IRC Logs for #io Wednesday, 2013-11-13

*** nisstyre has quit IRC00:56
*** Raimondii has joined #io01:42
*** Raimondi has quit IRC01:43
*** hashmal has quit IRC01:56
*** Raimondi has joined #io01:59
*** Raimondii has quit IRC02:02
*** mkroehnert has quit IRC02:10
*** nisstyre has joined #io02:48
*** nisstyre has quit IRC03:01
*** nisstyre has joined #io03:01
prologicjer:  In an Io-like language (mio) -- Would it be possible to sensibly construct a dictionary/mapping using syntactic sugar like Python? e.g: d = {"a": 1, "b": 2} -- Would this require special parsing or could it be done with normal messages?03:27
jerprologic, io has this support, { ... } is parsed as curlyBrackets(...)03:28
jerlikewise [...] is parsed as squareBrackets(...)03:29
jershort story, yeah it requires special parsing rules03:29
jerbut it's not a major thing03:29
jerio only had () being handled as getSlot("") call(...) when i started with the project03:30
jeri.e., a method with no name03:30
jeri added [] and {} support03:30
jerusing the same rule03:30
prologicI have [] () and {} support as well03:30
prologicwhere get parsed like03:30
prologic[](...) {}(...) etc03:30
prologicrather than the named curlyBackets, squarebackets, etc03:31
prologicso long story short, you'd still have to parse the arguments inside square/curly backers in a special way in the parser anyway?03:31
jer: is a problem, but it's a regular message name in io, you can write an operator in io that maps : to atPut03:34
jerthat's what i did for the example in 7 languages in 7 weeks by bruce tate; the telephone book example uses this trick03:35
prologicahh okay03:45
prologicand would it make sense to only define the : = atPut(...) block inside the {} block only?03:45
prologicso that it doesn't make sense unless you're constructing a dict literal in this way?03:45
*** gatesphere has quit IRC04:28
*** mkroehnert has joined #io08:05
*** mkroehnert has quit IRC08:23
*** hashmal has joined #io09:01
*** OpenSpace has quit IRC10:31
prologicjer, whoohoo11:34
prologicI have dict literals in the form of {k1: v1, k2: v2, k3: v3, ..., kn: vn}11:35
prologicI also have syntantic suguar for what you'd call getAt11:35
prologice.g: xs = [1, 2, 3]; xs[0]11:35
*** `fogus has joined #io13:26
locksmio is coming along quite nicely16:51
jeroh man i hate myself -- building a rails app =/17:01
jerand it's coming along quite nicely17:01
hashmaljer: it's not that bad17:03
hashmaland it's horrible at the same time.17:03
jerhashmal, don't make me gouge out your eyes =]17:04
hashmalwell you're the one using rails right now. I can help you with the gouging :V17:05
locksjer: rails <317:26
*** fredreichbier has joined #io18:47
*** nisstyre has quit IRC20:00
*** obihann has joined #io20:21
*** chaign_c has joined #io20:25
robonerdjer, you should be using sinatra or webmachine20:30
robonerdnice to see you back into ruby too20:30
jerrobonerd, i'm not back into ruby20:31
jerused it to prototype an app20:31
jerbut now i don't want to rewrite it20:31
robonerda web app?20:31
jernot directly no20:32
jermore of an api20:32
robonerdwhat kinda api?20:37
robonerdthe anchor woman can barely contain her 'womanhood' in her top20:39
robonerdoh, that poor poor fabric20:39
prologichashmal, yes ruby/rails is tha tbad :)20:48
prologicI've always said Ruby is a bastardized version of Python 12 years behind20:48
prologiclocks, thanks :)20:49
prologictrying :)20:49
prologicand learning :)20:49
hashmalprologic: some points on that list are questionable, and even not related to ruby20:51
hashmal> Idiomatic use of the '_' character for word separation (SnakeCase? EmbeddedUnderscore); it's just damn ugly.20:51
hashmalif you really want to cry with language design, just try shirka20:54
hashmal(oops, shameless plug)20:54
robonerdprologic what's an example of why you hate ruby?20:54
hashmal(1..1_000_000).each {|n| n.instance_eval {@ewww = n.to_s}}20:59
hashmalthat could be hated I guess20:59
prologicrobonerd, biggest and foremost is probably it's scoping rules21:04
prologicand it's block notation21:04
robonerddescribe the scoping rules issue as you see it please?21:04
prologicalso my pet peeve is that Ruby programmers tend to rewwrite everything in Ruby and reinvent the wheel all the time21:04
prologicin pure ruby21:04
prologicwhy not juse reuse existing libraires21:05
prologicor interface with other languages?21:05
prologicRuby's scoping rules are too complex21:05
prologicIIRC there are 6 of them or more21:05
prologicthreads can change the definition of classes by the very same scoping rules21:05
prologicwhich is just wtf21:05
prologic<hashmal> (1..1_000_000).each {|n| n.instance_eval {@ewww = n.to_s}} <-- indeed21:06
prologican example of how unclear and unreadable Ruby really is21:06
prologicif that particular example is doing what I htink it is21:07
prologicit's just:21:07
hashmalwell I could paste a haskell example and if you're not familiar with it you'll understand shit21:07
prologicmap(str, range(1000000))21:07
prologicboth in Python and mio :)21:07
prologicI'm familiar with Haskell :)21:07
prologicI like Haskell and it's semantics21:07
prologicbut not it's syntax :)21:07
prologicI'm a big fan of functional programming :)21:08
prologicanyway enough b'tching about langinges :)21:08
prologicno language is perfect :)21:08
hashmalthe point is that any language (programming or not) will seem unclear and unreadable if you don't know it21:08
prologicexcept maby assembler21:08
prologicthat may be so21:08
prologicbut if you're building a new langauge21:09
prologicit helps to build on familar constructs and patterns21:09
prologicnot reinventing ugly rubbish21:09
prologictake PHP for example21:09
prologicusing \ as namespacing?21:09
prologicPHP has traits, interfaces and MI21:09
prologicpick one!21:09
prologicPHP has no protocols21:10
prologicjust thousands of thousands of global functions21:10
prologicbut some folk like it :)21:10
hashmalPHP is a pretty bad example, as it wasn't designed at all :D :D :D21:11
prologicPHP == Personal Hacking Project21:19
prologicyeah no there was really no clear design at all21:19
locksRuby <321:19
prologicunless you call it's f'd up design a design21:19
prologicsame with Ruby really ihmo21:19
locks9:04 PM <prologic> also my pet peeve is that Ruby programmers tend to rewwrite everything in Ruby and reinvent the wheel all the time21:19
locksreally? that's the problem with Ruby?21:19
hashmalno, php wasn't designed, the author himself said it21:19
prologicbut it's a particular pet peeve of mine :)21:19
hashmalit was an agglutination of stuff from the beginning21:20
prologice.g: why write an ldap server in pure ruby?21:20
prologicwhen you can use 389?21:20
*** hashmal has quit IRC21:20
prologicanyway gotta get ready for work and get out f here :)21:20
prologiccyas at work :)21:20
*** hashmal has joined #io21:20
prologicI'll just say one thing21:20
prologicwith mio, I intend to interface with bothh Python and C21:21
prologicas much reuse as possible :)21:21
*** fredreichbier has quit IRC22:16
*** mkroehnert has joined #io22:20
*** mkroehnert has quit IRC22:20
*** mkroehnert has joined #io22:21
*** gatesphere has joined #io22:53
robonerdyea C is awesome23:05
prologicplain ol C99 C is okay23:23
prologicI don't mind it as such23:23
prologicprobably why I don't mind RPython either23:23
prologicit's almost exactly like C23:23
prologicbut with Python syntax23:23
prologicjer:  when you get around here there's a few things i wanted to get your ideas/opinions on23:24
prologicjer:  first there's how I implemented dict literals:
prologicAnd then there's the last point on the 0.0.9 change log -- I feel this is a work around and would like to see it improved or generalized in some nicer way23:27
prologicI tried a few things last night re closure scoping rules but none worked very well23:27

Generated by 2.11.0 by Marius Gedminas - find it at!