Skip Menu |
 

This queue is for tickets about the Encode-ISO2022 CPAN distribution.

Report information
The Basics
Id: 117157
Status: open
Priority: 0/
Queue: Encode-ISO2022

People
Owner: Nobody in particular
Requestors: KENTNL [...] cpan.org
Cc:
AdminCc:

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



Subject: Runs forever on 5.25.4, stuck encoding chr(720 ) ( U+2D0 )

This is likely to be due to either changes in Perl itself, or changes in Encoding.pm

There are no other smoke reports listed for 5.25.4, I suspect, due to them running forever.

I forced a report by tripping a failure on purpose by sending a SIGALARM after it had been idling for 20 minutes on the first run, and 2 minutes on the second run, because after consulting previous PASSes, it seemed normal times were in seconds, not minutes.

http://www.cpantesters.org/cpan/report/43fcc7de-6850-11e6-bba0-2276043e2754

The failure appears to be in :

eval { $txt = Encode::encode($charset, Encode::decode_utf8($uni), 1) };
 

And that line of code runs indefinately, and strace only indicates loads of calls to brk()

 

Reducing it further I found that the following code also produces the same "block"


require Encode::ISO2022JP2;
my $text = Encode::encode( 'iso2022jp-2', chr(720) );

And this test case gets stuck allocating memory as well...

brk(0x1347000)                          = 0x1347000
mremap(0x7f67a1605000, 163840, 204800, MREMAP_MAYMOVE) = 0x7f67a15d3000
mremap(0x7f67a15d3000, 204800, 253952, MREMAP_MAYMOVE) = 0x7f67a15d3000
mremap(0x7f67a15d3000, 253952, 319488, MREMAP_MAYMOVE) = 0x7f67a15d3000
mremap(0x7f67a15d3000, 319488, 397312, MREMAP_MAYMOVE) = 0x7f67a1568000
mremap(0x7f67a1568000, 397312, 495616, MREMAP_MAYMOVE) = 0x7f67a14ef000
mremap(0x7f67a14ef000, 495616, 618496, MREMAP_MAYMOVE) = 0x7f67a14ef000
mremap(0x7f67a14ef000, 618496, 774144, MREMAP_MAYMOVE) = 0x7f67a14ef000
mremap(0x7f67a14ef000, 774144, 966656, MREMAP_MAYMOVE) = 0x7f67982f4000
mremap(0x7f67982f4000, 966656, 1208320, MREMAP_MAYMOVE) = 0x7f67a14a2000
mremap(0x7f67a14a2000, 1208320, 1507328, MREMAP_MAYMOVE) = 0x7f6798270000
mremap(0x7f6798270000, 1507328, 1884160, MREMAP_MAYMOVE) = 0x7f67980a4000
 

 

 

 

 

 

 

Subject: Runs forever on Encode >=2.85 (perl-5.25.4), stuck encoding chr(720 ) ( U+2D0 )

Bit of downgrading indicates Encode 2.85 and 2.86 are to blame for this.

 

Encode 2.84 fails its own tests on 5.25.4, however, Encode-ISO2202 works on Encode 2.84 and perl 5.25.4

Download (untitled) / with headers
text/plain 1.7k
On 2016-08-23 05:04:17, KENTNL wrote: Show quoted text
> Bit of downgrading indicates Encode 2.85 and 2.86 are to blame for this. > > Encode 2.84 fails its own tests on 5.25.4, however, Encode-ISO2202 works on > Encode 2.84 and perl 5.25.4
I can confirm the problem on my smoker machines (linux & freebsd) and I can confirm that it fails only with Encode 2.85 and up. Statistical analysis of reports generated on my smokers (negative theta is bad): **************************************************************** Regression 'mod:Encode' **************************************************************** Name Theta StdErr T-stat [0='const'] 1.0000 0.0000 5728387080189300.00 [1='eq_2.42_01'] 0.0000 0.0000 0.22 [2='eq_2.49'] -0.0000 0.0000 -1.65 [3='eq_2.60'] 0.0000 0.0000 0.61 [4='eq_2.62'] -0.0000 0.0000 -0.22 [5='eq_2.68'] 0.0000 0.0000 0.22 [6='eq_2.72'] 0.0000 0.0000 0.30 [7='eq_2.73'] 0.0000 0.0000 0.00 [8='eq_2.74'] 0.0000 0.0000 1.24 [9='eq_2.75'] 0.0000 0.0000 0.45 [10='eq_2.76'] 0.0000 0.0000 1.16 [11='eq_2.78'] 0.0000 0.0000 2.38 [12='eq_2.80'] -0.0000 0.0000 -3.60 [13='eq_2.82'] 0.0000 0.0000 0.00 [14='eq_2.83'] 0.0000 0.0000 1.35 [15='eq_2.84'] -0.0000 0.0000 -1.21 [16='eq_2.85'] -1.0000 0.0000 -4677208465205122.00 [17='eq_2.86'] -1.0000 0.0000 -5645959947169074.00 R^2= 1.000, N= 114, K= 18 ****************************************************************


This service is sponsored and maintained by Best Practical Solutions and runs on Perl.org infrastructure.

Please report any issues with rt.cpan.org to rt-cpan-admin@bestpractical.com.