArticle) {

tests/AuthorsContributionTest.php

@@ -5,7 +5,7 @@
 
 class AuthorsContributionTest extends DetectionOnDocumentTest
 {
-    private $patternContribution = array("contribuição", "dos", "autores");
+    private $patternContribution = ["contribuição", "dos", "autores"];
 
     public function setUp(): void
     {

tests/EthicsCommitteeTest.php

@@ -5,7 +5,7 @@
 
 class EthicsCommitteeTest extends DetectionOnDocumentTest
 {
-    private $patternCommittee = array("aprovação", "do", "comitê", "de", "ética");
+    private $patternCommittee = ["aprovação", "do", "comitê", "de", "ética"];
 
     public function setUp(): void
     {

tests/MetadataEnglishTest.php

@@ -6,8 +6,8 @@
 
 class MetadataEnglishTest extends DetectionOnDocumentTest
 {
-    private $patternKeywords = array("keywords");
-    private $patternAbstract = array("abstract");
+    private $patternKeywords = ["keywords"];
+    private $patternAbstract = ["abstract"];
     private $title = "A beautiful title";
     private $titleWithStyling = "<b>A</b> <i>beautiful</i> <u>title</u>";