1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace DataValues\Geo\Parsers; |
4
|
|
|
|
5
|
|
|
use DataValues\Geo\Values\LatLongValue; |
6
|
|
|
use ValueParsers\ParseException; |
7
|
|
|
use ValueParsers\ParserOptions; |
8
|
|
|
use ValueParsers\StringValueParser; |
9
|
|
|
use ValueParsers\ValueParser; |
10
|
|
|
|
11
|
|
|
/** |
12
|
|
|
* @since 0.1, renamed in 2.0 |
13
|
|
|
* |
14
|
|
|
* @license GPL-2.0+ |
15
|
|
|
* @author H. Snater < [email protected] > |
16
|
|
|
* @author Jeroen De Dauw < [email protected] > |
17
|
|
|
*/ |
18
|
|
|
abstract class LatLongParserBase implements ValueParser { |
19
|
|
|
|
20
|
|
|
const FORMAT_NAME = 'geo-coordinate'; |
21
|
|
|
|
22
|
|
|
/** |
23
|
|
|
* The symbols representing the different directions for usage in directional notation. |
24
|
|
|
*/ |
25
|
|
|
const OPT_NORTH_SYMBOL = 'north'; |
26
|
|
|
const OPT_EAST_SYMBOL = 'east'; |
27
|
|
|
const OPT_SOUTH_SYMBOL = 'south'; |
28
|
|
|
const OPT_WEST_SYMBOL = 'west'; |
29
|
|
|
|
30
|
|
|
/** |
31
|
|
|
* The symbol to use as separator between latitude and longitude. |
32
|
|
|
*/ |
33
|
|
|
const OPT_SEPARATOR_SYMBOL = 'separator'; |
34
|
|
|
|
35
|
|
|
/** |
36
|
|
|
* Delimiters used to split a coordinate string when unable to split by using the separator. |
37
|
|
|
* @var string[] |
38
|
|
|
*/ |
39
|
|
|
protected $defaultDelimiters; |
40
|
|
|
|
41
|
|
|
/** |
42
|
|
|
* @var ParserOptions |
43
|
|
|
*/ |
44
|
|
|
protected $options; |
45
|
|
|
|
46
|
|
|
public function __construct( ParserOptions $options = null ) { |
47
|
|
|
$this->options = $options ?: new ParserOptions(); |
48
|
|
|
|
49
|
|
|
$this->options->defaultOption( ValueParser::OPT_LANG, 'en' ); |
50
|
|
|
|
51
|
|
|
$this->options->defaultOption( self::OPT_NORTH_SYMBOL, 'N' ); |
52
|
|
|
$this->options->defaultOption( self::OPT_EAST_SYMBOL, 'E' ); |
53
|
|
|
$this->options->defaultOption( self::OPT_SOUTH_SYMBOL, 'S' ); |
54
|
|
|
$this->options->defaultOption( self::OPT_WEST_SYMBOL, 'W' ); |
55
|
|
|
|
56
|
|
|
$this->options->defaultOption( self::OPT_SEPARATOR_SYMBOL, ',' ); |
57
|
|
|
} |
58
|
|
|
|
59
|
|
|
/** |
60
|
|
|
* Parses a single coordinate segment (either latitude or longitude) and returns it as a float. |
61
|
|
|
* |
62
|
|
|
* @param string $coordinateSegment |
63
|
|
|
* |
64
|
|
|
* @throws ParseException |
65
|
|
|
* @return float |
66
|
|
|
*/ |
67
|
|
|
abstract protected function getParsedCoordinate( $coordinateSegment ); |
68
|
|
|
|
69
|
|
|
/** |
70
|
|
|
* Returns whether a coordinate split into its two segments is in the representation expected by |
71
|
|
|
* this parser. |
72
|
|
|
* |
73
|
|
|
* @param string[] $normalizedCoordinateSegments |
74
|
|
|
* |
75
|
|
|
* @return boolean |
76
|
|
|
*/ |
77
|
|
|
abstract protected function areValidCoordinates( array $normalizedCoordinateSegments ); |
78
|
|
|
|
79
|
|
|
/** |
80
|
|
|
* @see ValueParser::parse |
81
|
|
|
* |
82
|
|
|
* @param string $value |
83
|
|
|
* |
84
|
|
|
* @throws ParseException |
85
|
|
|
* @return LatLongValue |
86
|
|
|
*/ |
87
|
|
|
public function parse( $value ) { |
88
|
|
|
if ( !is_string( $value ) ) { |
89
|
|
|
throw new ParseException( 'Not a string' ); |
90
|
|
|
} |
91
|
|
|
|
92
|
|
|
$rawValue = $value; |
93
|
|
|
|
94
|
|
|
$value = $this->removeInvalidChars( $value ); |
95
|
|
|
|
96
|
|
|
$normalizedCoordinateSegments = $this->splitString( $value ); |
97
|
|
|
|
98
|
|
|
if ( !$this->areValidCoordinates( $normalizedCoordinateSegments ) ) { |
99
|
|
|
throw new ParseException( 'Not a valid geographical coordinate', $rawValue, static::FORMAT_NAME ); |
100
|
|
|
} |
101
|
|
|
|
102
|
|
|
list( $latitude, $longitude ) = $normalizedCoordinateSegments; |
103
|
|
|
|
104
|
|
|
return new LatLongValue( |
105
|
|
|
$this->getParsedCoordinate( $latitude ), |
106
|
|
|
$this->getParsedCoordinate( $longitude ) |
107
|
|
|
); |
108
|
|
|
} |
109
|
|
|
|
110
|
|
|
/** |
111
|
|
|
* Returns a string trimmed and with control characters and characters with ASCII values above |
112
|
|
|
* 126 removed. SPACE characters within the string are not removed to retain the option to split |
113
|
|
|
* the string using that character. |
114
|
|
|
* |
115
|
|
|
* @param string $string |
116
|
|
|
* |
117
|
|
|
* @return string |
118
|
|
|
*/ |
119
|
|
|
protected function removeInvalidChars( $string ) { |
120
|
|
|
$filtered = []; |
121
|
|
|
|
122
|
|
|
foreach ( str_split( $string ) as $character ) { |
123
|
|
|
$asciiValue = ord( $character ); |
124
|
|
|
|
125
|
|
|
if ( |
126
|
|
|
( $asciiValue >= 32 && $asciiValue < 127 ) |
127
|
|
|
|| $asciiValue == 194 |
128
|
|
|
|| $asciiValue == 176 |
129
|
|
|
) { |
130
|
|
|
$filtered[] = $character; |
131
|
|
|
} |
132
|
|
|
} |
133
|
|
|
|
134
|
|
|
return trim( implode( '', $filtered ) ); |
135
|
|
|
} |
136
|
|
|
|
137
|
|
|
/** |
138
|
|
|
* Splits a string into two strings using the separator specified in the options. If the string |
139
|
|
|
* could not be split using the separator, the method will try to split the string by analyzing |
140
|
|
|
* the used symbols. If the string could not be split into two parts, an empty array is |
141
|
|
|
* returned. |
142
|
|
|
* |
143
|
|
|
* @param string $normalizedCoordinateString |
144
|
|
|
* |
145
|
|
|
* @throws ParseException if unable to split input string into two segments |
146
|
|
|
* @return string[] |
147
|
|
|
*/ |
148
|
|
|
protected function splitString( $normalizedCoordinateString ) { |
149
|
|
|
$separator = $this->getOption( self::OPT_SEPARATOR_SYMBOL ); |
150
|
|
|
|
151
|
|
|
$normalizedCoordinateSegments = explode( $separator, $normalizedCoordinateString ); |
152
|
|
|
|
153
|
|
|
if ( count( $normalizedCoordinateSegments ) !== 2 ) { |
154
|
|
|
// Separator not present within the string, trying to figure out the segments by |
155
|
|
|
// splitting after the first direction character or degree symbol: |
156
|
|
|
$delimiters = $this->defaultDelimiters; |
157
|
|
|
|
158
|
|
|
$ns = [ |
159
|
|
|
$this->getOption( self::OPT_NORTH_SYMBOL ), |
160
|
|
|
$this->getOption( self::OPT_SOUTH_SYMBOL ) |
161
|
|
|
]; |
162
|
|
|
|
163
|
|
|
$ew = [ |
164
|
|
|
$this->getOption( self::OPT_EAST_SYMBOL ), |
165
|
|
|
$this->getOption( self::OPT_WEST_SYMBOL ) |
166
|
|
|
]; |
167
|
|
|
|
168
|
|
|
foreach ( $ns as $delimiter ) { |
169
|
|
|
if ( mb_strpos( $normalizedCoordinateString, $delimiter ) === 0 ) { |
170
|
|
|
// String starts with "north" or "west" symbol: Separation needs to be done |
171
|
|
|
// before the "east" or "west" symbol. |
172
|
|
|
$delimiters = array_merge( $ew, $delimiters ); |
173
|
|
|
break; |
174
|
|
|
} |
175
|
|
|
} |
176
|
|
|
|
177
|
|
|
if ( count( $delimiters ) !== count( $this->defaultDelimiters ) + 2 ) { |
178
|
|
|
$delimiters = array_merge( $ns, $delimiters ); |
179
|
|
|
} |
180
|
|
|
|
181
|
|
|
foreach ( $delimiters as $delimiter ) { |
182
|
|
|
$delimiterPos = mb_strpos( $normalizedCoordinateString, $delimiter ); |
183
|
|
|
if ( $delimiterPos !== false ) { |
184
|
|
|
$adjustPos = ( in_array( $delimiter, $ew ) ) ? 0 : mb_strlen( $delimiter ); |
185
|
|
|
$normalizedCoordinateSegments = [ |
186
|
|
|
mb_substr( $normalizedCoordinateString, 0, $delimiterPos + $adjustPos ), |
187
|
|
|
mb_substr( $normalizedCoordinateString, $delimiterPos + $adjustPos ) |
188
|
|
|
]; |
189
|
|
|
break; |
190
|
|
|
} |
191
|
|
|
} |
192
|
|
|
} |
193
|
|
|
|
194
|
|
|
if ( count( $normalizedCoordinateSegments ) !== 2 ) { |
195
|
|
|
throw new ParseException( __CLASS__ . ': Unable to split string ' |
196
|
|
|
. $normalizedCoordinateString . ' into two coordinate segments' ); |
197
|
|
|
} |
198
|
|
|
|
199
|
|
|
return $normalizedCoordinateSegments; |
200
|
|
|
} |
201
|
|
|
|
202
|
|
|
/** |
203
|
|
|
* Turns directional notation (N/E/S/W) of a single coordinate into non-directional notation |
204
|
|
|
* (+/-). |
205
|
|
|
* This method assumes there are no preceding or tailing spaces. |
206
|
|
|
* |
207
|
|
|
* @param string $coordinateSegment |
208
|
|
|
* |
209
|
|
|
* @return string |
210
|
|
|
*/ |
211
|
|
|
protected function resolveDirection( $coordinateSegment ) { |
212
|
|
|
$n = $this->getOption( self::OPT_NORTH_SYMBOL ); |
213
|
|
|
$e = $this->getOption( self::OPT_EAST_SYMBOL ); |
214
|
|
|
$s = $this->getOption( self::OPT_SOUTH_SYMBOL ); |
215
|
|
|
$w = $this->getOption( self::OPT_WEST_SYMBOL ); |
216
|
|
|
|
217
|
|
|
// If there is a direction indicator, remove it, and prepend a minus sign for south and west |
218
|
|
|
// directions. If there is no direction indicator, the coordinate is already non-directional |
219
|
|
|
// and no work is required. |
220
|
|
|
foreach ( [ $n, $e, $s, $w ] as $direction ) { |
221
|
|
|
// The coordinate segment may either start or end with a direction symbol. |
222
|
|
|
preg_match( |
223
|
|
|
'/^(' . $direction . '|)([^' . $direction . ']+)(' . $direction . '|)$/i', |
224
|
|
|
$coordinateSegment, |
225
|
|
|
$matches |
226
|
|
|
); |
227
|
|
|
|
228
|
|
|
if ( $matches[1] === $direction || $matches[3] === $direction ) { |
229
|
|
|
$coordinateSegment = $matches[2]; |
230
|
|
|
|
231
|
|
|
if ( in_array( $direction, [ $s, $w ] ) ) { |
232
|
|
|
$coordinateSegment = '-' . $coordinateSegment; |
233
|
|
|
} |
234
|
|
|
|
235
|
|
|
return $coordinateSegment; |
236
|
|
|
} |
237
|
|
|
} |
238
|
|
|
|
239
|
|
|
// Coordinate segment does not include a direction symbol. |
240
|
|
|
return $coordinateSegment; |
241
|
|
|
} |
242
|
|
|
|
243
|
|
|
protected function getOption( $optionName ) { |
244
|
|
|
return $this->options->getOption( $optionName ); |
245
|
|
|
} |
246
|
|
|
|
247
|
|
|
} |
248
|
|
|
|