mcndave/spout
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'master' into custom-column-widths

Browse Source
This commit is contained in:
Alexander Hofstede 2021-05-26 20:04:33 +02:00 committed by GitHub
commit 77189d72e6
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
19 changed files with 209 additions and 44 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@ -8,9 +8,9 @@
[![Total Downloads](https://poser.pugx.org/box/spout/downloads)](https://packagist.org/packages/box/spout)
Spout is a PHP library to read and write spreadsheet files (CSV, XLSX and ODS), in a fast and scalable way.
Contrary to other file readers or writers, it is capable of processing very large files while keeping the memory usage really low (less than 3MB).
Unlike other file readers or writers, it is capable of processing very large files, while keeping the memory usage really low (less than 3MB).
Join the community and come discuss about Spout: [![Gitter](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/box/spout?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge)
Join the community and come discuss Spout: [![Gitter](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/box/spout?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge)
## Documentation
@ -43,7 +43,7 @@ For information, the performance tests take about 10 minutes to run (processing
## Support
You can ask questions, submit new features ideas or discuss about Spout in the chat room:<br>
You can ask questions, submit new features ideas or discuss Spout in the chat room:<br>
[![Gitter](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/box/spout?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge)
## Copyright and License

3
composer.json
View File

@ -14,7 +14,8 @@
"require": {
"php": ">=7.2.0",
"ext-zip": "*",
"ext-xmlreader" : "*"
"ext-xmlreader": "*",
"ext-dom": "*"
},
"require-dev": {
"phpunit/phpunit": "^8",

25
docs/_pages/documentation.md
View File

@ -116,18 +116,19 @@ $reader->setShouldPreserveEmptyRows(true);
For fonts and alignments, {{ site.spout_html }} does not support all the possible formatting options yet. But you can find the most important ones:
| Category | Property | API
|:----------|:---------------|:--------------------------------------
| Font | Bold | `StyleBuilder::setFontBold()`
| | Italic | `StyleBuilder::setFontItalic()`
| | Underline | `StyleBuilder::setFontUnderline()`
| | Strikethrough | `StyleBuilder::setFontStrikethrough()`
| | Font name | `StyleBuilder::setFontName('Arial')`
| | Font size | `StyleBuilder::setFontSize(14)`
| | Font color | `StyleBuilder::setFontColor(Color::BLUE)`<br>`StyleBuilder::setFontColor(Color::rgb(0, 128, 255))`
| Alignment | Cell alignment | `StyleBuilder::setCellAlignment(CellAlignment::CENTER)`
| | Wrap text | `StyleBuilder::setShouldWrapText(true)`
| Category | Property | API
|:---------------------|:---------------|:--------------------------------------
| Font | Bold | `StyleBuilder::setFontBold()`
| | Italic | `StyleBuilder::setFontItalic()`
| | Underline | `StyleBuilder::setFontUnderline()`
| | Strikethrough | `StyleBuilder::setFontStrikethrough()`
| | Font name | `StyleBuilder::setFontName('Arial')`
| | Font size | `StyleBuilder::setFontSize(14)`
| | Font color | `StyleBuilder::setFontColor(Color::BLUE)`<br>`StyleBuilder::setFontColor(Color::rgb(0, 128, 255))`
| Alignment | Cell alignment | `StyleBuilder::setCellAlignment(CellAlignment::CENTER)`
| | Wrap text | `StyleBuilder::setShouldWrapText(true)`
| Format _(XLSX only)_ | Number format | `StyleBuilder::setFormat('0.000')`
| | Date format | `StyleBuilder::setFormat('m/d/yy h:mm')`
### Styling rows

2
src/Spout/Common/Entity/Cell.php
View File

@ -65,7 +65,7 @@ class Cell
protected $style;
/**
* @param $value mixed
* @param mixed|null $value
* @param Style|null $style
*/
public function __construct($value, Style $style = null)

34
src/Spout/Common/Helper/StringHelper.php
View File

@ -13,12 +13,20 @@ class StringHelper
/** @var bool Whether the mbstring extension is loaded */
protected $hasMbstringSupport;
/** @var bool Whether the code is running with PHP7 or older versions */
private $isRunningPhp7OrOlder;
/** @var array Locale info, used for number formatting */
private $localeInfo;
/**
*
*/
public function __construct()
{
$this->hasMbstringSupport = \extension_loaded('mbstring');
$this->isRunningPhp7OrOlder = \version_compare(PHP_VERSION, '8.0.0') < 0;
$this->localeInfo = \localeconv();
}
/**
@ -68,4 +76,30 @@ class StringHelper
return ($position !== false) ? $position : -1;
}
/**
* Formats a numeric value (int or float) in a way that's compatible with the expected spreadsheet format.
*
* Formatting of float values is locale dependent in PHP < 8.
* Thousands separators and decimal points vary from locale to locale (en_US: 12.34 vs pl_PL: 12,34).
* However, float values must be formatted with no thousands separator and a "." as decimal point
* to work properly. This method can be used to convert the value to the correct format before storing it.
*
* @see https://wiki.php.net/rfc/locale_independent_float_to_string for the changed behavior in PHP8.
*
* @param int|float $numericValue
* @return string
*/
public function formatNumericValue($numericValue)
{
if ($this->isRunningPhp7OrOlder && is_float($numericValue)) {
return str_replace(
[$this->localeInfo['thousands_sep'], $this->localeInfo['decimal_point']],
['', '.'],
$numericValue
);
}
return $numericValue;
}
}

13
src/Spout/Reader/Common/Manager/RowManager.php
View File

@ -56,12 +56,25 @@ class RowManager
$rowCells = $row->getCells();
$maxCellIndex = $numCells;
// If the row has empty cells, calling "setCellAtIndex" will add the cell
// but in the wrong place (the new cell is added at the end of the array).
// Therefore, we need to sort the array using keys to have proper order.
// @see https://github.com/box/spout/issues/740
$needsSorting = false;
for ($cellIndex = 0; $cellIndex < $maxCellIndex; $cellIndex++) {
if (!isset($rowCells[$cellIndex])) {
$row->setCellAtIndex($this->entityFactory->createCell(''), $cellIndex);
$needsSorting = true;
}
}
if ($needsSorting) {
$rowCells = $row->getCells();
ksort($rowCells);
$row->setCells($rowCells);
}
return $row;
}
}

21
src/Spout/Reader/XLSX/Helper/CellValueFormatter.php
View File

@ -31,14 +31,6 @@ class CellValueFormatter
/** Constants used for date formatting */
const NUM_SECONDS_IN_ONE_DAY = 86400;
const NUM_SECONDS_IN_ONE_HOUR = 3600;
const NUM_SECONDS_IN_ONE_MINUTE = 60;
/**
* February 29th, 1900 is NOT a leap year but Excel thinks it is...
* @see https://en.wikipedia.org/wiki/Year_1900_problem#Microsoft_Excel
*/
const ERRONEOUS_EXCEL_LEAP_YEAR_DAY = 60;
/** @var SharedStringsManager Manages shared strings */
protected $sharedStringsManager;
@ -130,10 +122,15 @@ class CellValueFormatter
*/
protected function formatInlineStringCellValue($node)
{
// 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);
$cellValue = $this->escaper->unescape($tNode->nodeValue);
// inline strings are formatted this way (they can contain any number of <t> nodes):
// <c r="A1" t="inlineStr"><is><t>[INLINE_STRING]</t><t>[INLINE_STRING_2]</t></is></c>
$tNodes = $node->getElementsByTagName(self::XML_NODE_INLINE_STRING_VALUE);
$cellValue = '';
for ($i = 0; $i < $tNodes->count(); $i++) {
$tNode = $tNodes->item($i);
$cellValue .= $this->escaper->unescape($tNode->nodeValue);
}
return $cellValue;
}

3
src/Spout/Reader/XLSX/Manager/SharedStringsManager.php
View File

@ -16,9 +16,6 @@ use Box\Spout\Reader\XLSX\Manager\SharedStringsCaching\CachingStrategyInterface;
*/
class SharedStringsManager
{
/** Main namespace for the sharedStrings.xml file */
const MAIN_NAMESPACE_FOR_SHARED_STRINGS_XML = 'http://schemas.openxmlformats.org/spreadsheetml/2006/main';
/** Definition of XML nodes names used to parse data */
const XML_NODE_SST = 'sst';
const XML_NODE_SI = 'si';

17
src/Spout/Reader/XLSX/Manager/WorkbookRelationshipsManager.php
View File

@ -17,10 +17,11 @@ class WorkbookRelationshipsManager
/** Path of workbook relationships XML file inside the XLSX file */
const WORKBOOK_RELS_XML_FILE_PATH = 'xl/_rels/workbook.xml.rels';
/** Relationships types */
/** Relationships types - For Transitional and Strict OOXML */
const RELATIONSHIP_TYPE_SHARED_STRINGS = 'http://schemas.openxmlformats.org/officeDocument/2006/relationships/sharedStrings';
const RELATIONSHIP_TYPE_STYLES = 'http://schemas.openxmlformats.org/officeDocument/2006/relationships/styles';
const RELATIONSHIP_TYPE_WORKSHEET = 'http://schemas.openxmlformats.org/officeDocument/2006/relationships/worksheet';
const RELATIONSHIP_TYPE_SHARED_STRINGS_STRICT = 'http://purl.oclc.org/ooxml/officeDocument/relationships/sharedStrings';
const RELATIONSHIP_TYPE_STYLES_STRICT = 'http://purl.oclc.org/ooxml/officeDocument/relationships/styles';
/** Nodes and attributes used to find relevant information in the workbook relationships XML file */
const XML_NODE_RELATIONSHIP = 'Relationship';
@ -52,7 +53,8 @@ class WorkbookRelationshipsManager
public function getSharedStringsXMLFilePath()
{
$workbookRelationships = $this->getWorkbookRelationships();
$sharedStringsXMLFilePath = $workbookRelationships[self::RELATIONSHIP_TYPE_SHARED_STRINGS];
$sharedStringsXMLFilePath = $workbookRelationships[self::RELATIONSHIP_TYPE_SHARED_STRINGS]
?? $workbookRelationships[self::RELATIONSHIP_TYPE_SHARED_STRINGS_STRICT];
// the file path can be relative (e.g. "styles.xml") or absolute (e.g. "/xl/styles.xml")
$doesContainBasePath = (\strpos($sharedStringsXMLFilePath, self::BASE_PATH) !== false);
@ -71,7 +73,8 @@ class WorkbookRelationshipsManager
{
$workbookRelationships = $this->getWorkbookRelationships();
return isset($workbookRelationships[self::RELATIONSHIP_TYPE_SHARED_STRINGS]);
return isset($workbookRelationships[self::RELATIONSHIP_TYPE_SHARED_STRINGS])
|| isset($workbookRelationships[self::RELATIONSHIP_TYPE_SHARED_STRINGS_STRICT]);
}
/**
@ -81,7 +84,8 @@ class WorkbookRelationshipsManager
{
$workbookRelationships = $this->getWorkbookRelationships();
return isset($workbookRelationships[self::RELATIONSHIP_TYPE_STYLES]);
return isset($workbookRelationships[self::RELATIONSHIP_TYPE_STYLES])
|| isset($workbookRelationships[self::RELATIONSHIP_TYPE_STYLES_STRICT]);
}
/**
@ -90,7 +94,8 @@ class WorkbookRelationshipsManager
public function getStylesXMLFilePath()
{
$workbookRelationships = $this->getWorkbookRelationships();
$stylesXMLFilePath = $workbookRelationships[self::RELATIONSHIP_TYPE_STYLES];
$stylesXMLFilePath = $workbookRelationships[self::RELATIONSHIP_TYPE_STYLES]
?? $workbookRelationships[self::RELATIONSHIP_TYPE_STYLES_STRICT];
// the file path can be relative (e.g. "styles.xml") or absolute (e.g. "/xl/styles.xml")
$doesContainBasePath = (\strpos($stylesXMLFilePath, self::BASE_PATH) !== false);

5
src/Spout/Writer/ODS/Manager/WorksheetManager.php
View File

@ -231,8 +231,9 @@ class WorksheetManager implements WorksheetManagerInterface
$data .= '<text:p>' . $cell->getValue() . '</text:p>';
$data .= '</table:table-cell>';
} elseif ($cell->isNumeric()) {
$data .= ' office:value-type="float" calcext:value-type="float" office:value="' . $cell->getValue() . '">';
$data .= '<text:p>' . $cell->getValue() . '</text:p>';
$cellValue = $this->stringHelper->formatNumericValue($cell->getValue());
$data .= ' office:value-type="float" calcext:value-type="float" office:value="' . $cellValue . '">';
$data .= '<text:p>' . $cellValue . '</text:p>';
$data .= '</table:table-cell>';
} elseif ($cell->isError() && is_string($cell->getValueEvenIfError())) {
// only writes the error value if it's a string

2
src/Spout/Writer/XLSX/Manager/WorksheetManager.php
View File

@ -265,7 +265,7 @@ EOD;
} elseif ($cell->isBoolean()) {
$cellXML .= ' t="b"><v>' . (int) ($cell->getValue()) . '</v></c>';
} elseif ($cell->isNumeric()) {
$cellXML .= '><v>' . $cell->getValue() . '</v></c>';
$cellXML .= '><v>' . $this->stringHelper->formatNumericValue($cell->getValue()) . '</v></c>';
} elseif ($cell->isError() && is_string($cell->getValueEvenIfError())) {
// only writes the error value if it's a string
$cellXML .= ' t="e"><v>' . $cell->getValueEvenIfError() . '</v></c>';

8
tests/Spout/Reader/XLSX/Helper/CellValueFormatterTest.php
View File

@ -186,18 +186,22 @@ class CellValueFormatterTest extends TestCase
public function testFormatInlineStringCellValue($value, $expectedFormattedValue)
{
$nodeListMock = $this->createMock(\DOMNodeList::class);
$nodeListMock
->expects($this->atLeastOnce())
->method('count')
->willReturn(1);
$nodeListMock
->expects($this->atLeastOnce())
->method('item')
->with(0)
->will($this->returnValue((object) ['nodeValue' => $value]));
->willReturn((object) ['nodeValue' => $value]);
$nodeMock = $this->createMock(\DOMElement::class);
$nodeMock
->expects($this->atLeastOnce())
->method('getElementsByTagName')
->with(CellValueFormatter::XML_NODE_INLINE_STRING_VALUE)
->will($this->returnValue($nodeListMock));
->willReturn($nodeListMock);
$formatter = new CellValueFormatter(null, null, false, false, new Escaper\XLSX());
$formattedValue = \ReflectionHelper::callMethodOnObject($formatter, 'formatInlineStringCellValue', $nodeMock);

43
tests/Spout/Reader/XLSX/ReaderTest.php
View File

@ -72,6 +72,19 @@ class ReaderTest extends TestCase
}
}
/**
* @return void
*/
public function testReadShouldSupportInlineStringsWithMultipleValueNodes()
{
$allRows = $this->getAllRowsForFile('sheet_with_multiple_value_nodes_in_inline_strings.xlsx');
$expectedRows = [
['VALUE 1 VALUE 2 VALUE 3 VALUE 4', 's1 - B1'],
];
$this->assertEquals($expectedRows, $allRows);
}
/**
* @return void
*/
@ -679,13 +692,29 @@ class ReaderTest extends TestCase
$allRows = $this->getAllRowsForFile('sheet_with_empty_cells.xlsx');
$expectedRows = [
['A', 'B', 'C'],
['A', '', 'C'],
['0', '', ''],
['1', '1', ''],
];
$this->assertEquals($expectedRows, $allRows, 'There should be 3 rows, with equal length');
}
/**
* https://github.com/box/spout/issues/184
* @return void
*/
public function testReadShouldCreateOutputEmptyCellPreservedWhenNoDimensionsSpecified()
{
$allRows = $this->getAllRowsForFile('sheet_with_empty_cells_without_dimensions.xlsx');
$expectedRows = [
['A', '', 'C'],
['0'],
['1', '1'],
];
$this->assertEquals($expectedRows, $allRows);
}
/**
* https://github.com/box/spout/issues/195
* @return void
@ -703,6 +732,18 @@ class ReaderTest extends TestCase
$this->assertEquals($expectedRows, $allRows, 'Cell values should not be trimmed');
}
/**
* https://github.com/box/spout/issues/726
* @return void
*/
public function testReaderShouldSupportStrictOOXML()
{
$allRows = $this->getAllRowsForFile('sheet_with_strict_ooxml.xlsx');
$this->assertEquals('UNIQUE_ACCOUNT_IDENTIFIER', $allRows[0][0]);
$this->assertEquals('A2Z34NJA7N2ESJ', $allRows[1][0]);
}
/**
* @param string $fileName
* @param bool $shouldFormatDates

35
tests/Spout/Writer/ODS/WriterTest.php
View File

@ -279,6 +279,41 @@ class WriterTest extends TestCase
$this->assertValueWasWrittenToSheet($fileName, 1, 10.2);
}
/**
* @return void
*/
public function testAddRowShouldSupportFloatValuesInDifferentLocale()
{
$previousLocale = \setlocale(LC_ALL, 0);
try {
// Pick a supported locale whose decimal point is a comma.
// Installed locales differ from one system to another, so we can't pick
// a given locale.
$supportedLocales = explode("\n", shell_exec('locale -a'));
foreach ($supportedLocales as $supportedLocale) {
\setlocale(LC_ALL, $supportedLocale);
if (\localeconv()['decimal_point'] === ',') {
break;
}
}
$this->assertEquals(',', \localeconv()['decimal_point']);
$fileName = 'test_add_row_should_support_float_values_in_different_locale.xlsx';
$dataRows = $this->createRowsFromValues([
[1234.5],
]);
$this->writeToODSFile($dataRows, $fileName);
$this->assertValueWasNotWrittenToSheet($fileName, 1, "1234,5");
$this->assertValueWasWrittenToSheet($fileName, 1, "1234.5");
} finally {
// reset locale
\setlocale(LC_ALL, $previousLocale);
}
}
/**
* @return array
*/

36
tests/Spout/Writer/XLSX/WriterTest.php
View File

@ -393,6 +393,42 @@ class WriterTest extends TestCase
$this->assertInlineDataWasWrittenToSheet($fileName, 1, 't="e"><v>#DIV/0</v>');
}
/**
* @return void
*/
public function testAddRowShouldSupportFloatValuesInDifferentLocale()
{
$previousLocale = \setlocale(LC_ALL, 0);
$valueToWrite = 1234.5; // needs to be defined before changing the locale as PHP8 would expect 1234,5
try {
// Pick a supported locale whose decimal point is a comma.
// Installed locales differ from one system to another, so we can't pick
// a given locale.
$supportedLocales = explode("\n", shell_exec('locale -a'));
foreach ($supportedLocales as $supportedLocale) {
\setlocale(LC_ALL, $supportedLocale);
if (\localeconv()['decimal_point'] === ',') {
break;
}
}
$this->assertEquals(',', \localeconv()['decimal_point']);
$fileName = 'test_add_row_should_support_float_values_in_different_locale.xlsx';
$dataRows = $this->createRowsFromValues([
[$valueToWrite],
]);
$this->writeToXLSXFile($dataRows, $fileName, $shouldUseInlineStrings = false);
$this->assertInlineDataWasNotWrittenToSheet($fileName, 1, "1234,5");
$this->assertInlineDataWasWrittenToSheet($fileName, 1, "1234.5");
} finally {
// reset locale
\setlocale(LC_ALL, $previousLocale);
}
}
/**
* @return void
*/

BIN
tests/resources/xlsx/sheet_with_empty_cells.xlsx
View File

Binary file not shown.

BIN
tests/resources/xlsx/sheet_with_empty_cells_without_dimensions.xlsx Normal file
View File

Binary file not shown.

BIN
tests/resources/xlsx/sheet_with_multiple_value_nodes_in_inline_strings.xlsx Normal file
View File

Binary file not shown.

BIN
tests/resources/xlsx/sheet_with_strict_ooxml.xlsx Normal file
View File

Binary file not shown.