After over two decades of working in PHP, I'm now working in Java. PHP is basically Java-lite. I am absolutely loving the compile-time safety of Java, but I dearly miss PHP's maps and arrays. In Java, the amount of verbosity for defining a map/list and operating on it is overwhelming.
Modern PHP is great. Many powerful language features, excellent performance, great community and package ecosystem, and decent enough safety with modern static analysis tools.
I'm not too sure I agree with the author's complaints here. When using something like array_filter, you're typically mapping from collection to collection (i.e. you don't care about the first element--you care about the whole thing) and so this problem is really a non-issue. The next follow up step would usually be foreach, or another operation like array_map, in which case it's a non-issue.
If you really do need the first element, you can use array_first. And if you really do need a fixed-sized collection, you can use SplFixedArray.
The point on properties is valid to an extent, but IMO not really an issue you commonly run into in the real world (regardless of language, your constructors should generally return an object in a usable state).
If I have an “array” and can do array[0] to get first item, but when I filter this array and array[0] throws an error, that’s super weird. What is the meaning of [] or what is an array even? The language forces me to understand how it is implemented under the hood. That’s exactly what the author says: leaky abstraction.
All PHP needs is python's flexible and convenient manipulation of lists/dicts/objects. Plus dropping the end semi colon and it would riiiiiiiiiiiiiiiip the fabric of spacetime.
> Modern PHP is great. Many powerful language features, excellent performance, great community and package ecosystem
I heard this a long time ago about perl. CPAN is great.
Well ... perl entered the fossilized era. I think people do not really observe things correctly. I am noticing the same with ruby right now - everyone sees that ruby is in decline, very strongly so, in the last 3 years. Yet you have blog posts such as "ruby is not dying - it is aging like fine wine". And these are all NOT BASED ON FACTUAL ANALYSIS. I still think ruby is a great language, but if people are not realistic in their assessment of a situation, what does this tell us about people's evaluation in general? People seem to shy away from criticism. You can see this on reddit too, where moderators ban and censor willy-nilly, or even on github, where you can also quickly get eliminated for not conforming to xyz. It's as if some people are very afraid of strong opinions. I don't understand why - an opinion that is objectively false, can be shown to be false.
> This lax behaviour for property definitions makes writing code around them harder. Especially when you take into account that any object can have properties dynamically added to them:
Doing so now raises a deprecation warning, unless you add #[AllowDynamicProperties], and PHP 9 will convert it to an error. I'm told this will simplify internals and unlock optimizations.
Arrays are still fairly awful, but generics may become a reality sooner rather than later, and on that could be built Vec and Dict types, à la Hack. PHP is going to be stuck with arrays as they are now for forever, but they'll at least become optional for new code.
> Not a warning—a FATAL error occurs if you try to access an uninitialized property. This comes up a lot in cases where you try to deserialize data into a PHP object. If a field's data isn't present you might not initialize the property at all.
I don't think that is an issue, except in interpretet type-unsafe languages it is harder to anticipate when writing code whether that value is NULL or undefined/uninitialized. E.g. it is basically the same in C#, but here the compiler warns you that the value is not initialized and forbids some actions (like reading the value of it).
It may work in your IDE (autocompletion, etc.) but there is no standard on this side. Some IDE have their own parsers, others use one of the LSPs for PHP.
PHP has quite a lot of oddities such as how loose comparisons (`==`) are made, numeric-strings, and type coercion. But the two oddities mentioned in the article are not that "odd" with a bit of context.
- PHP has `SplFixedArray`[^1] that work similar to the standard arrays you expect from other languages. SPL extension is always available in PHP 5.3+, it is not even possible to compile PHP without it anymore. There is no specific type for list-arrays and associative arrays, but there is an `array_is_list` function to quickly check it.
- For typed properties, if a property is not typed, it is effectively considered `mixed $var = null`. If the property is typed, and has no default value, then it is considered uninitialized, and not allowed to access.
Arrays in PHP are mostly confusing if you are used to them in other languages.
Instead of $array[0] - the first element is accessed via array_first().
Having them as key-value means, that you can easily just remove some items in the middle, during iteration etc. No automatic shifting happening.
The thing which bites me is when some internal functions actually reindex the array. array_filter does not, but for example array_reverse, array_slice etc. do (preserve_keys always defaults to false). And for array_merge too, but there it's no array_merge(preserve_keys: false), but instead the + operator. (Why is this operator overloaded?!)
On the topic of the uninitialized state, as co-author of that RFC:
I agree with the author that nullable properties should have been auto-initialized to NULL. I haven't ever seen any benefit of an uninitialized state for these. Some co-authors of that RFC disagreed and wished for consistency with the other typed properties. The good thing probably is, that we still could opt to change this with a relatively minor BC break.
For non-nullable properties, I do think there is value. Not every value is actually available/ready in a constructor. Sure you can assign dummy values to properties. But it's requiring you to then manually guard/assert that the property is actually initialized. If you happen to access a non-nullable typed property without isset(), then your code is likely broken anyway and I'm grateful for the Error exception thrown.
Also, PHP has this peculiar feature of ReflectionClass::newInstanceWithoutConstructor(). This is forcibly having an object in an uninitialized state. Whether that feature should exist or not is a good question, but in practice it's helpful for object hydration for example. This was one further motivation to introduce the uninitialized state.
The author of the post suggests checking at the constructor boundary. But this doesn't inhibit objects leaking / not finishing the initialization properly. (class Foo { public stdClass $object; function __construct() { global $foo; $foo = $this; } } new Foo; $foo->object ... is now still existing? PHP doesn't have mechanisms to invalidate objects at a distance. That would be the alternative, but also spooky.) Some choices need to be made, and all choices will have some rough edges.
Side note: I personally never use is_null(), but nearly always isset(). This nicely checks for the uninitialized state too. Static analysis tells me anyway, when I access a variable or property name which can never exist.
There's one problem with arrays that I haven't seen mentioned here or by the OP: when inserting a key-value, the type of the key may change.
For instance ["4" => "four"] === [4 => "Four"]
This can lead to some unexpected behaviors. For example, I've already been bitten by `array_merge()` whose result is different if its parameters are arrays with numeric indexes.
To note, it is surprisingly refreshing to completely forgo instanciable classes on a modern codebase.
Phpstan deals well with type definitions, arrays are powerful enough to contain whatever needed, and functions can be stored and passed around easily enough.
Array shapes are still second-class citizens defined in phpdoc, with an inferior editor UX, and lack of any run-time enforcement. A proper record type for PHP with value semantics would be an ideal solution for me. Would go nicely with the pattern matching proposal that's still incubating.
I’d love to see a post like this for JS that actually talks about things people run into. Usually when people make a post like this in js, its about archaic things nobody actually uses.
For me, it's the exact opposite. Every time I work with PHP, I wish I could have TypeScript's properly typed arrays and dictionaries instead of the janky untyped 2-in-1 mess it actually has.
I like the arrays, they make it feel like a bit of a low level language and there are some weird tricks one can do with them. As for class property behaviour, I kind of get nervous if my properties or the constructor don't enforce default values I can then fit into downstream logic.
One of the best things with PHP is PsySH, or "Tinker" as the laravelists call it. It's not a REPL in the Common Lisp sense, but it is quite nice for an interactive programming shell. I've spent countless hours solving problems very, very quickly in it, and alongside Picolisp pil + and Elixir iex it's one of the earliest tools I install on a new system.
The thing I miss the most is a nice concurrency story. It has become better but it's still a bit of a mess, often it's nicest to just implement workers as PHP and then implement control somewhere else, e.g. Elixir, or grab one of the application servers that are nowadays a thing in PHP.
> This example exposes the "uninitialized" state that a property can be in, which is NOT the same as NULL. This distinction frustratingly comes up when you try to do a null check on these properties:
If you're accessing an uninitialized property or checking if a property is uninitialized, you're probably already doing something wrong.
The point of class properties with no default value is that you're supposed to set them either in the constructor, immediately after creating an instance, or via some other method that guarantees they'll have a value by the time you need to read them (such as deserialization with validation).
If you want your properties to have a default "unset" value that you can trivially check for, that's what null is for. The author doesn't make it clear whether they are aware that you can declare a nullable string and give it the default value of null, but I hope they are.
I've written PHP off and on since the .php3 extension was a thing, and I can say that PHP very much deserved the bad rap it had for some time. It's evolved beyond most of that, but a lot of that is due to the composer ecosystem making up for it while the behavior of many builtins remains beyond repair. Which is fine, every language has baggage and warts. PHP's warts are sometimes heinously ugly, and they're on full display in many legacy codebases, but modern PHP is something I actually find to be fairly pleasant to develop with, far more than Go or vanilla JavaScript.
A lot of it came from the rather harmful "php a fractal of bad design" article that used to get posted everywhere despite being highly inaccurate and out of date. Thankfully its fairly rare to see someone daft enough to still try using it in a discussion. PHP's come a very long way since then.
Modern PHP is pretty good. I’ve been using it for a while so it’s comfortable.
I’ve been doing so Perl maintenance lately and I miss PHP. Perl is a lot weirder than PHP. If I didn’t know C or had dabbled in Perl before I would be completely confused. There is More Than One Way to do it (the Perl rallying cry) causes a lot of confusion. The one nice thing about Perl is that it doesn’t really change anymore, and you can see where it positively influenced others.
After over two decades of working in PHP, I'm now working in Java. PHP is basically Java-lite. I am absolutely loving the compile-time safety of Java, but I dearly miss PHP's maps and arrays. In Java, the amount of verbosity for defining a map/list and operating on it is overwhelming.
Modern PHP is great. Many powerful language features, excellent performance, great community and package ecosystem, and decent enough safety with modern static analysis tools.
I'm not too sure I agree with the author's complaints here. When using something like array_filter, you're typically mapping from collection to collection (i.e. you don't care about the first element--you care about the whole thing) and so this problem is really a non-issue. The next follow up step would usually be foreach, or another operation like array_map, in which case it's a non-issue.
If you really do need the first element, you can use array_first. And if you really do need a fixed-sized collection, you can use SplFixedArray.
The point on properties is valid to an extent, but IMO not really an issue you commonly run into in the real world (regardless of language, your constructors should generally return an object in a usable state).
If I have an “array” and can do array[0] to get first item, but when I filter this array and array[0] throws an error, that’s super weird. What is the meaning of [] or what is an array even? The language forces me to understand how it is implemented under the hood. That’s exactly what the author says: leaky abstraction.
All PHP needs is python's flexible and convenient manipulation of lists/dicts/objects. Plus dropping the end semi colon and it would riiiiiiiiiiiiiiiip the fabric of spacetime.
Have you tried Lua or one of its variants?
> Modern PHP is great. Many powerful language features, excellent performance, great community and package ecosystem
I heard this a long time ago about perl. CPAN is great.
Well ... perl entered the fossilized era. I think people do not really observe things correctly. I am noticing the same with ruby right now - everyone sees that ruby is in decline, very strongly so, in the last 3 years. Yet you have blog posts such as "ruby is not dying - it is aging like fine wine". And these are all NOT BASED ON FACTUAL ANALYSIS. I still think ruby is a great language, but if people are not realistic in their assessment of a situation, what does this tell us about people's evaluation in general? People seem to shy away from criticism. You can see this on reddit too, where moderators ban and censor willy-nilly, or even on github, where you can also quickly get eliminated for not conforming to xyz. It's as if some people are very afraid of strong opinions. I don't understand why - an opinion that is objectively false, can be shown to be false.
> This lax behaviour for property definitions makes writing code around them harder. Especially when you take into account that any object can have properties dynamically added to them:
Doing so now raises a deprecation warning, unless you add #[AllowDynamicProperties], and PHP 9 will convert it to an error. I'm told this will simplify internals and unlock optimizations.
Arrays are still fairly awful, but generics may become a reality sooner rather than later, and on that could be built Vec and Dict types, à la Hack. PHP is going to be stuck with arrays as they are now for forever, but they'll at least become optional for new code.
About undefined vs. NULL:
> Not a warning—a FATAL error occurs if you try to access an uninitialized property. This comes up a lot in cases where you try to deserialize data into a PHP object. If a field's data isn't present you might not initialize the property at all.
I don't think that is an issue, except in interpretet type-unsafe languages it is harder to anticipate when writing code whether that value is NULL or undefined/uninitialized. E.g. it is basically the same in C#, but here the compiler warns you that the value is not initialized and forbids some actions (like reading the value of it).
I've been writing PHP for 20 years now. It's my bread and butter.
The one thing I really wish PHP would add is structurally typed objects. I really miss it when moving back and forth between PHP and TypeScript.
They could call them anonymous objects if they want to (that would be a more culturally correct analogue to anonymous classes).
Like, I wish it was possible to do
and have it be equivalent to and then be able to typehint it like and honestly, why not go all the way and allow type definitions/aliases, something like That'd be great.You can do that. Of course, PHP's native types are quite limited, but a phpdoc syntax should work with static analysis tools. For instance:
Here are some documentation and examples:- For Psalm, see https://psalm.dev/docs/annotating_code/type_syntax/utility_t...
- For PHPstan, see https://phpstan.org/writing-php-code/phpdoc-types
It may work in your IDE (autocompletion, etc.) but there is no standard on this side. Some IDE have their own parsers, others use one of the LSPs for PHP.
PHP has quite a lot of oddities such as how loose comparisons (`==`) are made, numeric-strings, and type coercion. But the two oddities mentioned in the article are not that "odd" with a bit of context.
- PHP has `SplFixedArray`[^1] that work similar to the standard arrays you expect from other languages. SPL extension is always available in PHP 5.3+, it is not even possible to compile PHP without it anymore. There is no specific type for list-arrays and associative arrays, but there is an `array_is_list` function to quickly check it.
- For typed properties, if a property is not typed, it is effectively considered `mixed $var = null`. If the property is typed, and has no default value, then it is considered uninitialized, and not allowed to access.
[^1]: https://www.php.net/manual/en/class.splfixedarray.php
I feel like somebody needs to be reminded of “PHP: a fractal of bad design” :D
https://eev.ee/blog/2012/04/09/php-a-fractal-of-bad-design/
Good old empty() and isset() work on uninitialized properties though.
The last part about "$" is epic.
petition to use Euro symbol when coding PHP in EU.
Arrays in PHP are mostly confusing if you are used to them in other languages. Instead of $array[0] - the first element is accessed via array_first().
Having them as key-value means, that you can easily just remove some items in the middle, during iteration etc. No automatic shifting happening.
The thing which bites me is when some internal functions actually reindex the array. array_filter does not, but for example array_reverse, array_slice etc. do (preserve_keys always defaults to false). And for array_merge too, but there it's no array_merge(preserve_keys: false), but instead the + operator. (Why is this operator overloaded?!)
On the topic of the uninitialized state, as co-author of that RFC:
I agree with the author that nullable properties should have been auto-initialized to NULL. I haven't ever seen any benefit of an uninitialized state for these. Some co-authors of that RFC disagreed and wished for consistency with the other typed properties. The good thing probably is, that we still could opt to change this with a relatively minor BC break.
For non-nullable properties, I do think there is value. Not every value is actually available/ready in a constructor. Sure you can assign dummy values to properties. But it's requiring you to then manually guard/assert that the property is actually initialized. If you happen to access a non-nullable typed property without isset(), then your code is likely broken anyway and I'm grateful for the Error exception thrown.
Also, PHP has this peculiar feature of ReflectionClass::newInstanceWithoutConstructor(). This is forcibly having an object in an uninitialized state. Whether that feature should exist or not is a good question, but in practice it's helpful for object hydration for example. This was one further motivation to introduce the uninitialized state.
The author of the post suggests checking at the constructor boundary. But this doesn't inhibit objects leaking / not finishing the initialization properly. (class Foo { public stdClass $object; function __construct() { global $foo; $foo = $this; } } new Foo; $foo->object ... is now still existing? PHP doesn't have mechanisms to invalidate objects at a distance. That would be the alternative, but also spooky.) Some choices need to be made, and all choices will have some rough edges.
Side note: I personally never use is_null(), but nearly always isset(). This nicely checks for the uninitialized state too. Static analysis tells me anyway, when I access a variable or property name which can never exist.
There's one problem with arrays that I haven't seen mentioned here or by the OP: when inserting a key-value, the type of the key may change. For instance ["4" => "four"] === [4 => "Four"]
This can lead to some unexpected behaviors. For example, I've already been bitten by `array_merge()` whose result is different if its parameters are arrays with numeric indexes.
surprised not to see http://phpsadness.com/ here
How could you leave out left-associative ternary operators?
To note, it is surprisingly refreshing to completely forgo instanciable classes on a modern codebase.
Phpstan deals well with type definitions, arrays are powerful enough to contain whatever needed, and functions can be stored and passed around easily enough.
Array shapes are still second-class citizens defined in phpdoc, with an inferior editor UX, and lack of any run-time enforcement. A proper record type for PHP with value semantics would be an ideal solution for me. Would go nicely with the pattern matching proposal that's still incubating.
I’d love to see a post like this for JS that actually talks about things people run into. Usually when people make a post like this in js, its about archaic things nobody actually uses.
https://www.destroyallsoftware.com/talks/wat
Archaic maybe, but still brings a smile
Every time I work in another language I miss PHP’s arrays.
Basically every other language has the same functionality (or better) as a hashmap.
That array keys are auto-coerced to integers has bit me multiple times.
They are an incredibly versatile tool for sure. Even more so wrapped in a Laravel Collection
For me, it's the exact opposite. Every time I work with PHP, I wish I could have TypeScript's properly typed arrays and dictionaries instead of the janky untyped 2-in-1 mess it actually has.
Absolutely. If you don’t know PHP arrays aren’t actually arrays, the other languages feel inferior.
if("0") {} being equivalent to if(false) {} still gives me nightmares even though I've stopped using PHP for at least 6 years now :)
I knew this in and out, but as a Full-Stack PHP/Symfony/Frontend/JS guy who pivoted to mainly TS for b2b stuff, I still have to occasionally enter
into the browser console just to be sure, during code reviews :DAlso `empty("0") === true` is a common gotcha.
I like the arrays, they make it feel like a bit of a low level language and there are some weird tricks one can do with them. As for class property behaviour, I kind of get nervous if my properties or the constructor don't enforce default values I can then fit into downstream logic.
One of the best things with PHP is PsySH, or "Tinker" as the laravelists call it. It's not a REPL in the Common Lisp sense, but it is quite nice for an interactive programming shell. I've spent countless hours solving problems very, very quickly in it, and alongside Picolisp pil + and Elixir iex it's one of the earliest tools I install on a new system.
https://psysh.org/
The thing I miss the most is a nice concurrency story. It has become better but it's still a bit of a mess, often it's nicest to just implement workers as PHP and then implement control somewhere else, e.g. Elixir, or grab one of the application servers that are nowadays a thing in PHP.
> This example exposes the "uninitialized" state that a property can be in, which is NOT the same as NULL. This distinction frustratingly comes up when you try to do a null check on these properties:
If you're accessing an uninitialized property or checking if a property is uninitialized, you're probably already doing something wrong.
The point of class properties with no default value is that you're supposed to set them either in the constructor, immediately after creating an instance, or via some other method that guarantees they'll have a value by the time you need to read them (such as deserialization with validation).
If you want your properties to have a default "unset" value that you can trivially check for, that's what null is for. The author doesn't make it clear whether they are aware that you can declare a nullable string and give it the default value of null, but I hope they are.
I think the “bad rep” is coming from developers that stopped developing themselves.
I've written PHP off and on since the .php3 extension was a thing, and I can say that PHP very much deserved the bad rap it had for some time. It's evolved beyond most of that, but a lot of that is due to the composer ecosystem making up for it while the behavior of many builtins remains beyond repair. Which is fine, every language has baggage and warts. PHP's warts are sometimes heinously ugly, and they're on full display in many legacy codebases, but modern PHP is something I actually find to be fairly pleasant to develop with, far more than Go or vanilla JavaScript.
A lot of it came from the rather harmful "php a fractal of bad design" article that used to get posted everywhere despite being highly inaccurate and out of date. Thankfully its fairly rare to see someone daft enough to still try using it in a discussion. PHP's come a very long way since then.
No. Bad rep comes from developers who used other languages and never looked back.
I did PHP for 15 years. Modern PHP looks good but I still wouldn't go back.
All of PHP is an oddity. It is a practical oddity, but also ugly to no ends. I am glad to have abandoned it many years ago.
Surprisingly enough, I was more productive in PHP than I was in perl. Perhaps perl is even stranger than PHP.
Modern PHP is pretty good. I’ve been using it for a while so it’s comfortable.
I’ve been doing so Perl maintenance lately and I miss PHP. Perl is a lot weirder than PHP. If I didn’t know C or had dabbled in Perl before I would be completely confused. There is More Than One Way to do it (the Perl rallying cry) causes a lot of confusion. The one nice thing about Perl is that it doesn’t really change anymore, and you can see where it positively influenced others.
PHP is fundamentally broken, and unfixable. Its a toy language, and there is no reason to pick it in 2026