Merge branch 'master' into feature/num_format

README.md

@@ -8,7 +8,7 @@
 [![License](https://poser.pugx.org/box/spout/license)](https://packagist.org/packages/box/spout)
 
 Spout is a PHP library to read and write spreadsheet files (CSV, XLSX and ODS), in a fast and scalable way.
-Contrary to other file readers or writers, it is capable of processing very large files while keeping the memory usage really low (less than 10MB).
+Contrary to other file readers or writers, it is capable of processing very large files while keeping the memory usage really low (less than 3MB).
 
 Join the community and come discuss about Spout: [![Gitter](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/box/spout?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge)
 
@@ -301,7 +301,7 @@ For information, the performance tests take about 30 minutes to run (processing
 
 ## Frequently Asked Questions
 
-#### How can Spout handle such large data sets and still use less than 10MB of memory?
+#### How can Spout handle such large data sets and still use less than 3MB of memory?
 
 When writing data, Spout is streaming the data to files, one or few lines at a time. That means that it only keeps in memory the few rows that it needs to write. Once written, the memory is freed.
 

src/Spout/Reader/XLSX/Helper/SharedStringsHelper.php

@@ -100,11 +100,16 @@ class SharedStringsHelper
                 // removes nodes that should not be read, like the pronunciation of the Kanji characters
                 $cleanNode = $this->removeSuperfluousTextNodes($node);
 
-                // find all text nodes 't'; there can be multiple if the cell contains formatting
+                // find all text nodes "t"; there can be multiple if the cell contains formatting
                 $textNodes = $cleanNode->xpath('//ns:t');
 
                 $textValue = '';
-                foreach ($textNodes as $textNode) {
+                foreach ($textNodes as $nodeIndex => $textNode) {
+                    if ($nodeIndex !== 0) {
+                        // add a space between each "t" node
+                        $textValue .= ' ';
+                    }
+
                     if ($this->shouldPreserveWhitespace($textNode)) {
                         $textValue .= $textNode->__toString();
                     } else {
@@ -200,6 +205,8 @@ class SharedStringsHelper
     {
         $tagsToRemove = [
             'rPh', // Pronunciation of the text
+            'pPr', // Paragraph Properties / Previous Paragraph Properties
+            'rPr', // Run Properties for the Paragraph Mark / Previous Run Properties for the Paragraph Mark
         ];
 
         foreach ($tagsToRemove as $tagToRemove) {

src/Spout/Writer/ODS/Internal/Worksheet.php

@@ -134,6 +134,10 @@ class Worksheet implements WorksheetInterface
      */
     public function addRow($dataRow, $style)
     {
+        // $dataRow can be an associative array. We need to transform
+        // it into a regular array, as we'll use the numeric indexes.
+        $dataRowWithNumericIndexes = array_values($dataRow);
+
         $styleIndex = ($style->getId() + 1); // 1-based
         $cellsCount = count($dataRow);
         $this->maxNumColumns = max($this->maxNumColumns, $cellsCount);
@@ -144,10 +148,12 @@ class Worksheet implements WorksheetInterface
         $nextCellIndex = 1;
 
         for ($i = 0; $i < $cellsCount; $i++) {
-            $currentCellValue = $dataRow[$currentCellIndex];
+            $currentCellValue = $dataRowWithNumericIndexes[$currentCellIndex];
 
             // Using isset here because it is way faster than array_key_exists...
-            if (!isset($dataRow[$nextCellIndex]) || $currentCellValue !== $dataRow[$nextCellIndex]) {
+            if (!isset($dataRowWithNumericIndexes[$nextCellIndex]) ||
+                $currentCellValue !== $dataRowWithNumericIndexes[$nextCellIndex]) {
+
                 $numTimesValueRepeated = ($nextCellIndex - $currentCellIndex);
                 $data .= $this->getCellContent($currentCellValue, $styleIndex, $numTimesValueRepeated);
 

tests/Spout/Reader/XLSX/Helper/SharedStringsHelperTest.php

@@ -82,6 +82,22 @@ class SharedStringsHelperTest extends \PHPUnit_Framework_TestCase
         $sharedStringsHelper->cleanup();
     }
 
+    /**
+     * @return void
+     */
+    public function testGetStringAtIndexShouldWorkWithStringsContainingTextAndHyperlinkInSameCell()
+    {
+        $resourcePath = $this->getResourcePath('one_sheet_with_shared_strings_containing_text_and_hyperlink_in_same_cell.xlsx');
+        $sharedStringsHelper = new SharedStringsHelper($resourcePath);
+
+        $sharedStringsHelper->extractSharedStrings();
+
+        $sharedString = $sharedStringsHelper->getStringAtIndex(0);
+        $this->assertEquals('go to https://github.com please', $sharedString);
+
+        $sharedStringsHelper->cleanup();
+    }
+
     /**
      * @return void
      */

tests/Spout/Writer/CSV/WriterTest.php

@@ -74,6 +74,20 @@ class WriterTest extends \PHPUnit_Framework_TestCase
         $this->assertContains(EncodingHelper::BOM_UTF8, $writtenContent, 'The CSV file should contain a UTF-8 BOM');
     }
 
+    /**
+     * @return void
+     */
+    public function testWriteShouldSupportAssociativeArrays()
+    {
+        $allRows = [
+            ['foo' => 'csv--11', 'bar' => 'csv--12'],
+        ];
+        $writtenContent = $this->writeToCsvFileAndReturnWrittenContent($allRows, 'csv_from_associative_arrays.csv');
+        $writtenContent = $this->trimWrittenContent($writtenContent);
+
+        $this->assertEquals('csv--11,csv--12', $writtenContent, 'Values from associative arrays should be written');
+    }
+
     /**
      * @return void
      */

tests/Spout/Writer/ODS/WriterTest.php

@@ -23,7 +23,7 @@ class WriterTest extends \PHPUnit_Framework_TestCase
     public function testAddRowShouldThrowExceptionIfCannotOpenAFileForWriting()
     {
         $fileName = 'file_that_wont_be_written.ods';
-        $this->createUnwritableFolderIfNeeded($fileName);
+        $this->createUnwritableFolderIfNeeded();
         $filePath = $this->getGeneratedUnwritableResourcePath($fileName);
 
         $writer = WriterFactory::create(Type::ODS);
@@ -177,6 +177,25 @@ class WriterTest extends \PHPUnit_Framework_TestCase
         }
     }
 
+    /**
+     * @return void
+     */
+    public function testAddRowShouldSupportAssociativeArrays()
+    {
+        $fileName = 'test_add_row_should_support_associative_arrays.ods';
+        $dataRows = [
+            ['foo' => 'ods--11', 'bar' => 'ods--12'],
+        ];
+
+        $this->writeToODSFile($dataRows, $fileName);
+
+        foreach ($dataRows as $dataRow) {
+            foreach ($dataRow as $cellValue) {
+                $this->assertValueWasWritten($fileName, $cellValue);
+            }
+        }
+    }
+
     /**
      * @return void
      */

tests/Spout/Writer/XLSX/WriterTest.php

@@ -21,7 +21,7 @@ class WriterTest extends \PHPUnit_Framework_TestCase
     public function testAddRowShouldThrowExceptionIfCannotOpenAFileForWriting()
     {
         $fileName = 'file_that_wont_be_written.xlsx';
-        $this->createUnwritableFolderIfNeeded($fileName);
+        $this->createUnwritableFolderIfNeeded();
         $filePath = $this->getGeneratedUnwritableResourcePath($fileName);
 
         $writer = WriterFactory::create(Type::XLSX);
@@ -232,6 +232,25 @@ class WriterTest extends \PHPUnit_Framework_TestCase
         }
     }
 
+    /**
+     * @return void
+     */
+    public function testAddRowShouldSupportAssociativeArrays()
+    {
+        $fileName = 'test_add_row_should_support_associative_arrays.xlsx';
+        $dataRows = [
+            ['foo' => 'xlsx--11', 'bar' => 'xlsx--12'],
+        ];
+
+        $this->writeToXLSXFile($dataRows, $fileName, $shouldUseInlineStrings = true);
+
+        foreach ($dataRows as $dataRow) {
+            foreach ($dataRow as $cellValue) {
+                $this->assertInlineDataWasWrittenToSheet($fileName, 1, $cellValue);
+            }
+        }
+    }
+
     /**
      * @return void
      */