aboutsummaryrefslogtreecommitdiffstats
path: root/library/HTMLPurifier/UnitConverter.php
blob: 166f3bf306b9f55ab7c89ecbbd010bfa347e30eb (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
<?php

/**
 * Class for converting between different unit-lengths as specified by
 * CSS.
 */
class HTMLPurifier_UnitConverter
{

    const ENGLISH = 1;
    const METRIC = 2;
    const DIGITAL = 3;

    /**
     * Units information array. Units are grouped into measuring systems
     * (English, Metric), and are assigned an integer representing
     * the conversion factor between that unit and the smallest unit in
     * the system. Numeric indexes are actually magical constants that
     * encode conversion data from one system to the next, with a O(n^2)
     * constraint on memory (this is generally not a problem, since
     * the number of measuring systems is small.)
     */
    protected static $units = array(
        self::ENGLISH => array(
            'px' => 3, // This is as per CSS 2.1 and Firefox. Your mileage may vary
            'pt' => 4,
            'pc' => 48,
            'in' => 288,
            self::METRIC => array('pt', '0.352777778', 'mm'),
        ),
        self::METRIC => array(
            'mm' => 1,
            'cm' => 10,
            self::ENGLISH => array('mm', '2.83464567', 'pt'),
        ),
    );

    /**
     * Minimum bcmath precision for output.
     * @type int
     */
    protected $outputPrecision;

    /**
     * Bcmath precision for internal calculations.
     * @type int
     */
    protected $internalPrecision;

    /**
     * Whether or not BCMath is available.
     * @type bool
     */
    private $bcmath;

    public function __construct($output_precision = 4, $internal_precision = 10, $force_no_bcmath = false)
    {
        $this->outputPrecision = $output_precision;
        $this->internalPrecision = $internal_precision;
        $this->bcmath = !$force_no_bcmath && function_exists('bcmul');
    }

    /**
     * Converts a length object of one unit into another unit.
     * @param HTMLPurifier_Length $length
     *      Instance of HTMLPurifier_Length to convert. You must validate()
     *      it before passing it here!
     * @param string $to_unit
     *      Unit to convert to.
     * @return HTMLPurifier_Length|bool
     * @note
     *      About precision: This conversion function pays very special
     *      attention to the incoming precision of values and attempts
     *      to maintain a number of significant figure. Results are
     *      fairly accurate up to nine digits. Some caveats:
     *          - If a number is zero-padded as a result of this significant
     *            figure tracking, the zeroes will be eliminated.
     *          - If a number contains less than four sigfigs ($outputPrecision)
     *            and this causes some decimals to be excluded, those
     *            decimals will be added on.
     */
    public function convert($length, $to_unit)
    {
        if (!$length->isValid()) {
            return false;
        }

        $n = $length->getN();
        $unit = $length->getUnit();

        if ($n === '0' || $unit === false) {
            return new HTMLPurifier_Length('0', false);
        }

        $state = $dest_state = false;
        foreach (self::$units as $k => $x) {
            if (isset($x[$unit])) {
                $state = $k;
            }
            if (isset($x[$to_unit])) {
                $dest_state = $k;
            }
        }
        if (!$state || !$dest_state) {
            return false;
        }

        // Some calculations about the initial precision of the number;
        // this will be useful when we need to do final rounding.
        $sigfigs = $this->getSigFigs($n);
        if ($sigfigs < $this->outputPrecision) {
            $sigfigs = $this->outputPrecision;
        }

        // BCMath's internal precision deals only with decimals. Use
        // our default if the initial number has no decimals, or increase
        // it by how ever many decimals, thus, the number of guard digits
        // will always be greater than or equal to internalPrecision.
        $log = (int)floor(log(abs($n), 10));
        $cp = ($log < 0) ? $this->internalPrecision - $log : $this->internalPrecision; // internal precision

        for ($i = 0; $i < 2; $i++) {

            // Determine what unit IN THIS SYSTEM we need to convert to
            if ($dest_state === $state) {
                // Simple conversion
                $dest_unit = $to_unit;
            } else {
                // Convert to the smallest unit, pending a system shift
                $dest_unit = self::$units[$state][$dest_state][0];
            }

            // Do the conversion if necessary
            if ($dest_unit !== $unit) {
                $factor = $this->div(self::$units[$state][$unit], self::$units[$state][$dest_unit], $cp);
                $n = $this->mul($n, $factor, $cp);
                $unit = $dest_unit;
            }

            // Output was zero, so bail out early. Shouldn't ever happen.
            if ($n === '') {
                $n = '0';
                $unit = $to_unit;
                break;
            }

            // It was a simple conversion, so bail out
            if ($dest_state === $state) {
                break;
            }

            if ($i !== 0) {
                // Conversion failed! Apparently, the system we forwarded
                // to didn't have this unit. This should never happen!
                return false;
            }

            // Pre-condition: $i == 0

            // Perform conversion to next system of units
            $n = $this->mul($n, self::$units[$state][$dest_state][1], $cp);
            $unit = self::$units[$state][$dest_state][2];
            $state = $dest_state;

            // One more loop around to convert the unit in the new system.

        }

        // Post-condition: $unit == $to_unit
        if ($unit !== $to_unit) {
            return false;
        }

        // Useful for debugging:
        //echo "<pre>n";
        //echo "$n\nsigfigs = $sigfigs\nnew_log = $new_log\nlog = $log\nrp = $rp\n</pre>\n";

        $n = $this->round($n, $sigfigs);
        if (strpos($n, '.') !== false) {
            $n = rtrim($n, '0');
        }
        $n = rtrim($n, '.');

        return new HTMLPurifier_Length($n, $unit);
    }

    /**
     * Returns the number of significant figures in a string number.
     * @param string $n Decimal number
     * @return int number of sigfigs
     */
    public function getSigFigs($n)
    {
        $n = ltrim($n, '0+-');
        $dp = strpos($n, '.'); // decimal position
        if ($dp === false) {
            $sigfigs = strlen(rtrim($n, '0'));
        } else {
            $sigfigs = strlen(ltrim($n, '0.')); // eliminate extra decimal character
            if ($dp !== 0) {
                $sigfigs--;
            }
        }
        return $sigfigs;
    }

    /**
     * Adds two numbers, using arbitrary precision when available.
     * @param string $s1
     * @param string $s2
     * @param int $scale
     * @return string
     */
    private function add($s1, $s2, $scale)
    {
        if ($this->bcmath) {
            return bcadd($s1, $s2, $scale);
        } else {
            return $this->scale((float)$s1 + (float)$s2, $scale);
        }
    }

    /**
     * Multiples two numbers, using arbitrary precision when available.
     * @param string $s1
     * @param string $s2
     * @param int $scale
     * @return string
     */
    private function mul($s1, $s2, $scale)
    {
        if ($this->bcmath) {
            return bcmul($s1, $s2, $scale);
        } else {
            return $this->scale((float)$s1 * (float)$s2, $scale);
        }
    }

    /**
     * Divides two numbers, using arbitrary precision when available.
     * @param string $s1
     * @param string $s2
     * @param int $scale
     * @return string
     */
    private function div($s1, $s2, $scale)
    {
        if ($this->bcmath) {
            return bcdiv($s1, $s2, $scale);
        } else {
            return $this->scale((float)$s1 / (float)$s2, $scale);
        }
    }

    /**
     * Rounds a number according to the number of sigfigs it should have,
     * using arbitrary precision when available.
     * @param float $n
     * @param int $sigfigs
     * @return string
     */
    private function round($n, $sigfigs)
    {
        $new_log = (int)floor(log(abs($n), 10)); // Number of digits left of decimal - 1
        $rp = $sigfigs - $new_log - 1; // Number of decimal places needed
        $neg = $n < 0 ? '-' : ''; // Negative sign
        if ($this->bcmath) {
            if ($rp >= 0) {
                $n = bcadd($n, $neg . '0.' . str_repeat('0', $rp) . '5', $rp + 1);
                $n = bcdiv($n, '1', $rp);
            } else {
                // This algorithm partially depends on the standardized
                // form of numbers that comes out of bcmath.
                $n = bcadd($n, $neg . '5' . str_repeat('0', $new_log - $sigfigs), 0);
                $n = substr($n, 0, $sigfigs + strlen($neg)) . str_repeat('0', $new_log - $sigfigs + 1);
            }
            return $n;
        } else {
            return $this->scale(round($n, $sigfigs - $new_log - 1), $rp + 1);
        }
    }

    /**
     * Scales a float to $scale digits right of decimal point, like BCMath.
     * @param float $r
     * @param int $scale
     * @return string
     */
    private function scale($r, $scale)
    {
        if ($scale < 0) {
            // The f sprintf type doesn't support negative numbers, so we
            // need to cludge things manually. First get the string.
            $r = sprintf('%.0f', (float)$r);
            // Due to floating point precision loss, $r will more than likely
            // look something like 4652999999999.9234. We grab one more digit
            // than we need to precise from $r and then use that to round
            // appropriately.
            $precise = (string)round(substr($r, 0, strlen($r) + $scale), -1);
            // Now we return it, truncating the zero that was rounded off.
            return substr($precise, 0, -1) . str_repeat('0', -$scale + 1);
        }
        return sprintf('%.' . $scale . 'f', (float)$r);
    }
}

// vim: et sw=4 sts=4