Yukihiro "Matz" Matsumoto
2bda965258
should output debug info for child ireps; close #1575
2013-11-15 14:16:49 +09:00
Yukihiro "Matz" Matsumoto
13c2d65b02
should not free buffer that is still accessed
2013-11-15 11:47:08 +09:00
Yukihiro "Matz" Matsumoto
3286e63202
wrong size of filenames buffer
2013-11-15 11:46:30 +09:00
Yukihiro "Matz" Matsumoto
04a2c58e94
add new function mrb_load_irep_file_cxt() and simplifies mruby.c
2013-11-15 10:49:40 +09:00
Yukihiro "Matz" Matsumoto
71354b91cb
enum mrb_vtype varies on compile time configuration, namely MRB_NAN_BOXING
2013-11-15 02:45:52 +09:00
Yukihiro "Matz" Matsumoto
16b34d187a
remove warning on NAN_BOXING
2013-11-15 02:27:19 +09:00
fleuria
87fa03ebce
fix length check in read_section_debug(); fix #1572
2013-11-14 20:16:46 +08:00
Yukihiro "Matz" Matsumoto
c6cc92ac87
remove possible free-after-realloc problem; close #1570
2013-11-14 14:59:26 +09:00
Yukihiro "Matz" Matsumoto
fc83acfcff
remove unnecessary jump
2013-11-14 09:46:41 +09:00
bggd
0cbdbd81cb
Changed a variable declaration to top of block
2013-11-13 22:53:20 +09:00
Yukihiro "Matz" Matsumoto
edd6c17c1e
load_exec() no longer return fixnum value for no_exec; should close #1569
2013-11-13 15:48:13 +09:00
Yukihiro "Matz" Matsumoto
ca704e5438
pacify some warnings on OpenBSD
2013-11-13 10:33:10 +09:00
Yukihiro "Matz" Matsumoto
7078fcd9e4
fixnum in irep->pool may overflow
2013-11-13 10:15:57 +09:00
Yukihiro "Matz" Matsumoto
3bbf66f9f3
cancel #1565 since it breaks mirb; instead add special treatment to heredocs
2013-11-10 04:32:57 +09:00
Yukihiro "Matz" Matsumoto
7e64d7e531
check for corrupted mrb file data
2013-11-10 03:08:01 +09:00
Yukihiro "Matz" Matsumoto
29792d1725
Merge branch 'master' of github.com:mruby/mruby
2013-11-09 23:29:50 +09:00
Carson McDonald
0d0f3545f8
Change supress to suppress
2013-11-09 07:40:51 -05:00
Yukihiro "Matz" Matsumoto
ec0a94dae6
Merge pull request #1565 from Fleurer/fix1564
...
nextc(): always return an '\n' at end of input
2013-11-08 17:22:59 -08:00
Yukihiro "Matz" Matsumoto
7aabc657f1
protect returning irep (in proc) from GC
2013-11-09 04:08:24 +09:00
Yukihiro "Matz" Matsumoto
de790bdc27
allow irep to be GCed
2013-11-08 20:39:50 +09:00
fleuria
fa71403bc5
nextc(): always return an '\n' at end of input
...
fix #1564
2013-11-08 11:40:26 +08:00
Yukihiro "Matz" Matsumoto
cba07b0275
type check before pool scan
2013-11-07 09:14:44 +09:00
Yukihiro "Matz" Matsumoto
b6a73f99bf
call_iseq will never be NULL
2013-11-07 08:53:14 +09:00
Yukihiro "Matz" Matsumoto
c8936754e8
wrong return value from read_lineno_record()
2013-11-07 08:20:03 +09:00
Yukihiro "Matz" Matsumoto
465f633446
create Proc object before releasing memory pool
2013-11-07 08:16:41 +09:00
Yukihiro "Matz" Matsumoto
d8913dae3f
forgot to free irep->reps
2013-11-07 04:22:10 +09:00
Yukihiro "Matz" Matsumoto
677a2ac226
irep->pool not to be GCed
2013-11-07 04:20:46 +09:00
Yukihiro "Matz" Matsumoto
e92d4e2680
modified to use irep->reps to reference child ireps. preparation for
...
removing irep array from mrb_state. note that instructions OP_LAMBDA,
OP_EXEC and OP_EPUSH are incompatible, and dumped mrb format has changed.
2013-11-07 03:54:22 +09:00
Yukihiro "Matz" Matsumoto
f80401de6c
resolved conflicts and updated to latest return value change of mrb_generate_code()
2013-11-05 00:13:01 +09:00
Yukihiro "Matz" Matsumoto
e386760cc7
add Fiber.current
2013-11-04 23:58:32 +09:00
fleuria
a511957ec8
refactor mrb_context_run()
2013-11-03 12:05:25 +08:00
fleuria
d75a907edf
introduce mrb_context_run()
...
currently there are two scnenario to call mrb_run(), the first is
calling a proc, in this case mrb should create a new environment,
discarding all the variables except args, reciever and block.
the second is calling the newly generated irep, like in mirb.
in this case, the variables should be kept after mrb_run().
so we introduce mrb_context_run() to handle this seperately.
2013-11-03 11:32:59 +08:00
Yukihiro "Matz" Matsumoto
8af26752db
remove scope->idx
2013-11-02 23:40:36 +09:00
Yukihiro "Matz" Matsumoto
22f90d9f70
codedump() to take irep structure
2013-11-02 23:26:19 +09:00
Yukihiro "Matz" Matsumoto
0be572df62
change return value from mrb_generate_code()
2013-11-02 23:20:04 +09:00
Yukihiro "Matz" Matsumoto
92825660ea
changed argument type of mrb_ary_modify(); ref #1554
2013-11-01 09:57:20 +09:00
Yukihiro "Matz" Matsumoto
af03812877
add mrb_ary_modify(mrb,a); you have to ensure mrb_value a to be an array; ref #1554
2013-10-31 23:34:08 +09:00
Yukihiro "Matz" Matsumoto
0fe744f919
Merge branch 'master' of github.com:mruby/mruby
2013-10-31 11:55:06 +09:00
Yukihiro "Matz" Matsumoto
a228841668
restore context status to RUNNING on automatic yield at the end of the block; fix #1555
2013-10-31 11:54:57 +09:00
fleuria
c64cae4d2d
fix 3a4c8e2
...
unfortunately the previouse commit is incorrect, which still concate the
keywords, but like this:
method='"end\nb"' (201)
2013-10-30 17:17:13 +08:00
Yukihiro "Matz" Matsumoto
abe6db9454
resolve conflict #1552
2013-10-30 17:28:32 +09:00
Yukihiro "Matz" Matsumoto
ad4ce2ba4e
Merge pull request #1551 from Fleurer/fix1550
...
fix #1550
2013-10-30 01:25:41 -07:00
Yukihiro "Matz" Matsumoto
888ddf1c84
type check class/module in mrb_get_args(); close #1477
2013-10-30 13:24:55 +09:00
h2so5
12e00f9238
implement Class.new with block
2013-10-30 12:52:26 +09:00
Yukihiro "Matz" Matsumoto
f8fea6931c
preserve only arguments on stack; fix #1527
2013-10-30 12:01:18 +09:00
fleuria
3a4c8e2d9f
fix #1550
...
emit a "\n" as the first token for parser instead of taking the
first character from the next file for lexer, to prevent mruby
"concatenate" two keywords between files
it should be regarded as a work around, for it can not resolve
this case:
$ cat a.rb
"
$ cat b.rb
b"
$ bin/mrbc -o- -v a.rb b.rb
mruby - Embeddable Ruby Copyright (c) 2010-2013 mruby
developers
NODE_SCOPE:
NODE_BEGIN:
NODE_STR "
b " len 4
irep 0 nregs=2 nlocals=1 pools=1 syms=0
000 OP_STRING R1 "\n\nb "
001 OP_STOP
thanks @bovi 's idea
2013-10-30 10:34:28 +08:00
h2so5
88758e3630
Uniquify the results of Object#methods
2013-10-27 04:04:06 +09:00
Yukihiro "Matz" Matsumoto
3a668241be
nil/false should not be implicitly converted to integers in mrb_get_args(); close #1529
2013-10-26 17:11:09 +09:00
Yukihiro "Matz" Matsumoto
51213bfc7d
mrb_Integer() should not convert strings to integers
2013-10-26 17:06:50 +09:00
Yukihiro "Matz" Matsumoto
fcebd1243b
Merge pull request #1530 from h2so5/args-explicit-conversion
...
Disable implicit integer conversion in mrb_get_args
2013-10-26 00:47:03 -07:00