 Roberto Ierusalimschy
|
279c3a6961
A few changes in tests about number of bits in integers
|
il y a 6 ans |
|
Roberto Ierusalimschy
|
d12262068d
Small optimizations in range checks
|
il y a 6 ans |
|
Roberto Ierusalimschy
|
f99509581e
Removed extra information from RCS keyword strings
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
2c107e13a8
warning (comparison between signed and unsigned integers)
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
6e600695f8
field 'sizearray' in struct 'Table' changed to 'alimit', which can
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
fb8fa66136
no more 'luaH_emptyobject' and comparisons of addresses of global variables
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
34aa0c5bd7
new macros 'likely'/'unlikely' with hints for jump predictions
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
950fbcb971
detail ('l_castU2S' should only be used over lua_Unsigned values)
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
ef8263f81f
better names for macros for tags and types.
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
9243c414d9
first version of empty entries in tables
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
465b474899
small reorganization of 'luaV_flttointeger'/'luaV_tointeger'
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
1afd5a152d
more generic way to handle 'gclist'
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
e2b15aa21d
janitor work on casts
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
68af7cc81a
another try with table resize.
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
e663a24ab0
more freedom in handling memory-allocation errors (not all allocations
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
7622373033
using explicit tests for allocation overflow whenever possible
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
599f1742c6
detail (typo in comments)
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
26679ea35b
new function 'luaV_flttointeger' to convert floats to integers (without
|
il y a 7 ans |
|
Roberto Ierusalimschy
|
f96497397a
new type 'StackValue' for stack elements
|
il y a 8 ans |
|
Roberto Ierusalimschy
|
73ec04fcf3
no more 'DEADKEY'. Table traversals do not need to consider dead keys;
|
il y a 8 ans |
|
Roberto Ierusalimschy
|
b6f87491af
in hash nodes, keys are stored in separate pieces to avoid wasting
|
il y a 8 ans |
|
Roberto Ierusalimschy
|
e3d52da144
BUG: in 'computesizes', 'twotoi' overflows when a sequence has
|
il y a 8 ans |
|
Roberto Ierusalimschy
|
84910e04e2
better implementation for 'hash_search', without using 'size_t'
|
il y a 8 ans |
|
Roberto Ierusalimschy
|
3d879fbc5d
reimplementation of 'luaH_getn', trying to handle numeric limits
|
il y a 8 ans |
|
Roberto Ierusalimschy
|
b1b7790f7c
detail ('1' -> '1u' in unsigned operation)
|
il y a 8 ans |
|
Roberto Ierusalimschy
|
7b1fba69b7
using 'lastfree == NULL' to signal that table is using the dummy
|
il y a 8 ans |
|
Roberto Ierusalimschy
|
d103312661
details (typos in comments)
|
il y a 9 ans |
|
Roberto Ierusalimschy
|
5100bc8aa1
no need for a special case to get long strings (not that common)
|
il y a 9 ans |
|
Roberto Ierusalimschy
|
7485512384
added comment and assert about dead keys
|
il y a 9 ans |
|
Roberto Ierusalimschy
|
46de77b219
bug: despite its name, 'luaH_getstr' did not work for strings in
|
il y a 9 ans |
