Merge pull request #199 from box/check_valid_resource_on_close

Check file handle is valid before manipulating it
2016-04-12 11:03:15 -07:00 · 2016-04-12 10:36:59 -07:00 · 2016-04-09 13:20:48 -07:00 · 2016-04-07 08:58:54 +02:00 · 2016-04-06 22:00:47 +02:00 · 2016-03-23 17:02:42 -07:00
53 changed files with 1198 additions and 218 deletions
@@ -0,0 +1,11 @@
+# Ignore all test and documentation for archive
+/tests              export-ignore
+/.editorconfig      export-ignore
+/.gitattributes     export-ignore
+/.gitignore         export-ignore
+/.scrutinizer.yml   export-ignore
+/.travis.yml        export-ignore
+/composer.lock      export-ignore
+/CONTRIBUTING.md    export-ignore
+/logo.png           export-ignore
+/phpunit.xml        export-ignore
@@ -2,7 +2,7 @@

 [![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.svg?branch=master)](http://travis-ci.org/box/spout)
+[![Build Status](https://travis-ci.org/box/spout.svg?branch=master)](https://travis-ci.org/box/spout)
 [![Code Coverage](https://scrutinizer-ci.com/g/box/spout/badges/coverage.png?b=master)](https://scrutinizer-ci.com/g/box/spout/?branch=master)
 [![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)
@@ -98,6 +98,8 @@ For XLSX and ODS files, the number of rows per sheet is limited to 1,048,576. By

 ## Advanced usage

+If you are looking for  how to perform some common, more advanced tasks with Spout, please take a look at the [Wiki](https://github.com/box/spout/wiki). It contains code snippets, ready to be used.
+
 ### Configuring the CSV reader and writer

 It is possible to configure both the CSV reader and writer to specify the field separator as well as the field enclosure:
@@ -108,6 +110,7 @@ use Box\Spout\Common\Type;
 $reader = ReaderFactory::create(Type::CSV);
 $reader->setFieldDelimiter('|');
 $reader->setFieldEnclosure('@');
+$reader->setEndOfLineCharacter("\r");
 ```

 Additionally, if you need to read non UTF-8 files, you can specify the encoding of your file this way:
@@ -157,7 +160,7 @@ Font      | Bold          | `StyleBuilder::setFontBold()`
          | Strikethrough | `StyleBuilder::setFontStrikethrough()`
          | Font name     | `StyleBuilder::setFontName('Arial')`
          | Font size     | `StyleBuilder::setFontSize(14)`
-          | Font color    | `StyleBuilder::setFontSize(Color::BLUE)`<br>`StyleBuilder::setFontSize(Color::rgb(0, 128, 255))`
+          | Font color    | `StyleBuilder::setFontColor(Color::BLUE)`<br>`StyleBuilder::setFontColor(Color::rgb(0, 128, 255))`
 Alignment | Wrap text     | `StyleBuilder::setShouldWrapText()`


@@ -275,10 +278,12 @@ If you just want to check that everything is working as expected, executing the
 If you want to run performance tests, you will need to checkout the `perf-tests` branch. Multiple test suites can then be run, depending on the expected output:

 * `phpunit` - runs the whole test suite (unit + functional + performance tests)
-* `phpunit --testuite no-perf-tests` - only runs the unit and functional tests
-* `phpunit --testuite perf-tests` - only runs the performance tests
+* `phpunit --exclude-group perf-tests` - only runs the unit and functional tests
+* `phpunit --group perf-tests` - only runs the performance tests

-For information, the performance tests take about one hour to run (processing 2 million rows files is not a quick thing).
+For information, the performance tests take about 30 minutes to run (processing 1 million rows files is not a quick thing).
+
+> Performance tests status: [![Build Status](https://travis-ci.org/box/spout.svg?branch=perf-tests)](https://travis-ci.org/box/spout)


 ## Frequently Asked Questions
@@ -29,5 +29,11 @@
        "psr-4": {
            "Box\\Spout\\": "src/Spout"
        }
+    },
+    "extra": {
+        "branch-alias": {
+            "dev-master": "2.5.x-dev"
+        }
    }
+
 }
@@ -281,32 +281,14 @@ class GlobalFunctionsHelper
    }

    /**
-     * Wrapper around global function stream_get_line()
-     * @see stream_get_line()
+     * Wrapper around global function stream_get_wrappers()
+     * @see stream_get_wrappers()
     *
-     * @param resource $handle
-     * @param int $length
-     * @param string|void $ending
-     * @return string|bool
-     */
-    public function stream_get_line($handle, $length, $ending = null)
-    {
-        return stream_get_line($handle, $length, $ending);
-    }
-
-    /**
-     * Wrapper around global function str_getcsv()
-     * @see str_getcsv()
-     *
-     * @param string $input
-     * @param string|void $delimiter
-     * @param string|void $enclosure
-     * @param string|void $escape
     * @return array
     */
-    public function str_getcsv($input, $delimiter = null, $enclosure = null, $escape = null)
+    public function stream_get_wrappers()
    {
-        return str_getcsv($input, $delimiter, $enclosure, $escape);
+        return stream_get_wrappers();
    }

    /**
@@ -19,6 +19,13 @@ abstract class AbstractReader implements ReaderInterface
    /** @var \Box\Spout\Common\Helper\GlobalFunctionsHelper Helper to work with global functions */
    protected $globalFunctionsHelper;

+    /**
+     * Returns whether stream wrappers are supported
+     *
+     * @return bool
+     */
+    abstract protected function doesSupportStreamWrapper();
+
    /**
     * Opens the file at the given file path to make it ready to be read
     *
@@ -62,6 +69,10 @@ abstract class AbstractReader implements ReaderInterface
     */
    public function open($filePath)
    {
+        if ($this->isStreamWrapper($filePath) && (!$this->doesSupportStreamWrapper() || !$this->isSupportedStreamWrapper($filePath))) {
+            throw new IOException("Could not open $filePath for reading! Stream wrapper used is not supported for this type of file.");
+        }
+
        if (!$this->isPhpStream($filePath)) {
            // we skip the checks if the provided file path points to a PHP stream
            if (!$this->globalFunctionsHelper->file_exists($filePath)) {
@@ -72,8 +83,7 @@ abstract class AbstractReader implements ReaderInterface
        }

        try {
-            // Need to use realpath to fix "Can't open file" on some Windows setup
-            $fileRealPath = realpath($filePath);
+            $fileRealPath = $this->getFileRealPath($filePath);
            $this->openReader($fileRealPath);
            $this->isStreamOpened = true;
        } catch (\Exception $exception) {
@@ -81,6 +91,67 @@ abstract class AbstractReader implements ReaderInterface
        }
    }

+    /**
+     * Returns the real path of the given path.
+     * If the given path is a valid stream wrapper, returns the path unchanged.
+     *
+     * @param string $filePath
+     * @return string
+     */
+    protected function getFileRealPath($filePath)
+    {
+        if ($this->isSupportedStreamWrapper($filePath)) {
+            return $filePath;
+        }
+
+        // Need to use realpath to fix "Can't open file" on some Windows setup
+        return realpath($filePath);
+    }
+
+    /**
+     * Returns the scheme of the custom stream wrapper, if the path indicates a stream wrapper is used.
+     * For example, php://temp => php, s3://path/to/file => s3...
+     *
+     * @param string $filePath Path of the file to be read
+     * @return string|null The stream wrapper scheme or NULL if not a stream wrapper
+     */
+    protected function getStreamWrapperScheme($filePath)
+    {
+        $streamScheme = null;
+        if (preg_match('/^(\w+):\/\//', $filePath, $matches)) {
+            $streamScheme = $matches[1];
+        }
+        return $streamScheme;
+    }
+
+    /**
+     * Checks if the given path is an unsupported stream wrapper
+     * (like local path, php://temp, mystream://foo/bar...).
+     *
+     * @param string $filePath Path of the file to be read
+     * @return bool Whether the given path is an unsupported stream wrapper
+     */
+    protected function isStreamWrapper($filePath)
+    {
+        return ($this->getStreamWrapperScheme($filePath) !== null);
+    }
+
+    /**
+     * Checks if the given path is an supported stream wrapper
+     * (like php://temp, mystream://foo/bar...).
+     * If the given path is a local path, returns true.
+     *
+     * @param string $filePath Path of the file to be read
+     * @return bool Whether the given path is an supported stream wrapper
+     */
+    protected function isSupportedStreamWrapper($filePath)
+    {
+        $streamScheme = $this->getStreamWrapperScheme($filePath);
+        return ($streamScheme !== null) ?
+            in_array($streamScheme, $this->globalFunctionsHelper->stream_get_wrappers()) :
+            true;
+    }
+
    /**
     * Checks if a path is a PHP stream (like php://output, php://memory, ...)
     *
@@ -89,7 +160,8 @@ abstract class AbstractReader implements ReaderInterface
     */
    protected function isPhpStream($filePath)
    {
-        return (strpos($filePath, 'php://') === 0);
+        $streamScheme = $this->getStreamWrapperScheme($filePath);
+        return ($streamScheme === 'php');
    }

    /**
@@ -29,6 +29,12 @@ class Reader extends AbstractReader
    /** @var string Encoding of the CSV file to be read */
    protected $encoding = EncodingHelper::ENCODING_UTF8;

+    /** @var string Defines the End of line */
+    protected $endOfLineCharacter = "\n";
+
+    /** @var string */
+    protected $autoDetectLineEndings;
+
    /**
     * Sets the field delimiter for the CSV.
     * Needs to be called before opening the reader.
@@ -68,6 +74,29 @@ class Reader extends AbstractReader
        return $this;
    }

+    /**
+     * Sets the EOL for the CSV.
+     * Needs to be called before opening the reader.
+     *
+     * @param string $endOfLineCharacter used to properly get lines from the CSV file.
+     * @return Reader
+     */
+    public function setEndOfLineCharacter($endOfLineCharacter)
+    {
+        $this->endOfLineCharacter = $endOfLineCharacter;
+        return $this;
+    }
+
+    /**
+     * Returns whether stream wrappers are supported
+     *
+     * @return bool
+     */
+    protected function doesSupportStreamWrapper()
+    {
+        return true;
+    }
+
    /**
     * Opens the file at the given path to make it ready to be read.
     * If setEncoding() was not called, it assumes that the file is encoded in UTF-8.
@@ -78,6 +107,9 @@ class Reader extends AbstractReader
     */
    protected function openReader($filePath)
    {
+        $this->autoDetectLineEndings = ini_get('auto_detect_line_endings');
+        ini_set('auto_detect_line_endings', '1');
+
        $this->filePointer = $this->globalFunctionsHelper->fopen($filePath, 'r');
        if (!$this->filePointer) {
            throw new IOException("Could not open file $filePath for reading.");
@@ -88,6 +120,7 @@ class Reader extends AbstractReader
            $this->fieldDelimiter,
            $this->fieldEnclosure,
            $this->encoding,
+            $this->endOfLineCharacter,
            $this->globalFunctionsHelper
        );
    }
@@ -113,5 +146,7 @@ class Reader extends AbstractReader
        if ($this->filePointer) {
            $this->globalFunctionsHelper->fclose($this->filePointer);
        }
+
+        ini_set('auto_detect_line_endings', $this->autoDetectLineEndings);
    }
 }
@@ -14,7 +14,7 @@ use Box\Spout\Common\Helper\EncodingHelper;
 class RowIterator implements IteratorInterface
 {
    /**
-     * If no value is given to stream_get_line(), it defaults to 8192 (which may be too low).
+     * If no value is given to fgetcsv(), it defaults to 8192 (which may be too low).
     * Alignement with other functions like fgets() is discussed here: https://bugs.php.net/bug.php?id=48421
     */
    const MAX_READ_BYTES_PER_LINE = 32768;
@@ -49,6 +49,9 @@ class RowIterator implements IteratorInterface
    /** @var string End of line delimiter, encoded using the same encoding as the CSV */
    protected $encodedEOLDelimiter;

+    /** @var string End of line delimiter, given by the user as input. */
+    protected $inputEOLDelimiter;
+
    /**
     * @param resource $filePointer Pointer to the CSV file to read
     * @param string $fieldDelimiter Character that delimits fields
@@ -56,12 +59,13 @@ class RowIterator implements IteratorInterface
     * @param string $encoding Encoding of the CSV file to be read
     * @param \Box\Spout\Common\Helper\GlobalFunctionsHelper $globalFunctionsHelper
     */
-    public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $globalFunctionsHelper)
+    public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $endOfLineDelimiter, $globalFunctionsHelper)
    {
        $this->filePointer = $filePointer;
        $this->fieldDelimiter = $fieldDelimiter;
        $this->fieldEnclosure = $fieldEnclosure;
        $this->encoding = $encoding;
+        $this->inputEOLDelimiter = $endOfLineDelimiter;
        $this->globalFunctionsHelper = $globalFunctionsHelper;

        $this->encodingHelper = new EncodingHelper($globalFunctionsHelper);
@@ -124,16 +128,12 @@ class RowIterator implements IteratorInterface
        }

        do {
-            $lineData = false;
-            $utf8EncodedLineData = $this->getNextUTF8EncodedLine();
-            if ($utf8EncodedLineData !== false) {
-                $lineData = $this->globalFunctionsHelper->str_getcsv($utf8EncodedLineData, $this->fieldDelimiter, $this->fieldEnclosure);
-            }
+            $rowData = $this->getNextUTF8EncodedRow();
            $hasNowReachedEndOfFile = $this->globalFunctionsHelper->feof($this->filePointer);
-        } while (($lineData === false && !$hasNowReachedEndOfFile) || $this->isEmptyLine($lineData));
+        } while (($rowData === false && !$hasNowReachedEndOfFile) || $this->isEmptyLine($rowData));

-        if ($lineData !== false) {
-            $this->rowDataBuffer = $lineData;
+        if ($rowData !== false) {
+            $this->rowDataBuffer = $rowData;
            $this->numReadRows++;
        } else {
            // If we reach this point, it means end of file was reached.
@@ -143,24 +143,39 @@ class RowIterator implements IteratorInterface
    }

    /**
-     * Returns the next line, converted if necessary to UTF-8.
-     * Neither fgets nor fgetcsv don't work with non UTF-8 data... so we need to do some things manually.
+     * Returns the next row, converted if necessary to UTF-8.
+     * As fgetcsv() does not manage correctly encoding for non UTF-8 data,
+     * we remove manually whitespace with ltrim or rtrim (depending on the order of the bytes)
     *
-     * @return string|false The next line for the current file pointer, encoded in UTF-8 or FALSE if nothing to read
+     * @return array|false The row for the current file pointer, encoded in UTF-8 or FALSE if nothing to read
     * @throws \Box\Spout\Common\Exception\EncodingConversionException If unable to convert data to UTF-8
     */
-    protected function getNextUTF8EncodedLine()
+    protected function getNextUTF8EncodedRow()
    {
-        // Read until the EOL delimiter or EOF is reached. The delimiter's encoding needs to match the CSV's encoding.
-        $encodedEOLDelimiter = $this->getEncodedEOLDelimiter();
-        $encodedLineData = $this->globalFunctionsHelper->stream_get_line($this->filePointer, self::MAX_READ_BYTES_PER_LINE, $encodedEOLDelimiter);
+        $encodedRowData = fgetcsv($this->filePointer, self::MAX_READ_BYTES_PER_LINE, $this->fieldDelimiter, $this->fieldEnclosure);
+        if (false === $encodedRowData) {
+            return false;
+        }

-        // If the line could have been read, it can be converted to UTF-8
-        $utf8EncodedLineData = ($encodedLineData !== false) ?
-            $this->encodingHelper->attemptConversionToUTF8($encodedLineData, $this->encoding) :
-            false;
+        foreach ($encodedRowData as $cellIndex => $cellValue) {
+            switch($this->encoding) {
+                case EncodingHelper::ENCODING_UTF16_LE:
+                case EncodingHelper::ENCODING_UTF32_LE:
+                    // remove whitespace from the beginning of a string as fgetcsv() add extra whitespace when it try to explode non UTF-8 data
+                    $cellValue = ltrim($cellValue);
+                    break;

-        return $utf8EncodedLineData;
+                case EncodingHelper::ENCODING_UTF16_BE:
+                case EncodingHelper::ENCODING_UTF32_BE:
+                    // remove whitespace from the end of a string as fgetcsv() add extra whitespace when it try to explode non UTF-8 data
+                    $cellValue = rtrim($cellValue);
+                    break;
+            }
+
+            $encodedRowData[$cellIndex] = $this->encodingHelper->attemptConversionToUTF8($cellValue, $this->encoding);
+        }
+
+        return $encodedRowData;
    }

    /**
@@ -172,7 +187,7 @@ class RowIterator implements IteratorInterface
    protected function getEncodedEOLDelimiter()
    {
        if (!isset($this->encodedEOLDelimiter)) {
-            $this->encodedEOLDelimiter = $this->encodingHelper->attemptConversionFromUTF8("\n", $this->encoding);
+            $this->encodedEOLDelimiter = $this->encodingHelper->attemptConversionFromUTF8($this->inputEOLDelimiter, $this->encoding);
        }

        return $this->encodedEOLDelimiter;
@@ -21,9 +21,9 @@ class Sheet implements SheetInterface
     * @param string $encoding Encoding of the CSV file to be read
     * @param \Box\Spout\Common\Helper\GlobalFunctionsHelper $globalFunctionsHelper
     */
-    public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $globalFunctionsHelper)
+    public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $endOfLineCharacter, $globalFunctionsHelper)
    {
-        $this->rowIterator = new RowIterator($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $globalFunctionsHelper);
+        $this->rowIterator = new RowIterator($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $endOfLineCharacter, $globalFunctionsHelper);
    }

    /**
@@ -25,9 +25,9 @@ class SheetIterator implements IteratorInterface
     * @param string $encoding Encoding of the CSV file to be read
     * @param \Box\Spout\Common\Helper\GlobalFunctionsHelper $globalFunctionsHelper
     */
-    public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $globalFunctionsHelper)
+    public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $endOfLineCharacter, $globalFunctionsHelper)
    {
-        $this->sheet = new Sheet($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $globalFunctionsHelper);
+        $this->sheet = new Sheet($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $endOfLineCharacter, $globalFunctionsHelper);
    }

    /**
@@ -47,7 +47,6 @@ class CellValueFormatter

    /**
     * Returns the (unescaped) correctly marshalled, cell value associated to the given XML node.
-     * @TODO Add other types !!
     * @see http://docs.oasis-open.org/office/v1.2/os/OpenDocument-v1.2-os-part1.html#refTable13
     *
     * @param \DOMNode $node
@@ -119,7 +118,8 @@ class CellValueFormatter
    protected function formatFloatCellValue($node)
    {
        $nodeValue = $node->getAttribute(self::XML_ATTRIBUTE_VALUE);
-        $cellValue = is_int($nodeValue) ? intval($nodeValue) : floatval($nodeValue);
+        $nodeIntValue = intval($nodeValue);
+        $cellValue = ($nodeIntValue == $nodeValue) ? $nodeIntValue : floatval($nodeValue);
        return $cellValue;
    }

@@ -19,6 +19,16 @@ class Reader extends AbstractReader
    /** @var SheetIterator To iterator over the ODS sheets */
    protected $sheetIterator;

+    /**
+     * Returns whether stream wrappers are supported
+     *
+     * @return bool
+     */
+    protected function doesSupportStreamWrapper()
+    {
+        return false;
+    }
+
    /**
     * Opens the file at the given file path to make it ready to be read.
     *
@@ -20,6 +20,7 @@ class RowIterator implements IteratorInterface
    const XML_NODE_TABLE = 'table:table';
    const XML_NODE_ROW = 'table:table-row';
    const XML_NODE_CELL = 'table:table-cell';
+    const MAX_COLUMNS_EXCEL = 16384;

    /** Definition of XML attribute used to parse data */
    const XML_ATTRIBUTE_NUM_COLUMNS_REPEATED = 'table:number-columns-repeated';
@@ -127,22 +128,23 @@ class RowIterator implements IteratorInterface

                } else if ($this->xmlReader->isPositionedOnEndingNode(self::XML_NODE_ROW)) {
                    // End of the row description
-                    $isEmptyRow = ($numCellsRead <= 1 && empty($cellValue));
+                    $isEmptyRow = ($numCellsRead <= 1 && $this->isEmptyCellValue($cellValue));
                    if ($isEmptyRow) {
                        // skip empty rows
                        $this->next();
                        return;
                    }

-                    // Only add value if the last read cell is not empty or does not need to repeat cell values.
+                    // Only add the value if the last read cell is not a trailing empty cell repeater in Excel.
+                    // The current count of read columns is determined by counting the values in $rowData.
                    // This is to avoid creating a lot of empty cells, as Excel adds a last empty "<table:table-cell>"
                    // with a number-columns-repeated value equals to the number of (supported columns - used columns).
-                    // In Excel, the number of supported columns is 16384, but we don't want to returns rows with always 16384 cells.
-                    if (!empty($cellValue) || $numColumnsRepeated === 1) {
+                    // In Excel, the number of supported columns is 16384, but we don't want to returns rows with
+                    // always 16384 cells.
+                    if ((count($rowData) + $numColumnsRepeated) !== self::MAX_COLUMNS_EXCEL) {
                        for ($i = 0; $i < $numColumnsRepeated; $i++) {
                            $rowData[] = $cellValue;
                        }
-
                        $this->numReadRows++;
                    }
                    break;
@@ -181,6 +183,17 @@ class RowIterator implements IteratorInterface
        return $this->cellValueFormatter->extractAndFormatNodeValue($node);
    }

+    /**
+     * empty() replacement that honours 0 as a valid value
+     *
+     * @param $value The cell value
+     * @return bool
+     */
+    protected function isEmptyCellValue($value)
+    {
+        return (!isset($value) || trim($value) === '');
+    }
+
    /**
     * Return the current element, from the buffer.
     * @link http://php.net/manual/en/iterator.current.php
@@ -19,7 +19,7 @@ class Sheet implements SheetInterface
    /** @var int ID of the sheet */
    protected $id;

-    /** @var int Index of the sheet, based on order of creation (zero-based) */
+    /** @var int Index of the sheet, based on order in the workbook (zero-based) */
    protected $index;

    /** @var string Name of the sheet */
@@ -27,7 +27,7 @@ class Sheet implements SheetInterface

    /**
     * @param XMLReader $xmlReader XML Reader, positioned on the "<table:table>" element
-     * @param int $sheetIndex Index of the sheet, based on order of creation (zero-based)
+     * @param int $sheetIndex Index of the sheet, based on order in the workbook (zero-based)
     * @param string $sheetName Name of the sheet
     */
    public function __construct($xmlReader, $sheetIndex, $sheetName)
@@ -48,7 +48,7 @@ class Sheet implements SheetInterface

    /**
     * @api
-     * @return int Index of the sheet, based on order of creation (zero-based)
+     * @return int Index of the sheet, based on order in the workbook (zero-based)
     */
    public function getIndex()
    {
@@ -12,6 +12,14 @@ use Box\Spout\Common\Exception\InvalidArgumentException;
 */
 class CellHelper
 {
+    // Using ord() is super slow... Using a pre-computed hash table instead.
+    private static $columnLetterToIndexMapping = [
+        'A' => 0, 'B' => 1, 'C' => 2, 'D' => 3, 'E' => 4, 'F' => 5, 'G' => 6,
+        'H' => 7, 'I' => 8, 'J' => 9, 'K' => 10, 'L' => 11, 'M' => 12, 'N' => 13,
+        'O' => 14, 'P' => 15, 'Q' => 16, 'R' => 17, 'S' => 18, 'T' => 19, 'U' => 20,
+        'V' => 21, 'W' => 22, 'X' => 23, 'Y' => 24, 'Z' => 25,
+    ];
+
    /**
     * Fills the missing indexes of an array with a given value.
     * For instance, $dataArray = []; $a[1] = 1; $a[3] = 3;
@@ -50,34 +58,31 @@ class CellHelper
        }

        $columnIndex = 0;
-        $capitalAAsciiValue = ord('A');
-        $capitalZAsciiValue = ord('Z');
-        $step = $capitalZAsciiValue - $capitalAAsciiValue + 1;

        // Remove row information
-        $column = preg_replace('/\d/', '', $cellIndex);
-        $columnLength = strlen($column);
+        $columnLetters = preg_replace('/\d/', '', $cellIndex);

-        /*
-         * This is how the following loop will process the data:
-         * A   => 0
-         * Z   => 25
-         * AA  => 26   : (26^(2-1) * (0+1)) + 0
-         * AB  => 27   : (26^(2-1) * (0+1)) + 1
-         * BC  => 54   : (26^(2-1) * (1+1)) + 2
-         * BCZ => 1455 : (26^(3-1) * (1+1)) + (26^(2-1) * (2+1)) + 25
-         */
-        foreach (str_split($column) as $single_cell_index)
-        {
-            $currentColumnIndex = ord($single_cell_index) - $capitalAAsciiValue;
+        // strlen() is super slow too... Using isset() is way faster and not too unreadable,
+        // since we checked before that there are between 1 and 3 letters.
+        $columnLength = isset($columnLetters[1]) ? (isset($columnLetters[2]) ? 3 : 2) : 1;

-            if ($columnLength === 1) {
-                $columnIndex += $currentColumnIndex;
-            } else {
-                $columnIndex += pow($step, ($columnLength - 1)) * ($currentColumnIndex + 1);
-            }
-
-            $columnLength--;
+        // Looping over the different letters of the column is slower than this method.
+        // Also, not using the pow() function because it's slooooow...
+        switch ($columnLength) {
+            case 1:
+                $columnIndex = (self::$columnLetterToIndexMapping[$columnLetters]);
+                break;
+            case 2:
+                $firstLetterIndex = (self::$columnLetterToIndexMapping[$columnLetters[0]] + 1) * 26;
+                $secondLetterIndex = self::$columnLetterToIndexMapping[$columnLetters[1]];
+                $columnIndex = $firstLetterIndex + $secondLetterIndex;
+                break;
+            case 3:
+                $firstLetterIndex = (self::$columnLetterToIndexMapping[$columnLetters[0]] + 1) * 676;
+                $secondLetterIndex = (self::$columnLetterToIndexMapping[$columnLetters[1]] + 1) * 26;
+                $thirdLetterIndex = self::$columnLetterToIndexMapping[$columnLetters[2]];
+                $columnIndex = $firstLetterIndex + $secondLetterIndex + $thirdLetterIndex;
+                break;
        }

        return $columnIndex;
@@ -86,12 +91,13 @@ class CellHelper
    /**
     * Returns whether a cell index is valid, in an Excel world.
     * To be valid, the cell index should start with capital letters and be followed by numbers.
+     * There can only be 3 letters, as there can only be 16,384 rows, which is equivalent to 'XFE'.
     *
     * @param string $cellIndex The Excel cell index ('A1', 'BC13', ...)
     * @return bool
     */
    protected static function isValidCellIndex($cellIndex)
    {
-        return (preg_match('/^[A-Z]+\d+$/', $cellIndex) === 1);
+        return (preg_match('/^[A-Z]{1,3}\d+$/', $cellIndex) === 1);
    }
 }
@@ -28,7 +28,6 @@ class CellValueFormatter
    const XML_ATTRIBUTE_STYLE_ID = 's';

    /** Constants used for date formatting */
-    const NUM_DAYS_FROM_JAN_1_1900_TO_JAN_1_1970 = 25569; // actually 25567 but accommodating for an Excel bug with some bisextile years
    const NUM_SECONDS_IN_ONE_DAY = 86400;

    /**
@@ -119,8 +118,7 @@ class CellValueFormatter
        // inline strings are formatted this way:
        // <c r="A1" t="inlineStr"><is><t>[INLINE_STRING]</t></is></c>
        $tNode = $node->getElementsByTagName(self::XML_NODE_INLINE_STRING_VALUE)->item(0);
-        $escapedCellValue = trim($tNode->nodeValue);
-        $cellValue = $this->escaper->unescape($escapedCellValue);
+        $cellValue = $this->escaper->unescape($tNode->nodeValue);
        return $cellValue;
    }

@@ -170,7 +168,8 @@ class CellValueFormatter
        if ($shouldFormatAsDate) {
            return $this->formatExcelTimestampValue(floatval($nodeValue));
        } else {
-            return is_int($nodeValue) ? intval($nodeValue) : floatval($nodeValue);
+            $nodeIntValue = intval($nodeValue);
+            return ($nodeIntValue == $nodeValue) ? $nodeIntValue : floatval($nodeValue);
        }
    }

@@ -183,17 +182,26 @@ class CellValueFormatter
     */
    protected function formatExcelTimestampValue($nodeValue)
    {
-        $numDaysSince1970Jan1 = $nodeValue - self::NUM_DAYS_FROM_JAN_1_1900_TO_JAN_1_1970;
-
        // Fix for the erroneous leap year in Excel
-        if ($nodeValue < self::ERRONEOUS_EXCEL_LEAP_YEAR_DAY) {
-            $numDaysSince1970Jan1++;
+        if (ceil($nodeValue) > self::ERRONEOUS_EXCEL_LEAP_YEAR_DAY) {
+            --$nodeValue;
        }

-        $unixTimestamp = round($numDaysSince1970Jan1 * self::NUM_SECONDS_IN_ONE_DAY);
+        // The value 1.0 represents 1900-01-01. Numbers below 1.0 are not valid Excel dates.
+        if ($nodeValue < 1.0) {
+            return null;
+        }
+
+        // Do not use any unix timestamps for calculation to prevent
+        // issues with numbers exceeding 2^31.
+        $secondsRemainder = fmod($nodeValue, 1) * self::NUM_SECONDS_IN_ONE_DAY;
+        $secondsRemainder = round($secondsRemainder, 0);

        try {
-            $cellValue = (new \DateTime())->setTimestamp($unixTimestamp);
+            $cellValue = \DateTime::createFromFormat('|Y-m-d', '1899-12-31');
+            $cellValue->modify('+' . intval($nodeValue) . 'days');
+            $cellValue->modify('+' . $secondsRemainder . 'seconds');
+
            return $cellValue;
        } catch (\Exception $e) {
            return null;
@@ -143,7 +143,9 @@ class FileBasedStrategy implements CachingStrategyInterface
        }

        $sharedString = null;
-        if (array_key_exists($indexInFile, $this->inMemoryTempFileContents)) {
+
+        // Using isset here because it is way faster than array_key_exists...
+        if (isset($this->inMemoryTempFileContents[$indexInFile])) {
            $escapedSharedString = $this->inMemoryTempFileContents[$indexInFile];
            $sharedString = $this->unescapeLineFeed($escapedSharedString);
        }
@@ -13,9 +13,6 @@ use Box\Spout\Reader\XLSX\Sheet;
 */
 class SheetHelper
 {
-    /** Extension for XML files */
-    const XML_EXTENSION = '.xml';
-
    /** Paths of XML files relative to the XLSX file root */
    const CONTENT_TYPES_XML_FILE_PATH = '[Content_Types].xml';
    const WORKBOOK_XML_RELS_FILE_PATH = 'xl/_rels/workbook.xml.rels';
@@ -79,9 +76,15 @@ class SheetHelper
            $sheetNode = $sheetNodes[$i];
            $sheetDataXMLFilePath = $sheetNode->getAttribute('PartName');

-            $sheets[] = $this->getSheetFromXML($sheetDataXMLFilePath, $i);
+            $sheets[] = $this->getSheetFromXML($sheetDataXMLFilePath);
        }

+        // make sure the sheets are sorted by index
+        // (as the sheets are not necessarily in this order in the XML file)
+        usort($sheets, function ($sheet1, $sheet2) {
+            return ($sheet1->getIndex() - $sheet2->getIndex());
+        });
+
        return $sheets;
    }

@@ -91,60 +94,37 @@ class SheetHelper
     * Then we look at "xl/worbook.xml" to find the sheet entry associated to the found ID.
     * The entry contains the ID and name of the sheet.
     *
-     * If this piece of data can't be found by parsing the different XML files, the ID will default
-     * to the sheet index, based on order in [Content_Types].xml. Similarly, the sheet's name will
-     * default to the data sheet XML file name ("xl/worksheets/sheet2.xml" => "sheet2").
-     *
     * @param string $sheetDataXMLFilePath Path of the sheet data XML file as in [Content_Types].xml
-     * @param int $sheetIndexZeroBased Index of the sheet, based on order in [Content_Types].xml (zero-based)
     * @return \Box\Spout\Reader\XLSX\Sheet Sheet instance
     */
-    protected function getSheetFromXML($sheetDataXMLFilePath, $sheetIndexZeroBased)
+    protected function getSheetFromXML($sheetDataXMLFilePath)
    {
-        $sheetName = $this->getDefaultSheetName($sheetDataXMLFilePath);
-
-        /*
-         * In [Content_Types].xml, the path is "/xl/worksheets/sheet1.xml"
-         * In workbook.xml.rels, it is only "worksheets/sheet1.xml"
-         */
+        // In [Content_Types].xml, the path is "/xl/worksheets/sheet1.xml"
+        // In workbook.xml.rels, it is only "worksheets/sheet1.xml"
        $sheetDataXMLFilePathInWorkbookXMLRels = ltrim($sheetDataXMLFilePath, '/xl/');

        // find the node associated to the given file path
        $workbookXMLResElement = $this->getWorkbookXMLRelsAsXMLElement();
        $relationshipNodes = $workbookXMLResElement->xpath('//ns:Relationship[@Target="' . $sheetDataXMLFilePathInWorkbookXMLRels . '"]');
+        $relationshipNode = $relationshipNodes[0];

-        if (count($relationshipNodes) === 1) {
-            $relationshipNode = $relationshipNodes[0];
-            $sheetId = $relationshipNode->getAttribute('Id');
+        $relationshipSheetId = $relationshipNode->getAttribute('Id');

-            $workbookXMLElement = $this->getWorkbookXMLAsXMLElement();
-            $sheetNodes = $workbookXMLElement->xpath('//ns:sheet[@r:id="' . $sheetId . '"]');
+        $workbookXMLElement = $this->getWorkbookXMLAsXMLElement();
+        $sheetNodes = $workbookXMLElement->xpath('//ns:sheet[@r:id="' . $relationshipSheetId . '"]');
+        $sheetNode = $sheetNodes[0];

-            if (count($sheetNodes) === 1) {
-                $sheetNode = $sheetNodes[0];
-                $escapedSheetName = $sheetNode->getAttribute('name');
+        $escapedSheetName = $sheetNode->getAttribute('name');
+        $sheetIdOneBased = $sheetNode->getAttribute('sheetId');
+        $sheetIndexZeroBased = $sheetIdOneBased - 1;

-                /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-                $escaper = new \Box\Spout\Common\Escaper\XLSX();
-                $sheetName = $escaper->unescape($escapedSheetName);
-            }
-        }
+        /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
+        $escaper = new \Box\Spout\Common\Escaper\XLSX();
+        $sheetName = $escaper->unescape($escapedSheetName);

        return new Sheet($this->filePath, $sheetDataXMLFilePath, $this->sharedStringsHelper, $sheetIndexZeroBased, $sheetName);
    }

-    /**
-     * Returns the default name of the sheet whose data is located
-     * at the given path.
-     *
-     * @param string $sheetDataXMLFilePath Path of the sheet data XML file
-     * @return string The default sheet name
-     */
-    protected function getDefaultSheetName($sheetDataXMLFilePath)
-    {
-        return $this->globalFunctionsHelper->basename($sheetDataXMLFilePath, self::XML_EXTENSION);
-    }
-
    /**
     * Returns a representation of the workbook.xml.rels file, ready to be parsed.
     * The returned value is cached.
@@ -148,7 +148,8 @@ class StyleHelper

        // Default style (0) does not format numeric values as timestamps. Only custom styles do.
        // Also if the style ID does not exist in the styles.xml file, format as numeric value.
-        if ($styleId === self::DEFAULT_STYLE_ID || !array_key_exists($styleId, $stylesAttributes)) {
+        // Using isset here because it is way faster than array_key_exists...
+        if ($styleId === self::DEFAULT_STYLE_ID || !isset($stylesAttributes[$styleId])) {
            return false;
        }

@@ -193,7 +194,8 @@ class StyleHelper
    {
        $customNumberFormats = $this->getCustomNumberFormats();

-        if (!array_key_exists($numFmtId, $customNumberFormats)) {
+        // Using isset here because it is way faster than array_key_exists...
+        if (!isset($customNumberFormats[$numFmtId])) {
            return false;
        }

@@ -37,6 +37,16 @@ class Reader extends AbstractReader
        return $this;
    }

+    /**
+     * Returns whether stream wrappers are supported
+     *
+     * @return bool
+     */
+    protected function doesSupportStreamWrapper()
+    {
+        return false;
+    }
+
    /**
     * Opens the file at the given file path to make it ready to be read.
     * It also parses the sharedStrings.xml file to get all the shared strings available in memory
@@ -15,7 +15,7 @@ class Sheet implements SheetInterface
    /** @var \Box\Spout\Reader\XLSX\RowIterator To iterate over sheet's rows */
    protected $rowIterator;

-    /** @var int Index of the sheet, based on order of creation (zero-based) */
+    /** @var int Index of the sheet, based on order in the workbook (zero-based) */
    protected $index;

    /** @var string Name of the sheet */
@@ -25,7 +25,7 @@ class Sheet implements SheetInterface
     * @param string $filePath Path of the XLSX file being read
     * @param string $sheetDataXMLFilePath Path of the sheet data XML file as in [Content_Types].xml
     * @param Helper\SharedStringsHelper Helper to work with shared strings
-     * @param int $sheetIndex Index of the sheet, based on order of creation (zero-based)
+     * @param int $sheetIndex Index of the sheet, based on order in the workbook (zero-based)
     * @param string $sheetName Name of the sheet
     */
    public function __construct($filePath, $sheetDataXMLFilePath, $sharedStringsHelper, $sheetIndex, $sheetName)
@@ -46,7 +46,7 @@ class Sheet implements SheetInterface

    /**
     * @api
-     * @return int Index of the sheet, based on order of creation (zero-based)
+     * @return int Index of the sheet, based on order in the workbook (zero-based)
     */
    public function getIndex()
    {
@@ -67,7 +67,8 @@ class SheetIterator implements IteratorInterface
     */
    public function next()
    {
-        if (array_key_exists($this->currentSheetIndex, $this->sheets)) {
+        // Using isset here because it is way faster than array_key_exists...
+        if (isset($this->sheets[$this->currentSheetIndex])) {
            $currentSheet = $this->sheets[$this->currentSheetIndex];
            $currentSheet->getRowIterator()->end();

@@ -319,6 +319,11 @@ abstract class AbstractWriter implements WriterInterface
    public function close()
    {
        $this->closeWriter();
+
+        if (is_resource($this->filePointer)) {
+            $this->globalFunctionsHelper->fclose($this->filePointer);
+        }
+
        $this->isWriterOpened = false;
    }
 }
@@ -96,10 +96,6 @@ class Writer extends AbstractWriter
     */
    protected function closeWriter()
    {
-        if ($this->filePointer) {
-            $this->globalFunctionsHelper->fclose($this->filePointer);
-        }
-
        $this->lastWrittenRowIndex = 0;
    }
 }
@@ -56,7 +56,9 @@ abstract class AbstractStyleHelper
    protected function hasStyleAlreadyBeenRegistered($style)
    {
        $serializedStyle = $style->serialize();
-        return array_key_exists($serializedStyle, $this->serializedStyleToStyleIdMappingTable);
+
+        // Using isset here because it is way faster than array_key_exists...
+        return isset($this->serializedStyleToStyleIdMappingTable[$serializedStyle]);
    }

    /**
@@ -10,6 +10,9 @@ namespace Box\Spout\Writer\Common\Helper;
 */
 class CellHelper
 {
+    /** @var array Cache containing the mapping column index => cell index */
+    private static $columnIndexToCellIndexCache = [];
+
    /**
     * Returns the cell index (base 26) associated to the base 10 column index.
     * Excel uses A to Z letters for column indexing, where A is the 1st column,
@@ -21,18 +24,26 @@ class CellHelper
     */
    public static function getCellIndexFromColumnIndex($columnIndex)
    {
-        $cellIndex = '';
-        $capitalAAsciiValue = ord('A');
+        $originalColumnIndex = $columnIndex;

-        do {
-            $modulus = $columnIndex % 26;
-            $cellIndex = chr($capitalAAsciiValue + $modulus) . $cellIndex;
+        // Using isset here because it is way faster than array_key_exists...
+        if (!isset(self::$columnIndexToCellIndexCache[$originalColumnIndex])) {
+            $cellIndex = '';
+            $capitalAAsciiValue = ord('A');

-            // substracting 1 because it's zero-based
-            $columnIndex = intval($columnIndex / 26) - 1;
-        } while ($columnIndex >= 0);
+            do {
+                $modulus = $columnIndex % 26;
+                $cellIndex = chr($capitalAAsciiValue + $modulus) . $cellIndex;

-        return $cellIndex;
+                // substracting 1 because it's zero-based
+                $columnIndex = intval($columnIndex / 26) - 1;
+
+            } while ($columnIndex >= 0);
+
+            self::$columnIndexToCellIndexCache[$originalColumnIndex] = $cellIndex;
+        }
+
+        return self::$columnIndexToCellIndexCache[$originalColumnIndex];
    }

    /**
@@ -12,52 +12,142 @@ class ZipHelper
 {
    const ZIP_EXTENSION = '.zip';

+    /** Controls what to do when trying to add an existing file */
+    const EXISTING_FILES_SKIP = 'skip';
+    const EXISTING_FILES_OVERWRITE = 'overwrite';
+
+    /** @var string Path of the folder where the zip file will be created */
+    protected $tmpFolderPath;
+
+    /** @var \ZipArchive The ZipArchive instance */
+    protected $zip;
+
    /**
-     * Zips the root folder and streams the contents of the zip into the given stream
-     *
-     * @param string $folderPath Path to the folder to be zipped
-     * @param resource $streamPointer Pointer to the stream to copy the zip
-     * @return void
+     * @param string $tmpFolderPath Path of the temp folder where the zip file will be created
     */
-    public function zipFolderAndCopyToStream($folderPath, $streamPointer)
+    public function __construct($tmpFolderPath)
    {
-        $zipFilePath = $this->getZipFilePath($folderPath);
-        $this->zipFolder($folderPath, $zipFilePath);
-        $this->copyZipToStream($zipFilePath, $streamPointer);
+        $this->tmpFolderPath = $tmpFolderPath;
+    }
+
+    /**
+     * Returns the already created ZipArchive instance or
+     * creates one if none exists.
+     *
+     * @return \ZipArchive
+     */
+    protected function createOrGetZip()
+    {
+        if (!isset($this->zip)) {
+            $this->zip = new \ZipArchive();
+            $zipFilePath = $this->getZipFilePath();
+
+            $this->zip->open($zipFilePath, \ZipArchive::CREATE|\ZipArchive::OVERWRITE);
+        }
+
+        return $this->zip;
    }

    /**
-     * @param string $folderPathToZip Path to the folder to be zipped
     * @return string Path where the zip file of the given folder will be created
     */
-    public function getZipFilePath($folderPathToZip)
+    public function getZipFilePath()
    {
-        return $folderPathToZip . self::ZIP_EXTENSION;
+        return $this->tmpFolderPath . self::ZIP_EXTENSION;
    }

    /**
-     * Zips the given folder
+     * Adds the given file, located under the given root folder to the archive.
+     * The file will be compressed.
     *
-     * @param string $folderPath Path of the folder to be zipped
-     * @param string $destinationPath Path where the zip file will be created
+     * Example of use:
+     *   addFileToArchive('/tmp/xlsx/foo', 'bar/baz.xml');
+     *   => will add the file located at '/tmp/xlsx/foo/bar/baz.xml' in the archive, but only as 'bar/baz.xml'
+     *
+     * @param string $rootFolderPath Path of the root folder that will be ignored in the archive tree.
+     * @param string $localFilePath Path of the file to be added, under the root folder
+     * @param string|void $existingFileMode Controls what to do when trying to add an existing file
     * @return void
     */
-    public function zipFolder($folderPath, $destinationPath)
+    public function addFileToArchive($rootFolderPath, $localFilePath, $existingFileMode = self::EXISTING_FILES_OVERWRITE)
    {
-        $zip = new \ZipArchive();
-        if ($zip->open($destinationPath, \ZipArchive::CREATE)) {
-            $this->addFolderToZip($zip, $folderPath);
-            $zip->close();
+        $this->addFileToArchiveWithCompressionMethod(
+            $rootFolderPath,
+            $localFilePath,
+            $existingFileMode,
+            \ZipArchive::CM_DEFAULT
+        );
+    }
+
+    /**
+     * Adds the given file, located under the given root folder to the archive.
+     * The file will NOT be compressed.
+     *
+     * Example of use:
+     *   addUncompressedFileToArchive('/tmp/xlsx/foo', 'bar/baz.xml');
+     *   => will add the file located at '/tmp/xlsx/foo/bar/baz.xml' in the archive, but only as 'bar/baz.xml'
+     *
+     * @param string $rootFolderPath Path of the root folder that will be ignored in the archive tree.
+     * @param string $localFilePath Path of the file to be added, under the root folder
+     * @param string|void $existingFileMode Controls what to do when trying to add an existing file
+     * @return void
+     */
+    public function addUncompressedFileToArchive($rootFolderPath, $localFilePath, $existingFileMode = self::EXISTING_FILES_OVERWRITE)
+    {
+        $this->addFileToArchiveWithCompressionMethod(
+            $rootFolderPath,
+            $localFilePath,
+            $existingFileMode,
+            \ZipArchive::CM_STORE
+        );
+    }
+
+    /**
+     * Adds the given file, located under the given root folder to the archive.
+     * The file will NOT be compressed.
+     *
+     * Example of use:
+     *   addUncompressedFileToArchive('/tmp/xlsx/foo', 'bar/baz.xml');
+     *   => will add the file located at '/tmp/xlsx/foo/bar/baz.xml' in the archive, but only as 'bar/baz.xml'
+     *
+     * @param string $rootFolderPath Path of the root folder that will be ignored in the archive tree.
+     * @param string $localFilePath Path of the file to be added, under the root folder
+     * @param string $existingFileMode Controls what to do when trying to add an existing file
+     * @param int $compressionMethod The compression method
+     * @return void
+     */
+    protected function addFileToArchiveWithCompressionMethod($rootFolderPath, $localFilePath, $existingFileMode, $compressionMethod)
+    {
+        $zip = $this->createOrGetZip();
+
+        if (!$this->shouldSkipFile($zip, $localFilePath, $existingFileMode)) {
+            $normalizedFullFilePath = $this->getNormalizedRealPath($rootFolderPath . '/' . $localFilePath);
+            $zip->addFile($normalizedFullFilePath, $localFilePath);
+
+            if (self::canChooseCompressionMethod()) {
+                $zip->setCompressionName($localFilePath, $compressionMethod);
+            }
        }
    }

    /**
-     * @param \ZipArchive $zip
-     * @param string $folderPath Path of the folder to add to the zip
+     * @return bool Whether it is possible to choose the desired compression method to be used
+     */
+    public static function canChooseCompressionMethod()
+    {
+        // setCompressionName() is a PHP7+ method...
+        return (method_exists(new \ZipArchive(), 'setCompressionName'));
+    }
+
+    /**
+     * @param string $folderPath Path to the folder to be zipped
+     * @param string|void $existingFileMode Controls what to do when trying to add an existing file
     * @return void
     */
-    protected function addFolderToZip($zip, $folderPath)
+    public function addFolderToArchive($folderPath, $existingFileMode = self::EXISTING_FILES_OVERWRITE)
    {
+        $zip = $this->createOrGetZip();
+
        $folderRealPath = $this->getNormalizedRealPath($folderPath) . '/';
        $itemIterator = new \RecursiveIteratorIterator(new \RecursiveDirectoryIterator($folderPath, \RecursiveDirectoryIterator::SKIP_DOTS), \RecursiveIteratorIterator::SELF_FIRST);

@@ -65,14 +155,26 @@ class ZipHelper
            $itemRealPath = $this->getNormalizedRealPath($itemInfo->getPathname());
            $itemLocalPath = str_replace($folderRealPath, '', $itemRealPath);

-            if ($itemInfo->isFile()) {
+            if ($itemInfo->isFile() && !$this->shouldSkipFile($zip, $itemLocalPath, $existingFileMode)) {
                $zip->addFile($itemRealPath, $itemLocalPath);
-            } else if ($itemInfo->isDir()) {
-                $zip->addEmptyDir($itemLocalPath);
            }
        }
    }

+    /**
+     * @param \ZipArchive $zip
+     * @param string $itemLocalPath
+     * @param string $existingFileMode
+     * @return bool Whether the file should be added to the archive or skipped
+     */
+    protected function shouldSkipFile($zip, $itemLocalPath, $existingFileMode)
+    {
+        // Skip files if:
+        //   - EXISTING_FILES_SKIP mode chosen
+        //   - File already exists in the archive
+        return ($existingFileMode === self::EXISTING_FILES_SKIP && $zip->locateName($itemLocalPath) !== false);
+    }
+
    /**
     * Returns canonicalized absolute pathname, containing only forward slashes.
     *
@@ -85,16 +187,30 @@ class ZipHelper
        return str_replace(DIRECTORY_SEPARATOR, '/', $realPath);
    }

+    /**
+     * Closes the archive and copies it into the given stream
+     *
+     * @param resource $streamPointer Pointer to the stream to copy the zip
+     * @return void
+     */
+    public function closeArchiveAndCopyToStream($streamPointer)
+    {
+        $zip = $this->createOrGetZip();
+        $zip->close();
+        unset($this->zip);
+
+        $this->copyZipToStream($streamPointer);
+    }
+
    /**
     * Streams the contents of the zip file into the given stream
     *
-     * @param string $zipFilePath Path to the zip file
     * @param resource $pointer Pointer to the stream to copy the zip
     * @return void
     */
-    protected function copyZipToStream($zipFilePath, $pointer)
+    protected function copyZipToStream($pointer)
    {
-        $zipFilePointer = fopen($zipFilePath, 'r');
+        $zipFilePointer = fopen($this->getZipFilePath(), 'r');
        stream_copy_to_stream($zipFilePointer, $pointer);
        fclose($zipFilePointer);
    }
@@ -24,7 +24,7 @@ class Sheet
    /** @var array Associative array [SHEET_INDEX] => [SHEET_NAME] keeping track of sheets' name to enforce uniqueness */
    protected static $SHEETS_NAME_USED = [];

-    /** @var int Index of the sheet, based on order of creation (zero-based) */
+    /** @var int Index of the sheet, based on order in the workbook (zero-based) */
    protected $index;

    /** @var string Name of the sheet */
@@ -34,7 +34,7 @@ class Sheet
    protected $stringHelper;

    /**
-     * @param int $sheetIndex Index of the sheet, based on order of creation (zero-based)
+     * @param int $sheetIndex Index of the sheet, based on order in the workbook (zero-based)
     */
    public function __construct($sheetIndex)
    {
@@ -45,7 +45,7 @@ class Sheet

    /**
     * @api
-     * @return int Index of the sheet, based on order of creation (zero-based)
+     * @return int Index of the sheet, based on order in the workbook (zero-based)
     */
    public function getIndex()
    {
@@ -103,14 +103,12 @@ class FileSystemHelper extends \Box\Spout\Common\Helper\FileSystemHelper
    protected function createManifestFile()
    {
        $manifestXmlFileContents = <<<EOD
-<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
-<manifest:manifest xmlns:manifest="urn:oasis:names:tc:opendocument:xmlns:manifest:1.0">
+<?xml version="1.0" encoding="UTF-8"?>
+<manifest:manifest xmlns:manifest="urn:oasis:names:tc:opendocument:xmlns:manifest:1.0" manifest:version="1.2">
    <manifest:file-entry manifest:full-path="/" manifest:media-type="application/vnd.oasis.opendocument.spreadsheet"/>
    <manifest:file-entry manifest:full-path="styles.xml" manifest:media-type="text/xml"/>
    <manifest:file-entry manifest:full-path="content.xml" manifest:media-type="text/xml"/>
    <manifest:file-entry manifest:full-path="meta.xml" manifest:media-type="text/xml"/>
-    <manifest:file-entry manifest:full-path="mimetype" manifest:media-type="text/plain"/>
-    <manifest:file-entry manifest:full-path="META-INF/manifest.xml" manifest:media-type="text/xml"/>
 </manifest:manifest>
 EOD;

@@ -265,10 +263,17 @@ EOD;
     */
    public function zipRootFolderAndCopyToStream($streamPointer)
    {
-        $zipHelper = new ZipHelper();
-        $zipHelper->zipFolderAndCopyToStream($this->rootFolder, $streamPointer);
+        $zipHelper = new ZipHelper($this->rootFolder);
+
+        // In order to have the file's mime type detected properly, files need to be added
+        // to the zip file in a particular order.
+        // @see http://www.jejik.com/articles/2010/03/how_to_correctly_create_odf_documents_using_zip/
+        $zipHelper->addUncompressedFileToArchive($this->rootFolder, self::MIMETYPE_FILE_NAME);
+
+        $zipHelper->addFolderToArchive($this->rootFolder, ZipHelper::EXISTING_FILES_SKIP);
+        $zipHelper->closeArchiveAndCopyToStream($streamPointer);

        // once the zip is copied, remove it
-        $this->deleteFile($zipHelper->getZipFilePath($this->rootFolder));
+        $this->deleteFile($zipHelper->getZipFilePath());
    }
 }
@@ -146,7 +146,8 @@ class Worksheet implements WorksheetInterface
        for ($i = 0; $i < $cellsCount; $i++) {
            $currentCellValue = $dataRow[$currentCellIndex];

-            if (!array_key_exists($nextCellIndex, $dataRow) || $currentCellValue !== $dataRow[$nextCellIndex]) {
+            // Using isset here because it is way faster than array_key_exists...
+            if (!isset($dataRow[$nextCellIndex]) || $currentCellValue !== $dataRow[$nextCellIndex]) {
                $numTimesValueRepeated = ($nextCellIndex - $currentCellIndex);
                $data .= $this->getCellContent($currentCellValue, $styleIndex, $numTimesValueRepeated);

@@ -217,6 +218,10 @@ class Worksheet implements WorksheetInterface
     */
    public function close()
    {
+        if (!is_resource($this->sheetFilePointer)) {
+            return;
+        }
+
        fclose($this->sheetFilePointer);
    }
 }
@@ -293,7 +293,7 @@ class Style
        if (!$this->hasSetFontSize && $baseStyle->getFontSize() !== self::DEFAULT_FONT_SIZE) {
            $mergedStyle->setFontSize($baseStyle->getFontSize());
        }
-        if (!$this->hasSetFontSize && $baseStyle->getFontColor() !== self::DEFAULT_FONT_COLOR) {
+        if (!$this->hasSetFontColor && $baseStyle->getFontColor() !== self::DEFAULT_FONT_COLOR) {
            $mergedStyle->setFontColor($baseStyle->getFontColor());
        }
        if (!$this->hasSetFontName && $baseStyle->getFontName() !== self::DEFAULT_FONT_NAME) {
@@ -353,10 +353,19 @@ EOD;
     */
    public function zipRootFolderAndCopyToStream($streamPointer)
    {
-        $zipHelper = new ZipHelper();
-        $zipHelper->zipFolderAndCopyToStream($this->rootFolder, $streamPointer);
+        $zipHelper = new ZipHelper($this->rootFolder);
+
+        // In order to have the file's mime type detected properly, files need to be added
+        // to the zip file in a particular order.
+        // "[Content_Types].xml" then at least 2 files located in "xl" folder should be zipped first.
+        $zipHelper->addFileToArchive($this->rootFolder, self::CONTENT_TYPES_XML_FILE_NAME);
+        $zipHelper->addFileToArchive($this->rootFolder, self::XL_FOLDER_NAME . '/' . self::WORKBOOK_XML_FILE_NAME);
+        $zipHelper->addFileToArchive($this->rootFolder, self::XL_FOLDER_NAME . '/' . self::STYLES_XML_FILE_NAME);
+
+        $zipHelper->addFolderToArchive($this->rootFolder, ZipHelper::EXISTING_FILES_SKIP);
+        $zipHelper->closeArchiveAndCopyToStream($streamPointer);

        // once the zip is copied, remove it
-        $this->deleteFile($zipHelper->getZipFilePath($this->rootFolder));
+        $this->deleteFile($zipHelper->getZipFilePath());
    }
 }
@@ -88,6 +88,10 @@ EOD;
     */
    public function close()
    {
+        if (!is_resource($this->sharedStringsFilePointer)) {
+            return;
+        }
+
        fwrite($this->sharedStringsFilePointer, '</sst>');

        // Replace the default strings count with the actual number of shared strings in the file header
@@ -131,36 +131,38 @@ EOD;
        $rowIndex = $this->lastWrittenRowIndex + 1;
        $numCells = count($dataRow);

-        $data = '<row r="' . $rowIndex . '" spans="1:' . $numCells . '">';
+        $rowXML = '<row r="' . $rowIndex . '" spans="1:' . $numCells . '">';

        foreach($dataRow as $cellValue) {
            $columnIndex = CellHelper::getCellIndexFromColumnIndex($cellNumber);
-            $data .= '<c r="' . $columnIndex . $rowIndex . '"';
-            $data .= ' s="' . $style->getId() . '"';
+            $cellXML = '<c r="' . $columnIndex . $rowIndex . '"';
+            $cellXML .= ' s="' . $style->getId() . '"';

            if (CellHelper::isNonEmptyString($cellValue)) {
                if ($this->shouldUseInlineStrings) {
-                    $data .= ' t="inlineStr"><is><t>' . $this->stringsEscaper->escape($cellValue) . '</t></is></c>';
+                    $cellXML .= ' t="inlineStr"><is><t>' . $this->stringsEscaper->escape($cellValue) . '</t></is></c>';
                } else {
                    $sharedStringId = $this->sharedStringsHelper->writeString($cellValue);
-                    $data .= ' t="s"><v>' . $sharedStringId . '</v></c>';
+                    $cellXML .= ' t="s"><v>' . $sharedStringId . '</v></c>';
                }
            } else if (CellHelper::isBoolean($cellValue)) {
-                $data .= ' t="b"><v>' . $cellValue . '</v></c>';
+                    $cellXML .= ' t="b"><v>' . intval($cellValue) . '</v></c>';
            } else if (CellHelper::isNumeric($cellValue)) {
-                $data .= '><v>' . $cellValue . '</v></c>';
+                $cellXML .= '><v>' . $cellValue . '</v></c>';
            } else if (empty($cellValue)) {
-                $data .= '/>';
+                // don't write empty cells (not appending to $cellXML is the right behavior!)
+                $cellXML = '';
            } else {
                throw new InvalidArgumentException('Trying to add a value with an unsupported type: ' . gettype($cellValue));
            }

+            $rowXML .= $cellXML;
            $cellNumber++;
        }

-        $data .= '</row>';
+        $rowXML .= '</row>';

-        $wasWriteSuccessful = fwrite($this->sheetFilePointer, $data);
+        $wasWriteSuccessful = fwrite($this->sheetFilePointer, $rowXML);
        if ($wasWriteSuccessful === false) {
            throw new IOException("Unable to write data in {$this->worksheetFilePath}");
        }
@@ -176,6 +178,10 @@ EOD;
     */
    public function close()
    {
+        if (!is_resource($this->sheetFilePointer)) {
+            return;
+        }
+
        fwrite($this->sheetFilePointer, '</sheetData>');
        fwrite($this->sheetFilePointer, '</worksheet>');
        fclose($this->sheetFilePointer);
@@ -204,6 +204,15 @@ class ReaderTest extends \PHPUnit_Framework_TestCase
        $this->assertEquals('This is, a comma', $allRows[0][0]);
    }

+    /**
+     * @return void
+     */
+    public function testReadShouldNotTruncateLineBreak()
+    {
+        $allRows = $this->getAllRowsForFile('csv_with_line_breaks.csv', ',');
+        $this->assertEquals("This is,\na comma", $allRows[0][0]);
+    }
+
    /**
     * @return array
     */
@@ -343,6 +352,55 @@ class ReaderTest extends \PHPUnit_Framework_TestCase
        $this->assertEquals($expectedRows, $allRows);
    }

+    /**
+     * https://github.com/box/spout/issues/184
+     * @return void
+     */
+    public function testReadShouldInludeRowsWithZerosOnly()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_zeros_in_row.csv');
+
+        $expectedRows = [
+            ['A', 'B', 'C'],
+            ['1', '2', '3'],
+            ['0', '0', '0']
+        ];
+        $this->assertEquals($expectedRows, $allRows, 'There should be only 3 rows, because zeros (0) are valid values');
+    }
+
+    /**
+     * https://github.com/box/spout/issues/184
+     * @return void
+     */
+    public function testReadShouldCreateOutputEmptyCellPreserved()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_empty_cells.csv');
+
+        $expectedRows = [
+            ['A', 'B', 'C'],
+            ['0', '', ''],
+            ['1', '1', '']
+        ];
+        $this->assertEquals($expectedRows, $allRows, 'There should be 3 rows, with equal length');
+    }
+
+    /**
+     * https://github.com/box/spout/issues/195
+     * @return void
+     */
+    public function testReaderShouldNotTrimCellValues()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_untrimmed_strings.csv');
+
+        $expectedRows = [
+            ['A'],
+            [' A '],
+            ["\n\tA\n\t"],
+        ];
+
+        $this->assertEquals($expectedRows, $allRows, 'Cell values should not be trimmed');
+    }
+
    /**
     * @param string $fileName
     * @param string|void $fieldDelimiter
@@ -377,4 +435,96 @@ class ReaderTest extends \PHPUnit_Framework_TestCase

        return $allRows;
    }
+
+    /**
+     * @return array
+     */
+    public function dataProviderForTestReadCustomEOL()
+    {
+        return [
+            ['csv_with_CR_EOL.csv', "\r"],
+            ['csv_standard.csv', "\n"],
+        ];
+    }
+
+    /**
+     * @dataProvider dataProviderForTestReadCustomEOL
+     *
+     * @param string $fileName
+     * @param string $customEOL
+     * @return void
+     */
+    public function testReadCustomEOLs($fileName, $customEOL)
+    {
+        $allRows = [];
+        $resourcePath = $this->getResourcePath($fileName);
+
+        /** @var \Box\Spout\Reader\CSV\Reader $reader */
+        $reader = ReaderFactory::create(Type::CSV);
+        $reader
+            ->setEndOfLineCharacter($customEOL)
+            ->open($resourcePath);
+
+        foreach ($reader->getSheetIterator() as $sheet) {
+            foreach ($sheet->getRowIterator() as $row) {
+                $allRows[] = $row;
+            }
+        }
+
+        $reader->close();
+
+        $expectedRows = [
+            ['csv--11', 'csv--12', 'csv--13'],
+            ['csv--21', 'csv--22', 'csv--23'],
+            ['csv--31', 'csv--32', 'csv--33'],
+        ];
+        $this->assertEquals($expectedRows, $allRows);
+    }
+
+    /**
+     * @return void
+     */
+    public function testReadCustomStreamWrapper()
+    {
+        $allRows = [];
+        $resourcePath = 'spout://csv_standard';
+
+        // register stream wrapper
+        stream_wrapper_register('spout', SpoutTestStream::CLASS_NAME);
+
+        /** @var \Box\Spout\Reader\CSV\Reader $reader */
+        $reader = ReaderFactory::create(Type::CSV);
+        $reader->open($resourcePath);
+
+        foreach ($reader->getSheetIterator() as $sheet) {
+            foreach ($sheet->getRowIterator() as $row) {
+                $allRows[] = $row;
+            }
+        }
+
+        $reader->close();
+
+        $expectedRows = [
+            ['csv--11', 'csv--12', 'csv--13'],
+            ['csv--21', 'csv--22', 'csv--23'],
+            ['csv--31', 'csv--32', 'csv--33'],
+        ];
+        $this->assertEquals($expectedRows, $allRows);
+
+        // cleanup
+        stream_wrapper_unregister('spout');
+    }
+
+    /**
+     * @expectedException \Box\Spout\Common\Exception\IOException
+     *
+     * @return void
+     */
+    public function testReadWithUnsupportedCustomStreamWrapper()
+    {
+        /** @var \Box\Spout\Reader\CSV\Reader $reader */
+        $reader = ReaderFactory::create(Type::CSV);
+        $reader->open('unsupported://foobar');
+    }
+
 }
@@ -0,0 +1,120 @@
+<?php
+
+namespace Box\Spout\Reader\CSV;
+
+/**
+ * Class SpoutTestStream
+ * Custom stream that reads CSV files located in the tests/resources/csv folder.
+ * For example: spout://foobar will point to tests/resources/csv/foobar.csv
+ *
+ * @package Box\Spout\Reader\CSV
+ */
+class SpoutTestStream
+{
+    const CLASS_NAME = __CLASS__;
+
+    const PATH_TO_CSV_RESOURCES = 'tests/resources/csv/';
+    const CSV_EXTENSION = '.csv';
+
+    /** @var int $position */
+    private $position;
+
+    /** @var resource $fileHandle */
+    private $fileHandle;
+
+    /**
+     * @param string $path
+     * @param int $flag
+     * @return array
+     */
+    public function url_stat($path, $flag)
+    {
+        $filePath = $this->getFilePathFromStreamPath($path);
+        return stat($filePath);
+    }
+
+    /**
+     * @param string $streamPath
+     * @return string
+     */
+    private function getFilePathFromStreamPath($streamPath)
+    {
+        $fileName = parse_url($streamPath, PHP_URL_HOST);
+        return self::PATH_TO_CSV_RESOURCES . $fileName . self::CSV_EXTENSION;
+    }
+
+    /**
+     * @param string $path
+     * @param string $mode
+     * @param int $options
+     * @param string $opened_path
+     * @return bool
+     */
+    public function stream_open($path, $mode, $options, &$opened_path)
+    {
+        $this->position = 0;
+
+        // the path is like "spout://csv_name" so the actual file name correspond the name of the host.
+        $filePath = $this->getFilePathFromStreamPath($path);
+        $this->fileHandle = fopen($filePath, $mode);
+
+        return true;
+    }
+
+    /**
+     * @param int $numBytes
+     * @return string
+     */
+    public function stream_read($numBytes)
+    {
+        $this->position += $numBytes;
+        return fread($this->fileHandle, $numBytes);
+    }
+
+    /**
+     * @return int
+     */
+    public function stream_tell()
+    {
+        return $this->position;
+    }
+
+    /**
+     * @param int $offset
+     * @param int|void $whence
+     * @return bool
+     */
+    public function stream_seek($offset, $whence = SEEK_SET)
+    {
+        $result = fseek($this->fileHandle, $offset, $whence);
+        if ($result === -1) {
+            return false;
+        }
+
+        if ($whence === SEEK_SET) {
+            $this->position = $offset;
+        } else if ($whence === SEEK_CUR) {
+            $this->position += $offset;
+        } else {
+            // not implemented
+        }
+
+        return true;
+    }
+
+    /**
+     * @return bool
+     */
+    public function stream_close()
+    {
+        return fclose($this->fileHandle);
+    }
+
+    /**
+     * @return bool
+     */
+    public function stream_eof()
+    {
+        return feof($this->fileHandle);
+    }
+}
@@ -363,6 +363,79 @@ class ReaderTest extends \PHPUnit_Framework_TestCase
        $this->assertEquals($expectedRows, $allRows);
    }

+    /**
+     * @expectedException \Box\Spout\Common\Exception\IOException
+     *
+     * @return void
+     */
+    public function testReadWithUnsupportedCustomStreamWrapper()
+    {
+        /** @var \Box\Spout\Reader\ODS\Reader $reader */
+        $reader = ReaderFactory::create(Type::ODS);
+        $reader->open('unsupported://foobar');
+    }
+
+    /**
+     * @expectedException \Box\Spout\Common\Exception\IOException
+     *
+     * @return void
+     */
+    public function testReadWithSupportedCustomStreamWrapper()
+    {
+        /** @var \Box\Spout\Reader\ODS\Reader $reader */
+        $reader = ReaderFactory::create(Type::ODS);
+        $reader->open('php://memory');
+    }
+
+    /**
+     * https://github.com/box/spout/issues/184
+     * @return void
+     */
+    public function testReadShouldInludeRowsWithZerosOnly()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_zeros_in_row.ods');
+
+        $expectedRows = [
+            ['A', 'B', 'C'],
+            ['1', '2', '3'],
+            ['0', '0', '0']
+        ];
+        $this->assertEquals($expectedRows, $allRows, 'There should be only 3 rows, because zeros (0) are valid values');
+    }
+
+    /**
+     * https://github.com/box/spout/issues/184
+     * @return void
+     */
+    public function testReadShouldCreateOutputEmptyCellPreserved()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_empty_cells.ods');
+
+        $expectedRows = [
+            ['A', 'B', 'C'],
+            ['0', '', ''],
+            ['1', '1', '']
+        ];
+        $this->assertEquals($expectedRows, $allRows, 'There should be 3 rows, with equal length');
+    }
+
+    /**
+     * https://github.com/box/spout/issues/195
+     * @return void
+     */
+    public function testReaderShouldNotTrimCellValues()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_untrimmed_strings.ods');
+
+        $expectedRows = [
+            ['A'],
+            [' A '],
+            ["\n\tA\n\t"],
+        ];
+
+        $this->assertEquals($expectedRows, $allRows, 'Cell values should not be trimmed');
+    }
+
    /**
     * @param string $fileName
     * @return array All the read rows the given file
@@ -0,0 +1,168 @@
+<?php
+
+namespace Box\Spout\Reader\XLSX\Helper;
+
+/**
+ * Class CellValueFormatterTest
+ *
+ * @package Box\Spout\Reader\XLSX\Helper
+ */
+class CellValueFormatterTest extends \PHPUnit_Framework_TestCase
+{
+    /**
+     * @return array
+     */
+    public function dataProviderForTestExcelDate()
+    {
+        return [
+            [CellValueFormatter::CELL_TYPE_NUMERIC, 42429, '2016-02-29 00:00:00'],
+            [CellValueFormatter::CELL_TYPE_NUMERIC, '146098', '2299-12-31 00:00:00'],
+            [CellValueFormatter::CELL_TYPE_NUMERIC, -700, null],
+            [CellValueFormatter::CELL_TYPE_NUMERIC, 0, null],
+            [CellValueFormatter::CELL_TYPE_NUMERIC, 0.5, null],
+            [CellValueFormatter::CELL_TYPE_NUMERIC, 1, '1900-01-01 00:00:00'],
+            [CellValueFormatter::CELL_TYPE_NUMERIC, 59.999988425926, '1900-02-28 23:59:59'],
+            [CellValueFormatter::CELL_TYPE_NUMERIC, 60.458333333333, '1900-02-28 11:00:00'],
+        ];
+    }
+
+    /**
+     * @dataProvider dataProviderForTestExcelDate
+     *
+     * @param string $cellType
+     * @param int|float|string $nodeValue
+     * @param string|null $expectedDateAsString
+     * @return void
+     */
+    public function testExcelDate($cellType, $nodeValue, $expectedDateAsString)
+    {
+        $nodeListMock = $this->getMockBuilder('DOMNodeList')->disableOriginalConstructor()->getMock();
+
+        $nodeListMock
+            ->expects($this->atLeastOnce())
+            ->method('item')
+            ->with(0)
+            ->will($this->returnValue((object)['nodeValue' => $nodeValue]));
+
+        $nodeMock = $this->getMockBuilder('DOMElement')->disableOriginalConstructor()->getMock();
+
+        $nodeMock
+            ->expects($this->atLeastOnce())
+            ->method('getAttribute')
+            ->will($this->returnValueMap([
+                [CellValueFormatter::XML_ATTRIBUTE_TYPE, $cellType],
+                [CellValueFormatter::XML_ATTRIBUTE_STYLE_ID, 123],
+            ]));
+
+        $nodeMock
+            ->expects($this->atLeastOnce())
+            ->method('getElementsByTagName')
+            ->with(CellValueFormatter::XML_NODE_VALUE)
+            ->will($this->returnValue($nodeListMock));
+
+        $styleHelperMock = $this->getMockBuilder('Box\Spout\Reader\XLSX\Helper\StyleHelper')->disableOriginalConstructor()->getMock();
+
+        $styleHelperMock
+            ->expects($this->once())
+            ->method('shouldFormatNumericValueAsDate')
+            ->with(123)
+            ->will($this->returnValue(true));
+
+        $formatter = new CellValueFormatter(null, $styleHelperMock);
+        $result = $formatter->extractAndFormatNodeValue($nodeMock);
+
+        if ($expectedDateAsString === null) {
+            $this->assertNull($result);
+        } else {
+            $this->assertInstanceOf('DateTime', $result);
+            $this->assertSame($expectedDateAsString, $result->format('Y-m-d H:i:s'));
+        }
+    }
+
+    /**
+     * @return array
+     */
+    public function dataProviderForTestFormatNumericCellValueWithNumbers()
+    {
+        return [
+            [42, 42, 'integer'],
+            [42.5, 42.5, 'double'],
+            [-42, -42, 'integer'],
+            [-42.5, -42.5, 'double'],
+            ['42', 42, 'integer'],
+            ['42.5', 42.5, 'double'],
+            [865640023012945, 865640023012945, 'integer'],
+            ['865640023012945', 865640023012945, 'integer'],
+            [865640023012945.5, 865640023012945.5, 'double'],
+            ['865640023012945.5', 865640023012945.5, 'double'],
+            [PHP_INT_MAX, PHP_INT_MAX, 'integer'],
+            [~PHP_INT_MAX + 1, ~PHP_INT_MAX + 1, 'integer'], // ~PHP_INT_MAX === PHP_INT_MIN, PHP_INT_MIN being PHP7+
+            [PHP_INT_MAX + 1, PHP_INT_MAX + 1, 'double'],
+        ];
+    }
+
+    /**
+     * @dataProvider dataProviderForTestFormatNumericCellValueWithNumbers
+     *
+     * @param int|float|string $value
+     * @param int|float $expectedFormattedValue
+     * @param string $expectedType
+     * @return void
+     */
+    public function testFormatNumericCellValueWithNumbers($value, $expectedFormattedValue, $expectedType)
+    {
+        $styleHelperMock = $this->getMockBuilder('Box\Spout\Reader\XLSX\Helper\StyleHelper')->disableOriginalConstructor()->getMock();
+        $styleHelperMock
+            ->expects($this->once())
+            ->method('shouldFormatNumericValueAsDate')
+            ->will($this->returnValue(false));
+
+        $formatter = new CellValueFormatter(null, $styleHelperMock);
+        $formattedValue = \ReflectionHelper::callMethodOnObject($formatter, 'formatNumericCellValue', $value, 0);
+
+        $this->assertEquals($expectedFormattedValue, $formattedValue);
+        $this->assertEquals($expectedType, gettype($formattedValue));
+    }
+
+    /**
+     * @return array
+     */
+    public function dataProviderForTestFormatStringCellValue()
+    {
+        return [
+            ['A', 'A'],
+            [' A ', ' A '],
+            ["\n\tA\n\t", "\n\tA\n\t"],
+            [' ', ' '],
+        ];
+    }
+
+    /**
+     * @dataProvider dataProviderForTestFormatStringCellValue
+     *
+     * @param string $value
+     * @param string $expectedFormattedValue
+     * @return void
+     */
+    public function testFormatInlineStringCellValue($value, $expectedFormattedValue)
+    {
+        $nodeListMock = $this->getMockBuilder('DOMNodeList')->disableOriginalConstructor()->getMock();
+        $nodeListMock
+            ->expects($this->atLeastOnce())
+            ->method('item')
+            ->with(0)
+            ->will($this->returnValue((object)['nodeValue' => $value]));
+
+        $nodeMock = $this->getMockBuilder('DOMElement')->disableOriginalConstructor()->getMock();
+        $nodeMock
+            ->expects($this->atLeastOnce())
+            ->method('getElementsByTagName')
+            ->with(CellValueFormatter::XML_NODE_INLINE_STRING_VALUE)
+            ->will($this->returnValue($nodeListMock));
+
+        $formatter = new CellValueFormatter(null, null);
+        $formattedValue = \ReflectionHelper::callMethodOnObject($formatter, 'formatInlineStringCellValue', $nodeMock);
+
+        $this->assertEquals($expectedFormattedValue, $formattedValue);
+    }
+}
@@ -73,6 +73,28 @@ class ReaderTest extends \PHPUnit_Framework_TestCase
        }
    }

+    /**
+     * @return void
+     */
+    public function testReadShouldSupportSheetsDefinitionInRandomOrder()
+    {
+        $allRows = $this->getAllRowsForFile('two_sheets_with_sheets_definition_in_reverse_order.xlsx');
+
+        $expectedRows = [
+            ['s1 - A1', 's1 - B1', 's1 - C1', 's1 - D1', 's1 - E1'],
+            ['s1 - A2', 's1 - B2', 's1 - C2', 's1 - D2', 's1 - E2'],
+            ['s1 - A3', 's1 - B3', 's1 - C3', 's1 - D3', 's1 - E3'],
+            ['s1 - A4', 's1 - B4', 's1 - C4', 's1 - D4', 's1 - E4'],
+            ['s1 - A5', 's1 - B5', 's1 - C5', 's1 - D5', 's1 - E5'],
+            ['s2 - A1', 's2 - B1', 's2 - C1', 's2 - D1', 's2 - E1'],
+            ['s2 - A2', 's2 - B2', 's2 - C2', 's2 - D2', 's2 - E2'],
+            ['s2 - A3', 's2 - B3', 's2 - C3', 's2 - D3', 's2 - E3'],
+            ['s2 - A4', 's2 - B4', 's2 - C4', 's2 - D4', 's2 - E4'],
+            ['s2 - A5', 's2 - B5', 's2 - C5', 's2 - D5', 's2 - E5'],
+        ];
+        $this->assertEquals($expectedRows, $allRows);
+    }
+
    /**
     * @return void
     */
@@ -382,6 +404,81 @@ class ReaderTest extends \PHPUnit_Framework_TestCase
        $this->assertEquals($expectedRows, $allRows);
    }

+    /**
+     * @expectedException \Box\Spout\Common\Exception\IOException
+     *
+     * @return void
+     */
+    public function testReadWithUnsupportedCustomStreamWrapper()
+    {
+        /** @var \Box\Spout\Reader\XLSX\Reader $reader */
+        $reader = ReaderFactory::create(Type::XLSX);
+        $reader->open('unsupported://foobar');
+    }
+
+    /**
+     * @expectedException \Box\Spout\Common\Exception\IOException
+     *
+     * @return void
+     */
+    public function testReadWithSupportedCustomStreamWrapper()
+    {
+        /** @var \Box\Spout\Reader\XLSX\Reader $reader */
+        $reader = ReaderFactory::create(Type::XLSX);
+        $reader->open('php://memory');
+    }
+
+    /**
+     * https://github.com/box/spout/issues/184
+     * @return void
+     */
+    public function testReadShouldInludeRowsWithZerosOnly()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_zeros_in_row.xlsx');
+
+        $expectedRows = [
+            ['A', 'B', 'C'],
+            ['1', '2', '3'],
+            ['0', '0', '0']
+        ];
+        $this->assertEquals($expectedRows, $allRows, 'There should be only 3 rows, because zeros (0) are valid values');
+    }
+
+    /**
+     * https://github.com/box/spout/issues/184
+     * @return void
+     */
+    public function testReadShouldCreateOutputEmptyCellPreserved()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_empty_cells.xlsx');
+
+        $expectedRows = [
+            ['A', 'B', 'C'],
+            ['0', '', ''],
+            ['1', '1', '']
+        ];
+        $this->assertEquals($expectedRows, $allRows, 'There should be 3 rows, with equal length');
+    }
+
+
+    /**
+     * https://github.com/box/spout/issues/195
+     * @return void
+     */
+    public function testReaderShouldNotTrimCellValues()
+    {
+        $allRows = $this->getAllRowsForFile('sheet_with_untrimmed_inline_strings.xlsx');
+
+        $expectedRows = [
+            ['A'],
+            [' A '],
+            ["\n\tA\n\t"],
+        ];
+
+        $this->assertEquals($expectedRows, $allRows, 'Cell values should not be trimmed');
+    }
+
+
    /**
     * @param string $fileName
     * @return array All the read rows the given file
@@ -5,6 +5,7 @@ namespace Box\Spout\Writer\ODS;
 use Box\Spout\Common\Type;
 use Box\Spout\Reader\Wrapper\XMLReader;
 use Box\Spout\TestUsingResource;
+use Box\Spout\Writer\Common\Helper\ZipHelper;
 use Box\Spout\Writer\WriterFactory;

 /**
@@ -368,6 +369,30 @@ class WriterTest extends \PHPUnit_Framework_TestCase
        $this->assertValueWasWrittenToSheet($fileName, 1, 'a dream');
    }

+    /**
+     * @return void
+     */
+    public function testGeneratedFileShouldHaveTheCorrectMimeType()
+    {
+        // Only PHP7+ gives the correct mime type since it requires adding
+        // uncompressed files to the final archive (which support was added in PHP7)
+        if (!ZipHelper::canChooseCompressionMethod()) {
+            $this->markTestSkipped(
+                'The PHP version used does not support setting the compression method of archived files,
+                resulting in the mime type to be detected incorrectly.'
+            );
+        }
+
+        $fileName = 'test_mime_type.ods';
+        $resourcePath = $this->getGeneratedResourcePath($fileName);
+        $dataRow = ['foo'];
+
+        $this->writeToODSFile([$dataRow], $fileName);
+
+        $finfo = new \finfo(FILEINFO_MIME_TYPE);
+        $this->assertEquals('application/vnd.oasis.opendocument.spreadsheet', $finfo->file($resourcePath));
+    }
+
    /**
     * @param array $allRows
     * @param string $fileName
@@ -385,6 +385,21 @@ class WriterTest extends \PHPUnit_Framework_TestCase
        $this->assertInlineDataWasWrittenToSheet($fileName, 1, 'control&#039;s _x0015_ &quot;character&quot;');
    }

+    /**
+     * @return void
+     */
+    public function testGeneratedFileShouldHaveTheCorrectMimeType()
+    {
+        $fileName = 'test_mime_type.xlsx';
+        $resourcePath = $this->getGeneratedResourcePath($fileName);
+        $dataRows = [['foo']];
+
+        $this->writeToXLSXFile($dataRows, $fileName);
+
+        $finfo = new \finfo(FILEINFO_MIME_TYPE);
+        $this->assertEquals('application/vnd.openxmlformats-officedocument.spreadsheetml.sheet', $finfo->file($resourcePath));
+    }
+
    /**
     * @param array $allRows
     * @param string $fileName
@@ -3,6 +3,7 @@ require_once(dirname(__DIR__) . '/vendor/autoload.php');

 require_once(dirname(__DIR__) . '/tests/Spout/TestUsingResource.php');
 require_once(dirname(__DIR__) . '/tests/Spout/ReflectionHelper.php');
+require_once(dirname(__DIR__) . '/tests/Spout/Reader/CSV/SpoutTestStream.php');

 // Make sure a timezone is set to be able to work with dates
 date_default_timezone_set('UTC');
@@ -0,0 +1 @@
+csv--11,csv--12,csv--13
csv--21,csv--22,csv--23
csv--31,csv--32,csv--33
@@ -0,0 +1,2 @@
+"This is,
+a comma",csv--12
@@ -0,0 +1,3 @@
+"A","B","C"
+0,,
+1,1,
@@ -0,0 +1,5 @@
+A
+" A "
+"
+	A
+	"
@@ -0,0 +1,3 @@
+"A","B","C"
+1,2,3
+0,0,0
Author	SHA1	Message	Date
Adrien Loison	e9cd7a397e	Merge pull request #199 from box/check_valid_resource_on_close Check file handle is valid before manipulating it	2016-04-12 11:03:15 -07:00
Adrien Loison	8bb42ebc23	Check file handle is valid before manipulating it This fixes issues when something went wrong on reader/writer init and the developer wants to close the reader/writer. The file handle may not be defined so we need to add a check for it, before actually using it.	2016-04-12 10:36:59 -07:00
Adrien Loison	71a6f6a937	Merge pull request #196 from madflow/trim-inline-strings Fix #195	2016-04-09 13:20:48 -07:00
madflow	616925148e	Renamed xlsx file, #195	2016-04-07 08:58:54 +02:00
madflow	6f0f7c9690	Fix #195	2016-04-06 22:00:47 +02:00
Adrien Loison	b69e28050b	Merge pull request #187 from skeleton/issue-183 Fix line breaks on CSV reader	2016-03-23 17:02:42 -07:00
skeleton	d6e8fe4b54	Fix line breaks on CSV reader	2016-03-23 23:26:49 +01:00
Adrien Loison	e321f30c3b	Merge pull request #191 from box/close_file_pointer_when_done_writing Writers did not close written file pointer	2016-03-23 10:45:42 -07:00
Adrien Loison	c31373fb1a	Writers did not close written file pointer	2016-03-23 10:21:27 -07:00
Adrien Loison	6c57125c0c	Merge pull request #189 from madflow/ods-missing-values Fixes for #184	2016-03-21 10:24:38 -07:00
madflow	30837f869d	Coding style and typos	2016-03-20 08:46:30 +01:00
madflow	e60054f3c4	More explicit rule for ignoring empty placeholder cells in Excel ODS #184	2016-03-19 11:34:32 +01:00
madflow	3ee7099c95	Fix zeros treated as missing values #184	2016-03-19 11:34:32 +01:00
madflow	2b1160bb33	Tests for #184	2016-03-19 11:34:31 +01:00
Adrien Loison	049fd990b9	Merge pull request #188 from box/stream_wrapper_support Custom stream wrapper support	2016-03-18 18:09:22 -07:00
Adrien Loison	d2ac54c578	Custom stream wrapper support Added support for custom stream wrappers, such as "fly" or "s3". Support is determined per reader.	2016-03-18 17:09:13 -07:00
Adrien Loison	0c90d102ef	Merge pull request #179 from lichunqiang/gitattributes Add .gitattributes	2016-02-21 18:11:03 -08:00
lichunqiang	e39dcb3847	Add .gitattributes	2016-02-22 09:48:28 +08:00
Adrien Loison	10d1140a95	Merge pull request #175 from IlyaBakhlin/master Fixing the boolean bug.	2016-02-15 16:43:30 -08:00
Ilya Bakhlin	48debbcbca	Simplifying the fix.	2016-02-15 22:11:04 +01:00
Ilya Bakhlin	f4d6fb87ee	Fixing the boolean bug.	2016-02-15 15:55:46 +01:00
Adrien Loison	771afcb5f1	Merge pull request #173 from sfichera/master Support for variable EOL for CSV	2016-02-14 00:20:48 -08:00
Sebastian Fichera	86e26632f6	Added test case for custom EOL characters...	2016-02-12 16:30:18 -06:00
Sebastian Fichera	8614f79da3	Minor fixes in order to be ok with naming conventions and code documentation...	2016-02-11 17:51:24 -06:00
Sebastian Fichera	4827e56cac	Added new public function usage to docs...	2016-02-11 17:15:48 -06:00
Sebastian Fichera	03e85ffc21	Added EOL configuration support while reading CSV files... Enhancement for #172 issue…	2016-02-11 17:12:54 -06:00
Adrien Loison	e4cc8b4eaa	Merge pull request #169 from welcoMattic/patch-1 Update README.md	2016-01-27 13:26:18 -08:00
Mathieu Santo Stefano--Féron	73341d06a9	Update README.md Fix typo	2016-01-27 18:21:06 +01:00
Adrien Loison	209372462b	Merge pull request #167 from box/fix_int_float_value_formatters Fix CellValueFormatter for numeric values	2016-01-14 11:29:01 -08:00
Adrien Loison	4a5da2ad74	Fix CellValueFormatter for numeric values The value passed into the format() function is coming from an XML file and has never been coerced. Therefore, when checking is_int($value), the check always returns false - because it's a string. Changing the check fixes the issue and Spout now correctly parses large numbers.	2016-01-14 11:11:31 -08:00
Adrien Loison	c48c07db99	Merge pull request #165 from box/support_xlsx_sheets_random_order Support XLSX that are defined in random order	2016-01-08 08:50:49 -08:00
Adrien Loison	a804be4844	Support XLSX that are defined in random order Some software generate [Content_Types].xml file with sheets definition in random order. Instead of having the first sheet (id = 1) defined first, it may be defined in 3rd position. Therefore, to read the file in the correct order, sheets order need to be fixed.	2016-01-08 08:42:29 -08:00
Adrien Loison	4bfbb41c95	Merge pull request #162 from box/update_branch_alias Update dev-master branch alias	2015-12-22 12:22:15 +01:00
Adrien Loison	61b71bf379	Update dev-master branch alias	2015-12-22 12:17:15 +01:00
Adrien Loison	3c3294061a	Merge pull request #161 from garak/patch-1 support branch alias	2015-12-21 12:05:16 +01:00
Massimiliano Arione	1fde6d836a	support branch alias	2015-12-21 10:54:15 +01:00
Adrien Loison	05489cda88	Merge pull request #160 from box/reference-wiki Update README.md	2015-12-21 10:52:46 +01:00
Adrien Loison	128db45f22	Update README.md	2015-12-21 10:45:02 +01:00
Adrien Loison	9a85d84a2e	Merge pull request #155 from KiNgMaR/Fix-Dates-Beyond-2037 XLSX Date Support / Fix for years beyond 2037	2015-12-17 10:59:45 -08:00
Ingmar Runge	4407cffeff	XLSX Date Support / Test + Fix for years beyond 2037 This also fixes years < 1902 on 32-bit PHP systems.	2015-12-17 08:52:15 +01:00
Adrien Loison	8d27b3097d	Merge pull request #154 from box/fix_font_color_check Fix hasSetFontColor check	2015-12-08 10:35:49 -08:00
Adrien Loison	2c6cb1ffe5	Fix hasSetFontColor check	2015-12-08 10:31:16 -08:00
Adrien Loison	22daea5f9a	Merge pull request #153 from box/improve_zip_for_mime_detection Improve ZIP interface for better mime detection	2015-12-05 19:13:47 -08:00
Adrien Loison	ed0e8f79cc	Improve ZIP interface for better mime detection The ZipHelper interface is now more generic and allow single files to be added. It supports adding uncompressed files (for PHP7+), which is required to have the mime detection magic work with ODS files. Also fixed a few issues with the created ODS file (thanks to https://odf-validator.rhcloud.com/)	2015-12-05 18:06:13 -08:00
Adrien Loison	44d72d8245	Merge pull request #152 from box/proper_mime_type_detection Proper mime type detection for XLSX files	2015-12-05 00:30:28 -08:00
Adrien Loison	728dd3b399	Proper mime type detection for XLSX files Heuristics to detect proper mime type for XLSX files expect to see certain files at the beginning of the XLSX archive. The order in which the XML files are added therefore matters. Specifically, "[Content_Types].xml" should be added first, followed by the files located in the "xl" folder (at least 1 file).	2015-12-05 00:20:40 -08:00
Adrien Loison	05a9a1b60a	Merge pull request #146 from box/update_perf_tests_doc Update README.md	2015-11-13 10:36:40 -08:00
Adrien Loison	a76624a721	Update README.md	2015-11-13 10:14:10 -08:00
Adrien Loison	90cbb7b5a6	Merge pull request #145 from box/speed_up Various speed improvements	2015-11-12 14:09:52 -08:00
Adrien Loison	f55520661e	Various speed improvements	2015-11-12 13:55:25 -08:00
Adrien Loison	cb5dae22e4	Merge pull request #143 from box/xlsx_skip_empty_cells_on_write [XLSX] Skip empty cells on write	2015-11-10 17:26:26 -08:00
Adrien Loison	2f6193ce20	[XLSX] Skip empty cells on write Since cells are referenced by their coordinates (A2, B4...), it is not necessary to write empty cells. This will reduce the final size of the generated XML and therefore XLSX file.	2015-11-10 17:17:54 -08:00
				`@@ -0,0 +1 @@`
				`csv--11,csv--12,csv--13 csv--21,csv--22,csv--23 csv--31,csv--32,csv--33`