Commit Graph

337 Commits

Author SHA1 Message Date
Roberto Ierusalimschy 00744a2729 'luaO_arith' -> 'luaO_numarith' 2013-04-29 13:57:48 -03:00
Roberto Ierusalimschy 1294b09d8e first implementation of literal integers (no constant folding yet) 2013-04-16 15:46:28 -03:00
Roberto Ierusalimschy 07f861385e first steps in the support of integers: basic representation + table indexing + basic API ops (tointeger/pushinteger) + equality + a few extra stuff 2013-04-15 12:44:46 -03:00
Roberto Ierusalimschy 8f8665fffa 'ttypenv' -> 'ttnov' 2013-04-12 16:07:09 -03:00
Roberto Ierusalimschy cf160abcf9 no more macro 'changenvalue' 2012-09-11 15:21:44 -03:00
Roberto Ierusalimschy 77cbd817d1 better(?) handling of '#define's for IEEE-related tricks + avoid using
IEEE trick for 64-bit integer types (lua_Integer on 64-bit machines)
2012-05-11 11:10:50 -03:00
Roberto Ierusalimschy 3cadc37f47 no more 'Proto' objects on the stack. Protos are anchored on outer
Protos or on a Closure, which must be created before the Proto.
2012-05-08 10:53:33 -03:00
Roberto Ierusalimschy a4b96ce9a3 first implementation of long strings 2012-01-25 19:05:40 -02:00
Roberto Ierusalimschy 291f564485 field 'reserved' -> 'extra' (may be used for other purposes too) 2012-01-23 21:05:51 -02:00
Roberto Ierusalimschy 28f8ed9113 new macro 'checktype' 2012-01-20 20:47:11 -02:00
Roberto Ierusalimschy fd22ccd6d0 object tag keeps variant bits too -> no need for 'isC' field in
Closures + more strick typing for closure variants
2012-01-20 20:05:50 -02:00
Roberto Ierusalimschy 87cf2bf711 extra comments about what fields are optional debug information 2011-10-31 15:48:22 -02:00
Roberto Ierusalimschy 1350a2bcb5 more macros to try to make NaN trick work with other sizes of
IEEE float numbers. (It has not been tested with such different
sizes...)
2011-10-17 12:46:13 -02:00
Roberto Ierusalimschy 87912786af dead objects are not collectable. 2011-09-24 18:12:01 -03:00
Roberto Ierusalimschy 9e89a55a4f NILCONSTANT also depends no endianess (with NaN trick) 2011-07-04 17:29:02 -03:00
Roberto Ierusalimschy fd80e63468 configuration for NaN trick big-endian + macro 'luai_checknum' to
ensure numbers comming from C are not "improper" (some kinds of
signaling NaNs)
2011-06-13 11:13:06 -03:00
Roberto Ierusalimschy 35391d9b1a macro 'NNMARK' (mark for non-number values using the NaN trick) 2011-06-09 15:21:25 -03:00
Roberto Ierusalimschy 99b1b8e918 first version with 'NANTRICK' (packing all Lua values inside a 'double') 2011-06-07 16:02:33 -03:00
Roberto Ierusalimschy 3b44821334 stricter control (using tag variants) over closure kinds (Lua x C) 2011-06-02 16:31:40 -03:00
Roberto Ierusalimschy 0214dab989 yet another macro ('num_') 2011-05-31 16:15:01 -03:00
Roberto Ierusalimschy 9b7dddad7d no need for two different implementations for equality (one raw and
one with metamethods)
2011-05-31 15:24:36 -03:00
Roberto Ierusalimschy 2a59a999d4 yet another macro abstracting TValue representation 2011-05-26 14:12:31 -03:00
Roberto Ierusalimschy 3cf1729a02 new macro 'lua_longassert' that is equivalent to an assertion without
a stringfication of the condition, to avoid too long string literals
(limited by C90 to ~510 characters)
2011-05-05 16:43:14 -03:00
Roberto Ierusalimschy 25b4e35ee8 new macros abstracting TValue representation (to easy change to
other representations)
2011-05-05 12:53:23 -03:00
Roberto Ierusalimschy 7e1facaa4e detail in assertion 2011-05-04 14:04:06 -03:00
Roberto Ierusalimschy ad2531a0ee more complete (and hopefuly more correct) handling of 'sizeof(char)' 2011-05-03 13:01:57 -03:00
Roberto Ierusalimschy a958b711f5 cleaner scheme of bits for variant types (light C functions) and
collectable types
2011-04-07 13:11:57 -03:00
Roberto Ierusalimschy 83abbac946 all names used internally by macros end with '_' 2011-04-05 11:24:07 -03:00
Roberto Ierusalimschy 98816d0ce5 small problems with 'luaone.c' 2011-02-28 14:32:10 -03:00
Roberto Ierusalimschy b63b0928cf prototype for 'luaO_ceillog2' was different from definition 2011-01-07 13:46:27 -02:00
Roberto Ierusalimschy 1aead7f553 'hexafloat' moved to 'lobject.c' (hexa conversion needs it too) 2010-12-10 12:53:15 -02:00
Roberto Ierusalimschy ccc4fc9cf0 detection of erroneous numeric strings with \0 (such as "1\0") 2010-12-06 19:08:36 -02:00
Roberto Ierusalimschy 9b7a12c46d finalizers (__gc) for tables 2010-11-26 12:32:31 -02:00
Roberto Ierusalimschy 78f9635111 several new comments 2010-07-26 12:53:23 -03:00
Roberto Ierusalimschy 575074fd85 Lua closures are cached for reuse 2010-06-04 10:25:10 -03:00
Roberto Ierusalimschy 0bccf03d7e slightly better definition for 'changenvalue' 2010-05-07 15:44:46 -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 7dfa4cd655 first implementation of light C functions 2010-04-14 12:13:48 -03:00
Roberto Ierusalimschy d20ff60615 new macro LUA_NUMTAGS 2010-04-12 13:07:29 -03:00
Roberto Ierusalimschy 064e406f67 no more fenvs! 2010-03-26 17:58:11 -03:00
Roberto Ierusalimschy 25c557ec63 first version of _ENV; no more global variables 2010-03-12 16:14:06 -03:00
Roberto Ierusalimschy 1ebf109c53 'setobj' must assing fields one by one, to allow trick of using
pad space in table keys
2010-01-08 18:00:20 -02:00
Roberto Ierusalimschy 062e809e54 new macros 'LUAI_DDEC'/'LUAI_DDEF' to better control declarations and
definitions of non-static variables
2009-11-19 17:06:52 -02:00
Roberto Ierusalimschy 2583bac3d3 several small changes to simplify changing TValue if needed 2009-11-06 15:09:27 -02:00
Roberto Ierusalimschy 9756f56354 better control over accesses to TValue fields 2009-11-05 15:43:54 -02:00
Roberto Ierusalimschy 0ac3d07ea6 first implementation of lexical environments 2009-09-30 12:38:37 -03:00
Roberto Ierusalimschy 5938212748 information about upvalues (where they come from) kept in Proto structure,
instead of sequence of pseudo-opcodes after OP_CLOSURE
2009-09-28 13:32:50 -03:00
Roberto Ierusalimschy bd5fa9cc8f new macro 'getproto' 2009-07-15 15:37:19 -03:00
Roberto Ierusalimschy a36c8e1718 new macro 'changenvalue' 2009-06-18 13:36:40 -03:00
Roberto Ierusalimschy 6a024df440 new function 'luaO_arith' to perform basic arithmetic operations 2009-06-17 14:48:34 -03:00
Roberto Ierusalimschy 3dbf305408 no more support for old-style varargs 2009-03-26 09:57:01 -03:00
Roberto Ierusalimschy bb92ef23cb bug: string at the end of TString must consider the entire TString
entity, not only its 'tsv' part
2008-08-05 16:26:23 -03:00
Roberto Ierusalimschy a76fa25199 'ceillog2' now is exported (other modules may need it) 2008-01-30 16:05:23 -02:00
Roberto Ierusalimschy 94d40f3980 luaO_log2 can be private to ltable.c 2007-04-10 09:18:17 -03:00
Roberto Ierusalimschy a7c9e45c64 avoid trailing white spaces 2006-09-11 11:07:24 -03:00
Roberto Ierusalimschy 440113bbe8 unused macro removed 2006-01-18 09:37:34 -02:00
Roberto Ierusalimschy ffb798e1e2 avoids type punning for table keys 2006-01-10 10:51:53 -02:00
Roberto Ierusalimschy c2bd68d2c3 detail 2005-10-24 15:37:33 -02:00
Roberto Ierusalimschy 0d1685d751 bug: last tag is not the number of tags... 2005-06-13 11:19:00 -03:00
Roberto Ierusalimschy bf2b342ac1 better control for compatibility code for varargs 2005-06-13 11:15:54 -03:00
Roberto Ierusalimschy d987cf1f81 new mark LUAI_DATA for extern data 2005-06-06 10:30:25 -03:00
Roberto Ierusalimschy 90de38bf1f warnings in VS .Net 2005-05-31 11:25:18 -03:00
Roberto Ierusalimschy 1ab2b93462 debug information for last line of a function definition 2005-05-05 17:47:02 -03:00
Roberto Ierusalimschy 8718fda9b2 added LUAI_FUNC to functions not in the API 2005-04-25 16:24:10 -03:00
Roberto Ierusalimschy 7d45a5f48f C functions and userdata also have environments 2005-02-18 10:40:02 -02:00
Roberto Ierusalimschy 334ba8132b cleaner way to remark open upvalues 2005-01-18 15:18:09 -02:00
Roberto Ierusalimschy e2498e079e change in hash algorithm so that it does not need empty slot
(tables can be 100% full)
2005-01-05 16:20:51 -02:00
Roberto Ierusalimschy 531874f6ce better (correct?) way to handle `next' over removed keys 2004-12-04 16:10:22 -02:00
Roberto Ierusalimschy 737ec947d3 better implementation for `floating-point bytes' 2004-11-01 12:06:50 -03:00
Roberto Ierusalimschy bd38017ddf small optimization for table size in machines with double allignment 2004-10-06 15:34:16 -03:00
Roberto Ierusalimschy 616438fe9a new way to use `vararg' parameters (with `...') 2004-05-31 15:51:50 -03:00
Roberto Ierusalimschy a4e1230f95 better way to control open upvalues 2004-03-15 18:04:33 -03:00
Roberto Ierusalimschy 898e8a6794 added include to stdarg.h (as it uses vararg facilities) 2004-02-20 13:01:05 -03:00
Roberto Ierusalimschy 2aaf7394ad more and better tools (assertions & inspectors) to check incremental GC 2004-02-16 16:09:52 -03:00
Roberto Ierusalimschy 47fc57a252 `TObject' renamed to `TValue' + other name changes and better assertions
for incremental garbage collection
2003-12-10 10:13:36 -02:00
Roberto Ierusalimschy 366e4af3c9 towards incremental GC 2003-11-18 12:55:11 -02:00
Roberto Ierusalimschy 8332d5c8a5 parser fully reentrant(!) 2003-08-27 18:01:44 -03:00
Roberto Ierusalimschy 572a69df78 Lua does not need all those different types... 2003-04-28 16:26:16 -03:00
Roberto Ierusalimschy ab9e062928 name changes to avoid name collision between macros and variables 2003-03-18 09:50:04 -03:00
Roberto Ierusalimschy 60c83ded30 small optimization for sizes of array constructors 2003-02-18 13:02:56 -03:00
Roberto Ierusalimschy 7a40cdbda0 different variables for number of upvalues and size of upvalue array
(makes code clearer)
2003-02-11 08:46:24 -02:00
Roberto Ierusalimschy dd8edecae1 new functions to manipulate upvales (get/setupvalue) 2002-12-19 09:11:55 -02:00
Roberto Ierusalimschy b40c9f7a3c better code to signal write barriers 2002-11-14 14:16:21 -02:00
Roberto Ierusalimschy 5c5d9b2703 back to `__mode' metafield to specify weakness 2002-11-14 10:01:35 -02:00
Roberto Ierusalimschy 3010eb0536 all objects with several children (tables, closures, stacks, prototypes)
go to `gray' queue
2002-11-13 09:49:19 -02:00
Roberto Ierusalimschy dff9be4224 new macros to distinguish different types of object moves (for future GC
evolution).
2002-11-07 13:37:10 -02:00
Roberto Ierusalimschy 6820da5096 detail 2002-11-04 10:31:44 -02:00
Roberto Ierusalimschy 96e15b8501 threads now are real Lua objects, subject to garbage collection 2002-10-25 17:05:28 -03:00
Roberto Ierusalimschy 84e0b1bc97 small changes in type configuration facilities 2002-10-22 14:18:28 -03:00
Roberto Ierusalimschy c196348717 in case of memory allocation errors, sizecode and sizelineinfo can
be different
2002-10-16 17:40:58 -03:00
Roberto Ierusalimschy b3d0682fb9 use of different buffers for scanner and concatenation 2002-10-08 15:46:08 -03:00
Roberto Ierusalimschy 565e6d74e1 state's buffer is used only for chars 2002-09-19 10:03:53 -03:00
Roberto Ierusalimschy 4964e7c8a0 details 2002-09-02 16:54:49 -03:00
Roberto Ierusalimschy fdafd4f4a8 new structure for collectable objects, sharing a common header 2002-08-30 16:09:21 -03:00
Roberto Ierusalimschy da19c436cc cleaning the stage for generational collection 2002-08-16 11:45:55 -03:00
Roberto Ierusalimschy 634344d61f new API for weak mode 2002-08-06 14:06:56 -03:00
Roberto Ierusalimschy 5037196f6f new macros `ttis*' 2002-08-05 11:50:39 -03:00
Roberto Ierusalimschy 79c8edb6c4 new names for light userdata operations 2002-07-17 13:25:13 -03:00
Roberto Ierusalimschy 9f4b5b5232 weak keys are removed only after finalization 2002-07-01 14:06:58 -03:00
Roberto Ierusalimschy fdfd5b44ee TM_GETTABLE/TM_SETTABLE don't need fast access anymore 2002-06-24 17:18:38 -03:00
Roberto Ierusalimschy e34f282365 `luaV_gettable' returns element position 2002-06-24 10:08:45 -03:00
Roberto Ierusalimschy 3941af53ad first implementation of independent global table per function 2002-06-20 17:41:46 -03:00
Roberto Ierusalimschy 864c96f36c new fallback for equality `__eq' 2002-06-13 10:39:55 -03:00
Roberto Ierusalimschy eeab473fc8 new fallback __le (less equal), for partial order 2002-06-12 11:56:22 -03:00
Roberto Ierusalimschy 955def0348 new names for string formating functions 2002-05-16 15:39:46 -03:00
Roberto Ierusalimschy b7a0503c1d new format for error messages 2002-05-15 15:57:44 -03:00
Roberto Ierusalimschy dea6b6da94 new function `lua_vpushstr' to replace uses of `sprintf' 2002-05-07 14:36:56 -03:00
Roberto Ierusalimschy 0dbf0c5953 new format for test intructions (handle NaN correctly) 2002-05-06 12:51:41 -03:00
Roberto Ierusalimschy 237969724f support for `light' userdata + simpler support for `boxed' udata 2002-04-05 15:54:31 -03:00
Roberto Ierusalimschy 801aaf37b1 simpler implementation for line information 2002-03-25 14:47:14 -03:00
Roberto Ierusalimschy b7ed502dea configurable macros should live in llimits.h 2002-03-18 15:16:16 -03:00
Roberto Ierusalimschy 71c716e5a8 avoid C identifiers beginning with '_' 2002-03-11 09:45:00 -03:00
Roberto Ierusalimschy 8a1a512c64 lu_byte is enough for those fields (maxstack < 256) 2002-03-05 09:42:47 -03:00
Roberto Ierusalimschy e01f5e6809 better order of record fields for 64-bit machines 2002-02-08 20:42:41 -02:00
Roberto Ierusalimschy d6fd33e76f `eventtable' renamed to `metatable' 2002-01-30 15:26:44 -02:00
Roberto Ierusalimschy b3bb0f132b new interface for weak modes 2002-01-09 19:50:35 -02:00
Roberto Ierusalimschy e04f7ed450 first version of Lua "stackless" 2001-12-18 18:52:30 -02:00
Roberto Ierusalimschy 9aff171f3b new type `boolean' 2001-12-11 20:48:44 -02:00
Roberto Ierusalimschy 592a309177 tag system replaced by event tables 2001-12-05 18:15:18 -02:00
Roberto Ierusalimschy 413fc7334b new implementation for lua upvalues (sugested by E.T.): simpler and solves
a bug for multi-stacks
2001-11-29 18:22:22 -02:00
Roberto Ierusalimschy 72659a0605 no more explicit support for wide-chars; too much troble... 2001-11-28 18:13:13 -02:00
Roberto Ierusalimschy 26bf2adace optimizations for space in LClosures and time cleanning weak tables 2001-11-06 19:41:53 -02:00
Roberto Ierusalimschy 21aa7e55f2 optimization for array part of a Table 2001-10-25 17:14:14 -02:00
Roberto Ierusalimschy 15462edb0f new definitions for closure structures 2001-10-02 13:45:03 -03:00
Roberto Ierusalimschy 8e9b1e4ae9 details 2001-09-25 14:08:46 -03:00
Roberto Ierusalimschy abdbe883a8 first implementation of unrestricted static scoping 2001-09-07 14:39:10 -03:00
Roberto Ierusalimschy e1d072571e better syntax for type casts 2001-08-31 16:46:07 -03:00
Roberto Ierusalimschy c3d72096c4 use a table to find (and reuse) constants when parsing 2001-08-27 12:16:28 -03:00
Roberto Ierusalimschy 770954510f rename of `kproto' to `p' 2001-06-28 11:57:17 -03:00
Roberto Ierusalimschy b346834a09 new macros for changing numbers 2001-06-28 11:48:44 -03:00
Roberto Ierusalimschy 37f3a1c045 too much optimization to "break" keys in tables; keep them as TObjects... 2001-06-26 10:20:45 -03:00
Roberto Ierusalimschy 8e586c13fc cleaner way to ensure alignment for strings and userdata 2001-06-15 17:36:57 -03:00
Roberto Ierusalimschy ba11831d35 smaller structs for udata and for strings 2001-06-07 12:01:21 -03:00
Roberto Ierusalimschy d5b83ead90 new implementation for userdatas, without `keys' 2001-06-06 15:00:19 -03:00
Roberto Ierusalimschy 762d059a13 new implementation for the Virtual Machine 2001-06-05 15:17:01 -03:00
Roberto Ierusalimschy 0e0e4a480e first implementation for weak tables 2001-04-11 11:42:41 -03:00
Roberto Ierusalimschy 6048c4f74d better way to link callinfo's and stack 2001-03-07 15:09:25 -03:00
Roberto Ierusalimschy 7b84f9e65c lower-case for macros with arguments 2001-03-02 14:27:50 -03:00
Roberto Ierusalimschy 1e40b4dc61 buffer should be void *, as char now is not that neutral...
+ little bug in pointer arithmetic
2001-02-23 17:32:32 -03:00
Roberto Ierusalimschy 39b7978329 first (big) step to support wide chars 2001-02-23 14:17:25 -03:00
Roberto Ierusalimschy 888f91fa24 code check for upvalues 2001-02-20 15:28:11 -03:00
Roberto Ierusalimschy 099442c41f better separation between basic types 2001-02-20 15:15:33 -03:00
Roberto Ierusalimschy d2e340f467 string pointers are always fully aligned 2001-02-09 18:22:29 -02:00
Roberto Ierusalimschy 1f9e3731d1 back to the basics (well-behaved variant record...) 2001-02-02 14:32:00 -02:00
Roberto Ierusalimschy 426d3e43bd lock/unlock may use L + better structure for internal debug stuff 2001-02-02 13:13:05 -02:00
Roberto Ierusalimschy e506b864cd no need for tags in boxed values :-( 2001-02-01 15:40:48 -02:00
Roberto Ierusalimschy 63a822c8e1 all boxed types start with their tags 2001-01-29 17:34:02 -02:00
Roberto Ierusalimschy 6b71a9cfe5 smaller tables for machines with 8-bit alignment 2001-01-29 15:17:26 -02:00
Roberto Ierusalimschy b82242d4c4 detail 2001-01-26 13:58:50 -02:00
Roberto Ierusalimschy a53d9b66ca first implementation for type names 2001-01-25 14:45:36 -02:00