Roberto Ierusalimschy
|
77077b39d5
|
comment explaining OP_VARARG was wrong (and corresponding code was not
very clear)
|
15 years ago |
Roberto Ierusalimschy
|
5bc91c6405
|
no more one environment per thread: all threads share a single global
environment
|
15 years ago |
Roberto Ierusalimschy
|
f5073de0a7
|
'ipairs' goes until length of array instead of stopping at the first nil
|
15 years ago |
Roberto Ierusalimschy
|
f6ed285cf2
|
new hash for doubles based on frexp, to avoid low-level tricks
|
15 years ago |
Roberto Ierusalimschy
|
1448e736f0
|
better documentation for constructor syntax
|
15 years ago |
Roberto Ierusalimschy
|
6e415e9387
|
local function definition does not need to correct debug information
|
15 years ago |
Roberto Ierusalimschy
|
b114c99a60
|
two small bugs: 'debug.getinfo' did not consider negative indices as out
of range + 'debug.[gs]etlocal' crash on tail calls
|
15 years ago |
Roberto Ierusalimschy
|
916587508c
|
parser keeps list of active local variables in a single dynamic array,
therefore saving C stack space
|
15 years ago |
Roberto Ierusalimschy
|
a5382b763c
|
new function lua_copy
|
15 years ago |
Roberto Ierusalimschy
|
ba21aa8b2b
|
'cpcall' must ensure correct environment for called function
|
15 years ago |
Roberto Ierusalimschy
|
d119cf3035
|
a few more instructions to testC (func2udata and getfield)
|
15 years ago |
Roberto Ierusalimschy
|
0ac3d07ea6
|
first implementation of lexical environments
|
15 years ago |
Roberto Ierusalimschy
|
5938212748
|
information about upvalues (where they come from) kept in Proto structure,
instead of sequence of pseudo-opcodes after OP_CLOSURE
|
15 years ago |
Roberto Ierusalimschy
|
1829911d7c
|
some operations may shrink g->totalbytes so g->estimate must be
more flexible
|
15 years ago |
Roberto Ierusalimschy
|
bed2cb725a
|
initialize g->estimate (just in case)
|
15 years ago |
Roberto Ierusalimschy
|
ab75bd23ee
|
handle extended opcodes (with OP_EXTRAARG) when checking metamethod
calls
|
15 years ago |
Roberto Ierusalimschy
|
cee94df6e0
|
function list in luaL_register may be NULL for an empty list
|
15 years ago |
Roberto Ierusalimschy
|
fcc46467fa
|
limit of constants per function changed to 2^26 using extra arguments
to opcodes LOADK, GETGLOBAL, and SETGLOBAL
|
15 years ago |
Roberto Ierusalimschy
|
f8e354e240
|
detail (space between function name and its parameter list)
|
15 years ago |
Roberto Ierusalimschy
|
a650378822
|
'cpcall' reimplemented as a predefined value in the registry
|
15 years ago |
Roberto Ierusalimschy
|
99182c6872
|
references must start after predefined values in the registry
|
15 years ago |
Roberto Ierusalimschy
|
67cae2854c
|
'lua_mainthread' replaced by new preregistered value LUA_RIDX_MAINTHREAD
|
15 years ago |
Roberto Ierusalimschy
|
0e45ffb8e4
|
first implementation of 'lua_yieldk' (yield with continuation)
|
15 years ago |
Roberto Ierusalimschy
|
071e54ae0b
|
extra facilities to testC: comments + commands print, callk and yield +
Cfunc with extra upvalues
|
15 years ago |
Roberto Ierusalimschy
|
a68494fb32
|
'debug.[gs]etupvalue' should be able to access C upvalues too
|
15 years ago |
Roberto Ierusalimschy
|
4b44793dc5
|
"bug": lua_createtable with sizes changed (array <-> record)
|
15 years ago |
Roberto Ierusalimschy
|
90e2a716c8
|
details + comments
|
15 years ago |
Roberto Ierusalimschy
|
d9fbbe1f23
|
"file:write" returns "file" in case of success
|
15 years ago |
Roberto Ierusalimschy
|
4a714cebd1
|
API checks now have explanatory messages
|
15 years ago |
Roberto Ierusalimschy
|
5b6be84106
|
ensures that argument 'mode' to 'io.open' matches "[rwa]%+?b?", to
avoid passing invalid modes to 'fopen'.
|
15 years ago |
Roberto Ierusalimschy
|
b2bb2f7f59
|
better machinery to test continuations in the C API
|
15 years ago |
Roberto Ierusalimschy
|
f096ab5421
|
correct way to check arguments to 'strftime'
|
15 years ago |
Roberto Ierusalimschy
|
4e1ffc482a
|
test for 'fs' being NULL at the end of 'close_func' is useless ('fs'
cannot be NULl and the new ls->fs can only be NULL when the token
is EOS, in which case anchor_token has no effect).
|
15 years ago |
Roberto Ierusalimschy
|
bc439e21cd
|
avoid an unprotected call to 'lua_tostring' which theoretically may
cause a panicked exit
|
15 years ago |
Roberto Ierusalimschy
|
1b4480003b
|
"stange numbers" (-0 and NaN) also go to the constant table (as
strings with their binary representation). Therefore, constant
folding may produce these results.
|
15 years ago |
Roberto Ierusalimschy
|
42167804b8
|
luaH_[gs]etnum renamed to luaH_[gs]etint (as they only accept integers,
not generic numbers).
|
15 years ago |
Roberto Ierusalimschy
|
3135a6bbab
|
luaH_[gs]etnum renamed to luaH_[gs]etint (as they only accept integers,
not generic numbers)
|
15 years ago |
Roberto Ierusalimschy
|
95cbc402dc
|
"But" -> "Bug" (misspelling)
|
15 years ago |
Roberto Ierusalimschy
|
decada8dc2
|
'debug.getfenv' does not check whether it has an argument
|
15 years ago |
Roberto Ierusalimschy
|
ee5d03b42a
|
small bug: debug.getfenv should check whether it has an argument
|
15 years ago |
Roberto Ierusalimschy
|
323f33d014
|
'collectgarbage"count"' returns a second argument with the count%1024.
|
15 years ago |
Roberto Ierusalimschy
|
55dc7fb240
|
bug: stack must be cleared until its end (including extra size) +
control of stack size moved to 'ldo.c'
|
16 years ago |
Roberto Ierusalimschy
|
11d7ba79f2
|
tail calls do not need to move whole new frame down, only its slice
up to last parameter (the rest has not been used yet)
|
16 years ago |
Roberto Ierusalimschy
|
bd5fa9cc8f
|
new macro 'getproto'
|
16 years ago |
Roberto Ierusalimschy
|
2658c94925
|
detail (comment: 'correspoding' -> 'corresponding')
|
16 years ago |
Roberto Ierusalimschy
|
215885768c
|
'index2adr' -> 'index2addr' (correct spelling)
|
16 years ago |
Roberto Ierusalimschy
|
a5f68589c0
|
avoid using 'UNUSED' (not defined outside the kernel)
|
16 years ago |
Roberto Ierusalimschy
|
28419c4383
|
'luaL_tolstring' uses 'lua_tolstring' instead of 'lua_tostring' and
'lua_objlen'
|
16 years ago |
Roberto Ierusalimschy
|
5bff2aaf47
|
calls with LUA_MULTRET may leave no free slots in the stack
|
16 years ago |
Roberto Ierusalimschy
|
f76f4cb79d
|
new way to control stack overflow, controling only total size of the stack
|
16 years ago |