Bug #4346
closed
Sort_by! causes uniq! to crash on array of hashes
Description
=begin
Summary:
Sorting an array of hashes before doing a uniq! causes ruby to crash on uniq!.
Program to reproduce:
a = [
{ :color => "blue", :name => "water" },
{ :color => "red", :name => "fire" },
{ :color => "white", :name => "wind" },
{ :color => "green", :name => "earth" },
{ :color => "green", :name => "moss" },
{ :color => "white", :name => "snow" }
]
a.sort_by! { |e| e[:color] }
a.uniq! {|e| e[:color]}
puts a
A workaround is to do the uniq without the bang. Does not crash.
a = a.uniq {|e| e[:color]}
Changing the uniq! line as above will not crash and produce the expected result:
{:color=>"blue", :name=>"water"}
{:color=>"green", :name=>"moss"}
{:color=>"red", :name=>"fire"}
{:color=>"white", :name=>"wind"}
Taking out the sort_by! line also fixes the crashing.
Crash Happens on Ruby versions:
ruby 1.9.2p136 (2010-12-25 revision 30365) [x86_64-darwin10.5.0]
ruby 1.9.2p174 (2011-01-28 revision 30696) [x86_64-darwin10.6.0]
ruby 1.9.2p136 (2010-12-25 revision 30365) [x86_64-linux]
The method sort_by! is not available in 1.8.7. I did not test with any version of 1.8.7.
GDB on Mac yields:
(gdb) run
Starting program: /Users/user/.rvm/rubies/ruby-1.9.2-head/bin/ruby bug.rb
Reading symbols for shared libraries +++... done
Reading symbols for shared libraries . done
Reading symbols for shared libraries . done
Program received signal EXC_BAD_ACCESS, Could not access memory.
Reason: KERN_INVALID_ADDRESS at address: 0x000000000000001c
rb_ary_decrement_share [inlined] () at /Users/user/.rvm/src/ruby-1.9.2-head/array.c:195
195 long num = ARY_SHARED_NUM(shared) - 1;
(gdb) bt
#0 rb_ary_decrement_share [inlined] () at /Users/user/.rvm/src/ruby-1.9.2-head/array.c:195
#1 0x000000010000cc99 in rb_ary_modify (ary=4303797720) at array.c:260
#2 0x00000001000128fe in rb_ary_push [inlined] () at /Users/user/.rvm/src/ruby-1.9.2-head/array.c:721
#3 0x00000001000128fe in push_value (key=<value temporarily unavailable, due to optimizations>, val=4303798360, ary=4303797720) at array.c:260
#4 0x0000000100108990 in st_foreach (table=0x1003bb920, func=0x1000128e0 <push_value>, arg=4303797720) at st.c:778
#5 0x0000000100013573 in rb_ary_uniq_bang (ary=4303797720) at array.c:3435
#6 0x000000010017bc13 in vm_call_cfunc [inlined] () at /Users/user/.rvm/src/ruby-1.9.2-head/vm_insnhelper.c:402
#7 0x000000010017bc13 in vm_call_method (th=0x1003016b0, cfp=0x1004ffef8, num=0, blockptr=0x1004fff21, flag=0, id=<value temporarily unavailable, due to optimizations>, me=0x10033f350, recv=4303797720) at vm_insnhelper.c:260
#8 0x0000000100167dc4 in vm_exec_core (th=0x1003016b0, initial=<value temporarily unavailable, due to optimizations>) at insns.def:1006
#9 0x000000010016fa63 in vm_exec (th=0x1003016b0) at vm.c:1147
#10 0x000000010016fd6b in rb_iseq_eval_main (iseqval=4303800760) at vm.c:1388
#11 0x000000010003f4e2 in ruby_exec_internal (n=0x10086c9b8) at eval.c:214
#12 0x0000000100041e6c in ruby_exec_node [inlined] () at /Users/user/.rvm/src/ruby-1.9.2-head/eval.c:261
#13 0x0000000100041e6c in ruby_run_node (n=<value temporarily unavailable, due to optimizations>) at eval.c:260
#14 0x0000000100000ecf in main (argc=2, argv=0x7fff5fbff368) at main.c:35
GDB on Linux yields:
(gdb) run
Starting program: /home/user/.rvm/rubies/ruby-1.9.2-p136/bin/ruby bug.rb
[Thread debugging using libthread_db enabled]
[New Thread 0x7ffff67b5700 (LWP 1423)]
Program received signal SIGSEGV, Segmentation fault.
rb_ary_decrement_share (ary=6517200) at array.c:195
195 long num = ARY_SHARED_NUM(shared) - 1;
(gdb) bt
#0 rb_ary_decrement_share (ary=6517200) at array.c:195
#1 rb_ary_modify (ary=6517200) at array.c:260
#2 0x00007ffff79f90a3 in rb_ary_push (ary=6517216, item=7746848) at array.c:721
#3 0x00007ffff79f97ec in push_value (key=, val=7746848, ary=0) at array.c:3399
#4 0x00007ffff7ad4962 in st_foreach (table=0x7634e0, func=0x7ffff79f97e0 <push_value>, arg=6517200) at st.c:778
#5 0x00007ffff79fdb0c in rb_ary_uniq_bang (ary=6517200) at array.c:3435
#6 0x00007ffff7b35590 in vm_call_cfunc (th=0x603ce0, cfp=0x7ffff68b5f08, num=0, blockptr=, flag=, id=6517336,
me=0x6bd020, recv=6517200) at vm_insnhelper.c:402
#7 vm_call_method (th=0x603ce0, cfp=0x7ffff68b5f08, num=0, blockptr=, flag=, id=6517336, me=0x6bd020,
recv=6517200) at vm_insnhelper.c:524
#8 0x00007ffff7b285e9 in vm_exec_core (th=0x603ce0, initial=) at insns.def:1006
#9 0x00007ffff7b2edba in vm_exec (th=) at vm.c:1147
#10 0x00007ffff7b2f1f0 in rb_iseq_eval_main (iseqval=6520000) at vm.c:1388
#11 0x00007ffff7a278c2 in ruby_exec_internal (n=) at eval.c:214
#12 0x00007ffff7a278ed in ruby_exec_node (n=0x637cc0) at eval.c:261
#13 0x00007ffff7a2936e in ruby_run_node (n=0x637cc0) at eval.c:254
#14 0x000000000040095b in main (argc=2, argv=0x7fffffffe488) at main.c:35
The crash report (all 1.9's give similar output):
/tmp/bug.rb:64: [BUG] Segmentation fault
ruby 1.9.2p136 (2010-12-25 revision 30365) [x86_64-darwin10.5.0]
-- control frame ----------
c:0004 p:---- s:0010 b:0010 l:000009 d:000009 CFUNC :uniq!
c:0003 p:0090 s:0007 b:0007 l:0014e8 d:001a60 EVAL /tmp/bug.rb:64
c:0002 p:---- s:0004 b:0004 l:000003 d:000003 FINISH
c:0001 p:0000 s:0002 b:0002 l:0014e8 d:0014e8 TOP
-- Ruby level backtrace information ----------------------------------------
/tmp/bug.rb:64:in <main>' /tmp/bug.rb:64:in uniq!'
-- C level backtrace information -------------------------------------------
[NOTE]
You may have encountered a bug in the Ruby interpreter or extension libraries.
Bug reports are welcome.
For details: http://www.ruby-lang.org/bugreport.html
=end
Updated by naruse (Yui NARUSE) about 13 years ago
- Status changed from Open to Closed
- % Done changed from 0 to 100
=begin
This issue was solved with changeset r30738.
Chris, thank you for reporting this issue.
Your contribution to Ruby is greatly appreciated.
May Ruby be with you.
Add ML Reference and a test for r30736 [ruby-core:34997]
=end
Updated by squarism (Chris Dillon) about 13 years ago
=begin
Thank you. I tested with SVN revision stated above and the bug is gone.
=end