You can’t. Not because you can’t have a .irbrc file (You just need to
specify a directory in the HOME environment variable, and put it
there)… but because Microsoft got rid of color support in the command
prompt.
You could disable color in wirble and make it work though, I suspect.
I’m willing to bet that most of them don’t understand how to effectively
use a REPL loop, and that’s fine if you look at it from their
perspective.
That’s exactly right. I’m good using REPL for simple scripting, but I
can’t think of a good way to use it for bigger things. Could you point
me (us) out some reading on how to use it for serious development?
There is also the familiarity aspects. People coming from a non-REPL
background often look at it and ask, “Ok, now how do I save my program?”
or “Do I really have to retype my last 30 lines to fix that little bug
in method foo?”.
Those are the very first questions that come to my mind, indeed. What
are the answers?
well, if you’ve installed the gem, I think you can do it just by
entering a few commands:
load libraries
require ‘rubygems’
require ‘wirble’
start wirble (with color)
Wirble.init
Wirble.colorize
obviously that’s non-optimal, though, because it won’t automatically
load every time. I honestly don’t know how irb works on Windows boxes,
I have one but the irb on it is built into FreeRide or something. you
can probably use those commands to get started, though.
I’m willing to bet that most of them don’t understand how to
effectively
use a REPL loop, and that’s fine if you look at it from their
perspective.
That’s exactly right. I’m good using REPL for simple scripting, but I
can’t think of a good way to use it for bigger things. Could you point
me (us) out some reading on how to use it for serious development?
A big project is just a ton of little projects, once you chop it up.
Test driven development is all about that, I think.
A big project is just a ton of little projects, once you chop it up.
Test driven development is all about that, I think.
Yes, but sometimes you can’t chop it up so fine-grained that you have
self-contained, one-lined commands (that’s what I meant by
‘scripting’).
Say you write up a method and want to correct it, you have retype or
find through the history by pressing “Up” all the lines that “worked”
and re-entering them. You also can’t just read them from a file either,
AFAIK (not smthg like an incomplete method code, for instance).
Conversely, it’s not trivial to go through the lines that “worked” in
that awkward “Up”-key fashion, then select them and save them out to a
file.
Those are the questions Jonathan A. pointed out, and I don’t have
good answers for them.
Perhaps there’s some different mode in irb where you can quickly
inspect the aggregated command history as in a regular editor and
choose to append selected lines out to a file? That’d be very helpful,
especially with the ability to load them back as if you had typed them
in yourself, which would allow for loading of incomplete blocks of
code.
On Monday 22 January 2007 06:48 am, Helder R. wrote:
good answers for them.
Perhaps there’s some different mode in irb where you can quickly
inspect the aggregated command history as in a regular editor and
choose to append selected lines out to a file? That’d be very helpful,
especially with the ability to load them back as if you had typed them
in yourself, which would allow for loading of incomplete blocks of
code.
I am not a very experienced Ruby programmer, but what I find helps in
that
regard is that I do my irb work in a kde konsole. It displays all the
old
lines that I’ve entered (and results) with its “History” feature. I
then
use Linux style copy (select then paste with middle mouse button) to
grab the
lines I want, paste them into my editor, and edit out the stuff I don’t
want
(if I’ve copied more than I want).
Yes, but sometimes you can’t chop it up so fine-grained that you have
self-contained, one-lined commands (that’s what I meant by
‘scripting’).
I think it’s harder for some projects than others, but I’m not
willing to go all the way to “can’t”.
Say you write up a method and want to correct it, you have retype or
find through the history by pressing “Up” all the lines that “worked”
and re-entering them. You also can’t just read them from a file
either,
AFAIK (not smthg like an incomplete method code, for instance).
Conversely, it’s not trivial to go through the lines that “worked” in
that awkward “Up”-key fashion, then select them and save them out to a
file.
Here are some ideas:
def test
puts “This line is OK.”
puts “So is this one.”
Object.new.no_such_method # Oops!
end
=> nil
test
This line is OK.
So is this one.
NoMethodError: undefined method no_such_method' for #<Object:0x102f9ac> from (irb):4:in test’
from (irb):6
h 6
[0758] def test
[0759] puts “This line is OK.”
[0760] puts “So is this one.”
[0761] Object.new.no_such_method # Oops!
[0762] end
[0763] test
[0764] h 6
=> nil
get_lines(758…760) << “puts ‘Now this is better.’” << “end”
=> [“def test”, " puts “This line is OK.”“, " puts “So is this
one.””, “puts ‘Now this is better.’”, “end”]
eval((get_lines(758…760) << “puts ‘Now this is better.’” <<
“end”).join(“\n”))
=> nil
test
This line is OK.
So is this one.
Now this is better.
=> nil
You can also write to a file:
hw “test_method.rb”, 758…760
=> nil
File.read(“test_method.rb”)
=> “def test\n puts “This line is OK.”\n puts “So is this one.”\n”
File.open(“test_method.rb”, “a”) { |f| f << " puts ‘Fixed in the
file.’\nend\n" }
=> #<File:test_method.rb (closed)>
File.read(“test_method.rb”)=> “def test\n puts “This line is
OK.”\n puts “So is this one.”\n puts ‘Fixed in the file.’\nend\n”
load “test_method.rb”
=> true
test
This line is OK.
So is this one.
Fixed in the file.
=> nil
Uh … not really, for a number of reasons. To use the old saw about
nine women not being able to produce a baby in one month, a baby is not
a collection of small separately-developed pieces. A large software
project is a complex adaptive system, interacting with numerous
machines, users, creators and maintainers in many different ways. One
distinguishing characteristic of complex adaptive systems is that they
exhibit behavior that can’t be explained or predicted by examining
their components. Just like the baby, there are pieces you can remove
that degrade their functioning, pieces like hair and toenails that
return if removed (log files, for example ) and pieces you can’t
remove without destroying their functioning.
Irb-style “design-at-the-keyboard” doesn’t scale beyond a small, “agile”
project. I’m not sure test-driven development or continuous refactoring
do, either, because nobody has ever to my knowledge had the budget to
attempt it – we’re all too busy on the next “complex adaptive software
project that’s good enough”. Irb is part of the tool set you need
to work on a complex adaptive Ruby software project. I can’t imagine
working on such a project without irb. But if text-based tools are all
I’m allowed to work with, I’d seriously question my relationship with
management.
James Edward G. II
Thanks!! If I run out of queuing theory papers to buy, I might buy that
one. However:
"the coordination and communication grows exponentially with the number
of individual developers and sites. "
That’s not correct, either. The formula involves exponentials, but the
communication and coordination does not grow exponentially – nothing
large could be built if it did. I have the formulas somewhere; if I can
find them, I’ll post them here.
On Jan 22, 2007, at 8:59 AM, M. Edward (Ed) Borasky wrote:
Irb-style “design-at-the-keyboard” doesn’t scale beyond a small,
“agile” project. I’m not sure test-driven development or continuous
refactoring do, either, because nobody has ever to my knowledge had
the budget to attempt it
I am not a very experienced Ruby programmer, but what I find helps in that
regard is that I do my irb work in a kde konsole. It displays all the old
lines that I’ve entered (and results) with its “History” feature. I then
use Linux style copy (select then paste with middle mouse button) to grab the
lines I want, paste them into my editor, and edit out the stuff I don’t want
(if I’ve copied more than I want).
Hey, great idea! I’m so used to the default mode using bash that I had
never noticed I could use irb from inside konsole. I always ran it
from bash. I’ll try that out see how it works!
On Tuesday 23 January 2007 06:20 pm, Helder R. wrote:
Hey, great idea! I’m so used to the default mode using bash that I had
never noticed I could use irb from inside konsole. I always ran it
from bash. I’ll try that out see how it works!
) and pieces you can’t