TokenParser::next() - Code Metrics - Inspection of "Fix token parsing on latest PHP 8 version, fixes #..." - doctrine/annotations - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Pull Request — master (#344)

by Soner

created 2020-07-31 12:19 UTC

TokenParser::next() A

↳ Parent: TokenParser

Complexity

Conditions	6
Paths	3

Size

Total Lines	15
Code Lines	8

Duplication

Lines	0
Ratio	0 %

Importance

Changes

Metric	Value
eloc	8
dl	0
loc	15
rs	9.2222
c	0
b	0
f	0
cc	6
nc	3
nop	1

<?php

namespace Doctrine\Common\Annotations;

/**
 * Parses a file for namespaces/use/class declarations.
 *
 * @author Fabien Potencier <[email protected]>
 * @author Christian Kaps <[email protected]>
 */
class TokenParser
{
    /**
     * The token list.
     *
     * @var array
     */
    private $tokens;

    /**
     * The number of tokens.
     *
     * @var int
     */
    private $numTokens;

    /**
     * The current array pointer.
     *
     * @var int
     */
    private $pointer = 0;

    /**
     * @param string $contents
     */
    public function __construct($contents)
    {
        $this->tokens = token_get_all($contents);

        // The PHP parser sets internal compiler globals for certain things. Annoyingly, the last docblock comment it
        // saw gets stored in doc_comment. When it comes to compile the next thing to be include()d this stored
        // doc_comment becomes owned by the first thing the compiler sees in the file that it considers might have a
        // docblock. If the first thing in the file is a class without a doc block this would cause calls to
        // getDocBlock() on said class to return our long lost doc_comment. Argh.
        // To workaround, cause the parser to parse an empty docblock. Sure getDocBlock() will return this, but at least
        // it's harmless to us.
        token_get_all("<?php\n/**\n *\n */");

        $this->numTokens = count($this->tokens);

        // Compatibility defines for PHP < 8.0.
        if (!defined('T_NAME_QUALIFIED')) {
            \define('T_NAME_QUALIFIED', -2);
        }
        if (!defined('T_NAME_FULLY_QUALIFIED')) {
            \define('T_NAME_FULLY_QUALIFIED', -3);
        }
        if (!defined('T_NAME_RELATIVE')) {
            \define('T_NAME_RELATIVE', -4);
        }
    }

    /**
     * Gets the next non whitespace and non comment token.
     *
     * @param boolean $docCommentIsComment If TRUE then a doc comment is considered a comment and skipped.
     *                                     If FALSE then only whitespace and normal comments are skipped.
     *
     * @return array|null The token if exists, null otherwise.
     */
    public function next($docCommentIsComment = TRUE)
    {
        for ($i = $this->pointer; $i < $this->numTokens; $i++) {
            $this->pointer++;
            if ($this->tokens[$i][0] === T_WHITESPACE ||
                $this->tokens[$i][0] === T_COMMENT ||
                ($docCommentIsComment && $this->tokens[$i][0] === T_DOC_COMMENT)) {

                continue;
            }

            return $this->tokens[$i];
        }

        return null;
    }

    /**
     * Parses a single use statement.
     *
     * @return array A list with all found class names for a use statement.
     */
    public function parseUseStatement()
    {
        return PHP_VERSION_ID >=80000 ? $this->parseUseStatementV8() : $this->parseUseStatementV7();
    }

    private function parseUseStatementV7(): array
    {
        $groupRoot = '';
        $class = '';
        $alias = '';
        $statements = [];
        $explicitAlias = false;
        while (($token = $this->next())) {
            $isNameToken = $token[0] === T_STRING || $token[0] === T_NS_SEPARATOR;
            if (!$explicitAlias && $isNameToken) {
                $class .= $token[1];
                $alias = $token[1];
            } else if ($explicitAlias && $isNameToken) {
                $alias .= $token[1];
            } else if ($token[0] === T_AS) {
                $explicitAlias = true;
                $alias = '';
            } else if ($token === ',') {
                $statements[strtolower($alias)] = $groupRoot . $class;
                $class = '';
                $alias = '';
                $explicitAlias = false;
            } else if ($token === ';') {
                $statements[strtolower($alias)] = $groupRoot . $class;
                break;
            } else if ($token === '{' ) {
                $groupRoot = $class;
                $class = '';
            } else if ($token === '}' ) {
                continue;
            } else {
                break;
            }
        }

        return $statements;
    }

    private function parseUseStatementV8(): array
    {
        $groupRoot = '';
        $class = '';
        $alias = '';
        $statements = [];
        $explicitAlias = false;
        while (($token = $this->next())) {
            $isNameToken = $token[0] === T_STRING || $token[0] === T_NS_SEPARATOR;

            if ($token[0] === T_NAME_QUALIFIED || $token[0] === T_NAME_FULLY_QUALIFIED) {
                $class .= $token[1];

                $classSplit = explode('\\', $token[1]);
                $alias = $classSplit[count($classSplit) - 1];
            } else if($token[0] === T_NS_SEPARATOR) {
                $class .= '\\';
            } else if ($token[0] === T_AS) {
                $explicitAlias = true;
                $alias = '';
            } else if ($isNameToken && !$explicitAlias) {
                $class = $token[1];
                $alias = $token[1];
            } else if ($isNameToken && $explicitAlias) {
                $alias = $token[1];
            } else if ($token === ',') {
                $statements[strtolower($alias)] = $groupRoot . $class;
                $class = '';
                $alias = '';
                $explicitAlias = false;
            } else if ($token === ';') {
                $statements[strtolower($alias)] = $groupRoot . $class;
                break;
            } else if ($token === '{' ) {
                $groupRoot = $class;
                $class = '';
            } else if ($token === '}' ) {
                continue;
            } else {
                break;
            }
        }

        return $statements;
    }

    /**
     * Gets all use statements.
     *
     * @param string $namespaceName The namespace name of the reflected class.
     *
     * @return array A list with all found use statements.
     */
    public function parseUseStatements($namespaceName)
    {
        $statements = [];
        while (($token = $this->next())) {
            if ($token[0] === T_USE) {
                $statements = array_merge($statements, $this->parseUseStatement());
                continue;
            }
            if ($token[0] !== T_NAMESPACE || $this->parseNamespace() != $namespaceName) {
                continue;
            }

            // Get fresh array for new namespace. This is to prevent the parser to collect the use statements
            // for a previous namespace with the same name. This is the case if a namespace is defined twice
            // or if a namespace with the same name is commented out.
            $statements = [];
        }

        return $statements;
    }

    /**
     * Gets the namespace.
     *
     * @return string The found namespace.
     */
    public function parseNamespace()
    {
        $name = '';
        while (($token = $this->next()) && ($token[0] === T_STRING || $token[0] === T_NS_SEPARATOR || $token[0] === T_NAME_QUALIFIED)) {
            $name .= $token[1];
        }

        return $name;
    }

    /**
     * Gets the class name.
     *
     * @return string The found class name.
     */
    public function parseClass()
    {
        // Namespaces and class names are tokenized the same: T_STRINGs
        // separated by T_NS_SEPARATOR so we can use one function to provide
        // both.
        return $this->parseNamespace();
    }
}


1			<?php
2
3			namespace Doctrine\Common\Annotations;
4
5			/**
6			* Parses a file for namespaces/use/class declarations.
7			*
8			* @author Fabien Potencier <[email protected]>
9			* @author Christian Kaps <[email protected]>
10			*/
11			class TokenParser
12			{
13			/**
14			* The token list.
15			*
16			* @var array
17			*/
18			private $tokens;
19
20			/**
21			* The number of tokens.
22			*
23			* @var int
24			*/
25			private $numTokens;
26
27			/**
28			* The current array pointer.
29			*
30			* @var int
31			*/
32			private $pointer = 0;
33
34			/**
35			* @param string $contents
36			*/
37			public function __construct($contents)
38			{
39			$this->tokens = token_get_all($contents);
40
41			// The PHP parser sets internal compiler globals for certain things. Annoyingly, the last docblock comment it
42			// saw gets stored in doc_comment. When it comes to compile the next thing to be include()d this stored
43			// doc_comment becomes owned by the first thing the compiler sees in the file that it considers might have a
44			// docblock. If the first thing in the file is a class without a doc block this would cause calls to
45			// getDocBlock() on said class to return our long lost doc_comment. Argh.
46			// To workaround, cause the parser to parse an empty docblock. Sure getDocBlock() will return this, but at least
47			// it's harmless to us.
48			token_get_all("<?php\n/*\n \n */");
49
50			$this->numTokens = count($this->tokens);
51
52			// Compatibility defines for PHP < 8.0.
53			if (!defined('T_NAME_QUALIFIED')) {
54			\define('T_NAME_QUALIFIED', -2);
55			}
56			if (!defined('T_NAME_FULLY_QUALIFIED')) {
57			\define('T_NAME_FULLY_QUALIFIED', -3);
58			}
59			if (!defined('T_NAME_RELATIVE')) {
60			\define('T_NAME_RELATIVE', -4);
61			}
62			}
63
64			/**
65			* Gets the next non whitespace and non comment token.
66			*
67			* @param boolean $docCommentIsComment If TRUE then a doc comment is considered a comment and skipped.
68			* If FALSE then only whitespace and normal comments are skipped.
69			*
70			* @return array\|null The token if exists, null otherwise.
71			*/
72			public function next($docCommentIsComment = TRUE)
73			{
74			for ($i = $this->pointer; $i < $this->numTokens; $i++) {
75			$this->pointer++;
76			if ($this->tokens[$i][0] === T_WHITESPACE \|\|
77			$this->tokens[$i][0] === T_COMMENT \|\|
78			($docCommentIsComment && $this->tokens[$i][0] === T_DOC_COMMENT)) {
79
80			continue;
81			}
82
83			return $this->tokens[$i];
84			}
85
86			return null;
87			}
88
89			/**
90			* Parses a single use statement.
91			*
92			* @return array A list with all found class names for a use statement.
93			*/
94			public function parseUseStatement()
95			{
96			return PHP_VERSION_ID >=80000 ? $this->parseUseStatementV8() : $this->parseUseStatementV7();
97			}
98
99			private function parseUseStatementV7(): array
100			{
101			$groupRoot = '';
102			$class = '';
103			$alias = '';
104			$statements = [];
105			$explicitAlias = false;
106			while (($token = $this->next())) {
107			$isNameToken = $token[0] === T_STRING \|\| $token[0] === T_NS_SEPARATOR;
108			if (!$explicitAlias && $isNameToken) {
109			$class .= $token[1];
110			$alias = $token[1];
111			} else if ($explicitAlias && $isNameToken) {
112			$alias .= $token[1];
113			} else if ($token[0] === T_AS) {
114			$explicitAlias = true;
115			$alias = '';
116			} else if ($token === ',') {
117			$statements[strtolower($alias)] = $groupRoot . $class;
118			$class = '';
119			$alias = '';
120			$explicitAlias = false;
121			} else if ($token === ';') {
122			$statements[strtolower($alias)] = $groupRoot . $class;
123			break;
124			} else if ($token === '{' ) {
125			$groupRoot = $class;
126			$class = '';
127			} else if ($token === '}' ) {
128			continue;
129			} else {
130			break;
131			}
132			}
133
134			return $statements;
135			}
136
137			private function parseUseStatementV8(): array
138			{
139			$groupRoot = '';
140			$class = '';
141			$alias = '';
142			$statements = [];
143			$explicitAlias = false;
144			while (($token = $this->next())) {
145			$isNameToken = $token[0] === T_STRING \|\| $token[0] === T_NS_SEPARATOR;
146
147			if ($token[0] === T_NAME_QUALIFIED \|\| $token[0] === T_NAME_FULLY_QUALIFIED) {
148			$class .= $token[1];
149
150			$classSplit = explode('\\', $token[1]);
151			$alias = $classSplit[count($classSplit) - 1];
152			} else if($token[0] === T_NS_SEPARATOR) {
153			$class .= '\\';
154			} else if ($token[0] === T_AS) {
155			$explicitAlias = true;
156			$alias = '';
157			} else if ($isNameToken && !$explicitAlias) {
158			$class = $token[1];
159			$alias = $token[1];
160			} else if ($isNameToken && $explicitAlias) {
161			$alias = $token[1];
162			} else if ($token === ',') {
163			$statements[strtolower($alias)] = $groupRoot . $class;
164			$class = '';
165			$alias = '';
166			$explicitAlias = false;
167			} else if ($token === ';') {
168			$statements[strtolower($alias)] = $groupRoot . $class;
169			break;
170			} else if ($token === '{' ) {
171			$groupRoot = $class;
172			$class = '';
173			} else if ($token === '}' ) {
174			continue;
175			} else {
176			break;
177			}
178			}
179
180			return $statements;
181			}
182
183			/**
184			* Gets all use statements.
185			*
186			* @param string $namespaceName The namespace name of the reflected class.
187			*
188			* @return array A list with all found use statements.
189			*/
190			public function parseUseStatements($namespaceName)
191			{
192			$statements = [];
193			while (($token = $this->next())) {
194			if ($token[0] === T_USE) {
195			$statements = array_merge($statements, $this->parseUseStatement());
196			continue;
197			}
198			if ($token[0] !== T_NAMESPACE \|\| $this->parseNamespace() != $namespaceName) {
199			continue;
200			}
201
202			// Get fresh array for new namespace. This is to prevent the parser to collect the use statements
203			// for a previous namespace with the same name. This is the case if a namespace is defined twice
204			// or if a namespace with the same name is commented out.
205			$statements = [];
206			}
207
208			return $statements;
209			}
210
211			/**
212			* Gets the namespace.
213			*
214			* @return string The found namespace.
215			*/
216			public function parseNamespace()
217			{
218			$name = '';
219			while (($token = $this->next()) && ($token[0] === T_STRING \|\| $token[0] === T_NS_SEPARATOR \|\| $token[0] === T_NAME_QUALIFIED)) {
220			$name .= $token[1];
221			}
222
223			return $name;
224			}
225
226			/**
227			* Gets the class name.
228			*
229			* @return string The found class name.
230			*/
231			public function parseClass()
232			{
233			// Namespaces and class names are tokenized the same: T_STRINGs
234			// separated by T_NS_SEPARATOR so we can use one function to provide
235			// both.
236			return $this->parseNamespace();
237			}
238			}
239

doctrine / annotations

Pull Request — master (#344)

TokenParser::next() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like