RowIterator::getNextUTF8EncodedRow() - Code Metrics - Inspection of "Fix line breaks on CSV reader" - box/spout - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#187)

by Marie

created 2016-03-23 17:54 UTC

RowIterator::getNextUTF8EncodedRow() C

↳ Parent: RowIterator

Complexity

Conditions	7
Paths	7

Size

Total Lines	27
Code Lines	16

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	18
CRAP Score	7

Importance

Changes	1
Bugs	1	Features	0

Metric	Value
c	1
b	1
f	0
dl	0
loc	27
ccs	18
cts	18
cp	1
rs	6.7272
cc	7
eloc	16
nc	7
nop	0
crap	7

<?php

namespace Box\Spout\Reader\CSV;

use Box\Spout\Reader\IteratorInterface;
use Box\Spout\Common\Helper\EncodingHelper;

/**
 * Class RowIterator
 * Iterate over CSV rows.
 *
 * @package Box\Spout\Reader\CSV
 */
class RowIterator implements IteratorInterface
{
    /**
     * 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;

    /** @var resource Pointer to the CSV file to read */
    protected $filePointer;

    /** @var int Number of read rows */
    protected $numReadRows = 0;

    /** @var array|null Buffer used to store the row data, while checking if there are more rows to read */
    protected $rowDataBuffer = null;

    /** @var bool Indicates whether all rows have been read */
    protected $hasReachedEndOfFile = false;

    /** @var string Defines the character used to delimit fields (one character only) */
    protected $fieldDelimiter;

    /** @var string Defines the character used to enclose fields (one character only) */
    protected $fieldEnclosure;

    /** @var string Encoding of the CSV file to be read */
    protected $encoding;

    /** @var \Box\Spout\Common\Helper\GlobalFunctionsHelper Helper to work with global functions */
    protected $globalFunctionsHelper;

    /** @var \Box\Spout\Common\Helper\EncodingHelper Helper to work with different encodings */
    protected $encodingHelper;

    /** @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
     * @param string $fieldEnclosure Character that enclose fields
     * @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, $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);
    }

    /**
     * Rewind the Iterator to the first element
     * @link http://php.net/manual/en/iterator.rewind.php
     *
     * @return void
     */
    public function rewind()
    {
        $this->rewindAndSkipBom();

        $this->numReadRows = 0;
        $this->rowDataBuffer = null;

        $this->next();
    }

    /**
     * This rewinds and skips the BOM if inserted at the beginning of the file
     * by moving the file pointer after it, so that it is not read.
     *
     * @return void
     */
    protected function rewindAndSkipBom()
    {
        $byteOffsetToSkipBom = $this->encodingHelper->getBytesOffsetToSkipBOM($this->filePointer, $this->encoding);

        // sets the cursor after the BOM (0 means no BOM, so rewind it)
        $this->globalFunctionsHelper->fseek($this->filePointer, $byteOffsetToSkipBom);
    }

    /**
     * Checks if current position is valid
     * @link http://php.net/manual/en/iterator.valid.php
     *
     * @return boolean
     */
    public function valid()
    {
        return ($this->filePointer && !$this->hasReachedEndOfFile);
    }

    /**
     * Move forward to next element. Empty rows are skipped.
     * @link http://php.net/manual/en/iterator.next.php
     *
     * @return void
     * @throws \Box\Spout\Common\Exception\EncodingConversionException If unable to convert data to UTF-8
     */
    public function next()
    {
        $this->hasReachedEndOfFile = $this->globalFunctionsHelper->feof($this->filePointer);

        if ($this->hasReachedEndOfFile) {
            return;
        }

        do {
            $rowData = $this->getNextUTF8EncodedRow();
            $hasNowReachedEndOfFile = $this->globalFunctionsHelper->feof($this->filePointer);
        } while (($rowData === false && !$hasNowReachedEndOfFile) || $this->isEmptyLine($rowData));
<?php

function getDate($date)
{
    if ($date !== null) {
        return new DateTime($date);
    }

    return false;
}

        if ($rowData !== false) {
            $this->rowDataBuffer = $rowData;
            $this->numReadRows++;
        } else {
            // If we reach this point, it means end of file was reached.
            // This happens when the last lines are empty lines.
            $this->hasReachedEndOfFile = $hasNowReachedEndOfFile;
        }
    }

    /**
     * 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 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 getNextUTF8EncodedRow()
    {
        $encodedLineData = fgetcsv($this->filePointer, self::MAX_READ_BYTES_PER_LINE, $this->fieldDelimiter, $this->fieldEnclosure);
        if (false === $encodedLineData) {
            return false;
        }

        foreach ($encodedLineData as $i => $line) {
            switch($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
                    $line = ltrim($line);
                    break;

                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
                    $line = rtrim($line);
                    break;
            }

            $encodedLineData[$i] = $this->encodingHelper->attemptConversionToUTF8($line, $this->encoding);
        }

        return $encodedLineData;
    }

    /**
     * Returns the end of line delimiter, encoded using the same encoding as the CSV.
     * The return value is cached.
     *
     * @return string
     */
    protected function getEncodedEOLDelimiter()
    {
        if (!isset($this->encodedEOLDelimiter)) {
            $this->encodedEOLDelimiter = $this->encodingHelper->attemptConversionFromUTF8($this->inputEOLDelimiter, $this->encoding);
        }

        return $this->encodedEOLDelimiter;
    }

    /**
     * @param array $lineData Array containing the cells value for the line
     * @return bool Whether the given line is empty
     */
    protected function isEmptyLine($lineData)
    {
        return (is_array($lineData) && count($lineData) === 1 && $lineData[0] === null);
    }

    /**
     * Return the current element from the buffer
     * @link http://php.net/manual/en/iterator.current.php
     *
     * @return array|null
     */
    public function current()
    {
        return $this->rowDataBuffer;
    }

    /**
     * Return the key of the current element
     * @link http://php.net/manual/en/iterator.key.php
     *
     * @return int
     */
    public function key()
    {
        return $this->numReadRows;
    }

    /**
     * Cleans up what was created to iterate over the object.
     *
     * @return void
     */
    public function end()
    {
        // do nothing
    }
}


1		<?php
2
3		namespace Box\Spout\Reader\CSV;
4
5		use Box\Spout\Reader\IteratorInterface;
6		use Box\Spout\Common\Helper\EncodingHelper;
7
8		/**
9		* Class RowIterator
10		* Iterate over CSV rows.
11		*
12		* @package Box\Spout\Reader\CSV
13		*/
14		class RowIterator implements IteratorInterface
15		{
16		/**
17		* If no value is given to fgetcsv(), it defaults to 8192 (which may be too low).
18		* Alignement with other functions like fgets() is discussed here: https://bugs.php.net/bug.php?id=48421
19		*/
20		const MAX_READ_BYTES_PER_LINE = 32768;
21
22		/** @var resource Pointer to the CSV file to read */
23		protected $filePointer;
24
25		/** @var int Number of read rows */
26		protected $numReadRows = 0;
27
28		/** @var array\|null Buffer used to store the row data, while checking if there are more rows to read */
29		protected $rowDataBuffer = null;
30
31		/** @var bool Indicates whether all rows have been read */
32		protected $hasReachedEndOfFile = false;
33
34		/** @var string Defines the character used to delimit fields (one character only) */
35		protected $fieldDelimiter;
36
37		/** @var string Defines the character used to enclose fields (one character only) */
38		protected $fieldEnclosure;
39
40		/** @var string Encoding of the CSV file to be read */
41		protected $encoding;
42
43		/** @var \Box\Spout\Common\Helper\GlobalFunctionsHelper Helper to work with global functions */
44		protected $globalFunctionsHelper;
45
46		/** @var \Box\Spout\Common\Helper\EncodingHelper Helper to work with different encodings */
47		protected $encodingHelper;
48
49		/** @var string End of line delimiter, encoded using the same encoding as the CSV */
50		protected $encodedEOLDelimiter;
51
52		/** @var string End of line delimiter, given by the user as input. */
53		protected $inputEOLDelimiter;
54
55		/**
56		* @param resource $filePointer Pointer to the CSV file to read
57		* @param string $fieldDelimiter Character that delimits fields
58		* @param string $fieldEnclosure Character that enclose fields
59		* @param string $encoding Encoding of the CSV file to be read
60		* @param \Box\Spout\Common\Helper\GlobalFunctionsHelper $globalFunctionsHelper
61		*/
62	78	public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $endOfLineDelimiter, $globalFunctionsHelper)
63		{
64	78	$this->filePointer = $filePointer;
65	78	$this->fieldDelimiter = $fieldDelimiter;
66	78	$this->fieldEnclosure = $fieldEnclosure;
67	78	$this->encoding = $encoding;
68	78	$this->inputEOLDelimiter = $endOfLineDelimiter;
69	78	$this->globalFunctionsHelper = $globalFunctionsHelper;
70
71	78	$this->encodingHelper = new EncodingHelper($globalFunctionsHelper);
72	78	}
73
74		/**
75		* Rewind the Iterator to the first element
76		* @link http://php.net/manual/en/iterator.rewind.php
77		*
78		* @return void
79		*/
80	78	public function rewind()
81		{
82	78	$this->rewindAndSkipBom();
83
84	78	$this->numReadRows = 0;
85	78	$this->rowDataBuffer = null;
86
87	78	$this->next();
88	78	}
89
90		/**
91		* This rewinds and skips the BOM if inserted at the beginning of the file
92		* by moving the file pointer after it, so that it is not read.
93		*
94		* @return void
95		*/
96	78	protected function rewindAndSkipBom()
97		{
98	78	$byteOffsetToSkipBom = $this->encodingHelper->getBytesOffsetToSkipBOM($this->filePointer, $this->encoding);
99
100		// sets the cursor after the BOM (0 means no BOM, so rewind it)
101	78	$this->globalFunctionsHelper->fseek($this->filePointer, $byteOffsetToSkipBom);
102	78	}
103
104		/**
105		* Checks if current position is valid
106		* @link http://php.net/manual/en/iterator.valid.php
107		*
108		* @return boolean
109		*/
110	78	public function valid()
111		{
112	78	return ($this->filePointer && !$this->hasReachedEndOfFile);
113		}
114
115		/**
116		* Move forward to next element. Empty rows are skipped.
117		* @link http://php.net/manual/en/iterator.next.php
118		*
119		* @return void
120		* @throws \Box\Spout\Common\Exception\EncodingConversionException If unable to convert data to UTF-8
121		*/
122	78	public function next()
123		{
124	78	$this->hasReachedEndOfFile = $this->globalFunctionsHelper->feof($this->filePointer);
125
126	78	if ($this->hasReachedEndOfFile) {
127	57	return;
128		}
129
130		do {
131	78	$rowData = $this->getNextUTF8EncodedRow();
132	78	$hasNowReachedEndOfFile = $this->globalFunctionsHelper->feof($this->filePointer);
133	78	} while (($rowData === false && !$hasNowReachedEndOfFile) \|\| $this->isEmptyLine($rowData));
		0 ignored issues – show Security Bug introduced 2016-03-23 17:57 UTC by Report Bug Copy Issue Report It seems like `$rowData` defined by `$this->getNextUTF8EncodedRow()` on line `131` can also be of type `false`; however, `Box\Spout\Reader\CSV\RowIterator::isEmptyLine()` does only seem to accept `array`, did you maybe forget to handle an error condition? This check looks for type mismatches where the missing type is `false`. This is usually indicative of an error condtion. Consider the follow example <?php function getDate($date) { if ($date !== null) { return new DateTime($date); } return false; } This function either returns a new `DateTime` object or false, if there was an error. This is a typical pattern in PHP programming to show that an error has occurred without raising an exception. The calling code should check for this returned `false` before passing on the value to another function or method that may not be able to handle a `false`. Loading history...
134
135	78	if ($rowData !== false) {
136	72	$this->rowDataBuffer = $rowData;
137	72	$this->numReadRows++;
138	72	} else {
139		// If we reach this point, it means end of file was reached.
140		// This happens when the last lines are empty lines.
141	18	$this->hasReachedEndOfFile = $hasNowReachedEndOfFile;
142		}
143	78	}
144
145		/**
146		* Returns the next row, converted if necessary to UTF-8.
147		* As fgetcsv() does not manage correctly encoding for non UTF-8 data,
148		* we remove manually whitespace with ltrim or rtrim (depending on the order of the bytes)
149		*
150		* @return array\|false The row for the current file pointer, encoded in UTF-8 or FALSE if nothing to read
151		* @throws \Box\Spout\Common\Exception\EncodingConversionException If unable to convert data to UTF-8
152		*/
153	78	protected function getNextUTF8EncodedRow()
154		{
155	78	$encodedLineData = fgetcsv($this->filePointer, self::MAX_READ_BYTES_PER_LINE, $this->fieldDelimiter, $this->fieldEnclosure);
156	78	if (false === $encodedLineData) {
157	18	return false;
158		}
159
160	75	foreach ($encodedLineData as $i => $line) {
161	75	switch($this->encoding) {
162	75	case EncodingHelper::ENCODING_UTF16_LE:
163	75	case EncodingHelper::ENCODING_UTF32_LE:
164		// remove whitespace from the beginning of a string as fgetcsv() add extra whitespace when it try to explode non UTF-8 data
165	12	$line = ltrim($line);
166	12	break;
167
168	63	case EncodingHelper::ENCODING_UTF16_BE:
169	63	case EncodingHelper::ENCODING_UTF32_BE:
170		// remove whitespace from the end of a string as fgetcsv() add extra whitespace when it try to explode non UTF-8 data
171	6	$line = rtrim($line);
172	6	break;
173	75	}
174
175	75	$encodedLineData[$i] = $this->encodingHelper->attemptConversionToUTF8($line, $this->encoding);
176	75	}
177
178	75	return $encodedLineData;
179		}
180
181		/**
182		* Returns the end of line delimiter, encoded using the same encoding as the CSV.
183		* The return value is cached.
184		*
185		* @return string
186		*/
187		protected function getEncodedEOLDelimiter()
188		{
189		if (!isset($this->encodedEOLDelimiter)) {
190		$this->encodedEOLDelimiter = $this->encodingHelper->attemptConversionFromUTF8($this->inputEOLDelimiter, $this->encoding);
191		}
192
193		return $this->encodedEOLDelimiter;
194		}
195
196		/**
197		* @param array $lineData Array containing the cells value for the line
198		* @return bool Whether the given line is empty
199		*/
200	78	protected function isEmptyLine($lineData)
201		{
202	78	return (is_array($lineData) && count($lineData) === 1 && $lineData[0] === null);
203		}
204
205		/**
206		* Return the current element from the buffer
207		* @link http://php.net/manual/en/iterator.current.php
208		*
209		* @return array\|null
210		*/
211	72	public function current()
212		{
213	72	return $this->rowDataBuffer;
214		}
215
216		/**
217		* Return the key of the current element
218		* @link http://php.net/manual/en/iterator.key.php
219		*
220		* @return int
221		*/
222	48	public function key()
223		{
224	48	return $this->numReadRows;
225		}
226
227		/**
228		* Cleans up what was created to iterate over the object.
229		*
230		* @return void
231		*/
232		public function end()
233		{
234		// do nothing
235		}
236		}
237

box / spout

Pull Request — master (#187)

RowIterator::getNextUTF8EncodedRow() C

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like