nik/lush
1
1 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
3,287 Commits 1 Branch 0 Tags
4aaf0ef75c7b8f617f31a2c3794e97dee899dc08
Commit Graph
100 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Roberto Ierusalimschy 4aaf0ef75c "to-be-finalized" objects marked and sweeped like all other objects 2010-04-20 17:15:30 -03:00
Roberto Ierusalimschy ed720d09f3 upvalue barriers for Lua functions must act on the upvalue itself, 
not on its closure
 2010-04-20 17:14:50 -03:00
Roberto Ierusalimschy 376e939ef6 missing parentheses around 'luaL_pushresultsize' declaration 2010-04-19 15:52:15 -03:00
Roberto Ierusalimschy 9d28e2b3e5 bug with io.read(op, "*n") 2010-04-19 15:04:58 -03:00
Roberto Ierusalimschy 49ca1f7083 allows thread switches (when/if there are thread switches) in the same 
places that finalizers can run (so they should be safe...)
 2010-04-19 14:58:46 -03:00
Roberto Ierusalimschy 4db6f20770 ensure that 'luai_userstatethread' is always called (even if 
'stack_init' throws a memory error)
 2010-04-19 14:40:13 -03:00
Roberto Ierusalimschy cca71912e6 'gcinfo' was deprecated in version 5.0. 2010-04-19 14:02:02 -03:00
Roberto Ierusalimschy 8003e9544d 'string.format' may get buffer as an argument when there are 
missing arguments and format string is too long
 2010-04-19 13:39:25 -03:00
Roberto Ierusalimschy 973d81efb3 complete control over number of each kind of object allocated 2010-04-19 13:38:25 -03:00
Roberto Ierusalimschy 26d4a73962 with light C functions, 'pairs' does not need to keep 'next' as an 
upvalue.
 2010-04-19 13:36:06 -03:00
Roberto Ierusalimschy 99a39f6f4a tells the allocation function that the first block is a thread 2010-04-19 13:34:46 -03:00
Roberto Ierusalimschy 1bb407e494 in lua_gc/step, 'luaC_step' changes GCthreshold, so there was little 
control over the real step size.
 2010-04-19 13:33:19 -03:00
Roberto Ierusalimschy 53f9499f7f "light C function" is a better name than "C-function pointer" 2010-04-18 10:22:48 -03:00
Roberto Ierusalimschy 575befc394 comparsion tag methods follow the same rule as other binary TMs 2010-04-18 10:15:11 -03:00
Roberto Ierusalimschy e7ffb1b501 macro LUA_USE_LONGJMP allows choosing longjmp when compiling Lua as 
C++ code
 2010-04-18 09:41:35 -03:00
Roberto Ierusalimschy c066ac10a9 better line numbers for function calls and unary/binary operators + 
null statement + no more "ambiguous syntax" restriction
 2010-04-17 09:46:32 -03:00
Roberto Ierusalimschy 1e6940f291 'gcstate' now also runs collector until given state + small changes 
in 'testC' to test 'lua_topointer' and 'lua_tocfunction'
 2010-04-16 14:42:49 -03:00
Roberto Ierusalimschy d8d81ba891 new escape sequence '\*' + several comments + moving options from 
switch default into cases (as now locale is fixed)
 2010-04-16 09:31:07 -03:00
Roberto Ierusalimschy cf22133b69 no need to avoid calling ctype functions as now they are implemented 
by us (no inefficiencies due to accessing locale information)
 2010-04-15 16:44:43 -03:00
Roberto Ierusalimschy b7be05ad27 invalid instructions "cannot" happen 2010-04-15 16:43:43 -03:00
Roberto Ierusalimschy afdb19ac82 no more 'ccall' nor 'cpcall' functions. (With light C functions they 
are obsolete.)
 2010-04-14 12:14:21 -03:00
Roberto Ierusalimschy 7dfa4cd655 first implementation of light C functions 2010-04-14 12:13:48 -03:00
Roberto Ierusalimschy fc6203ee43 BUG: 'string.format' may get buffer when there are missing arguments 2010-04-12 13:13:02 -03:00
Roberto Ierusalimschy b2dd246b7a better control in 'totalmem' over choosing counters 2010-04-12 13:07:39 -03:00
Roberto Ierusalimschy d20ff60615 new macro LUA_NUMTAGS 2010-04-12 13:07:29 -03:00
Roberto Ierusalimschy d41b467320 check memory allows strings to live in the main GC list (it should 
work ok).
 2010-04-12 09:42:07 -03:00
Roberto Ierusalimschy 4541243355 patterns now accept '\0' as a regular character 2010-04-12 09:00:50 -03:00
Roberto Ierusalimschy 9100f7479a new implementation for Generic Buffer manipulation (using userdata as 
temporary buffer space)
 2010-04-09 13:14:46 -03:00
Roberto Ierusalimschy 055104f5b6 keep memory-error message in the global state, so that its use 
does not depend on Lua internalizing strings to avoid a string
creation on memory errors
 2010-04-08 14:16:46 -03:00
Roberto Ierusalimschy 28aa733c15 macro 'eqstr' was being used to compare non-string entities 2010-04-08 14:06:33 -03:00
Roberto Ierusalimschy 6abde1b05a no need to keep "_ENV" name in global state (can be kept in lex state) 2010-04-05 13:35:37 -03:00
Roberto Ierusalimschy d394d5536a new macro 'eqstr' 2010-04-05 13:26:37 -03:00
Roberto Ierusalimschy a2f98314a3 'lua_pushstring' now uses 'luaS_new' 2010-04-05 11:21:38 -03:00
Roberto Ierusalimschy a09c8d7beb cannot sweep main thread in sweepstring phase 2010-04-05 11:15:35 -03:00
Roberto Ierusalimschy f80ff7a336 'luaS_new' changed from macro to function 2010-04-03 17:24:18 -03:00
Roberto Ierusalimschy a08fc34ee4 avoid using 'luaS_new' when can use 'luaS_newlstr' 2010-04-02 12:39:07 -03:00
Roberto Ierusalimschy d00d2eaf51 small changes in 'luaO_pushvfstring' 2010-04-02 12:30:27 -03:00
Roberto Ierusalimschy 11126422d9 option to return GC to normal (incremental, non generational) mode 2010-04-02 12:19:19 -03:00
Roberto Ierusalimschy e54668b696 added proper code to trace garbage collection 2010-04-02 11:37:41 -03:00
Roberto Ierusalimschy 9eb832de54 work related to hooks moved from 'luaV_execute' to 'traceexec' 2010-03-29 17:45:49 -03:00
Roberto Ierusalimschy 82a1f5d354 comments 2010-03-29 14:44:31 -03:00
Roberto Ierusalimschy a8d3aa14fd global table now is only kept in the registry 2010-03-29 14:43:14 -03:00
Roberto Ierusalimschy 064e406f67 no more fenvs! 2010-03-26 17:58:11 -03:00
Roberto Ierusalimschy 5c87f61e6b major collections in generational mode 2010-03-25 16:37:23 -03:00
Roberto Ierusalimschy 3aa9598177 'mainthread' is not inserted in the 'allgc' list anymore, but swept 
separately.
 2010-03-25 10:06:36 -03:00
Roberto Ierusalimschy 64d39ed1b6 generational mode no longer sweep old objects 2010-03-24 12:51:10 -03:00
Roberto Ierusalimschy 4433dbb5f5 userdata with finalizers are kept in a separated list ('udgc'), instead 
of at the end of 'rootgc' (which was renamed to 'allgc', as it is not
"root" in the usual meaning for collectors)
 2010-03-24 10:07:01 -03:00
Roberto Ierusalimschy 5cb128ea54 BUG: emergency collector might resize 'strt' (string table) when 
creating a new string
+ atomic and markroot steps has some cost
+ full collection must leave collector in proper state when mode
is generational
 2010-03-23 17:16:06 -03:00
Roberto Ierusalimschy 74123e9686 draft version of a generational mode for garbage collection. (Not well 
tested; no major collections; ...)
 2010-03-22 15:28:03 -03:00
Roberto Ierusalimschy 9c196bebad detail (wrong comment about LUAI_GCPAUSE) 2010-03-22 14:45:55 -03:00
Roberto Ierusalimschy 28b2566c97 'lua_assert' not visible from libraries 2010-03-22 14:28:31 -03:00
Roberto Ierusalimschy caf74dd731 'cpcall' renamed to 'ccall' as it does not do a protected call 2010-03-19 18:04:17 -03:00
Roberto Ierusalimschy 62840c5fad 'ipairs' is deprecated 2010-03-19 12:52:48 -03:00
Roberto Ierusalimschy 5ca7cdd709 for compatibility only: 'module' changes the first upvalue of 
calling function to affect its environment
 2010-03-19 12:02:34 -03:00
Roberto Ierusalimschy 1514e49d43 avoid using function environments in C libraries (as it probably will 
be deprecated)
 2010-03-17 18:37:37 -03:00
Roberto Ierusalimschy 22ef84b6c8 '_ENV' name permanently stored in global state for easier access 2010-03-13 12:55:42 -03:00
Roberto Ierusalimschy 63a2b62468 must include 'string.h' because a macro uses 'strlen' 2010-03-13 12:55:01 -03:00
Roberto Ierusalimschy 489253d753 better definitions for lua_[gs]etglobal + less uses of ENVIRONINDEX 2010-03-13 00:57:46 -03:00
Roberto Ierusalimschy 25c557ec63 first version of _ENV; no more global variables 2010-03-12 16:14:06 -03:00
Roberto Ierusalimschy f292760f12 small optimization in luaL_addlstring (avoid adding chars one by one) 
(suggested by Chuck Coffing)
 2010-03-12 15:59:32 -03:00
Roberto Ierusalimschy 1124cb1247 first step towards _ENV: all chunks have an puvalues _ENV with the 
global table
 2010-03-08 13:55:52 -03:00
Roberto Ierusalimschy c295147874 when finding a 'name' for a function, handle the case when the function 
is a for iterator
 2010-03-05 11:01:29 -03:00
Roberto Ierusalimschy e560ac4862 when searching for a variable name, look existing upvalues before 
goingg to upper levels
 2010-03-04 15:12:57 -03:00
Roberto Ierusalimschy 915a9a912c just in case, better to flush error messages 2010-03-03 15:53:02 -03:00
Roberto Ierusalimschy ae0a5e2142 new option '*L' for io.read + options for io.lines 2010-03-03 15:48:57 -03:00
Roberto Ierusalimschy 347f0c33d1 avoid 'continue' unless necessary 2010-02-27 18:16:24 -03:00
Roberto Ierusalimschy 622e448710 removed useless test (argv[i] cannot be NULL when i<argc) 2010-02-27 18:15:36 -03:00
Roberto Ierusalimschy 0fe2576a39 new instructions to optimize indexing on upvalues 2010-02-26 17:40:29 -03:00
Roberto Ierusalimschy d08d237a49 detail: in loadfile read function, no need to return NULL on EOF; 
size ==0 is enough to signal EOF.
 2010-02-18 17:37:57 -02:00
Roberto Ierusalimschy 64a7ec987c avoid using 'ungetc' in loadfile 2010-02-18 17:32:41 -02:00
Roberto Ierusalimschy 4274738e81 new macro 'luai_writestringerror' 2010-02-18 17:18:41 -02:00
Roberto Ierusalimschy 0d7d559dcc no need to flush either stderr or \n-terminated outputs 2010-02-11 15:12:27 -02:00
Roberto Ierusalimschy af512ad6da use of 'conventional' names for shift and rotate operations + 
right/left versions for them
 2010-02-11 13:55:29 -02:00
Roberto Ierusalimschy d27108ccd5 removed support for '#fist-line comment' on binary files (as binary 
files do not have lines...)
 2010-02-11 13:52:50 -02:00
Roberto Ierusalimschy a838b3b496 better usage messages, showing entire offending argument 2010-02-09 09:58:57 -02:00
Roberto Ierusalimschy 4c54cd3a10 when yielding, original 'func' value must be kept and restored so 
that 'poscall' puts results in the right slot.
 2010-02-09 09:56:29 -02:00
Roberto Ierusalimschy c8ff7de7f0 field 'oldtop' renamed to 'extra', as it can be used for other 
purposes
 2010-02-09 09:55:37 -02:00
Roberto Ierusalimschy 3bae8e047c typo (thanks to Gavin) 2010-02-05 17:09:09 -02:00
Roberto Ierusalimschy 4f5f2fe367 new function 'luaL_cpcall' 2010-01-21 14:49:21 -02:00
Roberto Ierusalimschy 01586d539e typo in comment 2010-01-21 14:31:24 -02:00
Roberto Ierusalimschy c8ab34ab7a better messages for invalid options 2010-01-21 14:31:06 -02:00
Roberto Ierusalimschy 36ab1ee10c __unm metamethod gets nil as its 2nd parameter 2010-01-15 14:23:58 -02:00
Roberto Ierusalimschy 193ce46970 table.pack was locking last result, avoiding its collection 2010-01-13 17:59:10 -02:00
Roberto Ierusalimschy b678f246a4 HINSTANCE -> HMODULE (they are the same thing, but the MS documentation 
uses the latter) + LoadLibrary -> LoadLibraryEx with optional arguments,
to allow the option LOAD_WITH_ALTERED_SEARCH_PATH
 2010-01-13 14:30:27 -02:00
Roberto Ierusalimschy f270e7f044 warnings that are not compatible with C++ separated from other warning 
options
 2010-01-13 14:19:54 -02:00
Roberto Ierusalimschy 070d3743a7 "no value" added to array luaT_typenames + occurrences of "userdata" 
in that array unified in a single address
 2010-01-13 14:18:25 -02:00
Roberto Ierusalimschy 0c7de97d47 correct error message when yielding from outside a coroutine 2010-01-13 14:17:32 -02:00
Roberto Ierusalimschy 900b7d4514 on 64-bit machines, an address may not fit into a 'long', so it 
is better to convert from pointer to void to pointer to function
directly, even if ANSI C does not like it.
 2010-01-13 14:09:05 -02:00
Roberto Ierusalimschy 91f0b20a0e missing standard defines 2010-01-12 17:40:02 -02:00
Roberto Ierusalimschy e4b4623abc 'searchpath' changes dots into directory separators 2010-01-11 15:55:25 -02:00
Roberto Ierusalimschy cef96b73e1 added casts from int to enumerations to follow C++ rules 2010-01-11 15:38:30 -02:00
Roberto Ierusalimschy 427e01eb63 removed useless initialization 2010-01-11 15:37:59 -02:00
Roberto Ierusalimschy d320c908d0 conversion from pointer to void to pointer to function seems more 
"correct" if done through an integral type (because conversions
between pointers and numbers are allowed, only "implementation
defined").
 2010-01-11 15:34:59 -02:00
Roberto Ierusalimschy 8bfc696676 missing 'external' in variable declaration 2010-01-11 15:33:09 -02:00
Roberto Ierusalimschy 97a35e70af pathes corrected to 5.2 2010-01-11 15:15:30 -02:00
Roberto Ierusalimschy 03b556b963 corrected copyright notice to 2010 2010-01-11 15:15:11 -02:00
Roberto Ierusalimschy 572b94bbcb include of 'windows.h' moved to where it is needed 2010-01-11 15:11:24 -02:00
Roberto Ierusalimschy 692209fa8d Dynamic C functions should be created with the global table as their 
initial environments
 2010-01-11 15:06:31 -02:00
Roberto Ierusalimschy 953d499ea2 changed macro lua_number2int for Windows (added brackets; some recent 
compilers complain about the old format) + added missing definition
for 'lua_number2int' in Windows case.
 2010-01-11 14:35:20 -02:00
Roberto Ierusalimschy a541ada194 "-"-trick in C packages now accept name-v13... 2010-01-11 14:10:47 -02:00