Merge pull request #44 from box/bump_to_v1.0.9

Bump to version 1.0.9

README.md

@@ -1,11 +1,13 @@
 # Spout
 
+[![Latest Stable Version](https://poser.pugx.org/box/spout/v/stable)](https://packagist.org/packages/box/spout)
+[![Project Status](http://opensource.box.com/badges/active.svg)](http://opensource.box.com/badges)
+[![Build Status](https://travis-ci.org/box/spout.png?branch=master)](http://travis-ci.org/box/spout)
+[![Total Downloads](https://poser.pugx.org/box/spout/downloads)](https://packagist.org/packages/box/spout) [![License](https://poser.pugx.org/box/spout/license)](https://packagist.org/packages/box/spout)
+
 Spout is a PHP library to read and write CSV and XLSX files, 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).
 
-[![Build Status](https://travis-ci.org/box/spout.png?branch=master)](http://travis-ci.org/box/spout)
-[![Project Status](http://opensource.box.com/badges/active.svg)](http://opensource.box.com/badges)
-
 
 ## Installation
 
@@ -248,6 +250,11 @@ Here are a few numbers regarding the performance of Spout:
 | Read XLSX (using shared strings) | 1 second                 | 1-2 minutes                  | 35-40 minutes                    |
 | Write XLSX                       | 1 second                 | 20-25 seconds                | 8-10 minutes                     |
 
+#### Does Spout support charts or formulas?
+
+No. This is a compromise to keep memory usage low. Charts and formulas requires data to be kept in memory in order to be used.
+So the larger the file would be, the more memory would be consumed, preventing your code to scale well.
+
 
 ## Support
 

composer.json

@@ -2,7 +2,7 @@
     "name": "box/spout",
     "description": "PHP Library to read and write CSV and XLSX files, in a fast and scalable way",
     "type": "library",
-    "version": "1.0.8",
+    "version": "1.0.9",
     "keywords": ["php","read","write","csv","xlsx","excel","spreadsheet","scale","memory","stream","ooxml"],
     "license": "Apache-2.0",
     "homepage": "https://www.github.com/box/spout",

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

@@ -71,6 +71,24 @@ class SharedStringsHelper
         $this->tempFolder = $this->fileSystemHelper->createFolder($rootTempFolder, uniqid('sharedstrings'));
     }
 
+    /**
+     * Returns whether the XLSX file contains a shared strings XML file
+     *
+     * @return bool
+     */
+    public function hasSharedStrings()
+    {
+        $hasSharedStrings = false;
+        $zip = new \ZipArchive();
+
+        if ($zip->open($this->filePath) === true) {
+            $hasSharedStrings = ($zip->locateName(self::SHARED_STRINGS_XML_FILE_PATH) !== false);
+            $zip->close();
+        }
+
+        return $hasSharedStrings;
+    }
+
     /**
      * Builds an in-memory array containing all the shared strings of the worksheet.
      * All the strings are stored in a XML file, located at 'xl/sharedStrings.xml'.

src/Spout/Reader/XLSX.php

@@ -72,8 +72,12 @@ class XLSX extends AbstractReader
         $this->zip = new \ZipArchive();
 
         if ($this->zip->open($filePath) === true) {
-            // Extracts all the strings from the worksheets for easy access in the future
-            $this->extractSharedStrings($filePath);
+            $this->sharedStringsHelper = new SharedStringsHelper($filePath, $this->tempFolder);
+
+            if ($this->sharedStringsHelper->hasSharedStrings()) {
+                // Extracts all the strings from the worksheets for easy access in the future
+                $this->sharedStringsHelper->extractSharedStrings();
+            }
 
             // Fetch all available worksheets
             $this->worksheetHelper = new WorksheetHelper($filePath, $this->globalFunctionsHelper);
@@ -87,19 +91,6 @@ class XLSX extends AbstractReader
         }
     }
 
-    /**
-     * Builds an in-memory array containing all the shared strings of the worksheets.
-     *
-     * @param  string $filePath Path of the XLSX file to be read
-     * @return void
-     * @throws \Box\Spout\Common\Exception\IOException If sharedStrings XML file can't be read
-     */
-    protected function extractSharedStrings($filePath)
-    {
-        $this->sharedStringsHelper = new SharedStringsHelper($filePath, $this->tempFolder);
-        $this->sharedStringsHelper->extractSharedStrings();
-    }
-
     /**
      * Returns whether another worksheet exists after the current worksheet.
      *
@@ -215,13 +206,25 @@ class XLSX extends AbstractReader
                 $currentColumnIndex = CellHelper::getColumnIndexFromCellIndex($currentCellIndex);
                 $node = $this->xmlReader->expand();
 
+                $hasInlineString = ($this->xmlReader->getAttribute('t') === 'inlineStr');
                 $hasSharedString = ($this->xmlReader->getAttribute('t') === 's');
-                if ($hasSharedString) {
-                    $sharedStringIndex = intval($node->nodeValue);
+
+                if ($hasInlineString) {
+                    // inline strings are formatted this way:
+                    // <c r="A1" t="inlineStr"><is><t>[INLINE_STRING]</t></is></c>
+                    $tNode = $node->getElementsByTagName('t')->item(0);
+                    $rowData[$currentColumnIndex] = trim($tNode->nodeValue);
+                } else if ($hasSharedString) {
+                    // shared strings are formatted this way:
+                    // <c r="A1" t="s"><v>[SHARED_STRING_INDEX]</v></c>
+                    $vNode = $node->getElementsByTagName('v')->item(0);
+                    $sharedStringIndex = intval($vNode->nodeValue);
                     $rowData[$currentColumnIndex] = $this->sharedStringsHelper->getStringAtIndex($sharedStringIndex);
                 } else {
-                    // for inline strings or numbers, just get the value
-                    $rowData[$currentColumnIndex] = trim($node->nodeValue);
+                    // other values are formatted this way:
+                    // <c r="A1"><v>[VALUE]</v></c>
+                    $vNode = $node->getElementsByTagName('v')->item(0);
+                    $rowData[$currentColumnIndex] = intval($vNode->nodeValue);
                 }
             } else if ($this->xmlReader->nodeType == \XMLReader::END_ELEMENT && $this->xmlReader->name === 'row') {
                 // End of the row description

tests/Spout/Reader/XLSXTest.php

@@ -100,6 +100,20 @@ class XLSXTest extends \PHPUnit_Framework_TestCase
         }
     }
 
+    /**
+     * @return void
+     */
+    public function testReadShouldSupportFilesWithoutSharedStringsFile()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_no_shared_strings_file.xlsx');
+
+        $expectedRows = [
+            [10, 11],
+            [20, 21],
+        ];
+        $this->assertEquals($expectedRows, $allRows);
+    }
+
     /**
      * @return void
      */
@@ -231,6 +245,21 @@ class XLSXTest extends \PHPUnit_Framework_TestCase
         $this->assertEquals([], $allRows, 'Sheet with no cells should be correctly processed.');
     }
 
+    /**
+     * @return void
+     */
+    public function testReadShouldSkipFormulas()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_formulas.xlsx');
+
+        $expectedRows = [
+            ['val1', 'val2', 'total1', 'total2'],
+            [10, 20, 30, 21],
+            [11, 21, 32, 41],
+        ];
+        $this->assertEquals($expectedRows, $allRows);
+    }
+
     /**
      * @param string $fileName
      * @return array All the read rows the given file