PHP: rfc:precise_float_value

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== PHP RFC: More precise float value handling ======
  * Version: 1.1
  * Date: 2015-07-30
  * Author: Yasuo Ohgaki <yohgaki@ohgaki.net>, Jakub Zelenka <bukka@php.net>
  * Status: Implemented (PHP 7.1)
  * First Published at: http://wiki.php.net/rfc/precise_float_value

===== Introduction =====

This RFC is based on the discussion about displaying float values in json_encode and proposes more precise float value handling overall.

JSON is used to exchange data between systems. Although [[https://tools.ietf.org/html/rfc7159|JSON RFC "6 Numbers"]] does not require specific implementation for float/int type, a float value should be handled as precise as possible by default.

Currently json_encode() uses EG(precision) which is set to 14. That means that 14 digits at most are used for displaying (printing) the number. IEEE 754 double supports higher precision and serialize()/var_export() uses PG(serialize_precision) which set to 17 be default to be more precise. Since json_encode() uses EG(precision), json_encode() removes lower digits of fraction parts and destroys original value even if PHP's float could hold more precise float value.

<code php>
<?php
$j = '{ "v": 0.1234567890123456789 }';
var_dump(json_decode($j));
var_dump(json_encode(json_decode($j)));
ini_set('precision', 20);
var_dump(json_decode($j));
var_dump(json_encode(json_decode($j)));
var_dump(0.1234567890123456789);
?>
</code>

<code>
object(stdClass)#1 (1) {
  ["v"]=>
  float(0.12345678901235)
}
string(22) "{"v":0.12345678901235}"
object(stdClass)#1 (1) {
  ["v"]=>
  float(0.12345678901234567737)
}
string(28) "{"v":0.12345678901234567737}"
float(0.12345678901234567737)
</code>

PHP's float type stores "raw" IEEE 754 double and could display accurate fraction value up to 17 digits.

Current PHP outputs meaningless values for oversized EG(precision)/PG(serialize_precision).

<code php>
<?php
$v = 0.12345678901234567890;
var_dump($v);
ini_set('precision', 100);
var_dump($v);
?>
</code>

<code>
float(0.12345678901235)
float(0.12345678901234567736988623209981597028672695159912109375)
</code>

That is caused by used mode for double to string conversion.

===== Proposal =====

This RFC proposes to introduce a new setting EG(precision)=-1 and PG(serialize_precision)=-1 that uses zend_dtoa()'s mode 0 which uses better algorigthm for rounding float numbers (-1 is used to indicate 0 mode).

The RFC also proposes changing ini for JSON precision to PG(serialize_precision).

Followings are sample codes and outputs of the proposed patch.

NEW behavior
<code php>
<?php
$v = 10.0000000000001;

ini_set('precision', -1);
ini_set('serialize_precision', -1);

var_dump($v);
echo var_export($v, true), PHP_EOL;
echo json_encode($v), PHP_EOL;
echo $v, PHP_EOL;
?>
</code>

  float(10.0000000000001)
  10.0000000000001
  10.0000000000001
  10.0000000000001



OLD behavior
<code php>
<?php
$v = 10.00000000000001;

ini_set('precision', 14);
ini_set('serialize_precision', 17);

var_dump($v);
echo var_export($v, true), PHP_EOL;
ini_set('serialize_precision', 14);
echo json_encode($v), PHP_EOL;
ini_set('serialize_precision', 17);
echo $v, PHP_EOL;
?>
</code>

  float(10)
  10.000000000000011
  10
  10


Please note that IEEE float cannot store exactly precise values. e.g. Result of "10/3" - see phpt of the patch. Even with this proposal, there will be rounding errors, but the behavior becomes similar to other languages and values are more precise in many cases.


===== Backward Incompatible Changes =====

Setting mode 0 as default can mean that the rounding will be more precise which also means that the rounding might be different in var_export()/serialize().

The BC break could happen only if someone would rely on exact output but that shouldn't be the case. All our existing tests passes when 0 mode is used.

None when old INI value is used.

===== Proposed PHP Version(s) =====

  * PHP 7.1

===== RFC Impact =====
==== To SAPIs ====
None.

==== To Existing Extensions ====
PHP overall
  * 0 mode (EG(precision)= -1) float outputs values rounded to nearest.

Standard module and JSON
  * serialize(), var_export(), json_encode - Use PG(serialize_precision) and 0 mode by default.

==== To Opcache ====

Not affected.

==== New Constants ====

None.

==== php.ini Defaults ====

precision
  * hardcoded default values : 14 Unmodified
  * php.ini-development values : 14 Unmodified
  * php.ini-production values : 14 Unmodified

serialize_precision
  * hardcoded default values : -1
  * php.ini-development values : -1
  * php.ini-production values : -1


===== Open Issues =====

None.

===== Unaffected PHP Functionality =====

PHP uses "raw" IEEE 754 value internally regardless of precision settings. Therefore, this RFC does not affect internal computation.


===== Future Scope =====

WDDX
  * wddx_serialize_vars/value() - Use PG(serialize_precision) and 0 mode. It uses EG(precision) currently.

XML_RPC
  * xmlrpc_encode() - Use PG(serialize_precision) and 0 mode. It uses EG(precision) currently.


===== Voting =====

Requires a 2/3 majority for the first vote (mode 0 to become default for serialize precision) and 50%+1 majority for the second vote (changing default for json).

The both votes are straight Yes/No votes.

<doodle title="Should mode 0 be introduced and set as default for PG(serialize_precision)" auth="bukka" voteType="single" closed="true">
   * Yes
   * No
</doodle>

<doodle title="Should PG(serialize_precision) be used instead of EG(precision) in json_encode" auth="bukka" voteType="single" closed="true">
   * Yes
   * No
</doodle>

The votes started on 2016-06-12 at 19:00 UTC and ended on 2016-06-19 at 19:00 UTC.

===== Patches and Tests =====

The initial PR can be found here:

  * https://github.com/php/php-src/pull/1455

Note that the PR is currently outdated but it will be updated if the RFC is accepted.

===== Implementation =====

Merged into php-src for PHP 7.1: https://github.com/php/php-src/commit/75b86a2

After the project is implemented, this section should contain 
  - a link to the PHP manual entry for the feature

===== References =====

  * https://marc.info/?l=php-internals&m=143786181229714&w=2
  * https://bugs.php.net/bug.php?id=71473
  * http://blog.reverberate.org/2016/02/06/floating-point-demystified-part2.html

===== Rejected Features =====

None
Page Tools
