RowIterator::next() - 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 2016-03-15 15:54 UTC

RowIterator::next() B

↳ Parent: RowIterator

Complexity

Conditions	6
Paths	3

Size

Total Lines	22
Code Lines	13

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	15
CRAP Score	6

Importance

Changes	6
Bugs	4	Features	0

Metric	Value
c	6
b	4
f	0
dl	0
loc	22
ccs	15
cts	15
cp	1
rs	8.6737
cc	6
eloc	13
nc	3
nop	0
crap	6

<?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 stream_get_line(), 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 {
            $lineData = $this->getNextUTF8EncodedLine();
            $hasNowReachedEndOfFile = $this->globalFunctionsHelper->feof($this->filePointer);
        } while (($lineData === false && !$hasNowReachedEndOfFile) || $this->isEmptyLine($lineData));
<?php

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

    return false;
}

        if ($lineData !== false) {
            $this->rowDataBuffer = $lineData;
            $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 line, 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

     * @throws \Box\Spout\Common\Exception\EncodingConversionException If unable to convert data to UTF-8
     */
    protected function getNextUTF8EncodedLine()
    {
        $encodedLineData = fgetcsv($this->filePointer, 0, $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:
                    $line = ltrim($line);
                    break;

                case EncodingHelper::ENCODING_UTF16_BE:
                case EncodingHelper::ENCODING_UTF32_BE:
                    $line = rtrim($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 stream_get_line(), 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 $globalFunctionsHelper
61		*/
62	66	public function __construct($filePointer, $fieldDelimiter, $fieldEnclosure, $encoding, $endOfLineDelimiter, $globalFunctionsHelper)
63		{
64	66	$this->filePointer = $filePointer;
65	66	$this->fieldDelimiter = $fieldDelimiter;
66	66	$this->fieldEnclosure = $fieldEnclosure;
67	66	$this->encoding = $encoding;
68	66	$this->inputEOLDelimiter = $endOfLineDelimiter;
69	66	$this->globalFunctionsHelper = $globalFunctionsHelper;
70
71	66	$this->encodingHelper = new EncodingHelper($globalFunctionsHelper);
72	66	}
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	66	public function rewind()
81		{
82	66	$this->rewindAndSkipBom();
83
84	66	$this->numReadRows = 0;
85	66	$this->rowDataBuffer = null;
86
87	66	$this->next();
88	66	}
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	66	protected function rewindAndSkipBom()
97		{
98	66	$byteOffsetToSkipBom = $this->encodingHelper->getBytesOffsetToSkipBOM($this->filePointer, $this->encoding);
99
100		// sets the cursor after the BOM (0 means no BOM, so rewind it)
101	66	$this->globalFunctionsHelper->fseek($this->filePointer, $byteOffsetToSkipBom);
102	66	}
103
104		/**
105		* Checks if current position is valid
106		* @link http://php.net/manual/en/iterator.valid.php
107		*
108		* @return boolean
109		*/
110	66	public function valid()
111		{
112	66	return ($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	66	public function next()
123		{
124	66	$this->hasReachedEndOfFile = $this->globalFunctionsHelper->feof($this->filePointer);
125
126	66	if ($this->hasReachedEndOfFile) {
127	54	return;
128		}
129
130		do {
131	66	$lineData = $this->getNextUTF8EncodedLine();
132	66	$hasNowReachedEndOfFile = $this->globalFunctionsHelper->feof($this->filePointer);
133	66	} while (($lineData === false && !$hasNowReachedEndOfFile) \|\| $this->isEmptyLine($lineData));
		0 ignored issues – show Security Bug introduced 2016-03-15 16:05 UTC by Report Bug Copy Issue Report It seems like `$lineData` defined by `$this->getNextUTF8EncodedLine()` 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($date); } return false; } This function either returns a new `DateTime` object or false, if there was an error. This is a typical pattern in PHP programming to show that an error has occurred without raising an exception. The calling code should check for this returned `false` before passing on the value to another function or method that may not be able to handle a `false`. Loading history...
134	63
135	63	if ($lineData !== false) {
136	66	$this->rowDataBuffer = $lineData;
137	66	$this->numReadRows++;
138		} else {
139	66	// If we reach this point, it means end of file was reached.
140	60	// This happens when the last lines are empty lines.
141	60	$this->hasReachedEndOfFile = $hasNowReachedEndOfFile;
142	60	}
143		}
144
145	9	/**
146		* Returns the next line, converted if necessary to UTF-8.
147	66	* 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 string\|false The next line for the current file pointer, encoded in UTF-8 or FALSE if nothing to read
		0 ignored issues – show Documentation introduced 2016-03-15 16:05 UTC by Report Bug Copy Issue Report Should the return type not be `false\|array`? This check compares the return type specified in the `@return` annotation of a function or method doc comment with the types returned by the function and raises an issue if they mismatch. Loading history...
151		* @throws \Box\Spout\Common\Exception\EncodingConversionException If unable to convert data to UTF-8
152		*/
153		protected function getNextUTF8EncodedLine()
154		{
155		$encodedLineData = fgetcsv($this->filePointer, 0, $this->fieldDelimiter, $this->fieldEnclosure);
156	66	if (false === $encodedLineData) {
157		return false;
158		}
159	66
160	66	foreach ($encodedLineData as $i => $line) {
161		switch($this->encoding) {
162		case EncodingHelper::ENCODING_UTF16_LE:
163	66	case EncodingHelper::ENCODING_UTF32_LE:
164	66	$line = ltrim($line);
165	66	break;
166
167	66	case EncodingHelper::ENCODING_UTF16_BE:
168		case EncodingHelper::ENCODING_UTF32_BE:
169		$line = rtrim($line);
170		break;
171		}
172
173		$encodedLineData[$i] = $this->encodingHelper->attemptConversionToUTF8($line, $this->encoding);
174		}
175
176	66	return $encodedLineData;
177		}
178	66
179	66	/**
180	66	* Returns the end of line delimiter, encoded using the same encoding as the CSV.
181		* The return value is cached.
182	66	*
183		* @return string
184		*/
185		protected function getEncodedEOLDelimiter()
186		{
187		if (!isset($this->encodedEOLDelimiter)) {
188		$this->encodedEOLDelimiter = $this->encodingHelper->attemptConversionFromUTF8($this->inputEOLDelimiter, $this->encoding);
189	66	}
190
191	66	return $this->encodedEOLDelimiter;
192		}
193
194		/**
195		* @param array $lineData Array containing the cells value for the line
196		* @return bool Whether the given line is empty
197		*/
198		protected function isEmptyLine($lineData)
199		{
200	60	return (is_array($lineData) && count($lineData) === 1 && $lineData[0] === null);
201		}
202	60
203		/**
204		* Return the current element from the buffer
205		* @link http://php.net/manual/en/iterator.current.php
206		*
207		* @return array\|null
208		*/
209		public function current()
210		{
211	39	return $this->rowDataBuffer;
212		}
213	39
214		/**
215		* Return the key of the current element
216		* @link http://php.net/manual/en/iterator.key.php
217		*
218		* @return int
219		*/
220		public function key()
221		{
222		return $this->numReadRows;
223		}
224
225		/**
226		* Cleans up what was created to iterate over the object.
227		*
228		* @return void
229		*/
230		public function end()
231		{
232		// do nothing
233		}
234		}
235

box / spout

Pull Request — master (#187)

RowIterator::next() B

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like