Skip Menu |

This queue is for tickets about the Perlmazing CPAN distribution.

Report information
The Basics
Id: 112430
Status: resolved
Worked: 35 min
Priority: 0/
Queue: Perlmazing

People
Owner: ZARABOZO [...] cpan.org
Requestors: SREZIC [...] cpan.org
Cc:
AdminCc:

Bug Information
Severity: (no value)
Broken in: (no value)
Fixed in: 1.2808



Subject: Test fails (directory order; 1.2806)
It seems that you cannot rely on a specific directory or File::Find order: # Failed test 'callback' # at t/10-dir.t line 29. # got: '( # "t/test_dir/aAa", # "t/test_dir/aAa/bBb", # "t/test_dir/aAa/bBb/cCc", # "t/test_dir/aAa/bBb/cCc/level_three.txt", # "t/test_dir/aAa/bBb/level_two.txt", # "t/test_dir/aAa/leve_one.txt", # "t/test_dir/aAa_Copy", # "t/test_dir/aAa_Copy/bBb", # "t/test_dir/aAa_Copy/bBb/cCc", # "t/test_dir/aAa_Copy/bBb/cCc/level_three.txt", # "t/test_dir/aAa_Copy/bBb/level_two.txt", # "t/test_dir/aAa_Copy/bBb_Copy", # "t/test_dir/aAa_Copy/bBb_Copy/cCc", # "t/test_dir/aAa_Copy/bBb_Copy/cCc/level_three.txt", # "t/test_dir/aAa_Copy/bBb_Copy/cCc_Copy", # "t/test_dir/aAa_Copy/bBb_Copy/cCc_Copy/level_three.txt", # "t/test_dir/aAa_Copy/bBb_Copy/level_two.txt", # "t/test_dir/aAa_Copy/leve_one.txt", # "t/test_dir/level_zero.txt", # )' # expected: '( # "t/test_dir/level_zero.txt", # "t/test_dir/aAa_Copy", # "t/test_dir/aAa_Copy/leve_one.txt", # "t/test_dir/aAa_Copy/bBb_Copy", # "t/test_dir/aAa_Copy/bBb_Copy/level_two.txt", # "t/test_dir/aAa_Copy/bBb_Copy/cCc_Copy", # "t/test_dir/aAa_Copy/bBb_Copy/cCc_Copy/level_three.txt", # "t/test_dir/aAa_Copy/bBb_Copy/cCc", # "t/test_dir/aAa_Copy/bBb_Copy/cCc/level_three.txt", # "t/test_dir/aAa_Copy/bBb", # "t/test_dir/aAa_Copy/bBb/level_two.txt", # "t/test_dir/aAa_Copy/bBb/cCc", # "t/test_dir/aAa_Copy/bBb/cCc/level_three.txt", # "t/test_dir/aAa", # "t/test_dir/aAa/leve_one.txt", # "t/test_dir/aAa/bBb", # "t/test_dir/aAa/bBb/level_two.txt", # "t/test_dir/aAa/bBb/cCc", # "t/test_dir/aAa/bBb/cCc/level_three.txt", # )' # Looks like you failed 1 test of 3. t/10-dir.t ..........................
On 2016-02-25 14:33:11, SREZIC wrote: Show quoted text
> It seems that you cannot rely on a specific directory or File::Find order: > > # Failed test 'callback' > # at t/10-dir.t line 29. > # got: '( > # "t/test_dir/aAa", > # "t/test_dir/aAa/bBb", > # "t/test_dir/aAa/bBb/cCc", > # "t/test_dir/aAa/bBb/cCc/level_three.txt", > # "t/test_dir/aAa/bBb/level_two.txt", > # "t/test_dir/aAa/leve_one.txt", > # "t/test_dir/aAa_Copy", > # "t/test_dir/aAa_Copy/bBb", > # "t/test_dir/aAa_Copy/bBb/cCc", > # "t/test_dir/aAa_Copy/bBb/cCc/level_three.txt", > # "t/test_dir/aAa_Copy/bBb/level_two.txt", > # "t/test_dir/aAa_Copy/bBb_Copy", > # "t/test_dir/aAa_Copy/bBb_Copy/cCc", > # "t/test_dir/aAa_Copy/bBb_Copy/cCc/level_three.txt", > # "t/test_dir/aAa_Copy/bBb_Copy/cCc_Copy", > # "t/test_dir/aAa_Copy/bBb_Copy/cCc_Copy/level_three.txt", > # "t/test_dir/aAa_Copy/bBb_Copy/level_two.txt", > # "t/test_dir/aAa_Copy/leve_one.txt", > # "t/test_dir/level_zero.txt", > # )' > # expected: '( > # "t/test_dir/level_zero.txt", > # "t/test_dir/aAa_Copy", > # "t/test_dir/aAa_Copy/leve_one.txt", > # "t/test_dir/aAa_Copy/bBb_Copy", > # "t/test_dir/aAa_Copy/bBb_Copy/level_two.txt", > # "t/test_dir/aAa_Copy/bBb_Copy/cCc_Copy", > # "t/test_dir/aAa_Copy/bBb_Copy/cCc_Copy/level_three.txt", > # "t/test_dir/aAa_Copy/bBb_Copy/cCc", > # "t/test_dir/aAa_Copy/bBb_Copy/cCc/level_three.txt", > # "t/test_dir/aAa_Copy/bBb", > # "t/test_dir/aAa_Copy/bBb/level_two.txt", > # "t/test_dir/aAa_Copy/bBb/cCc", > # "t/test_dir/aAa_Copy/bBb/cCc/level_three.txt", > # "t/test_dir/aAa", > # "t/test_dir/aAa/leve_one.txt", > # "t/test_dir/aAa/bBb", > # "t/test_dir/aAa/bBb/level_two.txt", > # "t/test_dir/aAa/bBb/cCc", > # "t/test_dir/aAa/bBb/cCc/level_three.txt", > # )' > # Looks like you failed 1 test of 3. > t/10-dir.t ..........................
This is on a Linux system. On a FreeBSD system this test seems to pass...
On Thu Feb 25 14:33:11 2016, SREZIC wrote: Show quoted text
> It seems that you cannot rely on a specific directory or File::Find order:
Thank you very much for this report. That's a very strange failure. Both results are supposed to be built up with the same function, so I would expect the same exact order. The only difference is one of them having a callback attached to build the list. Can I please have the details of this test so I can try to replicate it? (OS/Perl versions) Thanks in advance, Francisco Zarabozo
On 2016-02-25 14:48:34, ZARABOZO wrote: Show quoted text
> On Thu Feb 25 14:33:11 2016, SREZIC wrote:
> > It seems that you cannot rely on a specific directory or File::Find > > order:
> > Thank you very much for this report. > > That's a very strange failure. Both results are supposed to be built > up with the same function, so I would expect the same exact order. The > only difference is one of them having a callback attached to build the > list. > > Can I please have the details of this test so I can try to replicate > it? (OS/Perl versions) >
See http://fast-matrix.cpantesters.org/?dist=Perlmazing%201.2806 By now the pattern is clear: it fails on all linux systems and passes on all freebsd systems, regardless of the perl version (so it is not a hash randomization issue) On the linux systems I checked if it depends on the build file system type, but it makes no change whether the tests run on ext4 or tmpfs.
On Thu Feb 25 15:20:11 2016, SREZIC wrote: Show quoted text
> On 2016-02-25 14:48:34, ZARABOZO wrote:
> > On Thu Feb 25 14:33:11 2016, SREZIC wrote:
> > > It seems that you cannot rely on a specific directory or File::Find > > > order:
> > > > Thank you very much for this report. > > > > That's a very strange failure. Both results are supposed to be built > > up with the same function, so I would expect the same exact order. > > The > > only difference is one of them having a callback attached to build > > the > > list. > > > > Can I please have the details of this test so I can try to replicate > > it? (OS/Perl versions) > >
> > See http://fast-matrix.cpantesters.org/?dist=Perlmazing%201.2806 > > By now the pattern is clear: it fails on all linux systems and passes > on all freebsd systems, regardless of the perl version (so it is not a > hash randomization issue) > > On the linux systems I checked if it depends on the build file system > type, but it makes no change whether the tests run on ext4 or tmpfs.
Fixed in version 1.2808. Thanks again for your report! :-)