Hi All,
C:\family\ruby>cat test.rb
def f(command_here)
`#{command_here}`
end
a = { "Local copy"=>"copy test.txt test.txt2" }
require 'benchmark'
Benchmark.bmbm do |x|
a.each do |k,v|
x.report(k) { f v }
end
end
C:\family\ruby>ruby test.rb
c:/ruby/lib/ruby/1.8/benchmark.rb:334:in `concat': can't modify frozen string (T
ypeError)
from c:/ruby/lib/ruby/1.8/benchmark.rb:334:in `report'
from test.rb:12
from test.rb:11:in `each'
from test.rb:11
from c:/ruby/lib/ruby/1.8/benchmark.rb:250:in `bmbm'
from test.rb:10
C:\family\ruby>
if i comment out line 334 in benchmark.rb like so,
#label.concat ' '
It works.
It also works for the ff cases
x.report(k) { f v }
x.report() { f v }
x.report("") { f v }
x.report("test") { f v }
What is the relevance of line 334 in benchmark.rb?
If the line is important, how can I fix my program so I can pass a var in x.report?
thank you and kind regards -botp
Peña wrote:
Hi All,
C:\family\ruby>cat test.rb
def f(command_here)
`#{command_here}`
end
a = { "Local copy"=>"copy test.txt test.txt2" }
require 'benchmark'
Benchmark.bmbm do |x|
a.each do |k,v|
x.report(k) { f v }
end
end
C:\family\ruby>ruby test.rb
c:/ruby/lib/ruby/1.8/benchmark.rb:334:in `concat': can't modify frozen string (T
ypeError)
from c:/ruby/lib/ruby/1.8/benchmark.rb:334:in `report'
from test.rb:12
from test.rb:11:in `each'
from test.rb:11
from c:/ruby/lib/ruby/1.8/benchmark.rb:250:in `bmbm'
from test.rb:10
C:\family\ruby>
if i comment out line 334 in benchmark.rb like so,
#label.concat ' '
It works.
It also works for the ff cases
x.report(k) { f v }
x.report() { f v }
x.report("") { f v }
x.report("test") { f v }
What is the relevance of line 334 in benchmark.rb?
If the line is important, how can I fix my program so I can pass a var in x.report?
thank you and kind regards -botp
A quick workaround would be to use k.dup:
x.report(k.dup) { f v }
Apparently, benchmark is being bad and munging its inputs (and falling down when one of them happens to be frozen because it is also a hash key).
···
--
vjoel : Joel VanderWerf : path berkeley edu : 510 665 3407
# A quick workaround would be to use k.dup:
···
From: Joel VanderWerf [mailto:vjoel@path.berkeley.edu] :
#
# x.report(k.dup) { f v }
brilliant insight. I never even tested that, thinking k.dup would just be another var/object. And besides, it does not even look *right
# Apparently, benchmark is being bad and munging its inputs
# (and falling
# down when one of them happens to be frozen because it is also
# a hash key).
it is really weird.
thank you and kind regards -botp