| --- Log opened Wed Mar 12 00:00:34 2008 |
| 00:12 | | You're now known as TheWatcher[T-2] |
| 00:17 | | You're now known as TheWatcher[zZzZ] |
| 01:25 | | Vornicus [~vorn@ServicesOp.Nightstar.Net] has quit [Ping Timeout] |
| 01:26 | | Vornotron [~vorn@Admin.Nightstar.Net] has joined #code |
| 02:42 | <@ToxicFrog> | Moltare: if it's accepting, say, 1=3, and it shouldn't, then this is a problem with the parse stage that should be fixed |
| 02:52 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Quit: Z?] |
| 04:23 | < C_tiger> | On the other hand, if it's accepting foo = (bar = 5) ... that's legal in some languages. |
| 04:24 | <@ToxicFrog> | More generally, my point is "if it's accepting things it shouldn't, this is a flaw in either your parser, or your formalization of the language grammar you used to construct the parser" |
| 04:25 | < C_tiger> | I'm never sure what the difference between parse and compile is. |
| 04:25 | <@ToxicFrog> | A parser takes a stream of tokens and turns them into an abstract representation of a program (or some other data, such as an XML tree) |
| 04:26 | <@ToxicFrog> | A compiler typically contains a parser. |
| 04:26 | < C_tiger> | Ok, but it does some of the substitution that happens at "compile time" like global replacement. |
| 04:26 | <@ToxicFrog> | It first lexes the code (turning a character stream into a token stream); then parses the token stream; and then uses a code generator on the intermediate representation to generate object code or bytecode. |
| 04:27 | <@ToxicFrog> | Not necessarily. |
| 04:27 | <@ToxicFrog> | C's #define, for example, happens during the first of two lex stages (the preprocessor) |
| 04:27 | < C_tiger> | Ah, right preprocess. |
| 04:27 | <@ToxicFrog> | Likewise #include. |
| 04:28 | <@ToxicFrog> | The resulting modified source then goes through a normal lex-parse-generate sequence. |
| 04:28 | < C_tiger> | I forgot about CPP. |
| 04:29 | <@ToxicFrog> | As a general rule, "parsing" is "the process of turning a linear stream of tokens into some useful data representation" |
| 04:29 | <@ToxicFrog> | In the case of compilers, this typically means turning tokenized source code into an abstract syntax tree. |
| 04:29 | < C_tiger> | Right, and compiling is turning the tree into? |
| 04:29 | <@ToxicFrog> | You missed what I said earlier. |
| 04:29 | <@ToxicFrog> | Compiling includes parsing. |
| 04:30 | < C_tiger> | Right. |
| 04:30 | <@ToxicFrog> | Compiling is turning source code into machine code. |
| 04:30 | < C_tiger> | Ok, that's why I was confused. |
| 04:30 | <@ToxicFrog> | Typically, this means lexing it, then parsing it, then doing code generation, although this isn't always the case. |
| 04:30 | <@ToxicFrog> | So a compiler will usually include as one of its subsystems a parser. |
| 04:30 | < C_tiger> | Right. |
| 04:32 | < C_tiger> | Man how did people write useful code with punchcards? |
| 04:34 | <@ToxicFrog> | Hand-assembling, basically. |
| 04:34 | <@ToxicFrog> | Write out the program in mnemonics, then translate into binary by hand. |
| 04:35 | < C_tiger> | creepy |
| 04:37 | <@ToxicFrog> | (concerning the stages of compiling: for a language where a lexer isn't necessary, see any of the one-character-per-instruction languages like n'funge or brainfuck; for one with no parser, see postscript; for one with no code generation, see ruby, which executes by walking the AST directly) |
| 04:38 | < C_tiger> | postscript is a language? |
| 04:38 | < C_tiger> | Ok, that came out sounding more stupid than it should have. |
| 04:38 | <@ToxicFrog> | Yes. What did you think it was? |
| 04:38 | < C_tiger> | I dunno something like HTML. |
| 04:38 | <@ToxicFrog> | Aah. |
| 04:38 | <@ToxicFrog> | No, it's a fully featured, stack-oriented, interpreted language. |
| 04:39 | <@ToxicFrog> | Page drawing and suchlike is done with function calls rather than declarative data. |
| 04:39 | <@ToxicFrog> | Makes a handy calculator. |
| 04:39 | < C_tiger> | Wow. |
| 04:39 | < C_tiger> | Cool. |
| 04:40 | < C_tiger> | Ok, I need bed... but I had a creepy thought. |
| 04:40 | < C_tiger> | so I'll have to muse it. |
| 04:40 | <@ToxicFrog> | Oh |
| 04:40 | <@ToxicFrog> | ? |
| 04:41 | < C_tiger> | Pretend DNA is the super binary and data is life. (All valid assumptions) |
| 04:41 | | Vornotron [~vorn@Admin.Nightstar.Net] has quit [Ping Timeout] |
| 04:41 | <@ToxicFrog> | DNA is quadary, actually |
| 04:41 | < C_tiger> | I know. |
| 04:41 | < C_tiger> | I'm not a biologist for nothing. |
| 04:42 | < C_tiger> | But I'm drawing an analogy. |
| 04:42 | < C_tiger> | so I'm analogizing the bases to binary encoding. |
| 04:43 | < C_tiger> | Then machine code WAS assembled by hand... (evidence for God?) but he got lazy and wrote a program that wrote more code. |
| 04:43 | < C_tiger> | So we extend upwards to have higher evolved languages. |
| 04:43 | < C_tiger> | *more highly* |
| 04:44 | < C_tiger> | I need sleep. |
| 04:44 | < C_tiger> | It's creepy parallel even if I'm too tired to explain it. |
| 04:46 | < C_tiger> | So you can extend it both ways: Either it's evidence for the existence of God and we're all just built as self-replicating code writers and compilers. OR eventually, code will evolve to the point where it writes itself and builds its own form of self-aware code-life. |
| 04:46 | < C_tiger> | Or something like that. |
| 04:47 | < C_tiger> | All this comes from translating your discussion of the parts of a compiler into terms I know. |
| 04:49 | < C_tiger> | So a lexer = RNA synthetase, parser = spliceosome, code generation = ribosome. |
| 04:49 | < C_tiger> | RNA polymerase. |
| 04:49 | < C_tiger> | I was thinking tRNA syntetase. |
| 04:49 | < C_tiger> | ugh, sleep. |
| 04:50 | | GeekSoldier|bed is now known as GeekSoldier|work |
| 06:29 | | Thaqui [~Thaqui@222.154.187.ns-2601] has joined #code |
| 06:29 | | mode/#code [+o Thaqui] by ChanServ |
| 06:35 | | AnnoDomini [AnnoDomini@83.21.31.ns-26029] has joined #Code |
| 06:35 | | mode/#code [+o AnnoDomini] by ChanServ |
| 07:59 | | GeekSoldier|work [~Rob@Nightstar-9987.dip.t-dialin.net] has quit [Ping Timeout] |
| 08:39 | | You're now known as TheWatcher |
| 10:22 | | Thaqui [~Thaqui@222.154.187.ns-2601] has left #code [Leaving] |
| 11:11 | | Vornucopia [~vorn@Admin.Nightstar.Net] has joined #code |
| 11:12 | | Vornucopia is now known as Vornicus |
| 11:13 | | Vornicus is now known as NSGuest-5514 |
| 11:30 | | AnnoDomini [AnnoDomini@83.21.31.ns-26029] has quit [Ping Timeout] |
| 11:37 | | AnnoDomini [AnnoDomini@83.21.42.ns-2659] has joined #Code |
| 11:37 | | mode/#code [+o AnnoDomini] by ChanServ |
| 11:55 | | NSGuest-5514 is now known as Vornicus |
| 12:18 | | AnnoDomini [AnnoDomini@83.21.42.ns-2659] has quit [Ping Timeout] |
| 12:25 | | AnnoDomini [AnnoDomini@83.21.45.ns-4389] has joined #Code |
| 12:25 | | mode/#code [+o AnnoDomini] by ChanServ |
| 12:42 | | ALPAY-ist [~KELEBEK@78.188.112.ns-21870] has joined #Code |
| 12:45 | | ALPAY-ist [~KELEBEK@78.188.112.ns-21870] has quit [Ping Timeout] |
| 13:38 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 13:38 | | mode/#code [+o gnolam] by ChanServ |
| 13:47 | | * ToxicFrog eyes python |
| 13:47 | <@ToxicFrog> | >>> 4e-7 |
| 13:47 | <@ToxicFrog> | 3.9999999999999998e-07 |
| 13:47 | < C_tiger> | snerk. |
| 13:48 | < Vornicus> | Yeah, meet doubles. |
| 13:51 | <@ToxicFrog> | Vornicus: AFAIK, doubles are capable of exactly representing that number, and postscript and lua have no problems with it |
| 13:51 | <@ToxicFrog> | I don't know what Python is using, but it's not doubles. |
| 13:55 | < Vornicus> | 4e-7 is... ...no, I don't think it is exact. |
| 13:55 | <@ToxicFrog> | For that matter, even if I make it e-1: |
| 13:55 | <@ToxicFrog> | > =0.4 |
| 13:55 | <@ToxicFrog> | 0.4 |
| 13:56 | <@ToxicFrog> | GS>0.4 = |
| 13:56 | <@ToxicFrog> | 0.4 |
| 13:56 | <@ToxicFrog> | >>> 0.4 |
| 13:56 | <@ToxicFrog> | 0.40000000000000002 |
| 13:56 | < Vornicus> | I /know/ that number isn't exact. |
| 13:57 | <@ToxicFrog> | ...hmm. It's not exact in gs, either, if you get it to actually display all the digits. |
| 13:57 | <@ToxicFrog> | ANAICT it's still exact in lua, though. Odd. |
| 14:00 | | * Vornicus tries to remember the exact gyrations you need to do subtraction in decimal. |
| 14:00 | < C_tiger> | There's a perl module for working with fractions. |
| 14:01 | <@McMartin> | 0.4 is not exactly representable in binary. |
| 14:01 | <@McMartin> | Lua's default double formatting is different. |
| 14:01 | <@McMartin> | And dividing or multiplying by some power of ten won't change this |
| 14:02 | < Vornicus> | Well, 4 is |
| 14:02 | < Vornicus> | but dividing by a power of ten I don't think /ever/ turns a number from unrepresentable to representable |
| 14:02 | | * Vornicus examines. Yeah, that's easily proven. |
| 14:02 | <@ToxicFrog> | Hmm. Weird. |
| 14:03 | <@ToxicFrog> | In Lua, 0.4e+N results in 4e+(N-1) exactly. |
| 14:03 | <@McMartin> | Any fraction in which the denominator is not a power of 2 is a repeating binary fraction. |
| 14:03 | <@McMartin> | Yeah, its printer is rounding. |
| 14:03 | <@ToxicFrog> | And it prints as "0.4" |
| 14:04 | <@ToxicFrog> | However, if I actually do %.20f, 0.4 I do see the wobble at the end. |
| 14:04 | <@McMartin> | You're seeing a difference in the format. |
| 14:04 | <@ToxicFrog> | Aah, yes |
| 14:04 | <@ToxicFrog> | Classtime |
| 14:04 | <@McMartin> | It's rounding the wobble away by default |
| 14:05 | | * Vornicus fiddles, decides the right answer is to get the Knuth and the "represent a number in a base, for non-integers" algorithm. |
| 15:15 | | Moltare [~moltare@Nightstar-29340.cable.ubr02.bath.blueyonder.co.uk] has quit [Ping Timeout] |
| 15:22 | | Moltare [~moltare@Nightstar-29340.cable.ubr02.bath.blueyonder.co.uk] has joined #code |
| 15:22 | | mode/#code [+o Moltare] by ChanServ |
| 16:21 | | Moltare [~moltare@Nightstar-29340.cable.ubr02.bath.blueyonder.co.uk] has quit [Quit: Twisted, yes, but strangely fluffy.] |
| 16:58 | | GeekSoldier|work [~Rob@Nightstar-9620.dip.t-dialin.net] has joined #code |
| 17:01 | | GeekSoldier|work is now known as GeekSoldier |
| 17:06 | | You're now known as TheWatcher[afk] |
| 18:32 | | You're now known as TheWatcher |
| 19:39 | | * Chalain prods ToxicFrog. |
| 19:39 | <@ToxicFrog> | Yers? |
| 19:39 | <@Chalain> | Or any Ludd^H^Haite around. |
| 19:40 | <@Chalain> | I'm up to my elbows in the guts of mysql-proxy, which is scripted with Lua |
| 19:40 | <@Chalain> | And I can't remember how to do a foreach on a table with non-numeric keys. |
| 19:41 | <@Chalain> | I just looked it up on the Lua reference site and remembered that a) I tried that years ago, b) it didn't work (i.e. I did not learn how to do it) and c) that I found out later that it WAS possible, so rather than decide it isn't possible, I figured |
| 19:41 | <@Chalain> | I'd ask. :-) |
| 19:43 | <@ToxicFrog> | pairs() |
| 19:44 | <@ToxicFrog> | Assuming this is lua 5.x |
| 19:44 | <@Chalain> | It is. |
| 19:44 | <@ToxicFrog> | for key,value in pairs(table) do ... end |
| 19:44 | <@Chalain> | bless you. |
| 19:45 | <@ToxicFrog> | (and for future reference: the generic for loop itself is covered in 2.4.5, and pairs() itself, in 5.1 with the rest of the luastdlib functions) |
| 19:45 | <@ToxicFrog> | (of the manual, that is) |
| 19:46 | <@ToxicFrog> | Depending on Lua version you may also have a map() or table.map(); if not, you can easily write your own, should it be useful. |
| 19:46 | <@Chalain> | Oka, um. |
| 19:46 | <@Chalain> | http://www.lua.org/manual/5.1/manual.html#2.4.5 <-- "pairs" not found. |
| 19:46 | <@Chalain> | Am I looking at the wrong manual? |
| 19:47 | <@ToxicFrog> | 2.4.5 is the for loop control structure |
| 19:47 | <@Chalain> | s/Am I (.*)\?$/Have I been $1 all this time?!?/ |
| 19:47 | <@ToxicFrog> | It doesn't cover pre-built iterators for use with it, as those are part of the standard library, not the language grammar |
| 19:48 | <@ToxicFrog> | And thus, as I said, are covered in chapter 5 |
| 19:49 | <@ToxicFrog> | "pairs (t) |
| 19:49 | <@ToxicFrog> | Returns three values: the next function, the table t, and nil, so that the construction |
| 19:49 | <@ToxicFrog> | for k,v in pairs(t) do body end |
| 19:49 | <@ToxicFrog> | will iterate over all key?value pairs of table t. |
| 19:49 | <@ToxicFrog> | See function next for the caveats of modifying the table during its traversal." |
| 19:49 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Ping Timeout] |
| 19:49 | <@Chalain> | Ah, there we go. |
| 19:49 | <@Chalain> | So basically the problem is Chalain can't read. |
| 19:49 | <@Chalain> | I can deal with this. |
| 19:49 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 19:49 | | mode/#code [+o gnolam] by ChanServ |
| 19:49 | <@Chalain> | Lord knows I've practiced for it enough. |
| 19:49 | <@Chalain> | Thanks! |
| 19:58 | <@ToxicFrog> | You're welcome. |
| 20:08 | <@jerith> | Ooh, we have a Chalain here. |
| 20:18 | | GeekSoldier is now known as GeekSoldier|bed |
| 20:40 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has quit [Ping Timeout] |
| 20:46 | | gnolam [lenin@Nightstar-10613.8.5.253.static.se.wasadata.net] has joined #Code |
| 20:46 | | mode/#code [+o gnolam] by ChanServ |
| 22:07 | | AnnoDomini [AnnoDomini@83.21.45.ns-4389] has quit [Quit: Gnothi seauton, malákas.] |
| 22:53 | | AFKSkull [~none@Nightstar-7066.dyn.optonline.net] has joined #code |
| 22:53 | < AFKSkull> | Hello, folks. |
| 22:53 | | AFKSkull is now known as ASCIISkull |
| 22:56 | < ASCIISkull> | I don't suppose anyone here knows enough about .net clickonce installations as to be ble to figure out, from a manifest, how to actually install the damn program? |
| --- Log closed Thu Mar 13 00:00:44 2008 |