In the existing implementation, `main_type` and `sub_type` would end up
being called multiple times potentially.
Instead of doing that, save the result so it can be re-used.
https://github.com/ruby/net-http/commit/179976f7ea
(https://github.com/ruby/irb/pull/575)
* Support native integration with ruby/debug
* Prevent using multi-irb and activating debugger at the same time
Multi-irb makes a few assumptions:
- IRB will manage all threads that host sub-irb sessions
- All IRB sessions will be run on the threads created by IRB itself
However, when using the debugger these assumptions are broken:
- `debug` will freeze ALL threads when it suspends the session (e.g. when
hitting a breakpoint, or performing step-debugging).
- Since the irb-debug integration runs IRB as the debugger's interface,
it will be run on the debugger's thread, which is not managed by IRB.
So we should prevent the 2 features from being used at the same time.
To do that, we check if the other feature is already activated when
executing the commands that would activate the other feature.
https://github.com/ruby/irb/commit/d8fb3246be
(https://github.com/ruby/irb/pull/652)
1. Dynamically including `HistorySavingAbility` makes things unnecessarily
complicated and should be avoided.
2. Because both `Reline` and `Readline` use a single `HISTORY` constant
to store history data. When nesting IRB sessions, only the first IRB
session should handle history loading and saving so we can avoid
duplicating history.
3. History saving callback should NOT be stored in `IRB.conf` as it's
recreated every time `IRB.setup` is called, which would happen when
nesting IRB sessions.
https://github.com/ruby/irb/commit/0fef0ae160
(https://github.com/ruby/irb/pull/635)
* Remove unused InputMethod#initialize
The constructor takes a `file_name` argument, but it is never used. The
only input method that needs a file is `FileInputMethod`, which has its
own constructor to take a file object directly.
So the constructor in `InputMethod` is not needed and its child classes
don't need to call `super` in their constructors.
* Remove unused FileInputMethod#file_name
https://github.com/ruby/irb/commit/153b1e9d1c
(https://github.com/ruby/irb/pull/658)
* Decouple `edit` command from `show_source`
2 commands should not depend on each other. If `edit` command also needs
to find a source, the source finding logic should be extracted into a
separate class.
* Return nil if is not an actual file path
* Refactor SourceFinder
https://github.com/ruby/irb/commit/9790517a0c
This reverts commit db80e947a3.
This patch uses `Thread.abort_on_exception = true` and it affects
all of threads. This is why CI systems fails.
How to modify:
- use `thread.abort_on_exception = true` for specific threads
- Run this code in a separated process
* Sync YARP gemspec
We were previously not syncing the YARP gemspec over which meant
that the Ruby YARP gem was out of sync.
* Sync YARP node / lex_compat files
(https://github.com/ruby/irb/pull/648)
IRB already has several features that rely on rdoc, such as:
- Autocompletion's document dialog
- Autocompletion's `PerfectMatchedProc`
- The `show_doc` command
- Easter egg
And we could use its pager more in the future too. So it makes sense to
declare rdoc as a dependency instead of relying on the one bundled with
Ruby.
https://github.com/ruby/irb/commit/4dffbb1dd3
Remove rb_control_frame_t::__bp__ and optimize bmethod calls
This commit removes the __bp__ field from rb_control_frame_t. It was
introduced to help MJIT, but since MJIT was replaced by RJIT, we can use
vm_base_ptr() to compute it from the SP of the previous control frame
instead. Removing the field avoids needing to set it up when pushing new
frames.
Simply removing __bp__ would cause crashes since RJIT and YJIT used a
slightly different stack layout for bmethod calls than the interpreter.
At the moment of the call, the two layouts looked as follows:
┌────────────┐ ┌────────────┐
│ frame_base │ │ frame_base │
├────────────┤ ├────────────┤
│ ... │ │ ... │
├────────────┤ ├────────────┤
│ args │ │ args │
├────────────┤ └────────────┘<─prev_frame_sp
│ receiver │
prev_frame_sp─>└────────────┘
RJIT & YJIT interpreter
Essentially, vm_base_ptr() needs to compute the address to frame_base
given prev_frame_sp in the diagrams. The presence of the receiver
created an off-by-one situation.
Make the interpreter use the layout the JITs use for iseq-to-iseq
bmethod calls. Doing so removes unnecessary argument shifting and
vm_exec_core() re-entry from the interpreter, yielding a speed
improvement visible through `benchmark/vm_defined_method.yml`:
patched: 7578743.1 i/s
master: 4796596.3 i/s - 1.58x slower
C-to-iseq bmethod calls now store one more VALUE than before, but that
should have negligible impact on overall performance.
Note that re-entering vm_exec_core() used to be necessary for firing
TracePoint events, but that's no longer the case since
9121e57a5f.
Closesruby/ruby#6428
(https://github.com/ruby/irb/pull/642)
* Define RelineInputMethod::HISTORY
The HistorySavingAbility module doesn't do anything if the input method
class doesn't define HISTORY.
- 3ac96be660/lib/irb/history.rb (L10)
- 3ac96be660/lib/irb/history.rb (L34)
This patch defines RelineInputMethod::HISTORY to avoid this.
* Improve history-saving's ability check
Instead of checking the existence of `input_method_class::HISTORY`, we should
make every input method class declare if it supports history saving or not.
Since the default value is `false`, it shouldn't break any custom input method
that inherits from `IRB::InputMethod`.
https://github.com/ruby/irb/commit/aec7a5b3f5
Previously, when sorting and comparing git Gemfile vs lockfile sources during
`bundler/setup` to figure out whether we need to re-resolve or not, we
would try to find the default branch if nothing more specific was
specified in the Gemfile.
If the git cache has been deleted thought, that would fail.
The error would still be swallowed (and the branch would simply not be
displayed), but trying to clone would still generate the side effect of
creating the parent folder for the clone.
That could affect non-writable systems that don't expect `bundler/setup`
to write to the filesystem at all.
To fix this, override `Bundler::Source::Git#identifier` to use
exclusively static information, so it does not even try to clone the
repo nor generate any side effects.
https://github.com/rubygems/rubygems/commit/582eb2ef39
(https://github.com/ruby/reline/pull/566)
* Avoid referencing IOGate from IOGate classes
The only time those classes being used is when themselves being the IOGate.
So when referencing to IOGate, it's better to use `self` instead.
* Avoid referencing to IOGate from LineEditor directly
* Avoid referencing to IOGate from Core directly
* Reference to Reline.core directly
* Replace Reline::IOGate with Reline.core.io_gate
This commit improves the behavior of showing the "toggle source" element on mouseover.
For example, when a method has one more signatures by using `:call-seq:`,
```ruby
# :call-seq:
# foo {|element| ... } -> self
# foo -> new_enumeration
def foo
end
```
The current CSS doesn't show "toggle source" even when hovering the second signature `foo -> new_enumeration`.
But this change will show "toggle source" always when hovering over any signature.
For details about the `.method-header` element, see `lib/rdoc/generator/template/darkfish/class.rhtml`:
0e060c69f5/lib/rdoc/generator/template/darkfish/class.rhtml (L101-L124)
For example, see https://docs.ruby-lang.org/en/3.2/Array.html#method-i-delete
(https://github.com/ruby/irb/pull/623)
* Rename `ext/history.rb` to `ext/eval_history.rb`
To confusion with `lib/irb/history.rb`
* Add eval_history tests
* Rename eval_history's History to EvalHistory to avoid confusion
* YJIT: Fix autosplat miscomp for blocks with optionals
When passing an array as the sole argument to `yield`, and the yieldee
takes more than 1 optional parameter, the array is expanded similar
to `*array` splat calls. This is called "autosplat" in
`setup_parameters_complex()`.
Previously, YJIT did not detect this autosplat condition. It passed the
array without expanding it, deviating from interpreter behavior.
Detect this conditon and refuse to compile it.
Fixes: Shopify/yjit#313
* RJIT: Fix autosplat miscomp for blocks with optionals
This is mirrors the same issue as YJIT. See previous commit.
(https://github.com/ruby/reline/pull/561)
* Use Reline::Core.encoding instead of directly referencing IOGate
* Set input/output based on the IOGate's interface
We have some flags that limit running git commit commands under certain
situations, for example, when running under `--local`. However, those
should only affect remote git operations, not local read-only operations
like `git --version`, or `git rev-parse --abbrev-ref HEAD`.
This commit refactors things to achieve that.
By doing this, the `#to_s` representation of a source is more
consistent, since we don't get any errors when reading the checked out
branch, and we avoid some flip-flop lockfile issues.
https://github.com/rubygems/rubygems/commit/4a529fce81
When dependencies in path sources have changed, we'll be re-resolving,
and we can't really know whether the resolution will be valid or invalid
for the Ruby platform, so skip the removal in that case.
https://github.com/rubygems/rubygems/commit/afc3b0956f
(https://github.com/ruby/irb/pull/613)
Since `IRB.conf[:SAVE_HISTORY]` is assigned with 1000 by default, history-saving
is a feature enabled by default. So it should not be treated as an extension,
which adds unnecessary complexity to the code.
(https://github.com/ruby/irb/pull/618)
* Test last value is assigned with measure enabled
* Remove unnecessary `result` variable
`Context#evaluate` always assigns the result of the evaluation to `_` so
we don't need to do it in `Irb#eval_input`.
* Move benchmarking logic into `Context#evaluate`
Current location of the benchmarking logic is too high up and includes
operations like command loading and argument transformation, which should
be excluded. So this commit moves it into `Context#evaluate` to reduce the
noise.
We don't move it further down to `Workspace#evaluate` because `Context`
is an argument of the measure block, which is not available in `Workspace`.
(https://github.com/ruby/irb/pull/615)
This makes sure `Context#evaluate` really just evaluates the input.
It will also make #575's implementation cleaner.
(https://github.com/ruby/timeout/pull/30)
throw/catch is used for non-local control flow, not for exceptional situations.
For exceptional situations, raise should be used instead. A timeout is an
exceptional situation, so it should use raise, not throw/catch.
Timeout's implementation that uses throw/catch internally causes serious problems.
Consider the following code:
```ruby
def handle_exceptions
yield
rescue Exception => exc
handle_error # e.g. ROLLBACK for databases
raise
ensure
handle_exit unless exc # e.g. COMMIT for databases
end
Timeout.timeout(1) do
handle_exceptions do
do_something
end
end
```
This kind of design ensures that all exceptions are handled as errors, and
ensures that all exits (normal exit, early return, throw/catch) are not
handled as errors. With Timeout's throw/catch implementation, this type of
code does not work, since a timeout triggers the normal exit path.
See https://github.com/rails/rails/pull/29333 for an example of the damage
Timeout's design has caused the Rails ecosystem.
This switches Timeout.timeout to use raise/rescue internally. It adds a
Timeout::ExitException subclass of exception for the internal raise/rescue,
which Timeout.timeout will convert to Timeout::Error for backwards
compatibility. Timeout::Error remains a subclass of RuntimeError.
This is how timeout used to work in Ruby 2.0. It was changed in Ruby 2.1,
after discussion in [Bug #8730] (commit
https://github.com/ruby/timeout/commit/238c003c921e in the timeout repository). I
think the change from using raise/rescue to using throw/catch has caused
significant harm to the Ruby ecosystem at large, and reverting it is
the most sensible choice.
From the translation of [Bug #8730], it appears the issue was that
someone could rescue Exception and not reraise the exception, causing
timeout errors to be swallowed. However, such code is broken anyway.
Using throw/catch causes far worse problems, because then it becomes
impossible to differentiate between normal control flow and exceptional
control flow.
Also related to this is [Bug #11344], which changed how
Thread.handle_interrupt interacted with Timeout.
https://github.com/ruby/timeout/commit/f16545abe6
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org>
(https://github.com/ruby/reline/pull/557)
`get_mbchar_byte_size_by_first_char` isn't used in Reline.
Also, this method implements the same functionality as `String#bytesize` and is unnecessary.
This commit is the initial sync of all files from ruby/yarp
into ruby/ruby. Notably, it does the following:
* Sync all ruby/yarp/lib/ files to ruby/ruby/lib/yarp
* Sync all ruby/yarp/src/ files to ruby/ruby/yarp/
* Sync all ruby/yarp/test/ files to ruby/ruby/test/yarp
(https://github.com/ruby/irb/pull/515)
* Implement heredoc embdoc and string indentation with bugfix
* Fix test_ruby_lex's indentation value
* Add embdoc indent test
* Add workaround for lines==[nil] passed to auto_indent when exit IRB with CTRL+d
add new features easily
(https://github.com/ruby/irb/pull/500)
* Add nesting level parser for multiple use (indent, prompt, termination check)
* Rewrite RubyLex using NestingParser
* Add nesting parser tests, fix some existing tests
* Add description comment, rename method to NestingParser
* Add comments and tweak code to RubyLex
* Update NestingParser test
* Extract list of ltype tokens to constants
This error message is also printed when using `bundler/setup` in frozen
model, so we're not necessarily installing any gems when it happens.
This new message play nicer with all situations.
https://github.com/rubygems/rubygems/commit/6874bbacce
Similarly to how the other ignored files are intended for local
development and not for production, the Gemfile and Gemfile.lock files
for a gem only relate to local development and aren't useful to people
installing the gem.
https://github.com/rubygems/rubygems/commit/59049c04be
* Unify length field for embedded and heap strings
The length field is of the same type and position in RString for both
embedded and heap allocated strings, so we can unify it.
* Remove RSTRING_EMBED_LEN
It's the only part that needs "root folder resultion" to figure out the
folder for the cache, but it's only needed for some things, so run that
logic lazily when needed.
https://github.com/rubygems/rubygems/commit/c7b9eae0bc
If we're in inline mode, Bundler first resolves using only local gems,
and if some gems are missing, then it re-resolves using remote gems.
However, "source resolution" from the initial "local" try was being
memoized, resulting in Bundler not looking for some gems remotely in the
second resolution.
This commit forces a proper re-resolve in this case.
https://github.com/rubygems/rubygems/commit/fdc631075e
(https://github.com/ruby/irb/pull/598)
* Rename conf to irb_context
* Drop Nop#irb method because it's only used by irb/ext/loader.rb
We don't need to expose this method to all command classes, especially
when it's just an alias of `irb_context.irb`.
(https://github.com/ruby/irb/pull/599)
I received a `RegexpError` when I typed `::Array[`.
::Array[/Users/mi/ghq/github.com/ruby/irb/lib/irb/completion.rb:236:in `retrieve_completion_data': premature end of char-class: /^Array[/ (RegexpError)
I cannot find where this is defined. I'm guessing neither this
nor the else branch is ever hit. I only found out when testing
VM changes, which had a bug that exposed this.
https://github.com/ruby/racc/commit/8983d0c134
This commit is to add an extra option to enable verbose mode (V=1) in the
generated `Makefile` at runtime of the Ruby to print compiler command lines by
the commands below when building native extensions. It's possible to enable the
verbose mode by setting the environment variable `MAKEFLAGS="V=1"`[1]
implemented in GNU make. However, I wanted to make a consistent user-interface
not depending on the specific make's implementation.
```
$ ruby /path/to/extconf.rb -- --with-verbose
```
You can also add the extra option via rake-compiler gem.
```
$ rake compiler -- --with-verbose
```
If the extra option is not given, the value of the
`RbConfig::CONFIG["MKMF_VERBOSE"]` enabled by the configure option below is
used.
```
$ ./configure --enable-mkmf-verbose
```
For the unit tests, updated the following files.
* The `test/mkmf/test_configuration.rb` was created to test the cases with the
`configuration` method and this implementation.
* Updated the `TestMkmf#assert_separately` to set the extra
arguments in `test/mkmf/base.rb`. Updated tests using the `assert_separately`.
* Added tests for `MakeMakefile#with_config` in the `test/mkmf/test_config.rb`.
[1] https://www.gnu.org/software/make/manual/html_node/Variables_002fRecursion.html
Fixes [Bug #19695]
Co-authored-by: Nobuyoshi Nakada <nobu@ruby-lang.org>
We can install RubyGems plugin by "gem install XXX". The installed
plugin is used from the NEXT "gem ...".
For example, "gem install gem-src kaminari" doesn't use gem-src plugin
for kaminari. "gem install gem-src && gem install kaminari" uses
gem-src plugin for kaminari.
How about loading a plugin immediately when the plugin is installed?
If this proposal is implemented, "gem install gem-src kaminari" works
like "gem install gem-src && gem install kaminari".
https://github.com/rubygems/rubygems/commit/4917d96f4c
They have version.rb files with same directory.
But version.rb have been removed at https://github.com/ruby/ruby/pull/3375
There is no reason to locate under the library name of directory.
(https://github.com/ruby/csv/pull/250)
Since PR #159, the minimum Ruby version is 2.5.0, a version which no
longer requires refinements for String#delete_suffix?, String#match? and
Regexp#match?.
(https://github.com/ruby/irb/pull/589)
* Allow `show_source` for private methods
Fix https://github.com/ruby/irb/pull/577
* Pend tests on TruffleRuby
It seems `eval(..., __LINE__ + 1)` does not work.
Other similar tests are also pended on TruffleRuby, so I think it
is acceptable.
* Use `private_method_defined?` instead of `defined?`
https://github.com/ruby/syntax_suggest/pull/187 Handle if/else with
empty/comment
line
Reported in #187 this code:
```
class Foo
def foo
if cond?
foo
else
# comment
end
end
# ...
def bar
if @recv
end_is_missing_here
end
end
```
Triggers an incorrect suggestion:
```
Unmatched keyword, missing `end' ?
1 class Foo
2 def foo
> 3 if cond?
> 5 else
8 end
16 end
```
Part of the issue is that while scanning we're using newlines to determine when to stop and pause. This is useful for determining logically smaller chunks to evaluate but in this case it causes us to pause before grabbing the "end" that is right below the newline. This problem is similar to https://github.com/ruby/syntax_suggest/pull/179.
However in the case of expanding same indentation "neighbors" I want to always grab all empty values at the end of the scan. I don't want to do that when changing indentation levels as it affects scan results.
There may be some way to normalize this behavior between the two, but the tests really don't like that change.
To fix this issue for expanding against different indentation I needed a way to first try and grab any additional newlines with the ability to rollback that guess. In #192 I experimented with decoupling scanning from the AroundBlockScan logic. It also added the ability to take a snapshot of the current scanner and rollback to prior changes.
With this ability in place now we:
- Grab extra empties before looking at the above/below line for the matching keyword/end statement
- If there's a match, grab it
- If there's no match, discard the newlines we picked up in the evaluation
That solves the issue.
https://github.com/ruby/syntax_suggest/commit/513646b912
AroundBlockScan started as a utility class that was meant to be used as a DSL for scanning and making new blocks. As logic got added to this class it became hard to reason about what exactly is being mutated when. I pulled the scanning logic out into it's own class which gives us a clean separation of concerns. This allowed me to remove a lot of accessors that aren't core to the logic provided by AroundBlockScan.
In addition to this refactor the ScanHistory class can snapshot a scan. This allows us to be more aggressive with scans in the future as we can now snapshot and rollback if it didn't turn out the way we wanted.
The change comes with a minor perf impact:
before: 5.092678 0.104299 5.196977 ( 5.226494)
after: 5.128536 0.099871 5.228407 ( 5.249542)
This represents a 0.996x change in speed (where 1x would be no change and 2x would be twice as fast). This is a 0.38% decrease in performance which is negligible. It's likely coming from the extra blocks being created while scanning. This is negligible and if the history feature works well we might be able to make better block decisions which is means fewer calls to ripper which is the biggest bottleneck.
While this doesn't fix https://github.com/ruby/syntax_suggest/issues/187 it's a good intermediate step that will hopefully make working on that issue easier.
https://github.com/ruby/syntax_suggest/commit/ad8487d8aa
The env var DEBUG does not work to produce detailed output. It is SYNTAX_SUGGEST_DEBUG. It was changed as part of the dead_end to syntax_suggest migration for Ruby 3.2
https://github.com/ruby/syntax_suggest/commit/c41da7aab7
(https://github.com/ruby/irb/pull/576)
* Simplify each_top_level_statement, reduce instance vars
* Update lib/irb/ruby-lex.rb
Co-authored-by: Stan Lo <stan001212@gmail.com>
* Remove unused ltype from TestRubyLex#check_state response
* Remove unnecessary const path of TerminateLineInput
* Combine duplicated code state check into method
---------
https://github.com/ruby/irb/commit/172453cec4
Co-authored-by: Stan Lo <stan001212@gmail.com>
(https://github.com/ruby/irb/pull/583)
1. Make `RubyLex#set_input` simply assign the input block. This matches
the behavior of `RubyLex#set_prompt`.
2. Merge `RubyLex#set_input`'s IO configuration logic with `#set_auto_indent`
into `#configure_io`.
(https://github.com/ruby/reline/pull/492)
* Rewrite dialog rendering
* Fix failing test of dialog with small screen
* Add multiple-dialog rendering test
* Add description comments for each part of render_dialog_changes
- Follow up RegexpLiteral at 9264d834215aa7ce14b0273032a7686c20141db9.
- Split the code to be generated so that `REGEXP` does not need
escapes.
- Use `REGEXP.match?` since support for Ruby 2.3 or earlier has been
dropped.
(https://github.com/ruby/irb/pull/574)
`MagicFile` was introduced around v0.9.6, which was like 14~15 years ago.
It was needed because back then we needed to read a file's magic comment
to determine the encoding of it, and read it with that encoding.
Commit: 3ee79e89
But now we expect files to be encoded in UTF-8 and don't specify encoding
through magic comments anymore, `MagicFile` can be retired.
(https://github.com/ruby/irb/pull/571)
* Simplify Locale#load
Instead of loading file content with `MagicFile` and then evaluting it,
we can just use `Kernel.load` to load the file.
* Remove unused optional argument
* Remove unused Locale#require and #toplevel_load
(https://github.com/ruby/irb/pull/573)
`MagicFile` was introduced around v0.9.6, which was like 14~15 years ago.
It was needed because back then we needed to read a file's magic comment
to determine the encoding of it, and read it with that encoding.
Commit: 3ee79e89ad
But now both EN and JA's help-message file are UTF-8 and have removed the
encoding comment, we don't need to open them with `MagicFile` anymore.
Josh Nichols
Stan Lo
Nobuyoshi Nakada
tomoya ishida
Takashi Kokubun
Benoit Daloze
Hiroshi SHIBATA
Kevin Newton
Mat Sadler
ko1
Jenny Shen
Xavier Noria
osyo-manga
Kentaro Takeyama
David Rodríguez
Jemma Issroff
Jean Boussier
Alan Wu
ima1zumi
Peter Zhu
B. Burt
Koichi ITO
ParadoxV5
Masafumi Koba
Vinicius Stock
gemmaro
Andy Waite
John Bachir
Burdette Lamar
Sutou Kouhei
Jeremy Evans
Yuta Kusuno
Petrik
Samuel Williams
Gareth Adams
Daniel Pepper
Vinicius Stock
Jun Aruga
Mau Magnaguagno
TSUYUSATO Kitsune
schneems
Adam Daniels
Samuel Giddins
Takumasa Ochi
zzak