Commit Graph

4369 Commits

Author SHA1 Message Date
Yukihiro "Matz" Matsumoto 2742ded32f Fixed wrong condition in #4926 fix. 2020-01-10 23:05:27 +09:00
Yukihiro "Matz" Matsumoto 70e5746896 Fixed wrong condition for copying arguments on stack; fix #4926
This bug was introduced in 694089f to address #4832
2020-01-10 17:58:15 +09:00
Yukihiro "Matz" Matsumoto dc86e35cac Add MRB_WITHOUT_FLOAT guard to <math.h>. 2020-01-08 10:53:11 +09:00
Yukihiro "Matz" Matsumoto 2124b9b4c9 Fix buffer overflow in mrb_str_len_to_dbl.
Issue 19902: mruby:mruby_fuzzer: Stack-buffer-overflow in mrb_str_len_to_dbl
2020-01-08 10:52:24 +09:00
Yukihiro "Matz" Matsumoto afad939173 Check memory boundary in mrb_str_len_to_dbl. 2020-01-07 08:58:06 +09:00
Yukihiro "Matz" Matsumoto 3e4e10ffb6 "0x10".to_f should be 0, not 16.0; fix #4924 2020-01-06 22:27:39 +09:00
Yukihiro "Matz" Matsumoto c60bccaa94 Check remaining string length before access to avoid OOB access. 2020-01-06 22:25:47 +09:00
Yukihiro "Matz" Matsumoto 59756aef14 Need to preserve the original input string in mrb_str_len_to_dbl. 2020-01-06 22:25:05 +09:00
Yukihiro "Matz" Matsumoto ccf28775b8 Fix mrb_str_len_to_dbl to support Hexadecimal like 0x10. 2020-01-06 15:52:39 +09:00
Yukihiro "Matz" Matsumoto 2532e625ed Refactor mrb_cstr_to_dbl; ref #4920 2020-01-06 15:52:39 +09:00
Yukihiro "Matz" Matsumoto 111045ecad Avoid creating temporary objects in read_irep_record_1; close #4920
The basic idea of this change is from @dearblue.

Note: the arguments of `mrb_str_pool()` have changed, but the function
is provided for internal use (No `MRB_API`). So basically you don't have
to worry about the change.
2020-01-06 15:52:39 +09:00
dearblue 00f5c92149 Call va_end() before return
The behavior when returning from a function without `va_end()` is
undefined.
2020-01-02 00:58:27 +09:00
Yukihiro "Matz" Matsumoto 9174b18f34 Rename mrb_num_args_error to mrb_argnum_error; ref #4863 2020-01-01 22:37:14 +09:00
Yukihiro "Matz" Matsumoto b9c78c3fcc Merge pull request #4863 from shuujii/add-mrb_num_args_error-for-wrong-number-of-arguments-error
Add `mrb_num_args_error()` for "wrong number of arguments" error
2020-01-01 22:34:22 +09:00
KOBAYASHI Shuji 40b8490fa4 Revert "SHARED string is not required when sharing POOL string" (75949836)
Because literal pool may be released by GC.

#### Example:

  ```ruby
  s1 = eval('"abcdefghijklmnopqrstuvwxyz01"')
  GC.start
  p s1  #=> "\x00\x00\x00\x00\x00\x00\x00\x90\x00\x00\x00\x00\x00\x00\x00\x90\x03\x00stuvwxyz01"
  ```
2020-01-01 19:16:05 +09:00
Yukihiro "Matz" Matsumoto 7247e4cf1b Merge pull request #4918 from dearblue/sync-vars
Integrate `i` and `arg_i` in `mrb_get_args()`
2020-01-01 15:31:16 +09:00
dearblue 7ed26ead9b Integrate i and arg_i in mrb_get_args()
The behavior of these two variables is the same.
2020-01-01 15:07:16 +09:00
dearblue 20d3882469 Do not include stdint.h before mruby.h; ref #4750 2020-01-01 14:17:11 +09:00
KOBAYASHI Shuji 0b2d54f4f1 Fix potentially use of wrong method cache
#### Example (with `MRB_METHOD_CACHE`)

  ```ruby
  GC.start
  c = Class.new
  p c            #=> #<Class:0x7fd6a180e790>
  c.new          #=> cache `c.new`
  c = nil
  GC.start       #=> `c` is GCed
  r = Range.dup
  p r            #=> #<Class:0x7fd6a180e790>
                 #   [same pointer as `c`]
  r.new(2, 3)    #=> ArgumentError: 'initialize':
                 #   wrong number of arguments (2 for 0)
                 #   [`c.new` is called instead of `r.new`]
  ```

#### Cause

  An entry of method cache is identified by class pointer and method
  id. However, reusing memory after GC may create a class with the same
  pointer as the cached class.

#### Treatment

  Cleared method caches of the class when the class is GCed.
2019-12-25 21:31:10 +09:00
KOBAYASHI Shuji e8299a89da Fix potentially crash in %n of mrb_vformat() with 64-bit int
If `mrb_sym` is smaller than `int`, it is promoted to `int`.
2019-12-20 23:04:46 +09:00
KOBAYASHI Shuji 6df6bd4ac7 Simplify print_backtrace() 2019-12-18 19:35:24 +09:00
Yukihiro "Matz" Matsumoto 815e0c400a Merge pull request #4875 from shuujii/remove-location-info-from-Exception-inspect
Remove location info from `Exception#inspect`
2019-12-18 01:55:23 +09:00
KOBAYASHI Shuji 29ecc3840a Refine output of mrb_print_error()
The following improvements are made according to Ruby's behavior:

- Match location number to index.
- Remove duplicate most recent call output.
- Fix that first call is not output when array (unpacked) backtrace.

### Example

  ```ruby
  def a; raise "error!" end
  def b; a end
  begin
    b
  rescue => e
    e.backtrace if ARGV[0] == "unpack"  # unpack backtrace
    raise e
  end
  ```

#### Before this patch:

  ```
  $ bin/mruby example.rb unpack
  trace (most recent call last):
    [0] example.rb:2:in b
    [1] example.rb:1:in a
  example.rb:1: error! (RuntimeError)
  ```

#### After this patch:

  ```
  $ bin/mruby example.rb unpack
  trace (most recent call last):
    [2] example.rb:4
    [1] example.rb:2:in b
  example.rb:1:in a: error! (RuntimeError)
  ```
2019-12-17 23:16:20 +09:00
KOBAYASHI Shuji 854c61c3ab Remove unneeded null checks to struct backtrace_location::filename
`struct backtrace_location` is created only in `each_backtrace()`, and
the `filename` field will never be null (it will be `(unknown)` if null).
2019-12-16 18:57:53 +09:00
KOBAYASHI Shuji d2f2f9db51 Remove location info from Exception#inspect
Because location info (file name and line number) is kept in the backtrace,
it should not be kept in the result of `inspect` (and the exception object
itself), I think.

### Example

  ```ruby
  # example.rb
  begin
    raise "err"
  rescue => e
    p e
  end
  ```

#### Before this patch:

  ```
  $ bin/mruby example.rb
  example.rb:2: err (RuntimeError)
  ```

#### After this patch:

  ```
  $ bin/mruby example.rb
  err (RuntimeError)
  ```
2019-12-14 22:29:37 +09:00
dearblue cc52fa66e8 Remove module only methods from class
The `#prepend_features` and `#module_function` methods are not haves for
class objects.
2019-12-14 22:12:32 +09:00
KOBAYASHI Shuji e18aa4d893 Fix arguments check to Array#each
#### Before this patch:

  ```
  $ mruby -e '[].each(1){}'  #=> no error
  ```

#### After this patch:

  ```
  $ mruby -e '[].each(1){}'  #=> ArgumentError: wrong number of arguments
  ```
2019-12-13 17:30:34 +09:00
KOBAYASHI Shuji 81de1f159c Add mrb_num_args_error() for "wrong number of arguments" error
To unify the style of messages.
2019-12-12 11:45:58 +09:00
KOBAYASHI Shuji 7192429e83 Fix behavior of Kernel#Integer to numbers ending with _ and spaces
#### Before this patch:

  ```ruby
  Integer("1_ ")  #=> 1
  ```

#### After this patch (same as Ruby):

  ```ruby
  Integer("1_ ")  #=> ArgumentError
  ```
2019-12-11 16:40:39 +09:00
KOBAYASHI Shuji bf431e77b8 Fix behavior of String#to_i/Kernel#Integer to numbers starting with _
#### Before this patch:

  ```ruby
  Integer("_1")  #=> 1
  "_1".to_i      #=> 1
  ```

#### After this patch (same as Ruby):

  ```ruby
  Integer("_1")  #=> ArgumentError
  "_1".to_i      #=> 0
  ```
2019-12-10 21:31:30 +09:00
Yukihiro "Matz" Matsumoto 543a9f84d1 Merge pull request #4858 from shuujii/fix-that-String-to_f-accepts-consecutive-_-as-a-numeric-expression
Fix that `String#to_f` accepts consecutive `_` as a numeric expression
2019-12-10 11:09:55 +09:00
KOBAYASHI Shuji 0893ee492c Fix that String#to_f accepts consecutive _ as a numeric expression
Consecutive `_` is not allowed as a numeric expression:

  1_2__3             #=> SyntaxError
  Float("1_2__3")    #=> ArgumentError
  Integer("1_2__3")  #=> ArgumentError
  "1_2__3".to_i      #=> 12

But `String#to_f` accept it, so I fixed the issue.

Before this patch:

  "1_2__3".to_f      #=> 123

After this patch:

  "1_2__3".to_f      #=> 12
2019-12-09 21:47:59 +09:00
Yukihiro "Matz" Matsumoto 694089fafe Fix mrb_get_argv() to return array pointer every time; fix #4832 2019-12-09 20:50:41 +09:00
Yukihiro "Matz" Matsumoto 8267993988 Merge pull request #4855 from dearblue/kwargs-uninit
Fix keyword arguments not be obtained with `mrb_get_args()`; Fix #4754
2019-12-09 17:31:53 +09:00
KOBAYASHI Shuji cf6ae0de6a Fix the error message of Kernel#Float
#### Before this patch:

  ```
  $ bin/mruby -e 'Float("1_a")'
  -e:1: invalid string for float(a) (ArgumentError)
  ```

#### After this patch:

  ```
  $ bin/mruby -e 'Float("1_a")'
  -e:1: invalid string for float("1_a") (ArgumentError)
  ```
2019-12-08 18:31:41 +09:00
dearblue 1ca4fdc4c7 Fix keyword arguments not be obtained with mrb_get_args(); Fix #4754
If ":" is after "|" and there is no "?" or "*", the keyword argument
could not be obtained and it was not initialized with `undef`.

For example: "|oo:"
2019-12-07 22:48:58 +09:00
KOBAYASHI Shuji a030446a94 Refine mrb_alloca()
* The allocated memory is guaranteed to be aligned for any data type (it was
  not guaranteed when string type is embed).
* Make allocation size exactly specified size (does not allocate space for a
  null byte).
2019-12-04 22:21:00 +09:00
Yukihiro "Matz" Matsumoto 190649e591 Merge pull request #4837 from shuujii/add-assertion-to-RVALUE-size
Add assertion to `RVALUE` size
2019-11-27 09:47:35 +09:00
KOBAYASHI Shuji 375c1ebc41 Rename BITSIZE to BIT and BIT to BIT_POS for consistency
The bit width terminology is unified to `BIT` according to `MRB_INT_BIT`
and `CHAR_BIT`. Also the bit position terminology is unified to `BIT_POS`.
2019-11-23 20:40:23 +09:00
KOBAYASHI Shuji 6cf99d12ac Add assertion to RVALUE size 2019-11-23 17:36:46 +09:00
KOBAYASHI Shuji a2df247063 Introduce mrb_ssize type for buffer size on memory; ref #4483
Previously, `mrb_int` was used as the type that represents the buffer size
on memory, but the sizes of `RString` and `RArray` exceed 6 words when
`MRB_INT64` is enabled on 32-bit CPU.

I don't think it is necessary to be able to represent the buffer size on
memory that exceeds the virtual address space. Therefore, for this purpose,
introduce `mrb_ssize` which doesn't exceed the sizes of `mrb_int` and
pointer.

I think all `mrb_int` used for this purpose should be changed to
`mrb_ssize`, but currently only the members of the structures (`RString`,
`mrb_shared_string`, `RArray` and `mrb_shared_array`) are changed.
2019-11-21 19:34:02 +09:00
KOBAYASHI Shuji 3241a7e5a8 Refactor mrb_string_value_cstr
- Keep `MRB_STR_ASCII` flag.
- Avoid a string object creation.
2019-11-19 01:07:58 +09:00
Yukihiro "Matz" Matsumoto bd87799673 Merge pull request #4825 from shuujii/fix-argument-specs-to-Kernel
Fix argument specs to `Kernel`
2019-11-18 08:15:58 +09:00
KOBAYASHI Shuji a367373fe3 Revert "Implement Ruby2.7's frozen strings from Symbol#to_s"
This feature was reverted from Ruby 2.7.
2019-11-16 19:47:31 +09:00
KOBAYASHI Shuji 04baaab311 Fix argument specs to Kernel 2019-11-15 19:07:42 +09:00
Yukihiro "Matz" Matsumoto 079aff1779 Revert a76dc04 to resolve #4820 2019-11-13 21:06:26 +09:00
KOBAYASHI Shuji edbc6d8ab5 Fix argument specs to Integer 2019-11-11 20:48:52 +09:00
Yukihiro "Matz" Matsumoto e50344b5da Avoid unnecessary Symbol#to_s call; fix #4812 2019-11-08 23:38:40 +09:00
Yukihiro "Matz" Matsumoto 8ff001ef30 Avoid crashing of Array#unshift; fix #4808
On cases like `a.unshift(*a)`.
2019-11-08 10:57:18 +09:00
KOBAYASHI Shuji 45aa5081e4 Clear MRB_STR_SHARED flag in mrb_str_modify_keep_ascii; close #4807 2019-11-07 22:16:59 +09:00