# [RFC brainstorm] Approximately equals operator - Thread: https://externals.io/message/126989 - Original archive: https://news-web.php.net/php.internals/126989 - Started by Niels Dossche on 2025-03-31T22:03:03+00:00 - Messages: 17 - Reply by email to internals@lists.php.net - More for agents: search with https://externals.io/search.md?q={query} · JSON API at https://externals.io/api · documentation at https://externals.io/llms.txt --- ## #126989 — Niels Dossche — 2025-03-31T22:03:03+00:00 URL: https://externals.io/message/126989#126989 Hi internals! I'm excited to share what I've been working on! I had an epiphany. I realized what we truly need to revolutionize PHP: a new operator. Hear me out. We live in an imperfect world, and we often approximate data, but neither `==` nor `===` are ideal comparison operators to deal with these kinds of data. Introducing: the "approximately equal" (or "approx-equal") operator `~=` (to immitate the maths symbol ≃). This combines the power of type coercion with approximating equality. Who cares if things are actually equal, close enough amirite? First of all, if `$a == $b` holds, then `$a ~= $b` obviously. The true power lies where the data is not exactly the same, but "close enough"! Here are some examples: We all had situations where we wanted to compare two floating point numbers and it turns out that due to the non-exact representation, seemingly-equal numbers don't match! Gone are those days because the `~=` operator nicely rounds the numbers for you before comparing them. This also means that the "Fundamental Theorem of Engineering" now holds! i.e. 2.7 ~= 3 and 3.14 ~= 3. Of course also 2.7 ~= 3.14. But this is false obviously: 2 ~= 1. Ever had trouble with users mistyping something? Say no more! "This is a tpyo" ~= "This is a typo". It's typo-resistant! However, if the strings are too different, then they're not approx-equal. For example: "vanilla" ~= "strawberry" gives false. How does this work? * The strings are equal if their levenshtein ratio is <= 50%, so it's adaptive to the length. * If the ratio is > 50%, then the shortest string comes first in the comparison, such that if we ever get a `~<` operator, then "vanilla" ~< "strawberry". There is of course a PoC implementation available at: https://github.com/php/php-src/pull/18214 You can see more examples on GitHub in the tests, here is a copy: ```php // Number compares var_dump(2 ~= 1); // false var_dump(1.4 ~= 1); // true var_dump(-1.4 ~= -1); // true var_dump(-1.5 ~= -1.8); // true var_dump(random_int(1, 1) ~= 1.1); // true // Array compares (just compares the lengths) var_dump([1, 2, 3] ~= [2, 3, 4]); // true var_dump([1, 2, 3] ~= [2, 3, 4, 5]); // false // String / string compares var_dump("This is a tpyo" ~= "This is a typo"); // true var_dump("something" ~= "different"); // false var_dump("Wtf bro" ~= "Wtf sis"); // true // String / different type compares var_dump(-1.5 ~= "-1.a"); // true var_dump(-1.5 ~= "-1.aaaaaaa"); // false var_dump(NULL ~= "blablabla"); // false ``` Note that this does not support all possible Opcache optimizations _yet_, nor does it support the JIT yet. However, there are no real blockers to add support for that. I look forward to hearing you! Have a nice first day of the month ;) Kind regards Niels --- ## #126990 — Bilge — 2025-03-31T22:14:48+00:00 URL: https://externals.io/message/126989#126990 · In reply to: #126989 > Hi internals! > > I'm excited to share what I've been working on! > I had an epiphany. I realized what we truly need to revolutionize PHP: a new operator. > > Hear me out. > We live in an imperfect world, and we often approximate data, but neither `==` nor `===` are ideal comparison operators to deal with these kinds of data. > > Introducing: the "approximately equal" (or "approx-equal") operator `~=` (to immitate the maths symbol ≃). > This combines the power of type coercion with approximating equality. > Who cares if things are actually equal, close enough amirite? > > First of all, if `$a == $b` holds, then `$a ~= $b` obviously. > The true power lies where the data is not exactly the same, but "close enough"! > > Here are some examples: > > We all had situations where we wanted to compare two floating point numbers and it turns out that due to the non-exact representation, seemingly-equal numbers don't match! Gone are those days because the `~=` operator nicely rounds the numbers for you before comparing them. > This also means that the "Fundamental Theorem of Engineering" now holds! > i.e. 2.7 ~= 3 and 3.14 ~= 3. Of course also 2.7 ~= 3.14. But this is false obviously: 2 ~= 1. > > Ever had trouble with users mistyping something? Say no more! > "This is a tpyo" ~= "This is a typo". It's typo-resistant! > However, if the strings are too different, then they're not approx-equal. > For example: "vanilla" ~= "strawberry" gives false. > How does this work? > * The strings are equal if their levenshtein ratio is <= 50%, so it's adaptive to the length. > * If the ratio is > 50%, then the shortest string comes first in the comparison, such that if we ever get a `~<` operator, then "vanilla" ~< "strawberry". > > There is of course a PoC implementation available at: https://github.com/php/php-src/pull/18214 > You can see more examples on GitHub in the tests, here is a copy: > ```php > // Number compares > var_dump(2 ~= 1); // false > var_dump(1.4 ~= 1); // true > var_dump(-1.4 ~= -1); // true > var_dump(-1.5 ~= -1.8); // true > var_dump(random_int(1, 1) ~= 1.1); // true > > // Array compares (just compares the lengths) > var_dump([1, 2, 3] ~= [2, 3, 4]); // true > var_dump([1, 2, 3] ~= [2, 3, 4, 5]); // false > > // String / string compares > var_dump("This is a tpyo" ~= "This is a typo"); // true > var_dump("something" ~= "different"); // false > var_dump("Wtf bro" ~= "Wtf sis"); // true > > // String / different type compares > var_dump(-1.5 ~= "-1.a"); // true > var_dump(-1.5 ~= "-1.aaaaaaa"); // false > var_dump(NULL ~= "blablabla"); // false > ``` > > Note that this does not support all possible Opcache optimizations _yet_, nor does it support the JIT yet. > However, there are no real blockers to add support for that. > > I look forward to hearing you! > > Have a nice first day of the month ;) > Kind regards > Niels For the float case it's fine (because Epsilon is well defined), but I think overloading for the string case is not fine, because the hard-coded 50% distance is subjective and users may well want to configure that, so an operator is thus not suitable, notwithstanding Levenshtein has very limited application. If there is any sense in doing string comparisons with this operator, I think the proposed case is not it. The array case is also not good in my view, where you're just comparing length; I see no use for that whatsoever. What it _should_ do instead is compare where order is indistinct, i.e. [1, 2, 3] ~= [3, 2, 1], similar to PHPUnit's assertEqualsCanonicalizing [1]. Cheers, Bilge [1]: https://github.com/sebastianbergmann/comparator/blob/d67eceae47e3956aa28ab0c6e43e5a6765f45779/src/ArrayComparator.php#L43-L46 --- ## #126992 — drealecs@gmail.com — 2025-04-01T02:45:16+00:00 URL: https://externals.io/message/126989#126992 · In reply to: #126989 > Hi internals! > > I'm excited to share what I've been working on! > I had an epiphany. I realized what we truly need to revolutionize PHP: a > new operator. > > Hear me out. > We live in an imperfect world, and we often approximate data, but neither > `==` nor `===` are ideal comparison operators to deal with these kinds of > data. > > Introducing: the "approximately equal" (or "approx-equal") operator `~=` > (to immitate the maths symbol ≃). > This combines the power of type coercion with approximating equality. > Who cares if things are actually equal, close enough amirite? > > Hi Niels, When I was reading it, I felt a bit unsure, but numbers related, it was making sense. When you got to strings with `"This is a tpyo" ~= "This is a typo"`, I also remembered today it's 1st of April, so there's that... -- Alex --- ## #126995 — Iliya Miroslavov Iliev — 2025-04-01T07:40:43+00:00 URL: https://externals.io/message/126989#126995 · In reply to: #126992 > var_dump(random_int(1, 1) ~= 1.1); // true This one cracked me :D Thanks Niels! On Tue, Apr 1, 2025 at 5:48 AM Alexandru Pătrănescu wrote: > > > > > > […] > Hi Niels, > > When I was reading it, I felt a bit unsure, but numbers related, it was > making sense. > When you got to strings with `"This is a tpyo" ~= "This is a typo"`, I > also remembered today it's 1st of April, so there's that... > > -- > Alex > -- Iliya Miroslavov Iliev i.miroslavov@gmail.com --- ## #126996 — tim@bastelstu.be — 2025-04-01T08:56:56+00:00 URL: https://externals.io/message/126989#126996 · In reply to: #126989 Hi Am 2025-04-01 00:03, schrieb Niels Dossche: > We all had situations where we wanted to compare two floating point > numbers and it turns out that due to the non-exact representation, > seemingly-equal numbers don't match! Gone are those days because the > `~=` operator nicely rounds the numbers for you before comparing them. > This also means that the "Fundamental Theorem of Engineering" now > holds! > i.e. 2.7 ~= 3 and 3.14 ~= 3. Of course also 2.7 ~= 3.14. But this is > false obviously: 2 ~= 1. > Thank you for your proposal. I've tried it and I believe I found some fundamental flaw in the current logic. Consider: Le 1 avr. 2025 à 00:03, Niels Dossche a écrit : > > Hi internals! > > I'm excited to share what I've been working on! > I had an epiphany. I realized what we truly need to revolutionize PHP: a new operator. > > [...] > First of all, if `$a == $b` holds, then `$a ~= $b` obviously. > The true power lies where the data is not exactly the same, but "close enough"! > Hi Niels, A major issue with the `==` operator, is that it is not transitive: https://3v4l.org/dISMi I firmly think that it should corrected this with the new `~=` operator: it will make approximate code easier to reason about. As a bonus, with this amendment, the principle of explosion could be used to severely optimise the implementation. —Claude --- ## #127010 — Iliya Miroslavov Iliev — 2025-04-01T13:06:26+00:00 URL: https://externals.io/message/126989#127010 · In reply to: #127009 Claude, in your example if `var_dump(false == true);` is `false` what is `true` in this world? It is `true` that `false` is not `true`. > > > Le 1 avr. 2025 à 00:03, Niels Dossche a écrit : > > Hi internals! > > I'm excited to share what I've been working on! > I had an epiphany. I realized what we truly need to revolutionize PHP: a > new operator. > > [...] > First of all, if `$a == $b` holds, then `$a ~= $b` obviously. > The true power lies where the data is not exactly the same, but "close > enough"! > > > Hi Niels, > > A major issue with the `==` operator, is that it is not transitive: > https://3v4l.org/dISMi > > I firmly think that it should corrected this with the new `~=` operator: > it will make approximate code easier to reason about. > > As a bonus, with this amendment, the principle of explosion could be used > to severely optimise the implementation. > > —Claude > -- Iliya Miroslavov Iliev i.miroslavov@gmail.com --- ## #127011 — Christoph M. Becker — 2025-04-01T16:13:00+00:00 URL: https://externals.io/message/126989#127011 · In reply to: #126989 > We live in an imperfect world, and we often approximate data, but neither `==` nor `===` are ideal comparison operators to deal with these kinds of data. > > Introducing: the "approximately equal" (or "approx-equal") operator `~=` (to immitate the maths symbol ≃). > This combines the power of type coercion with approximating equality. > Who cares if things are actually equal, close enough amirite? > > First of all, if `$a == $b` holds, then `$a ~= $b` obviously. > The true power lies where the data is not exactly the same, but "close enough"! IMO a step in the right direction, but it doesn't solve the problem that the developer might not even know which equality operator to apply. Thus, I proprose the whatever (?) equality (=) is right (->) here (!) operator, e.g. $value1 ?=->! $value2 I leave the trivial implementation as exercise to the reader, while I'm porting the even more powerful rmmadwim TCL command[1], which, incidentially, also had been proposed on an April 1st. [1] Christoph --- ## #127012 — Iliya Miroslavov Iliev — 2025-04-01T16:37:13+00:00 URL: https://externals.io/message/126989#127012 · In reply to: #127011 >IMO a step in the right direction, but it doesn't solve the problem that >the developer might not even know which equality operator to apply. >Thus, I proprose the whatever (?) equality (=) is right (->) here (!) >operator, e.g. You mean something like: ```php function str_aprox(ClientData $_, string $value1, string $value2) { if ($value1 ?=->! $value2) { return (bool)random_int(1,0); } else { return (bool)random_int(0,1); } } ``` This sounds reasonable. I approve! On Tue, Apr 1, 2025 at 7:16 PM Christoph M. Becker wrote: > > > > […] > neither `==` nor `===` are ideal comparison operators to deal with these > kinds of data. > > […] > (to immitate the maths symbol ≃). > > […] > enough"! > > IMO a step in the right direction, but it doesn't solve the problem that > the developer might not even know which equality operator to apply. > Thus, I proprose the whatever (?) equality (=) is right (->) here (!) > operator, e.g. > > $value1 ?=->! $value2 > > I leave the trivial implementation as exercise to the reader, while I'm > porting the even more powerful rmmadwim TCL command[1], which, > incidentially, also had been proposed on an April 1st. > > [1] > > Christoph > -- Iliya Miroslavov Iliev i.miroslavov@gmail.com --- ## #127014 — Rob Landers — 2025-04-01T18:13:48+00:00 URL: https://externals.io/message/126989#127014 · In reply to: #127010 > Claude, in your example if `var_dump(false == true);` is `false` what is `true` in this world? It is `true` that `false` is not `true`. > > > > […] > > > -- > Iliya Miroslavov Iliev > i.miroslavov@gmail.com "The sky is blue is only true when it is daytime, or to put another way: truth is relative" - A drunk guy on the beach, 2014 — Rob --- ## #127016 — Iliya Miroslavov Iliev — 2025-04-01T19:26:17+00:00 URL: https://externals.io/message/126989#127016 · In reply to: #127014 > > > Claude, in your example if `var_dump(false == true);` is `false` what is > `true` in this world? It is `true` that `false` is not `true`. > > On Tue, Apr 1, 2025 at 3:39 PM Claude Pache > wrote: > > > > Le 1 avr. 2025 à 00:03, Niels Dossche a écrit : > > Hi internals! > > I'm excited to share what I've been working on! > I had an epiphany. I realized what we truly need to revolutionize PHP: a > new operator. > > [...] > First of all, if `$a == $b` holds, then `$a ~= $b` obviously. > The true power lies where the data is not exactly the same, but "close > enough"! > > > Hi Niels, > > A major issue with the `==` operator, is that it is not transitive: > https://3v4l.org/dISMi > > I firmly think that it should corrected this with the new `~=` operator: > it will make approximate code easier to reason about. > > As a bonus, with this amendment, the principle of explosion could be used > to severely optimise the implementation. > > —Claude > > > > -- > Iliya Miroslavov Iliev > i.miroslavov@gmail.com > > > "The sky is blue is only true when it is daytime, or to put another way: > truth is relative" > - A drunk guy on the beach, 2014 > > — Rob > Rob, our sun is white not yellow. The atmosphere filters the blue color... but for more CSS lessons $100 -- Iliya Miroslavov Iliev i.miroslavov@gmail.com --- ## #127017 — Larry Garfield — 2025-04-01T19:52:32+00:00 URL: https://externals.io/message/126989#127017 · In reply to: #126989 > Hi internals! > > I'm excited to share what I've been working on! > I had an epiphany. I realized what we truly need to revolutionize PHP: > a new operator. > > Hear me out. > We live in an imperfect world, and we often approximate data, but > neither `==` nor `===` are ideal comparison operators to deal with > these kinds of data. > > Introducing: the "approximately equal" (or "approx-equal") operator > `~=` (to immitate the maths symbol ≃). > This combines the power of type coercion with approximating equality. > Who cares if things are actually equal, close enough amirite? > > First of all, if `$a == $b` holds, then `$a ~= $b` obviously. > The true power lies where the data is not exactly the same, but "close enough"! > > Here are some examples: > > We all had situations where we wanted to compare two floating point > numbers and it turns out that due to the non-exact representation, > seemingly-equal numbers don't match! Gone are those days because the > `~=` operator nicely rounds the numbers for you before comparing them. > This also means that the "Fundamental Theorem of Engineering" now holds! > i.e. 2.7 ~= 3 and 3.14 ~= 3. Of course also 2.7 ~= 3.14. But this is > false obviously: 2 ~= 1. > > Ever had trouble with users mistyping something? Say no more! > "This is a tpyo" ~= "This is a typo". It's typo-resistant! > However, if the strings are too different, then they're not > approx-equal. > For example: "vanilla" ~= "strawberry" gives false. > How does this work? > * The strings are equal if their levenshtein ratio is <= 50%, so it's > adaptive to the length. > * If the ratio is > 50%, then the shortest string comes first in the > comparison, such that if we ever get a `~<` operator, then "vanilla" ~< > "strawberry". > > There is of course a PoC implementation available at: > https://github.com/php/php-src/pull/18214 > You can see more examples on GitHub in the tests, here is a copy: > ```php > // Number compares > var_dump(2 ~= 1); // false > var_dump(1.4 ~= 1); // true > var_dump(-1.4 ~= -1); // true > var_dump(-1.5 ~= -1.8); // true > var_dump(random_int(1, 1) ~= 1.1); // true > > // Array compares (just compares the lengths) > var_dump([1, 2, 3] ~= [2, 3, 4]); // true > var_dump([1, 2, 3] ~= [2, 3, 4, 5]); // false > > // String / string compares > var_dump("This is a tpyo" ~= "This is a typo"); // true > var_dump("something" ~= "different"); // false > var_dump("Wtf bro" ~= "Wtf sis"); // true > > // String / different type compares > var_dump(-1.5 ~= "-1.a"); // true > var_dump(-1.5 ~= "-1.aaaaaaa"); // false > var_dump(NULL ~= "blablabla"); // false > ``` > > Note that this does not support all possible Opcache optimizations > _yet_, nor does it support the JIT yet. > However, there are no real blockers to add support for that. > > I look forward to hearing you! > > Have a nice first day of the month ;) > Kind regards > Niels Naturally, the degree of closeness for strings or for floats should be controlled by an ini setting. Maximum flexibility! --Larry Garfield --- ## #127019 — Derick Rethans — 2025-04-01T20:17:05+00:00 URL: https://externals.io/message/126989#127019 · In reply to: #127017 > > > […] > >Naturally, the degree of closeness for strings or for floats should be controlled by an ini setting. Maximum flexibility! > >--Larry Garfield You got to be joking! Everybody knows ini settings make things unportable. I suggest we introduce AI to determine the closeness instead. cheers Derick --- ## #127020 — Rob Landers — 2025-04-01T20:25:10+00:00 URL: https://externals.io/message/126989#127020 · In reply to: #127019 > > > […] > > You got to be joking! Everybody knows ini settings make things unportable. I suggest we introduce AI to determine the closeness instead. > > cheers > Derick > We have to be careful not to tie ourselves to a specific AI model. But we can use ini settings to allow the user to specify which model to use. — Rob --- ## #127021 — Iliya Miroslavov Iliev — 2025-04-01T21:33:23+00:00 URL: https://externals.io/message/126989#127021 · In reply to: #127020 > > > On 1 April 2025 20:52:32 BST, Larry Garfield > wrote: > > […] > enough"! > > […] > controlled by an ini setting. Maximum flexibility! > > […] > > You got to be joking! Everybody knows ini settings make things unportable. > I suggest we introduce AI to determine the closeness instead. > > cheers > Derick > > > We have to be careful not to tie ourselves to a specific AI model. But we > can use ini settings to allow the user to specify which model to use. > > — Rob > Rob, I don't trust users. This must be hardcoded inside PHP and we can build and support 100 different versions just to be sure no user makes mistakes. No hassle here -- Iliya Miroslavov Iliev i.miroslavov@gmail.com --- ## #127023 — mickmackusa — 2025-04-02T06:11:17+00:00 URL: https://externals.io/message/126989#127023 · In reply to: #127021 I love April Fools jokes. Yes, and let's add the null coalescing increment and decrement operators: $foo??++ and $bar??-- . I want to ensure that an undeclared or null variable defaults to 0 before becoming 1 or -1 without needing to manually instantiate the variable as 0. Pfft. --- Anyhow, why wouldn't rounding precision be decided by the precision of the float/integer with the lesser precision? 1. 1.55 ~= 1.5499 // true (because rounding 1.5499 to 2 decimal places is 1.55) 2. 3.1415 ~= 3.141 // false (because rounding 3.1415 to three decimal places is 3.142) 3. 1000 ~= 999.50000001 // true (rounding to integer comparison) mickmackusa --- ## #127024 — Jordi Boggiano — 2025-04-02T07:33:27+00:00 URL: https://externals.io/message/126989#127024 · In reply to: #126989 > We live in an imperfect world, and we often approximate data, but neither `==` nor `===` are ideal comparison operators to deal with these kinds of data. I am late to the party here, but in all seriousness when I read the subject my initial thought was that this was gonna be about adding a ~= operator to clearly show intent (does the same as == though), allowing us to simultaneously deprecate ==. Then in next major we can make == mean strict equal, deprecate ===, and you're left with == or ~=, which seems cleaner to me as you know nobody typed ~= accidentally while they meant ===. Anyway, while (maybe..) a less goofy idea it is probably just as unlikely to make it through BC concerns. Best, Jordi