2020-12-25 11:31:48 +03:00
# NEWS for Ruby 3.1.0
2020-01-15 06:42:47 +03:00
2021-12-15 19:24:31 +03:00
This document is a list of user-visible feature changes
2020-12-25 11:31:48 +03:00
since the **3.0.0** release, except for bug fixes.
2020-01-15 06:42:47 +03:00
2020-12-22 22:27:38 +03:00
Note that each entry is kept to a minimum, see links for details.
2020-01-15 06:42:47 +03:00
2020-01-18 11:46:52 +03:00
## Language changes
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-12-15 19:24:31 +03:00
* The block argument can now be anonymous if the block will
2021-12-05 13:08:25 +03:00
only be passed to another method. [[Feature #11256 ]]
2021-11-18 23:44:19 +03:00
```ruby
def foo(& )
bar(& )
end
```
2021-12-05 13:08:25 +03:00
* Pin operator now takes an expression. [[Feature #17411 ]]
2021-03-21 09:12:54 +03:00
```ruby
Prime.each_cons(2).lazy.find_all{_1 in [n, ^(n + 2)]}.take(3).to_a
#=> [[3, 5], [5, 7], [11, 13]]
```
2020-01-15 06:42:47 +03:00
2021-12-05 13:08:25 +03:00
* Pin operator now supports instance, class, and global variables.
[[Feature #17724 ]]
2021-05-14 01:31:46 +03:00
2021-07-16 05:23:36 +03:00
```ruby
2021-05-14 01:31:46 +03:00
@n = 5
Prime.each_cons(2).lazy.find{_1 in [n, ^@n]}
#=> [3, 5]
2021-07-16 05:23:36 +03:00
```
2021-05-14 01:31:46 +03:00
2021-12-05 13:08:25 +03:00
* One-line pattern matching is no longer experimental.
2021-07-17 05:13:52 +03:00
2021-12-19 15:37:08 +03:00
* Parentheses can be omitted in one-line pattern matching.
[[Feature #16182 ]]
```ruby
[0, 1] => _, x
{y: 2} => y:
x #=> 1
y #=> 2
```
2021-12-05 13:08:25 +03:00
* Multiple assignment evaluation order has been made consistent with
single assignment evaluation order. With single assignment, Ruby
uses a left-to-right evaluation order. With this code:
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
```ruby
foo[0] = bar
```
2021-09-11 13:18:22 +03:00
The following evaluation order is used:
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-09-11 13:18:22 +03:00
1. `foo`
2. `bar`
3. `[]=` called on the result of `foo`
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-09-11 13:18:22 +03:00
In Ruby before 3.1.0, multiple assignment did not follow this
evaluation order. With this code:
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-12-05 11:15:45 +03:00
```ruby
foo[0], bar.baz = a, b
```
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-09-11 13:18:22 +03:00
Versions of Ruby before 3.1.0 would evaluate in the following
order
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-09-11 13:18:22 +03:00
1. `a`
2. `b`
3. `foo`
4. `[]=` called on the result of `foo`
5. `bar`
6. `baz=` called on the result of `bar`
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-12-15 19:24:31 +03:00
Starting in Ruby 3.1.0, the evaluation order is now consistent with
single assignment, with the left-hand side being evaluated before
the right-hand side:
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-09-11 13:18:22 +03:00
1. `foo`
2. `bar`
3. `a`
4. `b`
5. `[]=` called on the result of `foo`
6. `baz=` called on the result of `bar`
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-09-11 13:18:22 +03:00
[[Bug #4443 ]]
Evaluate multiple assignment left hand side before right hand side
In regular assignment, Ruby evaluates the left hand side before
the right hand side. For example:
```ruby
foo[0] = bar
```
Calls `foo`, then `bar`, then `[]=` on the result of `foo`.
Previously, multiple assignment didn't work this way. If you did:
```ruby
abc.def, foo[0] = bar, baz
```
Ruby would previously call `bar`, then `baz`, then `abc`, then
`def=` on the result of `abc`, then `foo`, then `[]=` on the
result of `foo`.
This change makes multiple assignment similar to single assignment,
changing the evaluation order of the above multiple assignment code
to calling `abc`, then `foo`, then `bar`, then `baz`, then `def=` on
the result of `abc`, then `[]=` on the result of `foo`.
Implementing this is challenging with the stack-based virtual machine.
We need to keep track of all of the left hand side attribute setter
receivers and setter arguments, and then keep track of the stack level
while handling the assignment processing, so we can issue the
appropriate topn instructions to get the receiver. Here's an example
of how the multiple assignment is executed, showing the stack and
instructions:
```
self # putself
abc # send
abc, self # putself
abc, foo # send
abc, foo, 0 # putobject 0
abc, foo, 0, [bar, baz] # evaluate RHS
abc, foo, 0, [bar, baz], baz, bar # expandarray
abc, foo, 0, [bar, baz], baz, bar, abc # topn 5
abc, foo, 0, [bar, baz], baz, abc, bar # swap
abc, foo, 0, [bar, baz], baz, def= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz], baz, foo # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0 # topn 3
abc, foo, 0, [bar, baz], baz, foo, 0, baz # topn 2
abc, foo, 0, [bar, baz], baz, []= # send
abc, foo, 0, [bar, baz], baz # pop
abc, foo, 0, [bar, baz] # pop
[bar, baz], foo, 0, [bar, baz] # setn 3
[bar, baz], foo, 0 # pop
[bar, baz], foo # pop
[bar, baz] # pop
```
As multiple assignment must deal with splats, post args, and any level
of nesting, it gets quite a bit more complex than this in non-trivial
cases. To handle this, struct masgn_state is added to keep
track of the overall state of the mass assignment, which stores a linked
list of struct masgn_attrasgn, one for each assigned attribute.
This adds a new optimization that replaces a topn 1/pop instruction
combination with a single swap instruction for multiple assignment
to non-aref attributes.
This new approach isn't compatible with one of the optimizations
previously used, in the case where the multiple assignment return value
was not needed, there was no lhs splat, and one of the left hand side
used an attribute setter. This removes that optimization. Removing
the optimization allowed for removing the POP_ELEMENT and adjust_stack
functions.
This adds a benchmark to measure how much slower multiple
assignment is with the correct evaluation order.
This benchmark shows:
* 4-9% decrease for attribute sets
* 14-23% decrease for array member sets
* Basically same speed for local variable sets
Importantly, it shows no significant difference between the popped
(where return value of the multiple assignment is not needed) and
!popped (where return value of the multiple assignment is needed)
cases for attribute and array member sets. This indicates the
previous optimization, which was dropped in the evaluation
order fix and only affected the popped case, is not important to
performance.
Fixes [Bug #4443]
2021-04-21 20:49:19 +03:00
2021-12-05 13:08:25 +03:00
* Values in Hash literals and keyword arguments can be omitted.
[[Feature #14579 ]]
2021-09-16 11:31:39 +03:00
2021-12-05 13:08:25 +03:00
For example,
2021-09-11 13:13:53 +03:00
2021-09-16 11:28:31 +03:00
* `{x:, y:}` is a syntax sugar of `{x: x, y: y}` .
* `foo(x:, y:)` is a syntax sugar of `foo(x: x, y: y)` .
2021-12-05 13:08:25 +03:00
Constant names, local variable names, and method names are allowed as
key names. Note that a reserved word is considered as a local
variable or method name even if it's a pseudo variable name such as
`self` .
2021-09-11 13:13:53 +03:00
2021-12-15 19:24:31 +03:00
* Non main-Ractors can get instance variables (ivars) of classes/modules
2021-12-05 13:08:25 +03:00
if ivars refer to shareable objects.
[[Feature #17592 ]]
2021-10-22 19:38:05 +03:00
2021-12-15 19:24:31 +03:00
* A command syntax is allowed in endless method definitions, i.e.,
2021-12-13 08:01:40 +03:00
you can now write `def foo = puts "Hello"` .
Note that `private def foo = puts "Hello"` does not parse.
[[Feature #17398 ]]
2020-01-18 11:46:52 +03:00
## Command line options
2020-01-15 06:42:47 +03:00
2021-12-13 08:19:21 +03:00
* `--disable-gems` is now explicitly declared as "just for debugging".
2021-12-15 19:24:31 +03:00
Never use it in any real-world codebase.
2021-12-13 08:19:21 +03:00
[[Feature #17684 ]]
2020-03-03 11:37:51 +03:00
## Core classes updates
2021-12-15 19:24:31 +03:00
Note: We're only listing outstanding class updates.
2020-01-15 06:42:47 +03:00
2021-04-16 10:12:10 +03:00
* Array
* Array#intersect? is added. [[Feature #15198 ]]
2021-10-26 22:35:21 +03:00
* Class
2021-12-07 16:00:23 +03:00
* Class#subclasses, which returns an array of classes
directly inheriting from the receiver, not
including singleton classes.
[[Feature #18273 ]]
```ruby
class A; end
class B < A ; end
class C < B ; end
class D < A ; end
A.subclasses #=> [D, B]
B.subclasses #=> [C]
C.subclasses #=> []
```
2021-10-28 15:07:11 +03:00
2021-01-02 10:42:45 +03:00
* Enumerable
2021-12-07 16:00:23 +03:00
* Enumerable#compact is added. [[Feature #17312 ]]
2021-01-02 10:42:45 +03:00
2021-12-07 16:00:23 +03:00
* Enumerable#tally now accepts an optional hash to count. [[Feature #17744 ]]
2021-03-26 10:29:21 +03:00
2021-12-07 16:00:23 +03:00
* Enumerable#each_cons and each_slice to return a receiver. [[GH-1509]]
2021-10-28 12:21:27 +03:00
2021-12-07 16:00:23 +03:00
```ruby
[1, 2, 3].each_cons(2){}
# 3.0 => nil
# 3.1 => [1, 2, 3]
2021-10-28 12:21:27 +03:00
2021-12-07 16:00:23 +03:00
[1, 2, 3].each_slice(2){}
# 3.0 => nil
# 3.1 => [1, 2, 3]
```
2021-10-28 12:21:27 +03:00
2021-01-02 10:42:45 +03:00
* Enumerator::Lazy
2021-12-07 16:00:23 +03:00
* Enumerator::Lazy#compact is added. [[Feature #17312 ]]
2021-01-02 10:42:45 +03:00
2021-03-15 14:05:03 +03:00
* File
2021-12-07 16:00:23 +03:00
* File.dirname now accepts an optional argument for the level to
strip path components. [[Feature #12194 ]]
2021-03-15 14:05:03 +03:00
2021-12-06 11:36:01 +03:00
* GC
2021-12-07 16:00:23 +03:00
* "GC.measure_total_time = true" enables the measurement of GC.
Measurement can introduce overhead. It is enabled by default.
GC.measure_total_time returns the current setting.
GC.stat[:time] or GC.stat(:time) returns measured time
2021-12-21 20:49:19 +03:00
in milli-seconds. [[[Feature #10917 ]]]
2021-12-06 11:36:01 +03:00
2021-12-21 20:49:19 +03:00
* GC.total_time returns measured time in nano-seconds. [[[Feature #10917 ]]]
2021-12-06 11:36:01 +03:00
2018-10-07 07:02:46 +03:00
* Integer
2021-12-07 16:00:23 +03:00
* Integer.try_convert is added. [[Feature #15211 ]]
2018-10-07 07:02:46 +03:00
2021-11-18 21:51:14 +03:00
* Kernel
2021-12-07 16:00:23 +03:00
* Kernel#load now accepts a module as the second argument,
2021-12-15 19:24:31 +03:00
and will load the file using the given module as the
top-level module. [[Feature #6210 ]]
2021-11-18 21:51:14 +03:00
2021-12-13 11:45:29 +03:00
* Marshal
* Marshal.load now accepts a `freeze: true` option.
All returned objects are frozen except for `Class` and
`Module` instances. Strings are deduplicated. [[Feature #18148 ]]
2021-09-16 13:55:31 +03:00
* MatchData
2021-12-07 16:00:23 +03:00
* MatchData#match is added [[Feature #18172 ]]
2021-09-16 13:55:31 +03:00
2021-12-07 16:00:23 +03:00
* MatchData#match_length is added [[Feature #18172 ]]
2021-09-16 13:55:31 +03:00
2021-12-25 11:42:27 +03:00
* Method / UnboundMethod
2021-10-28 01:36:00 +03:00
2021-12-13 08:13:46 +03:00
* Method#public?, Method#private?, Method#protected?,
UnboundMethod#public?, UnboundMethod#private?,
UnboundMethod#protected? have been added. [[Feature #11689 ]]
2021-10-28 01:36:00 +03:00
2021-01-14 22:59:25 +03:00
* Module
2021-12-07 16:00:23 +03:00
* Module#prepend now modifies the ancestor chain if the receiver
already includes the argument. Module#prepend still does not
modify the ancestor chain if the receiver has already prepended
the argument. [[Bug #17423 ]]
2021-01-14 22:59:25 +03:00
2021-12-07 16:00:23 +03:00
* Module#private, #public , #protected , and #module_function will
now return their arguments. If a single argument is given, it
is returned. If no arguments are given, nil is returned. If
multiple arguments are given, they are returned as an array.
[[Feature #12495 ]]
2021-11-18 21:51:14 +03:00
2021-11-08 21:38:42 +03:00
* Process
2021-12-07 16:00:23 +03:00
* Process.\_fork is added. This is a core method for fork(2).
Do not call this method directly; it is called by existing
fork methods: Kernel.#fork, Process.fork, and IO.popen("-").
2021-12-15 19:24:31 +03:00
Application monitoring libraries can overwrite this method to
hook fork events. [[Feature #17795 ]]
2021-11-08 21:38:42 +03:00
2021-01-17 12:35:54 +03:00
* Struct
2021-12-07 16:00:23 +03:00
* Passing only keyword arguments to Struct#initialize is warned.
You need to use a Hash literal to set a Hash to a first member.
[[Feature #16806 ]]
2021-01-17 12:35:54 +03:00
2021-12-07 16:00:23 +03:00
* StructClass#keyword_init? is added [[Feature #18008 ]]
2021-07-15 12:21:49 +03:00
2021-08-17 11:07:01 +03:00
* String
2021-12-07 16:00:23 +03:00
* Update Unicode version to 13.0.0 [[Feature #17750 ]]
and Emoji version to 13.0 [[Feature #18029 ]]
2021-08-17 11:07:01 +03:00
2021-12-15 19:24:31 +03:00
* String#unpack and String#unpack1 now accept an `offset:` keyword
2021-12-13 11:45:29 +03:00
argument to start the unpacking after an arbitrary number of bytes
have been skipped. If `offset` is outside of the string bounds
`ArgumentError` is raised. [[Feature #18254 ]]
2021-05-26 09:18:16 +03:00
* Thread
2021-12-07 16:00:23 +03:00
* Thread#native_thread_id is added. [[Feature #17853 ]]
2021-05-26 09:18:16 +03:00
2021-02-15 10:06:43 +03:00
* Thread::Backtrace
2021-12-07 16:00:23 +03:00
* Thread::Backtrace.limit, which returns the value to limit backtrace
length set by `--backtrace-limit` command line option, is added.
[[Feature #17479 ]]
2021-02-15 10:06:43 +03:00
2021-12-25 11:44:23 +03:00
* Thread::Queue
* Thread::Queue.new now accepts an Enumerable of initial values.
[[Feature #17327 ]]
2021-12-13 08:19:31 +03:00
* Time
* Time.new now accepts optional `in:` keyword argument for the
timezone, as well as `Time.at` and `Time.now` , so that is now
you can omit minor arguments to `Time.new` . [[Feature #17485 ]]
2021-12-25 07:14:54 +03:00
```ruby
Time.new(2021, 12, 25, in: "+07:00")
#=> 2021-12-25 00:00:00 +0700
```
At the same time, time component strings are converted to
integers more strictly now.
```ruby
2021-12-25 12:24:39 +03:00
Time.new(2021, 12, 25, "+07:30")
#=> invalid value for Integer(): "+07:30" (ArgumentError)
2021-12-25 07:14:54 +03:00
```
2021-12-25 12:24:39 +03:00
Ruby 3.0 or earlier returned probably unexpected result
`2021-12-25 07:00:00` , not `2021-12-25 07:30:00` nor
`2021-12-25 00:00:00 +07:30` .
2021-12-13 08:28:59 +03:00
* Time#strftime supports RFC 3339 UTC for unknown offset local
time, `-0000` , as `%-z` . [[Feature #17544 ]]
2021-12-13 08:02:07 +03:00
* TracePoint
2021-12-21 11:44:48 +03:00
* TracePoint.allow_reentry is added to allow reenter while TracePoint
2021-12-13 08:02:07 +03:00
callback.
2021-12-13 15:20:12 +03:00
[[Feature #15912 ]]
2021-12-13 08:02:07 +03:00
2021-02-16 12:12:47 +03:00
* $LOAD_PATH
2021-12-07 16:00:23 +03:00
* $LOAD_PATH.resolve_feature_path does not raise. [[Feature #16043 ]]
2021-02-16 12:12:47 +03:00
2021-06-22 14:14:45 +03:00
* Fiber Scheduler
2021-12-07 16:00:23 +03:00
* Add support for `Addrinfo.getaddrinfo` using `address_resolve` hook.
[[Feature #17370 ]]
2021-06-22 14:14:45 +03:00
2021-12-07 16:00:23 +03:00
* Introduce non-blocking `Timeout.timeout` using `timeout_after` hook.
[[Feature #17470 ]]
2021-06-22 14:14:45 +03:00
2021-12-07 16:00:23 +03:00
* Introduce new scheduler hooks `io_read` and `io_write` along with a
low level `IO::Buffer` for zero-copy read/write. [[Feature #18020 ]]
2021-07-02 13:41:16 +03:00
2021-12-07 16:00:23 +03:00
* IO hooks `io_wait` , `io_read` , `io_write` , receive the original IO object
where possible. [[Bug #18003 ]]
2021-06-22 14:14:45 +03:00
2021-12-07 16:00:23 +03:00
* Make `Monitor` fiber-safe. [[Bug #17827 ]]
2021-06-22 14:14:45 +03:00
2021-12-07 16:00:23 +03:00
* Replace copy coroutine with pthread implementation. [[Feature #18015 ]]
2021-06-26 01:17:26 +03:00
2021-10-21 10:21:08 +03:00
* Refinement
2021-12-07 16:00:23 +03:00
* New class which represents a module created by Module#refine.
`include` and `prepend` are deprecated, and `import_methods` is added
instead. [[Bug #17429 ]]
2021-10-21 10:21:08 +03:00
2020-03-03 11:37:51 +03:00
## Stdlib updates
2021-12-05 13:08:25 +03:00
* The following default gem are updated.
2021-12-25 01:45:36 +03:00
* RubyGems 3.3.3
2021-12-15 12:00:47 +03:00
* base64 0.1.1
* benchmark 0.2.0
2021-12-23 20:30:04 +03:00
* bigdecimal 3.1.1
2021-12-25 01:45:36 +03:00
* bundler 2.3.3
2021-12-15 12:00:47 +03:00
* cgi 0.3.1
2021-12-24 08:41:11 +03:00
* csv 3.2.2
2021-12-15 12:00:47 +03:00
* date 3.2.2
2021-12-24 17:21:17 +03:00
* did_you_mean 1.6.1
2021-12-24 14:10:13 +03:00
* digest 3.1.0
2021-12-15 12:00:47 +03:00
* drb 2.1.0
* erb 2.2.3
2021-12-23 07:01:26 +03:00
* error_highlight 0.3.0
2021-12-15 12:00:47 +03:00
* etc 1.3.0
* fcntl 1.0.1
* fiddle 1.1.0
* fileutils 1.6.0
* find 0.1.1
2021-12-25 12:43:37 +03:00
* io-console 0.5.10
2021-12-15 12:00:47 +03:00
* io-wait 0.2.1
* ipaddr 1.2.3
2021-12-25 07:13:21 +03:00
* irb 1.4.1
2021-12-15 12:00:47 +03:00
* json 2.6.1
2021-12-20 12:02:19 +03:00
* logger 1.5.0
2021-12-15 12:00:47 +03:00
* net-http 0.2.0
* net-protocol 0.1.2
* nkf 0.1.1
* open-uri 0.2.0
2021-12-24 21:38:28 +03:00
* openssl 3.0.0
2021-12-15 12:00:47 +03:00
* optparse 0.2.0
* ostruct 0.5.2
* pathname 0.2.0
2021-12-20 06:46:11 +03:00
* pp 0.3.0
2021-12-15 12:00:47 +03:00
* prettyprint 0.1.1
2021-12-20 12:26:42 +03:00
* psych 4.0.3
2021-12-15 12:00:47 +03:00
* racc 1.6.0
2021-12-24 11:12:56 +03:00
* rdoc 6.4.0
2021-12-24 12:04:07 +03:00
* readline 0.0.3
* readline-ext 0.1.4
2021-12-25 05:04:23 +03:00
* reline 0.3.0
2021-12-15 12:00:47 +03:00
* resolv 0.2.1
* rinda 0.1.1
* ruby2_keywords 0.0.5
* securerandom 0.1.1
* set 1.0.2
* stringio 3.0.1
* strscan 3.0.1
* tempfile 0.1.2
* time 0.2.0
* timeout 0.2.0
* tmpdir 0.1.2
* un 0.2.0
* uri 0.11.0
* yaml 0.2.0
* zlib 2.1.1
2021-12-05 13:08:25 +03:00
* The following bundled gems are updated.
2021-12-15 12:53:16 +03:00
* minitest 5.15.0
2021-12-05 13:08:25 +03:00
* power_assert 2.0.1
* rake 13.0.6
2021-12-20 10:05:44 +03:00
* test-unit 3.5.3
2021-12-17 19:18:25 +03:00
* rexml 3.2.5
2021-12-24 16:54:45 +03:00
* rbs 2.0.0
2021-12-23 11:48:32 +03:00
* typeprof 0.21.1
2021-12-05 13:08:25 +03:00
* The following default gems are now bundled gems.
2021-12-17 19:18:25 +03:00
* net-ftp 0.1.3
* net-imap 0.2.2
* net-pop 0.1.1
* net-smtp 0.3.1
* matrix 0.4.2
* prime 0.1.2
2021-12-17 22:21:56 +03:00
* debug 1.4.0
2020-01-15 06:42:47 +03:00
2021-12-17 14:34:14 +03:00
* Coverage measurement now supports suspension. You can use `Coverage.suspend`
2021-12-13 07:53:34 +03:00
to stop the measurement temporarily, and `Coverage.resume` to restart it.
See [[Feature #18176 ]] in detail.
2021-12-13 13:13:21 +03:00
* Random::Formatter is moved to random/formatter.rb, so that you can
2021-12-15 19:24:31 +03:00
use `Random#hex` , `Random#base64` , and so on without SecureRandom.
2021-12-13 13:13:21 +03:00
[[Feature #18190 ]]
2020-03-03 11:37:51 +03:00
## Compatibility issues
2021-12-15 19:24:31 +03:00
Note: Excluding feature bug fixes.
2020-01-15 06:42:47 +03:00
2021-06-22 14:14:45 +03:00
* `rb_io_wait_readable` , `rb_io_wait_writable` and `rb_wait_for_single_fd` are
deprecated in favour of `rb_io_maybe_wait_readable` ,
`rb_io_maybe_wait_writable` and `rb_io_maybe_wait` respectively.
`rb_thread_wait_fd` and `rb_thread_fd_writable` are deprecated. [[Bug #18003 ]]
2020-08-24 09:38:03 +03:00
## Stdlib compatibility issues
2020-06-18 15:11:19 +03:00
2021-01-21 09:02:24 +03:00
* `ERB#initialize` warns `safe_level` and later arguments even without -w.
[[Feature #14256 ]]
2021-10-20 04:18:54 +03:00
* `lib/debug.rb` is replaced with `debug.gem`
2021-12-07 06:07:28 +03:00
* `Kernel#pp` in `lib/pp.rb` uses the width of `IO#winsize` by default.
This means that the output width is automatically changed depending on
your terminal size. [[Feature #12913 ]]
2021-12-21 13:02:41 +03:00
* Psych 4.0 changes `Psych.load` as `safe_load` by the default.
You may need to use Psych 3.3.2 for migrating to this behavior.
[[Bug #17866 ]]
2020-01-18 11:46:52 +03:00
## C API updates
2020-01-15 06:42:47 +03:00
2021-09-22 11:59:28 +03:00
* Documented. [[GH-4815]]
2021-11-09 00:33:04 +03:00
* `rb_gc_force_recycle` is deprecated and has been changed to a no-op.
[[Feature #18290 ]]
2020-01-18 11:46:52 +03:00
## Implementation improvements
2020-01-15 06:42:47 +03:00
2021-12-24 18:01:21 +03:00
* Inline cache mechanism is introduced for reading class variables.
[[Feature #17763 ]]
2021-12-06 11:19:42 +03:00
* `instance_eval` and `instance_exec` now only allocate a singleton class when
required, avoiding extra objects and improving performance. [[GH-5146]]
* The performance of `Struct` accessors is improved. [[GH-5131]]
2021-12-06 20:03:06 +03:00
* `mandatory_only?` builtin special form to improve performance on
builtin methods. [[GH-5112]]
2021-12-13 17:46:34 +03:00
* Experimental feature Variable Width Allocation in the garbage collector.
2021-12-13 18:18:39 +03:00
This feature is turned off by default and can be enabled by compiling Ruby
with flag `USE_RVARGC=1` set. [[Feature #18045 ]] [[Feature #18239 ]]
2021-12-13 17:46:34 +03:00
2021-12-14 03:08:01 +03:00
## JIT
2020-06-26 10:54:05 +03:00
2021-12-14 03:08:01 +03:00
* Rename Ruby 3.0's `--jit` to `--mjit` , and alias `--jit` to `--yjit`
on non-Windows x86-64 platforms and to `--mjit` on others.
### MJIT
* The default `--mjit-max-cache` is changed from 100 to 10000.
2021-06-01 08:00:14 +03:00
2021-08-13 09:19:15 +03:00
* JIT-ed code is no longer cancelled when a TracePoint for class events
is enabled.
2021-06-03 08:25:48 +03:00
* The JIT compiler no longer skips compilation of methods longer than
1000 instructions.
2021-12-14 03:08:01 +03:00
* `--mjit-verbose` and `--mjit-warning` output "JIT cancel" when JIT-ed
2021-08-13 08:54:40 +03:00
code is disabled because TracePoint or GC.compact is used.
2021-10-20 22:51:07 +03:00
### YJIT: New experimental in-process JIT compiler
2021-10-21 19:49:58 +03:00
New JIT compiler available as an experimental feature. [[Feature #18229 ]]
See [this blog post](https://shopify.engineering/yjit-just-in-time-compiler-cruby
) introducing the project.
2021-10-20 22:51:07 +03:00
* Disabled by default, use `--yjit` command-line option to enable YJIT.
2021-12-24 19:59:57 +03:00
* Performance improvements on benchmarks based on real-world software,
up to 22% on railsbench, 39% on liquid-render.
2021-10-20 22:51:07 +03:00
* Fast warm-up times.
2021-12-24 20:35:24 +03:00
* Limited to Unix-like x86-64 platforms for now.
2021-10-20 22:51:07 +03:00
2020-11-10 20:15:57 +03:00
## Static analysis
2020-10-20 07:16:19 +03:00
### RBS
2020-09-25 11:08:39 +03:00
2021-12-24 16:54:45 +03:00
* Generics type parameters can be bounded ([PR](https://github.com/ruby/rbs/pull/844)).
```rbs
# `T` must be compatible with the `_Output` interface.
# `PrettyPrint[String]` is ok, but `PrettyPrint[Integer]` is a type error.
class PrettyPrint[T < _Output ]
interface _Output
def < < : ( String ) - > void
end
attr_reader output: T
def initialize: (T output) -> void
end
```
* Type aliases can be generic. ([PR](https://github.com/ruby/rbs/pull/823))
```rbs
# Defines a generic type `list` .
type list[T] = [ T, list[T] ]
| nil
type str_list = list[String]
type int_list = list[Integer]
```
2021-12-17 07:54:23 +03:00
* [rbs collection ](https://github.com/ruby/rbs/blob/master/docs/collection.md ) has been introduced to manage gems’
2021-12-24 16:54:45 +03:00
2021-12-17 07:46:09 +03:00
* Many signatures for built-in and standard libraries have been added/updated.
2021-12-24 16:54:45 +03:00
2021-12-17 07:46:09 +03:00
* It includes many bug fixes and performance improvements too.
See the [CHANGELOG.md ](https://github.com/ruby/rbs/blob/master/CHANGELOG.md ) for more information.
2020-10-20 07:16:19 +03:00
### TypeProf
2021-10-20 05:05:05 +03:00
* [Experimental IDE support ](https://github.com/ruby/typeprof/blob/master/doc/ide.md ) has been implemented.
* Many bug fixes and performance improvements since Ruby 3.0.0.
2021-10-20 04:18:54 +03:00
## Debugger
* A new debugger [debug.gem ](https://github.com/ruby/debug ) is bundled.
2021-12-15 19:24:31 +03:00
debug.gem is a fast debugger implementation, and it provides many features
like remote debugging, colorful REPL, IDE (VSCode) integration, and more.
2021-10-20 04:18:54 +03:00
It replaces `lib/debug.rb` standard library.
* `rdbg` command is also installed into `bin/` directory to start and control
debugging execution.
2021-10-20 05:05:05 +03:00
## error_highlight
2021-12-13 07:49:39 +03:00
A built-in gem called error_highlight has been introduced.
2021-12-15 19:24:31 +03:00
It shows fine-grained error locations in the backtrace.
2021-12-13 07:49:39 +03:00
Example: `title = json[:article][:title]`
If `json` is nil, it shows:
```
$ ruby test.rb
test.rb:2:in `<main>': undefined method ` []' for nil:NilClass (NoMethodError)
title = json[:article][:title]
^^^^^^^^^^
```
If `json[:article]` returns nil, it shows:
2021-10-20 05:05:05 +03:00
```
$ ruby test.rb
2021-12-13 07:49:39 +03:00
test.rb:2:in `<main>': undefined method ` []' for nil:NilClass (NoMethodError)
2021-10-20 05:05:05 +03:00
2021-12-13 07:49:39 +03:00
title = json[:article][:title]
^^^^^^^^
2021-10-20 05:05:05 +03:00
```
2021-12-13 07:49:39 +03:00
This feature is enabled by default.
2021-10-20 05:05:05 +03:00
You can disable it by using a command-line option `--disable-error_highlight` .
See [the repository ](https://github.com/ruby/error_highlight ) in detail.
2021-12-25 11:05:39 +03:00
## IRB Autocomplete and Document Display
The IRB now has an autocomplete feature, where you can just type in the code, and the completion candidates dialog will appear. You can use Tab and Shift+Tab to move up and down.
If documents are installed when you select a completion candidate, the documentation dialog will appear next to the completion candidates dialog, showing part of the content. You can read the full document by pressing Alt+d.
2020-01-18 11:46:52 +03:00
## Miscellaneous changes
2021-01-02 10:42:45 +03:00
2021-05-17 02:49:53 +03:00
* lib/objspace/trace.rb is added, which is a tool for tracing the object
allocation. Just by requiring this file, tracing is started *immediately* .
Just by `Kernel#p` , you can investigate where an object was created.
Note that just requiring this file brings a large performance overhead.
2021-12-15 19:24:31 +03:00
This is only for debugging purposes. Do not use this in production.
2021-07-23 06:41:01 +03:00
[[Feature #17762 ]]
2021-01-02 10:42:45 +03:00
2021-07-23 06:41:34 +03:00
* Now exceptions raised in finalizers will be printed to `STDERR` , unless
`$VERBOSE` is `nil` . [[Feature #17798 ]]
2021-12-13 08:57:28 +03:00
* `ruby -run -e httpd` displays URLs to access. [[Feature #17847 ]]
2021-12-14 03:28:09 +03:00
* Add `ruby -run -e colorize` to colorize Ruby code using
`IRB::Color.colorize_code` .
2021-07-23 06:41:01 +03:00
[Bug #4443 ]: https://bugs.ruby-lang.org/issues/4443
2021-11-18 21:51:14 +03:00
[Feature #6210 ]: https://bugs.ruby-lang.org/issues/6210
2021-12-21 20:49:19 +03:00
[Feature #10917 ]: https://bugs.ruby-lang.org/issues/10917
2021-11-18 23:44:19 +03:00
[Feature #11256 ]: https://bugs.ruby-lang.org/issues/11256
2021-10-28 01:36:00 +03:00
[Feature #11689 ]: https://bugs.ruby-lang.org/issues/11689
2021-03-16 08:07:27 +03:00
[Feature #12194 ]: https://bugs.ruby-lang.org/issues/12194
2021-11-18 21:51:14 +03:00
[Feature #12495 ]: https://bugs.ruby-lang.org/issues/12495
2021-12-07 09:02:08 +03:00
[Feature #12913 ]: https://bugs.ruby-lang.org/issues/12913
2021-01-21 09:02:24 +03:00
[Feature #14256 ]: https://bugs.ruby-lang.org/issues/14256
2021-09-11 13:13:53 +03:00
[Feature #14579 ]: https://bugs.ruby-lang.org/issues/14579
2021-04-16 10:12:10 +03:00
[Feature #15198 ]: https://bugs.ruby-lang.org/issues/15198
2018-10-07 07:02:46 +03:00
[Feature #15211 ]: https://bugs.ruby-lang.org/issues/15211
2021-12-13 08:02:07 +03:00
[Feature #15912 ]: https://bugs.ruby-lang.org/issues/15912
2021-02-16 12:12:47 +03:00
[Feature #16043 ]: https://bugs.ruby-lang.org/issues/16043
2021-12-19 15:37:08 +03:00
[Feature #16182 ]: https://bugs.ruby-lang.org/issues/16182
2021-12-19 17:43:39 +03:00
[Feature #16806 ]: https://bugs.ruby-lang.org/issues/16806
2021-01-02 10:42:45 +03:00
[Feature #17312 ]: https://bugs.ruby-lang.org/issues/17312
2021-02-11 13:14:18 +03:00
[Feature #17327 ]: https://bugs.ruby-lang.org/issues/17327
2021-10-27 10:55:15 +03:00
[Feature #17370 ]: https://bugs.ruby-lang.org/issues/17370
2021-12-13 08:01:40 +03:00
[Feature #17398 ]: https://bugs.ruby-lang.org/issues/17398
2021-03-21 09:12:54 +03:00
[Feature #17411 ]: https://bugs.ruby-lang.org/issues/17411
2021-07-23 06:41:01 +03:00
[Bug #17423 ]: https://bugs.ruby-lang.org/issues/17423
2021-10-27 10:55:15 +03:00
[Bug #17429 ]: https://bugs.ruby-lang.org/issues/17429
[Feature #17470 ]: https://bugs.ruby-lang.org/issues/17470
2021-02-15 10:06:43 +03:00
[Feature #17479 ]: https://bugs.ruby-lang.org/issues/17479
2021-12-13 08:19:31 +03:00
[Feature #17485 ]: https://bugs.ruby-lang.org/issues/17485
2021-12-13 08:28:59 +03:00
[Feature #17544 ]: https://bugs.ruby-lang.org/issues/17544
2021-10-22 19:38:05 +03:00
[Feature #17592 ]: https://bugs.ruby-lang.org/issues/17592
2021-12-13 08:19:21 +03:00
[Feature #17684 ]: https://bugs.ruby-lang.org/issues/17684
2021-05-14 01:31:46 +03:00
[Feature #17724 ]: https://bugs.ruby-lang.org/issues/17724
2021-03-26 10:29:21 +03:00
[Feature #17744 ]: https://bugs.ruby-lang.org/issues/17744
2021-10-27 10:55:15 +03:00
[Feature #17750 ]: https://bugs.ruby-lang.org/issues/17750
2021-05-17 02:49:53 +03:00
[Feature #17762 ]: https://bugs.ruby-lang.org/issues/17762
2021-12-24 18:01:21 +03:00
[Feature #17763 ]: https://bugs.ruby-lang.org/issues/17763
2021-11-09 05:30:03 +03:00
[Feature #17795 ]: https://bugs.ruby-lang.org/issues/17795
2021-07-23 06:41:34 +03:00
[Feature #17798 ]: https://bugs.ruby-lang.org/issues/17798
2021-07-23 06:41:01 +03:00
[Bug #17827 ]: https://bugs.ruby-lang.org/issues/17827
2021-12-13 08:57:28 +03:00
[Feature #17847 ]: https://bugs.ruby-lang.org/issues/17847
2021-10-27 10:55:15 +03:00
[Feature #17853 ]: https://bugs.ruby-lang.org/issues/17853
2021-12-21 13:02:41 +03:00
[Bug #17866 ]: https://bugs.ruby-lang.org/issues/17866
2021-10-27 10:55:15 +03:00
[Bug #18003 ]: https://bugs.ruby-lang.org/issues/18003
2021-08-05 11:25:17 +03:00
[Feature #18008 ]: https://bugs.ruby-lang.org/issues/18008
2021-06-26 01:17:26 +03:00
[Feature #18015 ]: https://bugs.ruby-lang.org/issues/18015
2021-07-02 13:41:16 +03:00
[Feature #18020 ]: https://bugs.ruby-lang.org/issues/18020
2021-08-22 16:27:29 +03:00
[Feature #18029 ]: https://bugs.ruby-lang.org/issues/18029
2021-12-13 18:03:37 +03:00
[Feature #18045 ]: https://bugs.ruby-lang.org/issues/18045
2021-12-13 11:45:29 +03:00
[Feature #18148 ]: https://bugs.ruby-lang.org/issues/18148
2021-09-16 13:55:31 +03:00
[Feature #18172 ]: https://bugs.ruby-lang.org/issues/18172
2021-12-13 07:53:34 +03:00
[Feature #18176 ]: https://bugs.ruby-lang.org/issues/18176
2021-12-13 13:13:21 +03:00
[Feature #18190 ]: https://bugs.ruby-lang.org/issues/18190
2021-10-25 02:38:44 +03:00
[Feature #18229 ]: https://bugs.ruby-lang.org/issues/18229
2021-12-13 18:03:37 +03:00
[Feature #18239 ]: https://bugs.ruby-lang.org/issues/18239
2021-12-13 11:45:29 +03:00
[Feature #18254 ]: https://bugs.ruby-lang.org/issues/18254
2021-10-28 15:07:11 +03:00
[Feature #18273 ]: https://bugs.ruby-lang.org/issues/18273
2021-12-05 14:31:58 +03:00
[Feature #18290 ]: https://bugs.ruby-lang.org/issues/18290
2021-12-13 11:45:29 +03:00
2021-10-28 12:21:27 +03:00
[GH-1509]: https://github.com/ruby/ruby/pull/1509
2021-09-22 11:59:28 +03:00
[GH-4815]: https://github.com/ruby/ruby/pull/4815
2021-12-06 20:03:06 +03:00
[GH-5112]: https://github.com/ruby/ruby/pull/5112
2021-12-06 11:19:42 +03:00
[GH-5131]: https://github.com/ruby/ruby/pull/5131
2021-12-03 23:58:09 +03:00
[GH-5146]: https://github.com/ruby/ruby/pull/5146
