| Age | Commit message (Collapse) | Author |
|---|
|
|
|
With some refactoring.
* `ret` argument is always non-nil, so no check needed.
* move allocation check right after malloc().
* simplify conditions.
|
|
|
|
Consistent naming: `integer` to represent integer packed in `mrb_value`
instead of `inum`.
|
|
* should not raise error for non-string arguments
* avoid allocating case converted string internally
|
|
They return the checking argument without modification, so the values
are already there. Maybe we should change the return type to `void` but
keep them unchanged for compatibility.
|
|
- String#__lines
- Array#__ary_eq
- Array#__ary_cmp
- Hash#__delete
- Kernel#__case_eqq
- Integer#__coerce_step_counter
|
|
Consistent number conversion function names:
* `mrb_value` to immediate (C) value
* `mrb_int()` -> `mrb_as_int()`
* `mrb_to_flo()` -> `mrb_as_float()`
* `mrb_value` to `mrb_value` (converted)
* `mrb_to_int()'
* `mrb_Integer()` - removed
* `mrb_Float()` -> `mrb_to_float`
Consistent function name (avoid `_flo` suffix):
* `mrb_div_flo()` -> `mrb_div_float`
|
|
This work was done as follows:
- check: `git grep $'\011' -- :^oss-fuzz`
- convert: `ruby -pi -e 'nil while $_.sub!(/^(.*?)\t/) { $1 + " " * (8 - $1.size % 8) }'`
The `doc/guide/{compile,mrbgems}.md` file adds and removes whitespace to make the directory tree look the same.
In `mrbgems/mruby-socket/src/socket.c`, there is a part where the indent is changed from 4 to 2 at the same time as the tab is changed.
|
|
The "a"/"*" specifier of the `mrb_get_args()` function will now return `const mrb_value *`.
This is because it is difficult for the caller to check if it is an array object and write-barrier if necessary.
And it requires calling `mrb_ary_modify()` on the unmodified array object, which is also difficult (this is similar to #5087).
|
|
- Integrate `Fixnum` and `Integer`
- Remove `Integral`
- `int / int -> int`
- Replace `mrb_fixnum()` to `mrb_int()`
- Replace `mrb_fixnum_value()` to `mrb_int_value()`.
- Use `mrb_integer_p()` instead of `mrb_fixnum_p()`
|
|
- `MRB_WITHOUT_FLOAT` => `MRB_NO_FLOAT`
- `MRB_USE_FLOAT` => `MRB_USE_FLOAT32`
The former is to use `USE_XXX` naming convention. The latter is to make
sure `float` is 32bit float and not floating point number in general.
|
|
- The `s` specifier is a string pointer obtained without performing `mrb_str_modify()`, so it cannot be changed.
- The `z` specifier cannot be changed because it is a string pointer obtained by `RSTRING_CSTR()` which returns `const char *`.
|
|
`mrb_get_arg1()` raises `ArgumentError` if the method does not receive one
argument.
And replaces all `mrb_get_args(mrb, "o", &arg)` by the new function.
|
|
|
|
|
|
|
|
|
|
|
|
Unlike CRuby, there's no way to process strings byte-wise by core
methods because there's no per string encoding in mruby, so that
we moved 3 byte-wise operation methods from `mruby-string-ext` gem.
|
|
- `mrb_str_index_m()` and `mrb_str_rindex()`
Make `mrb_get_args()` called only once from called twice.
- `mrb_str_byteslice()`
Replace `goto` with `if ~ else`.
|
|
The binary sizes (gems are only `mruby-bin-mruby`) are reduced slightly in
my environment than before the introduction of new specifiers/modifiers
(5116789a) with this change.
------------+-------------------+-------------------+--------
BINARY | BEFORE (5116789a) | AFTER (This PR) | RATIO
------------+-------------------+-------------------+--------
mruby | 593416 bytes | 593208 bytes | -0.04%
libmruby.a | 769048 bytes | 767264 bytes | -0.23%
------------+-------------------+-------------------+--------
BTW, I accidentally changed `tasks/toolchains/visualcpp.rake` at #4613,
so I put it back.
|
|
|
|
Currently, `Integral#chr` in mruby changes behavior by `MRB_UTF8_STRING`
setting.
before this patch:
$ bin/mruby -e 'p 171.chr' #=> "\xab" (`MRB_UTF8_STRING` is disabled)
$ bin/mruby -e 'p 171.chr' #=> "«" (`MRB_UTF8_STRING` is enabled)
This behavior is incompatible with Ruby, and a little inconvenient because
it can't be interpreted as ASCII-8BIT with `MRB_UTF8_STRING`, I think.
So add encoding argument according to Ruby.
after this patch:
$ bin/mruby -e 'p 171.chr' #=> "\xab"
$ bin/mruby -e 'p 171.chr("ASCII-8BIT")' #=> "\xab"
$ bin/mruby -e 'p 171.chr("UTF-8")' #=> "«"
Allow only `String` for encoding because mruby doesn't have `Encoding`
class, and `"ASCII-8BIT"` (`"BINARY"`) and `"UTF-8"` (only with
`MRB_UTF8_STRING`) are valid value (default is `"ASCII-8BIT"`).
|
|
Because they're defined in both `mruby-string-ext` and `mruby-numeric-ext`
(they seem more natural to define in N, but `mruby-string-ext` depends on
`Integral#chr`).
|
|
Example:
$ bin/mruby -e '
p "あa".byteslice(1)
p "bar".byteslice(3)
p "bar".byteslice(4..0)
'
Before this patch:
"a"
""
RangeError (4..0 out of range)
After this patch (same as Ruby):
"\x81"
nil
nil
|
|
|
|
|
|
We have added internal convenience method `__to_str` which
does string type check.
The issue #3854 was fixed but fundamental flaw of lack of stack
depth check along with fibers still remains. Use `MRB_GC_FIXED_ARENA`
for workaround.
|
|
|
|
The issue is reported by `https://hackerone.com/dgaletic`.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This reverts commit 7b04fcd092006b6e78cd63619fb7ae972f8e0c5d.
The issue was addressed by 9e3cbaa. No longer needed.
|
|
|
|
|
|
|
|
|
|
mruby restriction:
In mruby, `String#delete` only takes single pattern argument.
|
|
mruby restriction:
In mruby, `String#count` does not take multiple pattern arguments,
but only one pattern.
|
|
mruby restriction:
`String#squeeze` can take more than 1 pattern arguments in CRuby,
in that case, the intersection of patterns will be used to match.
But in mruby, it doesn't take multiple patterns.
|
|
|
|
This patch is based on `mruby/c` implementation by Hirohito Higashi.
We might need to add `#tr_s`, `#squeeze` and `#delete` as well.
Adding them should not be too hard using functions we implemented here.
|
|
|
|
Avoid using `mrb_yield` in C code. The function is not recommended.
Because it doesn't work well with fibers.
|
|
`String#lines` (with a block) is now implemented in Ruby.
|
