Parser::symbol() - Code Metrics - Inspection of "solved TODOs regarding WithName." - lechimp-p/dicto.php - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( cc2765...340ff8 )

by Richard

created 2016-08-01 18:27 UTC

Parser::symbol() A

↳ Parent: Parser

Complexity

Conditions	1
Paths	1

Size

Total Lines	3
Code Lines	2

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	2
CRAP Score	1

Importance

Changes	1
Bugs	0	Features	0

Metric	Value
dl	0
loc	3
ccs	2
cts	2
cp	1
rs	10
c	1
b	0
f	0
cc	1
eloc	2
nc	1
nop	2
crap	1

<?php
/******************************************************************************
 * An implementation of dicto (scg.unibe.ch/dicto) in and for PHP.
 *
 * Copyright (c) 2016 Richard Klees <[email protected]>
 * 
 * This software is licensed under The MIT License. You should have received
 * a copy of the license along with the code.
 */

namespace Lechimp\Dicto\Definition;

/**
 * Baseclass for Parsers.
 */
abstract class Parser {
    /**
     * @var SymbolTable
     */
    protected $symbol_table;

    /**
     * @var Tokenizer|null
     */
    protected $tokenizer = null;

    /**
     * @var array   (Symbol, array $matches)
     */
    protected $token;

    public function __construct() {
        $this->symbol_table = $this->create_symbol_table();
    }

    /**
     * Parse the string according to this parser.
     *
     * @return mixed
     */
    public function parse($source) {
        try {
            $this->tokenizer = $this->create_tokenizer($source);
            $this->token = $this->tokenizer->current();
            return $this->root();
        }
        finally {
            $this->tokenizer = null;
            $this->token = null;

        }
    }

    /**
     * The root for the parse tree.
     *
     * @return  mixed
     */
    abstract protected function root();

    // Factory Methods

    /**
     * Build the Tokenizer.
     *
     * @return  Tokenizer
     */
    public function create_tokenizer($source) {
        assert('is_string($source)');
        return new Tokenizer($this->symbol_table, $source);
    }

    /**
     * Build the SymbolTable
     *
     * @return SymbolTable
     */
    public function create_symbol_table() {
        // TODO: When symbol, operator and stuff were moved to
        //       Symbol table, there could be an add_symbols method
        //       postprocessing the table instead of using this->symbol etc.
        return new SymbolTable();
    }

    // Helpers for defining the grammar.

    /**
     * Add a symbol to the symbol table.
     *
     * TODO: This most probably should go to symbol table.
     *
     * @param   string  $regexp
     * @param   int     $binding_power
     * @throws  \InvalidArgumentException if %$regexp% is not a regexp
     * @throws  \LogicException if there already is a symbol with that $regexp.
     * @return  Symbol
     */
    protected function symbol($regexp, $binding_power = 0) {
        return $this->symbol_table->add_symbol($regexp, $binding_power);
    }

    /**
     * Add an operator to the symbol table.
     *
     * TODO: This most probably should go to symbol table.
     *
     * Convenience, will split the given string and wrap each char in []
     * before passing it to symbol.
     *
     * @param   string  $op
     * @param   int     $binding_power
     * @throws  \InvalidArgumentException if %$regexp% is not a regexp
     * @throws  \LogicException if there already is a symbol with that $regexp.
     * @return  Symbol
     */
    protected function operator($op, $binding_power = 0) {
        $regexp = $this->operator_regexp($op);
        return $this->symbol($regexp, $binding_power);
    }

    /**
     * Add a literal to the symbol table, where the matches are
     * transformed using the $converter.
     *
     * TODO: This most probably should go to symbol table.
     *
     * @param   string      $regexp
     * @param   \Closure    $converter
     * @throws  \InvalidArgumentException if %$regexp% is not a regexp
     * @throws  \LogicException if there already is a symbol with that $regexp.
     * @return  Symbol
     */
    protected function literal($regexp, $converter) {
        return $this->symbol($regexp)
            ->null_denotation_is($converter);
    }

    // Helpers for actual parsing.

    /**
     * Set the current token to the next token from the tokenizer.
     *
     * @return  null
     */
    protected function fetch_next_token() {
        assert('is_array($this->token)');
        assert('$this->tokenizer !== null');
        $this->tokenizer->next();
        $this->token = $this->tokenizer->current();
    }

    /**
     * Get the current symbol.
     *
     * @return  Symbol
     */
    protected function current_symbol() {
        return $this->token[0];
    }

    /**
     * Get the current match.
     *
     * @return  string[] 
     */
    protected function current_match() {
        return $this->token[1];
    }

    /**
     * Advance the tokenizer to the next token if current token
     * was matched by the given regexp.
     *
     * @param   string  $regexp
     * @return  null
     */
    protected function advance($regexp) {
        assert('is_string($regexp)');
        assert('is_array($this->token)');
        assert('$this->tokenizer !== null');
        if (!$this->is_current_token_matched_by($regexp)) {
            throw new ParserException("Syntax Error: Expected '$regexp'");
        }
        $this->tokenizer->next();
        $this->token = $this->tokenizer->current();
    }

    /**
     * Advance the tokenizer to the next token if current token
     * was matched by the given operator.
     *
     * @param   string  $op
     * @return  null
     */
    protected function advance_operator($op) {
        $this->advance($this->operator_regexp($op));
    }

    /**
     * Is the end of the file reached?
     *
     * @return  bool
     */
    public function is_end_of_file_reached() {
        return $this->is_current_token_matched_by("");
    }

    /**
     * Check if the current token was matched by the given regexp.
     *
     * @param   string  $regexp
     * @return  bool
     */
    protected function is_current_token_matched_by($regexp) {
        assert('is_string($regexp)');
        return $this->token[0]->regexp() == $regexp;
    }

    /**
     * Check if the current token is the given operator.
     *
     * @param   string  $operator
     * @return  bool
     */
    protected function is_current_token_operator($operator) {
        return $this->is_current_token_matched_by($this->operator_regexp($operator));
    }

    // Internal Helpers
    /**
     * "abc" -> "[a][b][c]"
     *
     * @param   string  $op
     * @return  string
     */
    protected function operator_regexp($op) {
        assert('is_string($op)');
        $regexp = array();
        foreach (str_split($op, 1) as $c) {
            $regexp[] = "[$c]";
        }
        return implode("", $regexp);
    }
}


1		<?php
2		/******************************************************************************
3		* An implementation of dicto (scg.unibe.ch/dicto) in and for PHP.
4		*
5		* Copyright (c) 2016 Richard Klees <[email protected]>
6		*
7		* This software is licensed under The MIT License. You should have received
8		* a copy of the license along with the code.
9		*/
10
11		namespace Lechimp\Dicto\Definition;
12
13		/**
14		* Baseclass for Parsers.
15		*/
16		abstract class Parser {
17		/**
18		* @var SymbolTable
19		*/
20		protected $symbol_table;
21
22		/**
23		* @var Tokenizer\|null
24		*/
25		protected $tokenizer = null;
26
27		/**
28		* @var array (Symbol, array $matches)
29		*/
30		protected $token;
31
32	37	public function __construct() {
33	37	$this->symbol_table = $this->create_symbol_table();
34	37	}
35
36		/**
37		* Parse the string according to this parser.
38		*
39		* @return mixed
40		*/
41	37	public function parse($source) {
42		try {
43	37	$this->tokenizer = $this->create_tokenizer($source);
44	37	$this->token = $this->tokenizer->current();
45	37	return $this->root();
46	1	}
47		finally {
48	37	$this->tokenizer = null;
49	37	$this->token = null;
		0 ignored issues – show Documentation Bug introduced 2016-08-01 18:35 UTC by Report Bug Copy Issue Report It seems like `null` of type `null` is incompatible with the declared type `array` of property `$token`. Our type inference engine has found an assignment to a property that is incompatible with the declared type of that property. Either this assignment is in error or the assigned type should be added to the documentation/type hint for that property.. Loading history...
50		}
51		}
52
53		/**
54		* The root for the parse tree.
55		*
56		* @return mixed
57		*/
58		abstract protected function root();
59
60		// Factory Methods
61
62		/**
63		* Build the Tokenizer.
64		*
65		* @return Tokenizer
66		*/
67	37	public function create_tokenizer($source) {
68	37	assert('is_string($source)');
69	37	return new Tokenizer($this->symbol_table, $source);
70		}
71
72		/**
73		* Build the SymbolTable
74		*
75		* @return SymbolTable
76		*/
77	37	public function create_symbol_table() {
78		// TODO: When symbol, operator and stuff were moved to
79		// Symbol table, there could be an add_symbols method
80		// postprocessing the table instead of using this->symbol etc.
81	37	return new SymbolTable();
82		}
83
84		// Helpers for defining the grammar.
85
86		/**
87		* Add a symbol to the symbol table.
88		*
89		* TODO: This most probably should go to symbol table.
90		*
91		* @param string $regexp
92		* @param int $binding_power
93		* @throws \InvalidArgumentException if %$regexp% is not a regexp
94		* @throws \LogicException if there already is a symbol with that $regexp.
95		* @return Symbol
96		*/
97	37	protected function symbol($regexp, $binding_power = 0) {
98	37	return $this->symbol_table->add_symbol($regexp, $binding_power);
99		}
100
101		/**
102		* Add an operator to the symbol table.
103		*
104		* TODO: This most probably should go to symbol table.
105		*
106		* Convenience, will split the given string and wrap each char in []
107		* before passing it to symbol.
108		*
109		* @param string $op
110		* @param int $binding_power
111		* @throws \InvalidArgumentException if %$regexp% is not a regexp
112		* @throws \LogicException if there already is a symbol with that $regexp.
113		* @return Symbol
114		*/
115	37	protected function operator($op, $binding_power = 0) {
116	37	$regexp = $this->operator_regexp($op);
117	37	return $this->symbol($regexp, $binding_power);
118		}
119
120		/**
121		* Add a literal to the symbol table, where the matches are
122		* transformed using the $converter.
123		*
124		* TODO: This most probably should go to symbol table.
125		*
126		* @param string $regexp
127		* @param \Closure $converter
128		* @throws \InvalidArgumentException if %$regexp% is not a regexp
129		* @throws \LogicException if there already is a symbol with that $regexp.
130		* @return Symbol
131		*/
132	37	protected function literal($regexp, $converter) {
133	37	return $this->symbol($regexp)
134	37	->null_denotation_is($converter);
135		}
136
137		// Helpers for actual parsing.
138
139		/**
140		* Set the current token to the next token from the tokenizer.
141		*
142		* @return null
143		*/
144	35	protected function fetch_next_token() {
145	35	assert('is_array($this->token)');
146	35	assert('$this->tokenizer !== null');
147	35	$this->tokenizer->next();
148	35	$this->token = $this->tokenizer->current();
149	35	}
150
151		/**
152		* Get the current symbol.
153		*
154		* @return Symbol
155		*/
156	31	protected function current_symbol() {
157	31	return $this->token[0];
158		}
159
160		/**
161		* Get the current match.
162		*
163		* @return string[]
164		*/
165	35	protected function current_match() {
166	35	return $this->token[1];
167		}
168
169		/**
170		* Advance the tokenizer to the next token if current token
171		* was matched by the given regexp.
172		*
173		* @param string $regexp
174		* @return null
175		*/
176	14	protected function advance($regexp) {
177	14	assert('is_string($regexp)');
178	14	assert('is_array($this->token)');
179	14	assert('$this->tokenizer !== null');
180	14	if (!$this->is_current_token_matched_by($regexp)) {
181		throw new ParserException("Syntax Error: Expected '$regexp'");
182		}
183	14	$this->tokenizer->next();
184	14	$this->token = $this->tokenizer->current();
185	14	}
186
187		/**
188		* Advance the tokenizer to the next token if current token
189		* was matched by the given operator.
190		*
191		* @param string $op
192		* @return null
193		*/
194	10	protected function advance_operator($op) {
195	10	$this->advance($this->operator_regexp($op));
196	10	}
197
198		/**
199		* Is the end of the file reached?
200		*
201		* @return bool
202		*/
203	28	public function is_end_of_file_reached() {
204	28	return $this->is_current_token_matched_by("");
205		}
206
207		/**
208		* Check if the current token was matched by the given regexp.
209		*
210		* @param string $regexp
211		* @return bool
212		*/
213	35	protected function is_current_token_matched_by($regexp) {
214	35	assert('is_string($regexp)');
215	35	return $this->token[0]->regexp() == $regexp;
216		}
217
218		/**
219		* Check if the current token is the given operator.
220		*
221		* @param string $operator
222		* @return bool
223		*/
224	7	protected function is_current_token_operator($operator) {
225	7	return $this->is_current_token_matched_by($this->operator_regexp($operator));
226		}
227
228		// Internal Helpers
229		/**
230		* "abc" -> "[a][b][c]"
231		*
232		* @param string $op
233		* @return string
234		*/
235	37	protected function operator_regexp($op) {
236	37	assert('is_string($op)');
237	37	$regexp = array();
238	37	foreach (str_split($op, 1) as $c) {
239	37	$regexp[] = "[$c]";
240	37	}
241	37	return implode("", $regexp);
242		}
243		}
244

lechimp-p / dicto.php

Push — master ( cc2765...340ff8 )

Parser::symbol() A

Complexity

Size

Duplication

Code Coverage

Importance

Duplication Side-by-Side

Filter issues like