Roberto Ierusalimschy
b91bc93fd3
'setobj2t' incorporated into 'luaV_fastset' + 'invalidateTMcache'
...
is not needed in the fast track (as it does not create new
entries)
2015-09-09 10:44:07 -03:00
Roberto Ierusalimschy
53be1451a8
new definition for macro 'checkliveness'
2015-09-08 13:55:43 -03:00
Roberto Ierusalimschy
eb0be12ceb
detail in macro 'checkliveness' + macro 'setobj2t' defined as an
...
expression (to be used in macro 'luaV_fastset')
2015-09-08 13:54:52 -03:00
Roberto Ierusalimschy
1198d82530
'lua_longassert' defined as an expression (instead of a command)
2015-09-08 13:53:56 -03:00
Roberto Ierusalimschy
6556fcfe5a
small janitor work
2015-09-08 12:49:25 -03:00
Roberto Ierusalimschy
41964648ee
long strings are created directly in final position when possible
...
(instead of using an auxiliar buffer to first create the string
and then allocate the final string and copy result there)
2015-09-08 12:41:05 -03:00
Roberto Ierusalimschy
502214f8a5
added assert for NULL pointer in 'lua_pushlstring'
2015-08-25 15:50:37 -03:00
Roberto Ierusalimschy
ed19fe766c
added ';' at the end of "expression lines" ("return exp;") so that
...
an extra ";" at the end is enough to stop Lua printing the result
("return exp;;" is not valid)
2015-08-14 16:11:20 -03:00
Roberto Ierusalimschy
8f25d08637
'invalidateTMcache' not needed in all 'settable' uses
2015-08-03 17:40:26 -03:00
Roberto Ierusalimschy
3b795541c4
fast track for 'settable'
2015-08-03 16:50:49 -03:00
Roberto Ierusalimschy
20b9e59441
barrier macros rewritten as expressions
2015-08-03 16:40:42 -03:00
Roberto Ierusalimschy
935d9f7f90
new entry should not be commented out
2015-07-23 11:32:17 -03:00
Roberto Ierusalimschy
e247c3ada3
implementation of fast track for gettable operations
2015-07-20 15:24:50 -03:00
Roberto Ierusalimschy
b5dc2f9b0c
bug: 'io.lines' does not check maximum number of options
2015-07-20 13:37:12 -03:00
Roberto Ierusalimschy
5273477cb9
in 'lua_number2strx', use the call to 'l_sprintf' to add exponent
...
to result directly
2015-07-20 13:30:22 -03:00
Roberto Ierusalimschy
b9db9bd0ef
'lua_Number' added to 'L_Umaxalign' union
2015-07-15 12:57:13 -03:00
Roberto Ierusalimschy
050e8536bb
bug: 'io.lines' does not check maximum number of options
2015-07-15 11:40:28 -03:00
Roberto Ierusalimschy
2b61360d82
avoid overflows (detected with 'clang -ftrapv')
2015-07-13 10:30:03 -03:00
Roberto Ierusalimschy
5b6ac971f9
detail (comment)
2015-07-07 14:03:34 -03:00
Roberto Ierusalimschy
403e181d81
'strftime' puts its result directly into 'lua_Buffer'
2015-07-06 12:16:51 -03:00
Roberto Ierusalimschy
4af03c5ae1
better treatment for integer overflows + all errors throw an error
...
(instead of returning nil)
2015-07-04 13:35:14 -03:00
Roberto Ierusalimschy
bde17a419d
because of debt, 'totalbytes' can be negative (and therefore its
...
type must be signed)
2015-07-04 13:33:17 -03:00
Roberto Ierusalimschy
8950e0c049
avoid possibility of subtle arith. overflow
2015-07-04 13:32:34 -03:00
Roberto Ierusalimschy
8217e0d4fe
avoid subtle possibility of arithmetic overflow
2015-07-04 13:31:42 -03:00
Roberto Ierusalimschy
319ccfefbc
computations in numerical for loop must avoid overflows too
2015-07-04 13:31:03 -03:00
Roberto Ierusalimschy
6a8400ba4f
using 'clang' by default + changes in warnings ('old-style-declaration'
...
removed because it is included in 'extra' + 'strict-aliasing' removed
because it is included in 'all' + 'aggregate-return' removed because
no one would do it by mistake)
2015-07-01 14:52:09 -03:00
Roberto Ierusalimschy
2de3361c6c
detail
2015-07-01 14:47:12 -03:00
Roberto Ierusalimschy
e89763e964
simpler code for 'luaO_fb2int'
2015-06-26 16:32:07 -03:00
Roberto Ierusalimschy
be63d995c1
tiny bug: 'randomseed' was calling 'rand' (instead of 'l_rand')
...
to discard first value
2015-06-26 16:30:32 -03:00
Roberto Ierusalimschy
2b1fc1b38c
with string cache, it is not that important for 'type' to avoid
...
'lua_pushstring'
2015-06-26 16:25:45 -03:00
Roberto Ierusalimschy
f01c12c891
detail (cast to avoid mixing types in conditional expression)
2015-06-25 11:00:01 -03:00
Roberto Ierusalimschy
5aabb7a891
buffer size changed from size_t to int (it is always small) +
...
comments + assert that printf result fits in given buffer
2015-06-24 15:25:10 -03:00
Roberto Ierusalimschy
cb1e451999
detail
2015-06-24 15:23:57 -03:00
Roberto Ierusalimschy
43cff79bf7
detail
2015-06-21 10:50:29 -03:00
Roberto Ierusalimschy
95c3144393
redefine 'l_sprintf' to test correctness of buffer sizes given
...
to 'snprintf'
2015-06-18 11:27:44 -03:00
Roberto Ierusalimschy
19eb6ae580
using 'snprintf' in C99 (both for documentation of buffer sizes
...
and some complains from tools)
2015-06-18 11:26:05 -03:00
Roberto Ierusalimschy
cbe05b48bb
using 'lua_longassert' to avoid warnings
2015-06-18 11:25:26 -03:00
Roberto Ierusalimschy
e7f34ad395
better implementation for buffers (reallocated memory directly
...
with allocation function; generates much less garbage)
2015-06-18 11:23:14 -03:00
Roberto Ierusalimschy
a5cbb7c3a7
detail (i + 1 > exp is simply i >= exp)
2015-06-18 11:20:32 -03:00
Roberto Ierusalimschy
dcad08b76d
details (use original type when saving variable's value)
2015-06-18 11:19:52 -03:00
Roberto Ierusalimschy
afbf5c5dab
missing ']]' in long string
2015-06-18 11:05:28 -03:00
Roberto Ierusalimschy
a1415c0d72
bug in order NaN x int (tests must ensure that NaN does not get
...
converted to integer)
2015-06-09 12:53:35 -03:00
Roberto Ierusalimschy
2ecaf18138
using macros ('rttype' and 'settt_') to access "private" field 'tt_'
2015-06-09 11:21:42 -03:00
Roberto Ierusalimschy
cbef15f3ea
comment (FALLTHROUGH position)
2015-06-09 11:21:13 -03:00
Roberto Ierusalimschy
16e8bb360b
comment
2015-06-09 11:21:00 -03:00
Roberto Ierusalimschy
6628012e9e
new release number (5.3.1)
2015-06-03 10:03:38 -03:00
Roberto Ierusalimschy
ed1fe28bb8
When comparing integers with floats, use float as common type when
...
possible (i.e., when integer fits in a float without losses); seems
to be more efficient in architectures with native support for used
Lua numbers
2015-06-02 16:11:24 -03:00
Roberto Ierusalimschy
6645bb2df4
'strcache' elements as arrays of 1 element hints that cache can
...
be n-way (instead of direct mapped)
2015-06-01 13:34:37 -03:00
Roberto Ierusalimschy
02aed045de
bug: return hook may not see correct values for active local variables
...
when function returns
2015-05-22 16:32:42 -03:00
Roberto Ierusalimschy
72bd2ef6c5
dependencies updated
2015-05-22 16:29:43 -03:00
