Bug #6291
closedBacktrace printout halts for an extremely long time when large amounts of data are allocated
Description
=begin
When large amounts of data are allocated, it can take an eternity for the backtrace printout to complete. Here's a simple example:
class SomeClass
def initialize
@a = Array.new(100000, "just some random string")
end
end
h = Hash.new
200.times {|i| h[i] = SomeClass.new}
h.asdf
This will halt with the following printout:
test.rb:9
And then, after a very long time (depending on how much data is in memory), it will eventually finish:
test.rb:9: undefined method `asdf' for #Hash:0x1001bc348 (NoMethodError)
For me, this has caused a lot of frustration when running a test suite on an application which loads a lot of data in memory, as in many cases I will never get to see the actual backtrace. This makes debugging much more difficult. Perhaps there is some fundamental property of the Ruby language itself that makes this an 'expected behaviour', but if not, I would really appreciate to see this adressed.
=end
Updated by mame (Yusuke Endoh) over 12 years ago
Hello,
I have no idea except adding a length limitation or timeout to #inspect.
Any other ideas?
--
Yusuke Endoh mame@tsg.ne.jp
Updated by mame (Yusuke Endoh) over 12 years ago
- Status changed from Open to Assigned
- Assignee set to matz (Yukihiro Matsumoto)
Updated by matz (Yukihiro Matsumoto) over 12 years ago
- Status changed from Assigned to Closed
This one is not easy to fix.
We try to address it by improving #inspect, but we cannot make promise.
Matz.
Updated by shyouhei (Shyouhei Urabe) about 8 years ago
- Has duplicate Feature #6783: Infinite loop in inspect, not overriding inspect, to_s, and no known circular references. Stepping into inspect in debugger locks it up with 100% CPU. added
Updated by Eregon (Benoit Daloze) about 3 years ago
- Related to Feature #18285: NoMethodError#message uses a lot of CPU/is really expensive to call added