Gus Wirth wrote:
Problem one: The examples immediately start to use function overloading
even when the functions are not related except for the fact they happen
to be in the same module. To me this is the epitome of stupid
instruction as well as bad programming. Not only does it introduce a
concept before its time but it serves to confuse and obscure the
operation of the program.
I presume you are referring to:
-module(tut2).
-export([convert/2]).
convert(M, inch) ->
M / 2.54;
convert(N, centimeter) ->
N * 2.54.
I'll somewhat defend this one (but not the next one, see below). This
appears as function overloading to those coming from a different
language. I will agree, though, that this should probably be flagged
out since it caused confusion.
This is really rule matching and you *MUST* get this concept immediately
or you cannot proceed with the language.
Those are *not* different functions. They are two clauses of the *same*
function. Note that the first ends with a ";" and the second ends with
a "."
I'm not sure how you make this better. The next step is tuple matching
followed by recursive matching. The progression is fairly logical.
This didn't really cause me any confusion. However, I don't know why.
Presumably because I saw the previous example with recursion clauses so
I never really thought about this as overloading. I figure that I
anticipated where they were going with this since Erlang is a functional
language.
Problem two: Using the same word for functions and atoms. In section
three, you can see they use the same words for the functions and the
atoms despite the fact they are used for different things and are
commingled. This causes unnecessary confusion and again is bad form.
Yeah, I can see your point, now that I went and looked at it again.
Somehow, I didn't really notice the issue with the "ping" atom and the
"ping" function. I'm used to using the same names for signals in
Verilog/VHDL as they pass up the chain through different forms so I
didn't really notice as it changed from atoms to functions.
I tend to segregate those in my code. I use StudlyCaps for variables,
studlyCaps for atoms, and under_scores for functions.
-a
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg
