lua-l archive

Search lua-l

This index contains 143,615 documents and 1,774,615 keywords. Last update on 2023-03-09 .

61. Re: Possible enhancements for Lua 5.2 (score: 18)

Author: steve donovan <steve.j.donovan@...>

Date: Sun, 26 Jun 2011 15:49:25 +0200

The anti-feeling was too strong; it was felt that it was getting too close to the dense symbol-heavy syntax of Perl (or even C). As Dirk puts it in that post, the Pascal keyword-driven model has serv

62. Re: Possible enhancements for Lua 5.2 (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Sun, 26 Jun 2011 09:31:19 -0400

<snip> While we're at it, what happened to the lightweight syntax for functions that was discussed in a long thread some time ago [1]? More generally, are token-filter hooks being considered in any

63. Re: question about cdata type identification (score: 18)

Author: Mike Pall <mikelu-1104@...>

Date: Mon, 18 Apr 2011 19:30:47 +0200

I guess he wants to e.g. multiply a matrix with either a number, a complex number or another matrix. And all of that from the same __mul metamethod. Currently there's no easy way to distinguish the f

64. Re: question about cdata type identification (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Mon, 18 Apr 2011 13:13:40 -0400

That's fine for getting, but not for setting (__call cannot be used as l-value). Moreover, you might want to reserve __call for other semantics: in Numlua, m(first, last, step) returns a slice of ma

65. Re: question about cdata type identification (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Mon, 18 Apr 2011 12:16:18 -0400

<snip> I might be missing something, but why don't you just store this information in the matrix struct? I mean, what do you need this information through a 'type' function? <snip> If getmetatable re

66. Re: Lua language extension for complex number and vector values (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Mon, 4 Apr 2011 14:42:26 -0400

Right: order matters for in-place operations. Cheers, Luis -- Computers are useless. They can only give you answers. -- Pablo Picasso -- Luis Carvalho (Kozure) lua -e 'print((("lexcarvalho@NO.gmail.

67. Re: Lua language extension for complex number and vector values (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Mon, 4 Apr 2011 14:32:53 -0400

What about an "operation mode" switch to in-place operations? false 0.54030230586814+0.8414709848079i 0+1i 0.54030230586814+0.8414709848079i 0.54030230586814+0.8414709848079i 1.1438356437916+1.27988

68. Re: Lua language extension for complex number and vector values (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Mon, 4 Apr 2011 10:15:00 -0400

One way of doing this in Numlua 0.3 is to use the optional "in-place" argument to complex functions: -1.1312043837568+2.4717266720048i 1+2i -1.1312043837568+2.4717266720048i -1.1312043837568+2.471726

69. Re: Possible bug with the length operator (score: 18)

Author: James Graves <ansible@...>

Date: Sat, 2 Apr 2011 10:48:36 -0600

Here's my answer: The length of a table which has holes in it has different possible meanings. If Lua's default length operation is not satisfactory for your application, then what you really need is

70. Re: NumLua (score: 332)

Author: Luis Carvalho <lexcarvalho@...>

Date: Fri, 1 Apr 2011 10:17:38 -0400

As they say, absence of evidence is not evidence of absence. :) I guess I tend to follow the Lua style of commiting to a personal repo before pushing to the main branch. Numlua 0.3 is not shelved, b

71. Re: anything wrong by not using "module" ? (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Sun, 27 Mar 2011 16:09:48 -0400

Since we're off-topic, I'd be more surprised to *not* use something like local abs, sqrt = math.abs, math.sqrt norm2 = function (V) local v1, v2 = abs(V[1]), abs(V[2]) if v1 < v2 then v1, v2 = v2, v

72. Re: Perhaps VimOutliner might use Lua (score: 86)

Author: Emmanuel Oga <emmanueloga@...>

Date: Thu, 24 Feb 2011 01:56:24 -0300

Sure, make sure you have vim 7.3 installed. Check it out doing: vim --version (displays lots of info) and then: vim --version | grep lua (it should output +lua). if you have 7.3 but it says "-lua", y

73. Re: LuaJIT2 performance for number crunching (score: 18)

Author: Mike Pall <mikelu-1102@...>

Date: Wed, 23 Feb 2011 11:25:27 +0100

Currently the multiply is neither strength-reduced nor narrowed, so it's kind of costly. If this code is performance-sensitive, you'll probably want to dispatch to a dynamically specialized version o

74. Re: LuaJIT2 performance for number crunching (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Tue, 22 Feb 2011 23:15:37 -0500

<snip> Talking about vectors, how does luajit fare when dealing with strided vectors? Simple example: local ffi = require "ffi" ffi.cdef"typedef struct { int size, stride; double *data; } Vector;" lo

75. [ANN] LuaPSQL (was: Lua and Postgres) (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Sat, 15 Jan 2011 22:28:19 -0500

<snip> I've had a libpq binding for some time, but haven't released it yet since I've had no time to put together some decent documentation. Given the recent need and demand, I just decided to finall

76. Re: Self-awareness of functions? (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Thu, 30 Dec 2010 11:57:12 -0500

<snip> I missed the "C" part. Here's one solution, just for completeness: /* self.c */ include <lua.h> static int self (lua_State *L) { int n = lua_gettop(L); lua_Debug ar; lua_getstack(L, 1, &ar); /

77. Re: Self-awareness of functions? (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Wed, 29 Dec 2010 10:41:07 -0500

<snip> What about this? fact = function (n) if n < 2 then return 1 end local self = debug.getinfo(1, "f").func return n * self(n - 1) end Cheers, Luis -- Computers are useless. They can only give yo

78. Re: Database of storing pure lua objects (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Tue, 28 Dec 2010 10:15:28 -0500

If you don't mind using PostgreSQL, one solution is to use PL/Lua: test=# create table beowulf (state text, tag text); CREATE TABLE test=# insert into beowulf values ('foo', '{name="beowulf"}'); INS

79. Short lambdas, promises, and lazy lists (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Sat, 25 Dec 2010 14:14:19 -0500

Hi, I took the laziness from this end of year holiday season to play a little with the token filter patch, especially with the short lambda syntax. Here's a short write-up of my findings; I'm sharing

80. Re: Speed of #t (score: 297)

Author: Luis Carvalho <lexcarvalho@...>

Date: Mon, 13 Dec 2010 10:05:45 -0500

If you have sparse arrays then it's probably better to use adjacency lists: w = true -- or any weight G = {a = {b = w}} 1. Any node in the graph: a = next(G) 2. Any neighbor of a: b = next(G[a]) Che