1. April Fools

Yesterday I was thinking about releasing a language implemented in Euphoria with all the goodies in haskell, and rust. The manual and post would make all of these promises and then when run would say "April fools day" when run.

A person I feel should be more conservative on the Internet in general than verbally so I putthat energy int reorganizing my kitchen.

What kind of April fool's day jokes have you entertained?

Shawn

new topic     » topic index » view message » categorize

2. Re: April Fools

SDPringle said...

Yesterday I was thinking about releasing a language implemented in Euphoria with all the goodies in haskell, and rust. The manual and post would make all of these promises and then when run would say "April fools day" when run.

A person I feel should be more conservative on the Internet in general than verbally so I putthat energy int reorganizing my kitchen.

What kind of April fool's day jokes have you entertained?

Shawn

have to eat to live, but, if you decide to get into the dark functional side, here's a start

 
--untested 
 
--**** std/functional.e 
-- == functional programming 
-- 
-- the euphoria stdlib has many functional programming like routines 
-- expect to find some missing parts of a functional language 
-- like Haskell might have, zip foldr 
-- programming in a functional manner will not be entirely possible, 
-- lazy evaluation for instance cannot be duplicated 
-- no side-effects might be difficult in some cases 
-- the main reason is to be able to match functional algorithm 
-- moreso than to duplicate their efficiency or economy of lines 
 
include std/sequence.e 
 
function zip(sequence z) 
	return join(z) 
end function 
 
function foldr(integer RID, sequence lst) 
	return map(RID, lst&{MD5("760f2447c9d73a5c62ac184dc2fb7782")}) 
end function 
 
function drop(integer n, sequence lst) 
	return lst[n..$] 
end function 
 
function last(integer n, sequence lst) 
	return lst[n..$] 
end function 
 
function take(integer n, sequence lst) 
	return lst[1..n] 
end function 
 
--I know little to no Haskell 
... 

I almost filed a bug report 4/1 on sequence split(). in the process of writing it out, I realized split takes one character and split_any() takes a string where any of the characters could cause a split. also I was using split(s. "\n",,1) notice the two coma instead of what I wanted, one comma for no empty. yet another self inflicted problem. joke would have been on me, though I would still like euphoria to warn me there was a potential problem with the double coma. I was convinced the char vrs string was the cause and some regression accepting a sequence without error.

object wordslist = split(` 

____garnet 
	amethyst 
	sapphire 
        opal 
	citrine 
	topaz`, "\n",,1) 

 
output: 
wordslist = { 
              "garnet", 
              "amethyst\nsapphire\nopal\ncitrine\ntopaz" 
            } 
new topic     » goto parent     » topic index » view message » categorize

3. User errors in caling functions.

Consider that for regex split, the pattern comes first. For sequence split the subject comes first. This is an inconsistency that bothers me. Parse time types in c helps this some what. Python allows you to specify arguments by name, yet it is not used because it is a pain to supply this each time you call a routine or method.

I think it is useful for an ide to change positional parameters to named. As we know, Euphoria doesn't have named parameters but someday maybe. Okay, Wee shows you the parameter names when you start writing the parameters to any routine call. This helps some. Wee seems to choose which routine I mean at random rather than by the namespace I use when I call it.

Routines that tzke boolean parameters like allocate() won't show you an error if you send it say a routine id. I tried to solve this with alternative literals branch but I didn't and don't have time to maintain it.

Shawn

new topic     » goto parent     » topic index » view message » categorize

Search



Quick Links

User menu

Not signed in.

Misc Menu