From 03866a6604949854cab60f3901ff0d3486e44373 Mon Sep 17 00:00:00 2001
From: Adrien Loison <adrilo@users.noreply.github.com>
Date: Sun, 29 May 2016 22:16:59 -0700
Subject: [PATCH 1/4] Support XLSX with prefixed XML files (#237)

While the standard is not to have prefixes, some XLSX files have XML files containing a prefix.
Microsoft has a tool that generates such files: https://msdn.microsoft.com/en-us/library/office/gg278316.aspx
---
 src/Spout/Reader/Wrapper/XMLReader.php        |  27 +++-
 src/Spout/Reader/XLSX/Helper/SheetHelper.php  | 128 +++++++-----------
 tests/Spout/Reader/Wrapper/XMLReaderTest.php  |  33 +++++
 tests/Spout/Reader/XLSX/ReaderTest.php        |  17 +++
 .../xlsx/sheet_with_prefixed_xml_files.xlsx   | Bin 0 -> 7084 bytes
 5 files changed, 119 insertions(+), 86 deletions(-)
 create mode 100644 tests/resources/xlsx/sheet_with_prefixed_xml_files.xlsx

diff --git a/src/Spout/Reader/Wrapper/XMLReader.php b/src/Spout/Reader/Wrapper/XMLReader.php
index 42bd92c..94b28eb 100644
--- a/src/Spout/Reader/Wrapper/XMLReader.php
+++ b/src/Spout/Reader/Wrapper/XMLReader.php
@@ -138,9 +138,10 @@ class XMLReader extends \XMLReader
      */
     public function readUntilNodeFound($nodeName)
     {
-        while (($wasReadSuccessful = $this->read()) && ($this->nodeType !== \XMLReader::ELEMENT || $this->name !== $nodeName)) {
-            // do nothing
-        }
+        do {
+            $wasReadSuccessful = $this->read();
+            $isNotPositionedOnStartingNode = !$this->isPositionedOnStartingNode($nodeName);
+        } while ($wasReadSuccessful && $isNotPositionedOnStartingNode);
 
         return $wasReadSuccessful;
     }
@@ -170,7 +171,7 @@ class XMLReader extends \XMLReader
      */
     public function isPositionedOnStartingNode($nodeName)
     {
-        return ($this->nodeType === XMLReader::ELEMENT && $this->name === $nodeName);
+        return $this->isPositionedOnNode($nodeName, XMLReader::ELEMENT);
     }
 
     /**
@@ -179,6 +180,22 @@ class XMLReader extends \XMLReader
      */
     public function isPositionedOnEndingNode($nodeName)
     {
-        return ($this->nodeType === XMLReader::END_ELEMENT && $this->name === $nodeName);
+        return $this->isPositionedOnNode($nodeName, XMLReader::END_ELEMENT);
+    }
+
+    /**
+     * @param string $nodeName
+     * @param int $nodeType
+     * @return bool Whether the XML Reader is currently positioned on the node with given name and type
+     */
+    private function isPositionedOnNode($nodeName, $nodeType)
+    {
+        // In some cases, the node has a prefix (for instance, "<sheet>" can also be "<x:sheet>").
+        // So if the given node name does not have a prefix, we need to look at the unprefixed name ("localName").
+        // @see https://github.com/box/spout/issues/233
+        $hasPrefix = (strpos($nodeName, ':') !== false);
+        $currentNodeName = ($hasPrefix) ? $this->name : $this->localName;
+
+        return ($this->nodeType === $nodeType && $currentNodeName === $nodeName);
     }
 }
diff --git a/src/Spout/Reader/XLSX/Helper/SheetHelper.php b/src/Spout/Reader/XLSX/Helper/SheetHelper.php
index 5f74f44..ae7b8e0 100644
--- a/src/Spout/Reader/XLSX/Helper/SheetHelper.php
+++ b/src/Spout/Reader/XLSX/Helper/SheetHelper.php
@@ -2,7 +2,7 @@
 
 namespace Box\Spout\Reader\XLSX\Helper;
 
-use Box\Spout\Reader\Wrapper\SimpleXMLElement;
+use Box\Spout\Reader\Wrapper\XMLReader;
 use Box\Spout\Reader\XLSX\Sheet;
 
 /**
@@ -17,10 +17,6 @@ class SheetHelper
     const WORKBOOK_XML_RELS_FILE_PATH = 'xl/_rels/workbook.xml.rels';
     const WORKBOOK_XML_FILE_PATH = 'xl/workbook.xml';
 
-    /** Namespaces for the XML files */
-    const MAIN_NAMESPACE_FOR_WORKBOOK_XML_RELS = 'http://schemas.openxmlformats.org/package/2006/relationships';
-    const MAIN_NAMESPACE_FOR_WORKBOOK_XML = 'http://schemas.openxmlformats.org/spreadsheetml/2006/main';
-
     /** @var string Path of the XLSX file being read */
     protected $filePath;
 
@@ -33,12 +29,6 @@ class SheetHelper
     /** @var bool Whether date/time values should be returned as PHP objects or be formatted as strings */
     protected $shouldFormatDates;
 
-    /** @var \Box\Spout\Reader\Wrapper\SimpleXMLElement XML element representing the workbook.xml.rels file */
-    protected $workbookXMLRelsAsXMLElement;
-
-    /** @var \Box\Spout\Reader\Wrapper\SimpleXMLElement XML element representing the workbook.xml file */
-    protected $workbookXMLAsXMLElement;
-
     /**
      * @param string $filePath Path of the XLSX file being read
      * @param \Box\Spout\Reader\XLSX\Helper\SharedStringsHelper Helper to work with shared strings
@@ -62,13 +52,21 @@ class SheetHelper
     public function getSheets()
     {
         $sheets = [];
+        $sheetIndex = 0;
 
-        // Starting from "workbook.xml" as this file is the source of truth for the sheets order
-        $workbookXMLElement = $this->getWorkbookXMLAsXMLElement();
-        $sheetNodes = $workbookXMLElement->xpath('//ns:sheet');
+        $xmlReader = new XMLReader();
+        if ($xmlReader->open('zip://' . $this->filePath . '#' . self::WORKBOOK_XML_FILE_PATH)) {
+            while ($xmlReader->read()) {
+                if ($xmlReader->isPositionedOnStartingNode('sheet')) {
+                    $sheets[] = $this->getSheetFromSheetXMLNode($xmlReader, $sheetIndex);
+                    $sheetIndex++;
+                } else if ($xmlReader->isPositionedOnEndingNode('sheets')) {
+                    // stop reading once all sheets have been read
+                    break;
+                }
+            }
 
-        foreach ($sheetNodes as $sheetIndex => $sheetNode) {
-            $sheets[] = $this->getSheetFromSheetXMLNode($sheetNode, $sheetIndex);
+            $xmlReader->close();
         }
 
         return $sheets;
@@ -79,88 +77,56 @@ class SheetHelper
      * We can find the XML file path describing the sheet inside "workbook.xml.res", by mapping with the sheet ID
      * ("r:id" in "workbook.xml", "Id" in "workbook.xml.res").
      *
-     * @param \Box\Spout\Reader\Wrapper\SimpleXMLElement $sheetNode XML Node describing the sheet, as defined in "workbook.xml"
+     * @param \Box\Spout\Reader\Wrapper\XMLReader $xmlReaderOnSheetNode XML Reader instance, pointing on the node describing the sheet, as defined in "workbook.xml"
      * @param int $sheetIndexZeroBased Index of the sheet, based on order of appearance in the workbook (zero-based)
      * @return \Box\Spout\Reader\XLSX\Sheet Sheet instance
      */
-    protected function getSheetFromSheetXMLNode($sheetNode, $sheetIndexZeroBased)
+    protected function getSheetFromSheetXMLNode($xmlReaderOnSheetNode, $sheetIndexZeroBased)
     {
-        // To retrieve namespaced attributes, some versions of LibXML will accept prefixing the attribute
-        // with the namespace directly (tested on LibXML 2.9.3). For older versions (tested on LibXML 2.7.8),
-        // attributes need to be retrieved without the namespace hint.
-        $sheetId = $sheetNode->getAttribute('r:id');
-        if ($sheetId === null) {
-            $sheetId = $sheetNode->getAttribute('id');
-        }
-
-        $escapedSheetName = $sheetNode->getAttribute('name');
+        $sheetId = $xmlReaderOnSheetNode->getAttribute('r:id');
+        $escapedSheetName = $xmlReaderOnSheetNode->getAttribute('name');
 
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
         $escaper = new \Box\Spout\Common\Escaper\XLSX();
         $sheetName = $escaper->unescape($escapedSheetName);
 
-        // find the file path of the sheet, by looking at the "workbook.xml.res" file
-        $workbookXMLResElement = $this->getWorkbookXMLRelsAsXMLElement();
-        $relationshipNodes = $workbookXMLResElement->xpath('//ns:Relationship[@Id="' . $sheetId . '"]');
-        $relationshipNode = $relationshipNodes[0];
-
-        // In workbook.xml.rels, it is only "worksheets/sheet1.xml"
-        // In [Content_Types].xml, the path is "/xl/worksheets/sheet1.xml"
-        $sheetDataXMLFilePath = '/xl/' . $relationshipNode->getAttribute('Target');
+        $sheetDataXMLFilePath = $this->getSheetDataXMLFilePathForSheetId($sheetId);
 
         return new Sheet($this->filePath, $sheetDataXMLFilePath, $this->sharedStringsHelper, $this->shouldFormatDates, $sheetIndexZeroBased, $sheetName);
     }
 
     /**
-     * Returns a representation of the workbook.xml.rels file, ready to be parsed.
-     * The returned value is cached.
-     *
-     * @return \Box\Spout\Reader\Wrapper\SimpleXMLElement XML element representating the workbook.xml.rels file
+     * @param string $sheetId The sheet ID, as defined in "workbook.xml"
+     * @return string The XML file path describing the sheet inside "workbook.xml.res", for the given sheet ID
      */
-    protected function getWorkbookXMLRelsAsXMLElement()
+    protected function getSheetDataXMLFilePathForSheetId($sheetId)
     {
-        if (!$this->workbookXMLRelsAsXMLElement) {
-            $this->workbookXMLRelsAsXMLElement = $this->getFileAsXMLElementWithNamespace(
-                self::WORKBOOK_XML_RELS_FILE_PATH,
-                self::MAIN_NAMESPACE_FOR_WORKBOOK_XML_RELS
-            );
+        $sheetDataXMLFilePath = '';
+
+        // find the file path of the sheet, by looking at the "workbook.xml.res" file
+        $xmlReader = new XMLReader();
+        if ($xmlReader->open('zip://' . $this->filePath . '#' . self::WORKBOOK_XML_RELS_FILE_PATH)) {
+            while ($xmlReader->read()) {
+                if ($xmlReader->isPositionedOnStartingNode('Relationship')) {
+                    $relationshipSheetId = $xmlReader->getAttribute('Id');
+
+                    if ($relationshipSheetId === $sheetId) {
+                        // In workbook.xml.rels, it is only "worksheets/sheet1.xml"
+                        // In [Content_Types].xml, the path is "/xl/worksheets/sheet1.xml"
+                        $sheetDataXMLFilePath = $xmlReader->getAttribute('Target');
+
+                        // sometimes, the sheet data file path already contains "/xl/"...
+                        if (strpos($sheetDataXMLFilePath, '/xl/') !== 0) {
+                            $sheetDataXMLFilePath = '/xl/' . $sheetDataXMLFilePath;
+                            break;
+                        }
+                    }
+                }
+            }
+
+            $xmlReader->close();
         }
 
-        return $this->workbookXMLRelsAsXMLElement;
-    }
-
-    /**
-     * Returns a representation of the workbook.xml file, ready to be parsed.
-     * The returned value is cached.
-     *
-     * @return \Box\Spout\Reader\Wrapper\SimpleXMLElement XML element representating the workbook.xml.rels file
-     */
-    protected function getWorkbookXMLAsXMLElement()
-    {
-        if (!$this->workbookXMLAsXMLElement) {
-            $this->workbookXMLAsXMLElement = $this->getFileAsXMLElementWithNamespace(
-                self::WORKBOOK_XML_FILE_PATH,
-                self::MAIN_NAMESPACE_FOR_WORKBOOK_XML
-            );
-        }
-
-        return $this->workbookXMLAsXMLElement;
-    }
-
-    /**
-     * Loads the contents of the given file in an XML parser and register the given XPath namespace.
-     *
-     * @param string $xmlFilePath The path of the XML file inside the XLSX file
-     * @param string $mainNamespace The main XPath namespace to register
-     * @return \Box\Spout\Reader\Wrapper\SimpleXMLElement The XML element representing the file
-     */
-    protected function getFileAsXMLElementWithNamespace($xmlFilePath, $mainNamespace)
-    {
-        $xmlContents = $this->globalFunctionsHelper->file_get_contents('zip://' . $this->filePath . '#' . $xmlFilePath);
-
-        $xmlElement = new SimpleXMLElement($xmlContents);
-        $xmlElement->registerXPathNamespace('ns', $mainNamespace);
-
-        return $xmlElement;
+        return $sheetDataXMLFilePath;
     }
 }
diff --git a/tests/Spout/Reader/Wrapper/XMLReaderTest.php b/tests/Spout/Reader/Wrapper/XMLReaderTest.php
index a4deacd..1f7ffc4 100644
--- a/tests/Spout/Reader/Wrapper/XMLReaderTest.php
+++ b/tests/Spout/Reader/Wrapper/XMLReaderTest.php
@@ -198,4 +198,37 @@ class XMLReaderTest extends \PHPUnit_Framework_TestCase
 
         unlink($tempFolder . '/test.xlsx');
     }
+
+    /**
+     * @return array
+     */
+    public function dataProviderForTestIsPositionedOnStartingAndEndingNode()
+    {
+        return [
+            ['<test></test>'], // not prefixed
+            ['<x:test xmlns:x="foo"></x:test>'], // prefixed
+        ];
+    }
+
+    /**
+     * @dataProvider dataProviderForTestIsPositionedOnStartingAndEndingNode
+     *
+     * @param string $testXML
+     * @return void
+     */
+    public function testIsPositionedOnStartingAndEndingNode($testXML)
+    {
+        $xmlReader = new XMLReader();
+        $xmlReader->XML($testXML);
+
+        // the first read moves the pointer to "<test>"
+        $xmlReader->read();
+        $this->assertTrue($xmlReader->isPositionedOnStartingNode('test'));
+        $this->assertFalse($xmlReader->isPositionedOnEndingNode('test'));
+
+        // the seconds read moves the pointer to "</test>"
+        $xmlReader->read();
+        $this->assertFalse($xmlReader->isPositionedOnStartingNode('test'));
+        $this->assertTrue($xmlReader->isPositionedOnEndingNode('test'));
+    }
 }
diff --git a/tests/Spout/Reader/XLSX/ReaderTest.php b/tests/Spout/Reader/XLSX/ReaderTest.php
index 8620ed5..2703799 100644
--- a/tests/Spout/Reader/XLSX/ReaderTest.php
+++ b/tests/Spout/Reader/XLSX/ReaderTest.php
@@ -95,6 +95,23 @@ class ReaderTest extends \PHPUnit_Framework_TestCase
         $this->assertEquals($expectedRows, $allRows);
     }
 
+    /**
+     * @return void
+     */
+    public function testReadShouldSupportPrefixedXMLFiles()
+    {
+        // The XML files of this spreadsheet are prefixed.
+        // For instance, they use "<x:sheet>" instead of "<sheet>", etc.
+        $allRows = $this->getAllRowsForFile('sheet_with_prefixed_xml_files.xlsx');
+
+        $expectedRows = [
+            ['s1 - A1', 's1 - B1', 's1 - C1'],
+            ['s1 - A2', 's1 - B2', 's1 - C2'],
+            ['s1 - A3', 's1 - B3', 's1 - C3'],
+        ];
+        $this->assertEquals($expectedRows, $allRows);
+    }
+
     /**
      * @return void
      */
diff --git a/tests/resources/xlsx/sheet_with_prefixed_xml_files.xlsx b/tests/resources/xlsx/sheet_with_prefixed_xml_files.xlsx
new file mode 100644
index 0000000000000000000000000000000000000000..d8bbc39f5fda8ba0679a47e3df108e0b5f529216
GIT binary patch
literal 7084
zcmd^EWmFVev>r-IX@N_3GlDb%0un<?hr~E^gOrrCAV>-VQUU`=qjX3~DJdX0NJw{w
z3JmZDYw+rIz4!a=AG2l+JI>kXJNxYYp{9g}P67Y`umDV<wIFPNc?dTO0MLLA01zX7
zHMrsE;A-yR3fA&;GIucodf40b#&@ft(UXLL&0mdNe3wjaD!MK`QGg2uO!r)5uXa+k
zJ+cG(FoC7Il7Pz4Pmh)^pu!hVeSy=*`}S82L=3p;GqD>Xi#}}}TO3xi^&#AO#icMi
zQO{fQ%^XVNf<>(1@1-+VS`_B!1N%AcL8=-wxvDbx3uS1$G?BpjIO)xrDwOzwY2(tI
zWC;cJ+UYdUYg4s_f*gWuk|G8unAxv9%y|!{bF(slN@7+z)dshAC<c|7cD<6huvO#{
zo>qfS`g!)A!-K`Htd_|(M?%6hoo${6IN0v%bM#=%s}u$DPnL}9*4T7(UIz2~;X(SM
z#bqO~vpZ<~Q|EIQRVhwW0J4!LY@+rt#m9{z9jDB%+rb!8w+xy!uSb|15-0C00&aNo
zH`Uk$WckPq;j4YmE&#CpumOZf=uf$T_y%@1w{yAjL+9HGMA1Z>(IHMC6%_!${iy>y
z)2w`?;?PP*AXgd)>kOiX>$bniO<5~oBWvan)-jQk9Ne=r+5<I6s}C21#e*If00R!v
zCwsO|4y*Lio}r0k;zI?fd->$BDXfRBLZWD(=>o2mhIU(YCIwja0~n7OSX^Dnpxyv1
z=g>H;CvH}Q5378Mc&U#8s;?W!32-_%wh}LT^)ii-g*MUODwYq6)}dkL-@ZV;MK!>t
z#kZ(A*-41TY%q-(L5*QPuUUsrm)cE~@~Cnp4_qk1yyd9+{xX^9BAS>%(`NC+Da=l>
z&-IPo$3;|6=xK(#9|{kIL{SE1(Cejm(Wa}UpW?}9KDxpxE-!A%c4uQw;7V1jaR22i
zHEjBiYkAn08xIQ=jrh53S3?$G7LWCxOszz7UkCh1<nPJGMkLbA(NxXZ(dpm$UW&V+
z^azpfw`c$W!Fe@^lhYa8D>b`STX_iJk6+(Su8_sj&L`vmQF&<D1nkb(Q9Nv`TC;z*
zV@vgg$roiThdE~j1N*7FWv6k!)#u$~#wI-hx4|qojY8hl&sP&Ni0Wc&crBDTZuX60
z^tVLh>QGiN4KSaCJ2Uc_>JUCrPZm{&pp`8C(?zCaPf7RmgKR;!2F}YtlhKSU*BEn4
z_7W`ZfcCJ_hBqeBN~SjK{&4V@G!mU_MjFlyM0$(MrCw7<mX>S@I)llUu>-QGrYsp9
zVUDdzdDL=!_@v}QSW;^w*=}Nv*6*_D?BmFsgu#?l<)lUEvRrp|dY_?vvX@ly6UcbW
z09#I%4$2U(m~aidV^5`jQnX%JNbqKU87~5oI2Jk2PAeCP|0!%)l{}het?*N`p+@9J
zW@=y#qeL6Noe}v-G%jUqc}Cqw?FKLf)s;*5FIVzTm<Jh&C&<-oeM^TegsUXkCKdNr
zXnGC7C=g5kPus#4m-V<_zT8M!e=clc3@vVLSHQ3dkPY?JJ2DP=)@!_Tl96~&J`uo8
zwfnST4*cK8jD<KqCy1#n#M1odt@<4dgy*ze%$?n>P0f*@_+6diPwMvOt`IYbE95U>
zIR0v9>ga6F?S#OixwEVFx%l6T05nv7o&f<$4nz@9|3CRL6%-M;YGPq3Y{tXOFJ#6K
z5fTwLHRt0s<G;!e5#SL7I=R@JnK~86s+^(vekqaXQ(;vFYev?~*G*&zi5^R?aan$l
zX>&Kb3Gx;sr3&MW3E?Mbdo<tUy*VV-+c|a(yskap5a^aOzz&fDiM<`1%bCo{ZYge}
zD~wPwot{jl;7XW!34|Iw>W(LQ;3jQ70ps9tmv>RfM8%+>Hb;S)?J?-hpIqGQS{ko+
zcCEChtSFx(yH5|jxjKfa?@#2LPmRM30YzfJ?*=ZZ4yTm~=wQ2wXOiy7u1$fGH8I$n
z0|IGM4)*mT&BPMJD(K$A*0Ne&>20dAP3?82<GVGI#&}N*VAI%u-Y9sc8W}k{fJpD#
zZG^s{ZKrfkXj@dM!Af{&^=4-2;B~4~V@ttNa}^wO2RnXSQP>cFXdd!mJ+WhXZ+evV
zzfp!L6aWvqznuJ8wLvKSsk+n7W8+L9{;Y~{at5rQYIhx-ZA}~<ZO=;1*J9s0B<_I9
zOMo#|s5fM3sV_|;OP<VM@c>tEa+LOktwB0{^Qgugs1|iZ16_&4T5S-vQ@*%t3qL-n
zG<@12bx&<>^4g{HEOgnv!@9f0Awm2%OP?EaSM!rAT&Y<_E99TiyTM8{NigcpPXY9y
zOLU;1m1kTgH@Mibp`xVscpkT74Da=|;SH)Kp~yySFQv^%8aZd@XXMcm8?}r?YS)Mj
zh%R|h_?mQBb87JBsT9*Ytcb9|ZPgx!75LF5TV{<i%5p6R2rn9z_dNH=mKx-m4`#0`
zc{1zieESop4%v_7Vga1O8bHcE@>_cdv@c-*0GGZc#l_Xr&fMi|{nhGr7^fpK5&{Rn
zglN1!<oy%LT98YTuc4u>NIGhQM#DryIu*;>G$JdJ-=y}u1*Slzo;}q&Z@eKQkLh!I
z;yF3F7Z^l55W=qi3hs=75?lF7S&(^+>ll7%at&B@+yUcb{VXrd<-|Q`cSuB2Y%ip5
zvBF6B+OQx2qzhHVHJi{8dSWdg1^Xh{Qf^XN*M)9Ev;mY5Q7*_9$k0ivU^YB5>p6Ve
zxqzKYaQBsSQoCVs&0w>r>I@G1l~}sjW$~G!L9WMaL=jIrO=1Uww%#%(X?bzkmov@W
zPjA~c;%l>1lUj09)V@Npn#(Z_(W7nVLwlHs4dm-%O>9yI*4xQg#a7C62Pp**Tu(Jv
zEYW(@QxTF-?Jh)-wmY+Wgc31x5kIgq-QsD)cqu#_XoedfJ#}<vy-8@ho#zXfJ#qq{
z)R8V)h-U%Hrka;lqE)_-<gQb>T=uw9QQ4qJ4azhV*G_Ep0sS9=Y53%6Tl8LKLVDo#
z)=ficA+(op^14J4>CvzttQ0P%9tiP3gz2#_-rFG=QaTge*t(t5;rIE*Tx~#Cye{w<
z?_<efUWX?UGciv^n&AGkN^ZEEs3;q8xYP0V%38G|bs28WNow!z9HpeiQ-?DvZx{Pr
zIv93cfdsNa!Z)!j(#VbT$l@0EH)wyr<M(3x{c5>dncE{v{I{B!ctCO!ia=03B>+J5
z4MpG7zJ8uTX<c8%aZHf>gfn=fkjub1EyIEEntOevbx_;tTSo37B3nfgkZ8W+4R0!*
z+j>H4S8rEShuCA>x*h5Ui+(c#S@nnCSFd1DCmj~Hufyk(fHCZgZ3=;7cebOigKIz0
zF-!q?qpZg(Q<W~^X26~iQ(+p}0&_QAi)?jEn^g$2umZ^C@@yxbsJ@79W=kvcCn4BX
zfmKS=KT;W6XiN=-6RD_iZK|+8%!s)`3M0ONpELCU)+-0KmnZ1d0s2VJ>2i69P7_yc
z>hg1}nJr{r_h~r_4nPCdCB%zlhBEF+>iJ6o=ORmn)o5{Hx}`9YhYu*q<fsbzv3X6E
zEmUO9T7)e{Z`_^TanH=GeW!+jcOmGcnVpFPeed3Ku_j7=Yt&G8j!oXYxwco+w7~e~
zz}!r2db>R@aYv}K)n)W*=_URT8#?96PPl=(`<d0lo7LUurF0dX!PE1RRrS(pc`;KS
zIswm~aBIYy=Ejfn%v>Y-_{KKVfAm%kH%41l$p(YR)3QwTJgyr7rgWs($<pANLmmpg
zi;pkVgk`vMrtEsU0ktt}`kff+M!c~jxG%ZD;bXIlH(R!MS>^GcQj5c<mj<BF(2pre
zLbs`wdJ@5YJBx|5f!EW0_jk4}1<*@~eUCSwVBf=v3+%3Q5Bs0OKdrractzsmIHi*X
zP8wuNef<^T%#@%gx;MY#F<+jfsk>YkTX2m`KxS}q3>slVOf^{yd4RZ9)U*-HSRkS2
zJ>Mp?P2blGs3|c*t$D%YnelsUd`GUD`9nU*ol=p$g6{*o324ZNKkJr>C9#&3y->Hh
zF2k&OH%wXVqFMwxXmj8nX_iUx2&L_b771voDn?DH6uybeZ8;CFo+3)&N7%QTlZz4=
zDqP6+*myZ$a+v0MhV(Jy_Zc+87u<x&c&-{;V$8bXC`CwdX#&jCMB)K78$^>~BUdI^
zlpCb3Vte94+(l|$;!j=npi7Nqs|TwAgMB!2%yk4G5}&l*aVIb6<2;}K3xYKbNL5h6
zdu7oC?0`G>7B)?o`f6YwJEg3E34491X|DSkeo;>F0t=PrYa?X6UOD=XZ&=CopUJ*Q
zJ%vVkQufUn#gm@Of*Y5*$?hFo=JZ*2Fyy~&@KKjDKz*IXI~;qI3^&l>&Tw6Ah4<C?
zIGYU6L9PV${Crp=5zlQkBQ&{oOugVEdgW{y`Ft!7mTOaVnT9khMYpC`?80-7Z2G$K
z<B9Ov@zS^!4RYcvfS^d!l0usth7s-eb@1GkRAy(Ddr`X3e(y>NgNONslz?1Eyi|8S
zzPO{93<Zi~<yQ*sed3y8!Nl=7UA#v#6(f;Gi+gXa4+U>I7C5FEF5^sy-k<bjDpXfV
zFfvMG0!BAy^7q(!k#0c-D|4M^nXf=)aOSL)0+}AvCtSJesA@D;DDj>oZz%6`Z8j6c
z(C!8e(0WDZM1uurXyxU=vEk#xv;RzB-O4T1?;W1EK&zFgY-8RN_K;C=QL+c?p>a2O
zSX`<wbLKb7x>%CB<rQ|#n&-US%2ByK2iS=hH6Feh!DdpxDyJD9;Jh8C`@WXYCVSNK
zJ^5*JV@2Np+2-k*(yc~k@3G`Om!xr89;`vp^;_{_F<rj#!7$pZwNSOLY=~sZrQtHr
zA<i`cQfe!1J9e|ttw6Q4X#fvbeLB`Z()A3xdnDGX<$LeMW8>XY$8V=|`E@P<V%6&&
zj9FuG5iHm<KfI4x!qO*L{NfG|Txc_Vykpg#5=VPO-<`eWgy3!ZB}zw|_J3%bdnu|@
zCJ*taWYV=H51A=NEF{=JH8a++76M38RP`Tix|Zc6-Hgf^rk-fh8X21~y&Sg0e-gEx
zPr9vR)J7RuO>(E7z-cqyV{911)>fWIpa4E0+xa<X$KEqHnYg9=vj+a$a)sZ{W-v+o
zvesOoAIgu)|E;7S;XyRi5sd*2qFciMR?^S+Nez#r&YD=+QWMXcX*DYB+Cz$1Om$MA
zj55LP>=!Gr*!38vuRSb}0yaAAdBEdSc*HTB)%TvSzWoP{IC&>y{kTC7B6U=ja*ZeV
z6jx)OaxGaj*d#`}g<{&`qaQ_rtJt}v^AaKsiu^UcgcUbWRgc$cpmR;6A9Mx8Gy<@4
zEX2YWib%TNZoD+KhEcze(HgE~qt7YMmBFP2=HSq2inb04ZnZnPxfyd#Tvp(J798n#
zclJTB;XBds>Ea83g&DrS(1nsa&oZ`W)6%3qc$;V2YlLfhQ0#GKW-sqIJj!ft?VFiv
ztw*iAf||S?C@=1Rk8$td`eojG@5WR+<M$>>hceui?!U5UUYSjraO|+#H8gjEl-HpX
zN4dCX$TlxEsy+7WKSt}9_mFtCd)jhS{>w7f^<y+K%F|HgrHf$2>b<zdH|#e<s-~ri
zKK^BZ#zSV`#md~=6*(|{JE{B8>=f$=9cRR8k^QFo)#U62wvYKfc9`_03v<uNt54uy
z)9caElderzr0PYA$BHx8QsPFnU4Blj9?<3>(eKUY2;bkX9qn}>*d)(lF1@WiO!t;H
zud-3o$;9OjZvDE20}D<kjIj)Kfi^{7s$Wb{qvJ-jYG=6IUcM!pa4+@^)>(43>h^^T
zwI!>z%N6@Em{lX0{RXm<&?yHD>4nc(zU9L0*O{FC#Rqs>H48I<UI(SqnPwi?S+dZZ
zXibBA52{v0j6C+e`x`78rt+@&jAWKegrlfvj0|6T@VrK`mp_3+1xCzQOS>`@OMcZT
zkVnQRgn9a4v+CooSU#U2@;_(j>i^>uEg|SJD&{Y1{s;i^A-cH#ND<#ziqhiy5%>OL
zNB<go)_C>F5J1KDStV8r3N~+;_eh<6g<w8W^!kMG7U(6af&q&g*)0cmyHogb+{|Ww
z>HdPD?i^bDf=2K=gPM4Mka>qJ;UMF}u6JFS#p><~)38(`0Mrxww$KHn)Mha+EXti}
z;YCkLkei^S&Bve5K{dg4TmPDlZq#O7wQ6ro#=u$!9iMT~^M_%9V%Y}KqDGn29*a(c
zEVN<-W=GUoF$`vMKxUm0<E|iXi4@jqQ(@g0aEB@9hIZ^lXZNBlmGpFfGG+>x^YfDc
z`qgSqPlhgs^4fbKdz(3;2^*ePLoLH*Jni?UDK#6?l&0C@S6L1EaatSOYxiJ0)XlTY
zTORykPw;M@1dcU-;dkw6a_({be5gsaD}POpnkG7gb!a;5l(sXFvomlx3Ij@=f)_2{
z8n?W~M7|E2_xs%RPLe{W8CMg7EP0JNtEiV(Z=9R`O^%G3AYZbgOEgdcL!`<mK5ql7
zg10O8sQ_k6N!@tk5~i9G3WAsTZ_*fn?sM`73K{9(_YVJz7e$IO0uDds=wE~Wn<quu
z1pc)7I!OO5*x5LNly1bX&as5Q#``x<h&0<dZ}y!v{9Dj7mI^6l2&Ns$S^bFldtT01
zE2O~AS)Jps{@eVFY(NSo;)eYQ`qTUzr}f`{&R8v^T>q<|-#M-ye*esJA^nE_pZxwW
zIj_H+$r-1E6mrb(49}1C{(c<4B1hz#!Jm=(Z_Wo9=k|HC^PSDVfcs~f1nI{O0VxEN
zgd|P=w$)!(9*^_Y^v6K`W5a(QsgV)u&zt=mu#sW^+`l31qYzX(V#Vk8j5Pi|+h^?u
zQZ6FK`|rj-n~`4^e*UpNYe$eG{G091<^;Kb-+P*~=Ljjc5dRwcbpc<Ue0{&qd->C&
ihV*g|Q9=LU<?9otriAge5MBmc2ACo)10U&`-Twf?d+R6w

literal 0
HcmV?d00001


From 251c0bebc1c6379a5620cb9d053bb48d609fa9ab Mon Sep 17 00:00:00 2001
From: Adrien Loison <adrilo@users.noreply.github.com>
Date: Sun, 29 May 2016 23:22:57 -0700
Subject: [PATCH 2/4] Adding open_file_in_zip() helper function to XMLReader
 (#238)

---
 src/Spout/Reader/ODS/SheetIterator.php        |  6 +-
 src/Spout/Reader/Wrapper/XMLReader.php        | 56 +++++-----------
 src/Spout/Reader/XLSX/Helper/SheetHelper.php  |  4 +-
 src/Spout/Reader/XLSX/Helper/StyleHelper.php  |  3 +-
 tests/Spout/Reader/Wrapper/XMLReaderTest.php  | 64 +++++--------------
 tests/Spout/Writer/ODS/WriterTest.php         |  3 +-
 .../Spout/Writer/ODS/WriterWithStyleTest.php  | 21 +++---
 .../Spout/Writer/XLSX/WriterWithStyleTest.php | 18 ++++--
 8 files changed, 64 insertions(+), 111 deletions(-)

diff --git a/src/Spout/Reader/ODS/SheetIterator.php b/src/Spout/Reader/ODS/SheetIterator.php
index d0010bd..4f2ec01 100644
--- a/src/Spout/Reader/ODS/SheetIterator.php
+++ b/src/Spout/Reader/ODS/SheetIterator.php
@@ -15,6 +15,8 @@ use Box\Spout\Reader\Wrapper\XMLReader;
  */
 class SheetIterator implements IteratorInterface
 {
+    const CONTENT_XML_FILE_PATH = 'content.xml';
+
     /** Definition of XML nodes name and attribute used to parse sheet data */
     const XML_NODE_TABLE = 'table:table';
     const XML_ATTRIBUTE_TABLE_NAME = 'table:name';
@@ -63,8 +65,8 @@ class SheetIterator implements IteratorInterface
     {
         $this->xmlReader->close();
 
-        $contentXmlFilePath = $this->filePath . '#content.xml';
-        if ($this->xmlReader->open('zip://' . $contentXmlFilePath) === false) {
+        if ($this->xmlReader->openFileInZip($this->filePath, self::CONTENT_XML_FILE_PATH) === false) {
+            $contentXmlFilePath = $this->filePath . '#' . self::CONTENT_XML_FILE_PATH;
             throw new IOException("Could not open \"{$contentXmlFilePath}\".");
         }
 
diff --git a/src/Spout/Reader/Wrapper/XMLReader.php b/src/Spout/Reader/Wrapper/XMLReader.php
index 94b28eb..c979819 100644
--- a/src/Spout/Reader/Wrapper/XMLReader.php
+++ b/src/Spout/Reader/Wrapper/XMLReader.php
@@ -14,66 +14,44 @@ class XMLReader extends \XMLReader
 {
     use XMLInternalErrorsHelper;
 
+    const ZIP_WRAPPER = 'zip://';
+
     /**
-     * Set the URI containing the XML to parse
-     * @see \XMLReader::open
+     * Opens the XML Reader to read a file located inside a ZIP file.
      *
-     * @param string $URI URI pointing to the document
-   	 * @param string|null|void $encoding The document encoding
-   	 * @param int $options A bitmask of the LIBXML_* constants
+     * @param string $zipFilePath Path to the ZIP file
+     * @param string $fileInsideZipPath Relative or absolute path of the file inside the zip
      * @return bool TRUE on success or FALSE on failure
      */
-    public function open($URI, $encoding = null, $options = 0)
+    public function openFileInZip($zipFilePath, $fileInsideZipPath)
     {
         $wasOpenSuccessful = false;
-        $realPathURI = $this->convertURIToUseRealPath($URI);
+        $realPathURI = $this->getRealPathURIForFileInZip($zipFilePath, $fileInsideZipPath);
 
         // HHVM does not check if file exists within zip file
         // @link https://github.com/facebook/hhvm/issues/5779
-        if ($this->isRunningHHVM() && $this->isZipStream($realPathURI)) {
+        if ($this->isRunningHHVM()) {
             if ($this->fileExistsWithinZip($realPathURI)) {
-                $wasOpenSuccessful = parent::open($realPathURI, $encoding, $options|LIBXML_NONET);
+                $wasOpenSuccessful = $this->open($realPathURI, null, LIBXML_NONET);
             }
         } else {
-            $wasOpenSuccessful = parent::open($realPathURI, $encoding, $options|LIBXML_NONET);
+            $wasOpenSuccessful = $this->open($realPathURI, null, LIBXML_NONET);
         }
 
         return $wasOpenSuccessful;
     }
 
     /**
-     * Updates the given URI to use a real path.
-     * This is to avoid issues on some Windows setup.
+     * Returns the real path for the given path components.
+     * This is useful to avoid issues on some Windows setup.
      *
-     * @param string $URI URI
-     * @return string The URI using a real path
+     * @param string $zipFilePath Path to the ZIP file
+     * @param string $fileInsideZipPath Relative or absolute path of the file inside the zip
+     * @return string The real path URI
      */
-    protected function convertURIToUseRealPath($URI)
+    public function getRealPathURIForFileInZip($zipFilePath, $fileInsideZipPath)
     {
-        $realPathURI = $URI;
-
-        if ($this->isZipStream($URI)) {
-            if (preg_match('/zip:\/\/(.*)#(.*)/', $URI, $matches)) {
-                $documentPath = $matches[1];
-                $documentInsideZipPath = $matches[2];
-                $realPathURI = 'zip://' . realpath($documentPath) . '#' . $documentInsideZipPath;
-            }
-        } else {
-            $realPathURI = realpath($URI);
-        }
-
-        return $realPathURI;
-    }
-
-    /**
-     * Returns whether the given URI is a zip stream.
-     *
-     * @param string $URI URI pointing to a document
-     * @return bool TRUE if URI is a zip stream, FALSE otherwise
-     */
-    protected function isZipStream($URI)
-    {
-        return (strpos($URI, 'zip://') === 0);
+        return (self::ZIP_WRAPPER . realpath($zipFilePath) . '#' . $fileInsideZipPath);
     }
 
     /**
diff --git a/src/Spout/Reader/XLSX/Helper/SheetHelper.php b/src/Spout/Reader/XLSX/Helper/SheetHelper.php
index ae7b8e0..175de5b 100644
--- a/src/Spout/Reader/XLSX/Helper/SheetHelper.php
+++ b/src/Spout/Reader/XLSX/Helper/SheetHelper.php
@@ -55,7 +55,7 @@ class SheetHelper
         $sheetIndex = 0;
 
         $xmlReader = new XMLReader();
-        if ($xmlReader->open('zip://' . $this->filePath . '#' . self::WORKBOOK_XML_FILE_PATH)) {
+        if ($xmlReader->openFileInZip($this->filePath, self::WORKBOOK_XML_FILE_PATH)) {
             while ($xmlReader->read()) {
                 if ($xmlReader->isPositionedOnStartingNode('sheet')) {
                     $sheets[] = $this->getSheetFromSheetXMLNode($xmlReader, $sheetIndex);
@@ -105,7 +105,7 @@ class SheetHelper
 
         // find the file path of the sheet, by looking at the "workbook.xml.res" file
         $xmlReader = new XMLReader();
-        if ($xmlReader->open('zip://' . $this->filePath . '#' . self::WORKBOOK_XML_RELS_FILE_PATH)) {
+        if ($xmlReader->openFileInZip($this->filePath, self::WORKBOOK_XML_RELS_FILE_PATH)) {
             while ($xmlReader->read()) {
                 if ($xmlReader->isPositionedOnStartingNode('Relationship')) {
                     $relationshipSheetId = $xmlReader->getAttribute('Id');
diff --git a/src/Spout/Reader/XLSX/Helper/StyleHelper.php b/src/Spout/Reader/XLSX/Helper/StyleHelper.php
index 462433c..6fdcd80 100644
--- a/src/Spout/Reader/XLSX/Helper/StyleHelper.php
+++ b/src/Spout/Reader/XLSX/Helper/StyleHelper.php
@@ -76,10 +76,9 @@ class StyleHelper
         $this->customNumberFormats = [];
         $this->stylesAttributes = [];
 
-        $stylesXmlFilePath = $this->filePath .'#' . self::STYLES_XML_FILE_PATH;
         $xmlReader = new XMLReader();
 
-        if ($xmlReader->open('zip://' . $stylesXmlFilePath)) {
+        if ($xmlReader->openFileInZip($this->filePath, self::STYLES_XML_FILE_PATH)) {
             while ($xmlReader->read()) {
                 if ($xmlReader->isPositionedOnStartingNode(self::XML_NODE_NUM_FMTS)) {
                     $numFmtsNode = new SimpleXMLElement($xmlReader->readOuterXml());
diff --git a/tests/Spout/Reader/Wrapper/XMLReaderTest.php b/tests/Spout/Reader/Wrapper/XMLReaderTest.php
index 1f7ffc4..b16c648 100644
--- a/tests/Spout/Reader/Wrapper/XMLReaderTest.php
+++ b/tests/Spout/Reader/Wrapper/XMLReaderTest.php
@@ -20,12 +20,11 @@ class XMLReaderTest extends \PHPUnit_Framework_TestCase
     public function testOpenShouldFailIfFileInsideZipDoesNotExist()
     {
         $resourcePath = $this->getResourcePath('one_sheet_with_inline_strings.xlsx');
-        $nonExistingXMLFilePath = 'zip://' . $resourcePath . '#path/to/fake/file.xml';
 
         $xmlReader = new XMLReader();
 
         // using "@" to prevent errors/warning to be displayed
-        $wasOpenSuccessful = @$xmlReader->open($nonExistingXMLFilePath);
+        $wasOpenSuccessful = @$xmlReader->openFileInZip($resourcePath, 'path/to/fake/file.xml');
 
         $this->assertTrue($wasOpenSuccessful === false);
     }
@@ -72,10 +71,9 @@ class XMLReaderTest extends \PHPUnit_Framework_TestCase
     public function testReadShouldThrowExceptionOnError()
     {
         $resourcePath = $this->getResourcePath('one_sheet_with_invalid_xml_characters.xlsx');
-        $sheetDataXMLFilePath = 'zip://' . $resourcePath . '#xl/worksheets/sheet1.xml';
 
         $xmlReader = new XMLReader();
-        if ($xmlReader->open($sheetDataXMLFilePath) === false) {
+        if ($xmlReader->openFileInZip($resourcePath, 'xl/worksheets/sheet1.xml') === false) {
             $this->fail();
         }
 
@@ -95,43 +93,13 @@ class XMLReaderTest extends \PHPUnit_Framework_TestCase
         // The sharedStrings.xml file in "attack_billion_laughs.xlsx" contains
         // a doctype element that causes read errors
         $resourcePath = $this->getResourcePath('attack_billion_laughs.xlsx');
-        $sheetDataXMLFilePath = 'zip://' . $resourcePath . '#xl/sharedStrings.xml';
 
         $xmlReader = new XMLReader();
-        if ($xmlReader->open($sheetDataXMLFilePath) !== false) {
+        if ($xmlReader->openFileInZip($resourcePath, 'xl/sharedStrings.xml') !== false) {
             @$xmlReader->next('sst');
         }
     }
 
-    /**
-     * @return array
-     */
-    public function dataProviderForTestIsZipStream()
-    {
-        return [
-            ['/absolute/path/to/file.xlsx', false],
-            ['relative/path/to/file.xlsx', false],
-            ['php://temp', false],
-            ['zip:///absolute/path/to/file.xlsx', true],
-            ['zip://relative/path/to/file.xlsx', true],
-        ];
-    }
-
-    /**
-     * @dataProvider dataProviderForTestIsZipStream
-     *
-     * @param string $URI
-     * @param bool $expectedResult
-     * @return void
-     */
-    public function testIsZipStream($URI, $expectedResult)
-    {
-        $xmlReader = new XMLReader();
-        $isZipStream = \ReflectionHelper::callMethodOnObject($xmlReader, 'isZipStream', $URI);
-
-        $this->assertEquals($expectedResult, $isZipStream);
-    }
-
     /**
      * @return array
      */
@@ -167,34 +135,34 @@ class XMLReaderTest extends \PHPUnit_Framework_TestCase
     /**
      * @return array
      */
-    public function dataProviderForTestConvertURIToUseRealPath()
+    public function dataProviderForTestGetRealPathURIForFileInZip()
     {
         $tempFolder = realpath(sys_get_temp_dir());
+        $expectedRealPathURI = 'zip://' . $tempFolder . '/test.xlsx#test.xml';
 
         return [
-            ['/../../../' . $tempFolder . '/test.xlsx', $tempFolder . '/test.xlsx'],
-            [$tempFolder . '/test.xlsx', $tempFolder . '/test.xlsx'],
-            ['zip://' . $tempFolder . '/test.xlsx#test.xml', 'zip://' . $tempFolder . '/test.xlsx#test.xml'],
-            ['zip:///../../../' . $tempFolder . '/test.xlsx#test.xml', 'zip://' . $tempFolder . '/test.xlsx#test.xml'],
+            [$tempFolder, "$tempFolder/test.xlsx", 'test.xml', $expectedRealPathURI],
+            [$tempFolder, "/../../../$tempFolder/test.xlsx", 'test.xml', $expectedRealPathURI],
         ];
     }
 
     /**
-     * @dataProvider dataProviderForTestConvertURIToUseRealPath
+     * @dataProvider dataProviderForTestGetRealPathURIForFileInZip
      *
-     * @param string $URI
-     * @param string $expectedConvertedURI
+     * @param string $tempFolder
+     * @param string $zipFilePath
+     * @param string $fileInsideZipPath
+     * @param string $expectedRealPathURI
      * @return void
      */
-    public function testConvertURIToUseRealPath($URI, $expectedConvertedURI)
+    public function testGetRealPathURIForFileInZip($tempFolder, $zipFilePath, $fileInsideZipPath, $expectedRealPathURI)
     {
-        $tempFolder = sys_get_temp_dir();
         touch($tempFolder . '/test.xlsx');
 
         $xmlReader = new XMLReader();
-        $convertedURI = \ReflectionHelper::callMethodOnObject($xmlReader, 'convertURIToUseRealPath', $URI);
+        $realPathURI = \ReflectionHelper::callMethodOnObject($xmlReader, 'getRealPathURIForFileInZip', $zipFilePath, $fileInsideZipPath);
 
-        $this->assertEquals($expectedConvertedURI, $convertedURI);
+        $this->assertEquals($expectedRealPathURI, $realPathURI);
 
         unlink($tempFolder . '/test.xlsx');
     }
@@ -230,5 +198,7 @@ class XMLReaderTest extends \PHPUnit_Framework_TestCase
         $xmlReader->read();
         $this->assertFalse($xmlReader->isPositionedOnStartingNode('test'));
         $this->assertTrue($xmlReader->isPositionedOnEndingNode('test'));
+
+        $xmlReader->close();
     }
 }
diff --git a/tests/Spout/Writer/ODS/WriterTest.php b/tests/Spout/Writer/ODS/WriterTest.php
index 4c67e4e..43383e4 100644
--- a/tests/Spout/Writer/ODS/WriterTest.php
+++ b/tests/Spout/Writer/ODS/WriterTest.php
@@ -538,10 +538,9 @@ class WriterTest extends \PHPUnit_Framework_TestCase
     private function moveReaderToCorrectTableNode($fileName, $sheetIndex)
     {
         $resourcePath = $this->getGeneratedResourcePath($fileName);
-        $pathToSheetFile = $resourcePath . '#content.xml';
 
         $xmlReader = new XMLReader();
-        $xmlReader->open('zip://' . $pathToSheetFile);
+        $xmlReader->openFileInZip($resourcePath, 'content.xml');
         $xmlReader->readUntilNodeFound('table:table');
 
         for ($i = 1; $i < $sheetIndex; $i++) {
diff --git a/tests/Spout/Writer/ODS/WriterWithStyleTest.php b/tests/Spout/Writer/ODS/WriterWithStyleTest.php
index cc3fc50..9888bfc 100644
--- a/tests/Spout/Writer/ODS/WriterWithStyleTest.php
+++ b/tests/Spout/Writer/ODS/WriterWithStyleTest.php
@@ -296,17 +296,18 @@ class WriterWithStyleTest extends \PHPUnit_Framework_TestCase
         $cellElements = [];
 
         $resourcePath = $this->getGeneratedResourcePath($fileName);
-        $pathToStylesXmlFile = $resourcePath . '#content.xml';
 
-        $xmlReader = new \XMLReader();
-        $xmlReader->open('zip://' . $pathToStylesXmlFile);
+        $xmlReader = new XMLReader();
+        $xmlReader->openFileInZip($resourcePath, 'content.xml');
 
         while ($xmlReader->read()) {
-            if ($xmlReader->nodeType === \XMLReader::ELEMENT && $xmlReader->name === 'table:table-cell' && $xmlReader->getAttribute('office:value-type') !== null) {
+            if ($xmlReader->isPositionedOnStartingNode('table:table-cell') && $xmlReader->getAttribute('office:value-type') !== null) {
                 $cellElements[] = $xmlReader->expand();
             }
         }
 
+        $xmlReader->close();
+        
         return $cellElements;
     }
 
@@ -319,17 +320,18 @@ class WriterWithStyleTest extends \PHPUnit_Framework_TestCase
         $cellStyleElements = [];
 
         $resourcePath = $this->getGeneratedResourcePath($fileName);
-        $pathToStylesXmlFile = $resourcePath . '#content.xml';
 
-        $xmlReader = new \XMLReader();
-        $xmlReader->open('zip://' . $pathToStylesXmlFile);
+        $xmlReader = new XMLReader();
+        $xmlReader->openFileInZip($resourcePath, 'content.xml');
 
         while ($xmlReader->read()) {
-            if ($xmlReader->nodeType === \XMLReader::ELEMENT && $xmlReader->name === 'style:style' && $xmlReader->getAttribute('style:family') === 'table-cell') {
+            if ($xmlReader->isPositionedOnStartingNode('style:style') && $xmlReader->getAttribute('style:family') === 'table-cell') {
                 $cellStyleElements[] = $xmlReader->expand();
             }
         }
 
+        $xmlReader->close();
+        
         return $cellStyleElements;
     }
 
@@ -341,10 +343,9 @@ class WriterWithStyleTest extends \PHPUnit_Framework_TestCase
     private function getXmlSectionFromStylesXmlFile($fileName, $section)
     {
         $resourcePath = $this->getGeneratedResourcePath($fileName);
-        $pathToStylesXmlFile = $resourcePath . '#styles.xml';
 
         $xmlReader = new XMLReader();
-        $xmlReader->open('zip://' . $pathToStylesXmlFile);
+        $xmlReader->openFileInZip($resourcePath, 'styles.xml');
         $xmlReader->readUntilNodeFound($section);
 
         return $xmlReader->expand();
diff --git a/tests/Spout/Writer/XLSX/WriterWithStyleTest.php b/tests/Spout/Writer/XLSX/WriterWithStyleTest.php
index 9a531b9..d43b566 100644
--- a/tests/Spout/Writer/XLSX/WriterWithStyleTest.php
+++ b/tests/Spout/Writer/XLSX/WriterWithStyleTest.php
@@ -293,13 +293,16 @@ class WriterWithStyleTest extends \PHPUnit_Framework_TestCase
     private function getXmlSectionFromStylesXmlFile($fileName, $section)
     {
         $resourcePath = $this->getGeneratedResourcePath($fileName);
-        $pathToStylesXmlFile = $resourcePath . '#xl/styles.xml';
 
         $xmlReader = new XMLReader();
-        $xmlReader->open('zip://' . $pathToStylesXmlFile);
+        $xmlReader->openFileInZip($resourcePath, 'xl/styles.xml');
         $xmlReader->readUntilNodeFound($section);
 
-        return $xmlReader->expand();
+        $xmlSection = $xmlReader->expand();
+
+        $xmlReader->close();
+
+        return $xmlSection;
     }
 
     /**
@@ -311,17 +314,18 @@ class WriterWithStyleTest extends \PHPUnit_Framework_TestCase
         $cellElements = [];
 
         $resourcePath = $this->getGeneratedResourcePath($fileName);
-        $pathToStylesXmlFile = $resourcePath . '#xl/worksheets/sheet1.xml';
 
-        $xmlReader = new \XMLReader();
-        $xmlReader->open('zip://' . $pathToStylesXmlFile);
+        $xmlReader = new XMLReader();
+        $xmlReader->openFileInZip($resourcePath, 'xl/worksheets/sheet1.xml');
 
         while ($xmlReader->read()) {
-            if ($xmlReader->nodeType === \XMLReader::ELEMENT && $xmlReader->name === 'c') {
+            if ($xmlReader->isPositionedOnStartingNode('c')) {
                 $cellElements[] = $xmlReader->expand();
             }
         }
 
+        $xmlReader->close();
+
         return $cellElements;
     }
 

From efebfb2bc21fee43bd86758a556ae31f78e92aae Mon Sep 17 00:00:00 2001
From: Ingmar Runge <ingmar@irsoft.de>
Date: Mon, 30 May 2016 19:25:30 +0200
Subject: [PATCH 3/4] CellValueFormatterTest: fix expectations for 32bit PHP
 (#234)

---
 tests/Spout/Reader/XLSX/Helper/CellValueFormatterTest.php | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/tests/Spout/Reader/XLSX/Helper/CellValueFormatterTest.php b/tests/Spout/Reader/XLSX/Helper/CellValueFormatterTest.php
index 92831ab..96c71a9 100644
--- a/tests/Spout/Reader/XLSX/Helper/CellValueFormatterTest.php
+++ b/tests/Spout/Reader/XLSX/Helper/CellValueFormatterTest.php
@@ -87,6 +87,10 @@ class CellValueFormatterTest extends \PHPUnit_Framework_TestCase
      */
     public function dataProviderForTestFormatNumericCellValueWithNumbers()
     {
+        // Some test values exceed PHP_INT_MAX on 32-bit PHP. They are
+        // therefore converted to as doubles automatically by PHP.
+        $expectedBigNumberType = (PHP_INT_SIZE < 8 ? 'double' : 'integer');
+
         return [
             [42, 42, 'integer'],
             [42.5, 42.5, 'double'],
@@ -94,8 +98,8 @@ class CellValueFormatterTest extends \PHPUnit_Framework_TestCase
             [-42.5, -42.5, 'double'],
             ['42', 42, 'integer'],
             ['42.5', 42.5, 'double'],
-            [865640023012945, 865640023012945, 'integer'],
-            ['865640023012945', 865640023012945, 'integer'],
+            [865640023012945, 865640023012945, $expectedBigNumberType],
+            ['865640023012945', 865640023012945, $expectedBigNumberType],
             [865640023012945.5, 865640023012945.5, 'double'],
             ['865640023012945.5', 865640023012945.5, 'double'],
             [PHP_INT_MAX, PHP_INT_MAX, 'integer'],

From 1d3a9f939c43bfa6f2827f21ebb1c6e625847dee Mon Sep 17 00:00:00 2001
From: Adrien Loison <adrilo@users.noreply.github.com>
Date: Mon, 30 May 2016 13:55:21 -0700
Subject: [PATCH 4/4] Convert escapers to singletons (#239)

---
 src/Spout/Common/Escaper/ODS.php              |  4 ++
 src/Spout/Common/Escaper/XLSX.php             |  8 +++-
 src/Spout/Common/Helper/EncodingHelper.php    |  2 +-
 src/Spout/Common/Singleton.php                | 41 +++++++++++++++++++
 src/Spout/Reader/CSV/RowIterator.php          |  1 +
 .../Reader/ODS/Helper/CellValueFormatter.php  |  2 +-
 src/Spout/Reader/ODS/SheetIterator.php        |  2 +-
 .../Reader/XLSX/Helper/CellValueFormatter.php |  2 +-
 .../XLSX/Helper/SharedStringsHelper.php       |  2 +-
 src/Spout/Reader/XLSX/Helper/SheetHelper.php  |  2 +-
 src/Spout/Writer/ODS/Internal/Worksheet.php   |  2 +-
 .../Writer/XLSX/Helper/FileSystemHelper.php   |  2 +-
 .../XLSX/Helper/SharedStringsHelper.php       |  2 +-
 src/Spout/Writer/XLSX/Internal/Worksheet.php  |  2 +-
 tests/Spout/Common/Escaper/XLSXTest.php       |  4 +-
 15 files changed, 64 insertions(+), 14 deletions(-)
 create mode 100644 src/Spout/Common/Singleton.php

diff --git a/src/Spout/Common/Escaper/ODS.php b/src/Spout/Common/Escaper/ODS.php
index 3e252a7..86caeb3 100644
--- a/src/Spout/Common/Escaper/ODS.php
+++ b/src/Spout/Common/Escaper/ODS.php
@@ -2,6 +2,8 @@
 
 namespace Box\Spout\Common\Escaper;
 
+use Box\Spout\Common\Singleton;
+
 /**
  * Class ODS
  * Provides functions to escape and unescape data for ODS files
@@ -10,6 +12,8 @@ namespace Box\Spout\Common\Escaper;
  */
 class ODS implements EscaperInterface
 {
+    use Singleton;
+
     /**
      * Escapes the given string to make it compatible with XLSX
      *
diff --git a/src/Spout/Common/Escaper/XLSX.php b/src/Spout/Common/Escaper/XLSX.php
index 6f5bd1f..b5719a0 100644
--- a/src/Spout/Common/Escaper/XLSX.php
+++ b/src/Spout/Common/Escaper/XLSX.php
@@ -2,6 +2,8 @@
 
 namespace Box\Spout\Common\Escaper;
 
+use Box\Spout\Common\Singleton;
+
 /**
  * Class XLSX
  * Provides functions to escape and unescape data for XLSX files
@@ -10,13 +12,15 @@ namespace Box\Spout\Common\Escaper;
  */
 class XLSX implements EscaperInterface
 {
+    use Singleton;
+
     /** @var string[] Control characters to be escaped */
     protected $controlCharactersEscapingMap;
 
     /**
-     *
+     * Initializes the singleton instance
      */
-    public function __construct()
+    protected function init()
     {
         $this->controlCharactersEscapingMap = $this->getControlCharactersEscapingMap();
     }
diff --git a/src/Spout/Common/Helper/EncodingHelper.php b/src/Spout/Common/Helper/EncodingHelper.php
index 3edd7a1..3a30aaa 100644
--- a/src/Spout/Common/Helper/EncodingHelper.php
+++ b/src/Spout/Common/Helper/EncodingHelper.php
@@ -59,7 +59,7 @@ class EncodingHelper
     {
         $byteOffsetToSkipBom = 0;
 
-        if ($this->hasBom($filePointer, $encoding)) {
+        if ($this->hasBOM($filePointer, $encoding)) {
             $bomUsed = $this->supportedEncodingsWithBom[$encoding];
 
             // we skip the N first bytes
diff --git a/src/Spout/Common/Singleton.php b/src/Spout/Common/Singleton.php
new file mode 100644
index 0000000..015ede8
--- /dev/null
+++ b/src/Spout/Common/Singleton.php
@@ -0,0 +1,41 @@
+<?php
+
+namespace Box\Spout\Common;
+
+/**
+ * Class Singleton
+ * Defines a class as a singleton.
+ *
+ * @package Box\Spout\Common
+ */
+trait Singleton
+{
+    protected static $instance;
+
+    /**
+     * @return static
+     */
+    final public static function getInstance()
+    {
+        return isset(static::$instance)
+            ? static::$instance
+            : static::$instance = new static;
+    }
+
+    /**
+     * Singleton constructor.
+     */
+    final private function __construct()
+    {
+        $this->init();
+    }
+
+    /**
+     * Initializes the singleton
+     * @return void
+     */
+    protected function init() {}
+
+    final private function __wakeup() {}
+    final private function __clone() {}
+}
diff --git a/src/Spout/Reader/CSV/RowIterator.php b/src/Spout/Reader/CSV/RowIterator.php
index 95b2596..5189dfc 100644
--- a/src/Spout/Reader/CSV/RowIterator.php
+++ b/src/Spout/Reader/CSV/RowIterator.php
@@ -57,6 +57,7 @@ class RowIterator implements IteratorInterface
      * @param string $fieldDelimiter Character that delimits fields
      * @param string $fieldEnclosure Character that enclose fields
      * @param string $encoding Encoding of the CSV file to be read
+     * @param string $endOfLineDelimiter End of line delimiter
      * @param \Box\Spout\Common\Helper\GlobalFunctionsHelper $globalFunctionsHelper
      */
     public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $endOfLineDelimiter, $globalFunctionsHelper)
diff --git a/src/Spout/Reader/ODS/Helper/CellValueFormatter.php b/src/Spout/Reader/ODS/Helper/CellValueFormatter.php
index b39af21..38c1c85 100644
--- a/src/Spout/Reader/ODS/Helper/CellValueFormatter.php
+++ b/src/Spout/Reader/ODS/Helper/CellValueFormatter.php
@@ -48,7 +48,7 @@ class CellValueFormatter
         $this->shouldFormatDates = $shouldFormatDates;
 
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $this->escaper = new \Box\Spout\Common\Escaper\ODS();
+        $this->escaper = \Box\Spout\Common\Escaper\ODS::getInstance();
     }
 
     /**
diff --git a/src/Spout/Reader/ODS/SheetIterator.php b/src/Spout/Reader/ODS/SheetIterator.php
index 4f2ec01..f6cfdbe 100644
--- a/src/Spout/Reader/ODS/SheetIterator.php
+++ b/src/Spout/Reader/ODS/SheetIterator.php
@@ -51,7 +51,7 @@ class SheetIterator implements IteratorInterface
         $this->xmlReader = new XMLReader();
 
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $this->escaper = new \Box\Spout\Common\Escaper\ODS();
+        $this->escaper = \Box\Spout\Common\Escaper\ODS::getInstance();
     }
 
     /**
diff --git a/src/Spout/Reader/XLSX/Helper/CellValueFormatter.php b/src/Spout/Reader/XLSX/Helper/CellValueFormatter.php
index 286d348..a272a2e 100644
--- a/src/Spout/Reader/XLSX/Helper/CellValueFormatter.php
+++ b/src/Spout/Reader/XLSX/Helper/CellValueFormatter.php
@@ -62,7 +62,7 @@ class CellValueFormatter
         $this->shouldFormatDates = $shouldFormatDates;
 
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $this->escaper = new \Box\Spout\Common\Escaper\XLSX();
+        $this->escaper = \Box\Spout\Common\Escaper\XLSX::getInstance();
     }
 
     /**
diff --git a/src/Spout/Reader/XLSX/Helper/SharedStringsHelper.php b/src/Spout/Reader/XLSX/Helper/SharedStringsHelper.php
index 1ba4e6a..750c53e 100644
--- a/src/Spout/Reader/XLSX/Helper/SharedStringsHelper.php
+++ b/src/Spout/Reader/XLSX/Helper/SharedStringsHelper.php
@@ -80,7 +80,7 @@ class SharedStringsHelper
         $xmlReader = new XMLReader();
         $sharedStringIndex = 0;
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $escaper = new \Box\Spout\Common\Escaper\XLSX();
+        $escaper = \Box\Spout\Common\Escaper\XLSX::getInstance();
 
         $sharedStringsFilePath = $this->getSharedStringsFilePath();
         if ($xmlReader->open($sharedStringsFilePath) === false) {
diff --git a/src/Spout/Reader/XLSX/Helper/SheetHelper.php b/src/Spout/Reader/XLSX/Helper/SheetHelper.php
index 175de5b..a6ff909 100644
--- a/src/Spout/Reader/XLSX/Helper/SheetHelper.php
+++ b/src/Spout/Reader/XLSX/Helper/SheetHelper.php
@@ -87,7 +87,7 @@ class SheetHelper
         $escapedSheetName = $xmlReaderOnSheetNode->getAttribute('name');
 
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $escaper = new \Box\Spout\Common\Escaper\XLSX();
+        $escaper = \Box\Spout\Common\Escaper\XLSX::getInstance();
         $sheetName = $escaper->unescape($escapedSheetName);
 
         $sheetDataXMLFilePath = $this->getSheetDataXMLFilePathForSheetId($sheetId);
diff --git a/src/Spout/Writer/ODS/Internal/Worksheet.php b/src/Spout/Writer/ODS/Internal/Worksheet.php
index 7a0df1a..cec30af 100644
--- a/src/Spout/Writer/ODS/Internal/Worksheet.php
+++ b/src/Spout/Writer/ODS/Internal/Worksheet.php
@@ -47,7 +47,7 @@ class Worksheet implements WorksheetInterface
     {
         $this->externalSheet = $externalSheet;
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $this->stringsEscaper = new \Box\Spout\Common\Escaper\ODS();
+        $this->stringsEscaper = \Box\Spout\Common\Escaper\ODS::getInstance();
         $this->worksheetFilePath = $worksheetFilesFolder . '/sheet' . $externalSheet->getIndex() . '.xml';
 
         $this->stringHelper = new StringHelper();
diff --git a/src/Spout/Writer/XLSX/Helper/FileSystemHelper.php b/src/Spout/Writer/XLSX/Helper/FileSystemHelper.php
index 59df98a..786e62e 100644
--- a/src/Spout/Writer/XLSX/Helper/FileSystemHelper.php
+++ b/src/Spout/Writer/XLSX/Helper/FileSystemHelper.php
@@ -284,7 +284,7 @@ EOD;
 EOD;
 
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $escaper = new \Box\Spout\Common\Escaper\XLSX();
+        $escaper = \Box\Spout\Common\Escaper\XLSX::getInstance();
 
         /** @var Worksheet $worksheet */
         foreach ($worksheets as $worksheet) {
diff --git a/src/Spout/Writer/XLSX/Helper/SharedStringsHelper.php b/src/Spout/Writer/XLSX/Helper/SharedStringsHelper.php
index 9d9639d..292b663 100644
--- a/src/Spout/Writer/XLSX/Helper/SharedStringsHelper.php
+++ b/src/Spout/Writer/XLSX/Helper/SharedStringsHelper.php
@@ -49,7 +49,7 @@ EOD;
         fwrite($this->sharedStringsFilePointer, $header);
 
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $this->stringsEscaper = new \Box\Spout\Common\Escaper\XLSX();
+        $this->stringsEscaper = \Box\Spout\Common\Escaper\XLSX::getInstance();
     }
 
     /**
diff --git a/src/Spout/Writer/XLSX/Internal/Worksheet.php b/src/Spout/Writer/XLSX/Internal/Worksheet.php
index af3fbc4..354554a 100644
--- a/src/Spout/Writer/XLSX/Internal/Worksheet.php
+++ b/src/Spout/Writer/XLSX/Internal/Worksheet.php
@@ -56,7 +56,7 @@ EOD;
         $this->shouldUseInlineStrings = $shouldUseInlineStrings;
 
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $this->stringsEscaper = new \Box\Spout\Common\Escaper\XLSX();
+        $this->stringsEscaper = \Box\Spout\Common\Escaper\XLSX::getInstance();
 
         $this->worksheetFilePath = $worksheetFilesFolder . '/' . strtolower($this->externalSheet->getName()) . '.xml';
         $this->startSheet();
diff --git a/tests/Spout/Common/Escaper/XLSXTest.php b/tests/Spout/Common/Escaper/XLSXTest.php
index 7c0b2db..289e2a2 100644
--- a/tests/Spout/Common/Escaper/XLSXTest.php
+++ b/tests/Spout/Common/Escaper/XLSXTest.php
@@ -35,7 +35,7 @@ class XLSXTest extends \PHPUnit_Framework_TestCase
     public function testEscape($stringToEscape, $expectedEscapedString)
     {
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $escaper = new \Box\Spout\Common\Escaper\XLSX();
+        $escaper = \Box\Spout\Common\Escaper\XLSX::getInstance();
         $escapedString = $escaper->escape($stringToEscape);
 
         $this->assertEquals($expectedEscapedString, $escapedString, 'Incorrect escaped string');
@@ -67,7 +67,7 @@ class XLSXTest extends \PHPUnit_Framework_TestCase
     public function testUnescape($stringToUnescape, $expectedUnescapedString)
     {
         /** @noinspection PhpUnnecessaryFullyQualifiedNameInspection */
-        $escaper = new \Box\Spout\Common\Escaper\XLSX();
+        $escaper = \Box\Spout\Common\Escaper\XLSX::getInstance();
         $unescapedString = $escaper->unescape($stringToUnescape);
 
         $this->assertEquals($expectedUnescapedString, $unescapedString, 'Incorrect escaped string');