Source display doesn't work inside buffered subtests

[petdance]

$ cat foo.t
#!/var/perl/bin/perl

use warnings;
use strict;

use Test2::V0;
use Test2::Plugin::SourceDiag;

plan 2;

is( 'war', 'peace', 'Is war peace?' );

subtest 'Freedom subtest' => sub {
    is( 'freedom', 'slavery' );
};

diag 'Got out of the subtest OK';
is( 'ignorance', 'strength' );

gives

$ prove -v foo.t
[15:54:59] foo.t .. 
# Seeded srand with seed '20170627' from local date.
1..2
not ok 1 - Is war peace?
# Failure source code:
# ------------
# 11: is( 'war', 'peace', 'Is war peace?' );
# ------------
# Failed test 'Is war peace?'
# at foo.t line 11.
# +-----+----+-------+
# | GOT | OP | CHECK |
# +-----+----+-------+
# | war | eq | peace |
# +-----+----+-------+
Can't call method "previous_sibling" on an undefined value at /var/perl5.20.3/lib/site_perl/5.20.3/Test2/Plugin/SourceDiag.pm line 136.
A context appears to have been destroyed without first calling release().
Based on $@ it does not look like an exception was thrown (this is not always
a reliable test)

This is a problem because the global error variables ($!, $@, and $?) will
not be restored. In addition some release callbacks will not work properly from
inside a DESTROY method.

Here are the context creation details, just in case a tool forgot to call
release():
  File: /var/perl5.20.3/lib/site_perl/5.20.3/Test2/Tools/Subtest.pm
  Line: 34
  Tool: Test2::API::run_subtest

Cleaning up the CONTEXT stack...
A context appears to have been destroyed without first calling release().
Based on $@ it does not look like an exception was thrown (this is not always
a reliable test)

This is a problem because the global error variables ($!, $@, and $?) will
not be restored. In addition some release callbacks will not work properly from
inside a DESTROY method.

Here are the context creation details, just in case a tool forgot to call
release():
  File: foo.t
  Line: 15
  Tool: Test2::Tools::Subtest::subtest_buffered

Cleaning up the CONTEXT stack...
# Looks like your test exited with 255 after test #1.
# Did not follow plan: expected 2, ran 1.
Dubious, test returned 255 (wstat 65280, 0xff00)
Failed 2/2 subtests 
[15:54:59]

Test Summary Report
-------------------
foo.t (Wstat: 65280 Tests: 1 Failed: 1)
  Failed test:  1
  Non-zero exit status: 255
  Parse errors: Bad plan.  You planned 2 tests but ran 1.
Files=1, Tests=1,  0 wallclock secs ( 0.02 usr  0.00 sys +  0.18 cusr  0.02 csys =  0.22 CPU)
Result: FAIL

However, if I change that subtest to subtest_streamed it is happy:

$ prove -v foo.t
[15:55:50] foo.t .. 
# Seeded srand with seed '20170627' from local date.
1..2
not ok 1 - Is war peace?
# Failure source code:
# ------------
# 11: is( 'war', 'peace', 'Is war peace?' );
# ------------
# Failed test 'Is war peace?'
# at foo.t line 11.
# +-----+----+-------+
# | GOT | OP | CHECK |
# +-----+----+-------+
# | war | eq | peace |
# +-----+----+-------+
# Subtest: Freedom subtest
    not ok 1 - is( 'freedom', 'slavery' );
    # Failed test 'is( 'freedom', 'slavery' );'
    # at foo.t line 14.
    # +---------+----+---------+
    # | GOT     | OP | CHECK   |
    # +---------+----+---------+
    # | freedom | eq | slavery |
    # +---------+----+---------+
    1..1
not ok 2 - Subtest: Freedom subtest
# Failure source code:
# ------------
# 13: Test2::Tools::Subtest::subtest_streamed 'Freedom subtest' => sub {
# 14:     is( 'freedom', 'slavery' );
# 15: };
# ------------
# Failed test 'Subtest: Freedom subtest'
# at foo.t line 15.
# Got out of the subtest OK
not ok 3 - is( 'ignorance', 'strength' );
# Failed test 'is( 'ignorance', 'strength' );'
# at foo.t line 18.
# +-----------+----+----------+
# | GOT       | OP | CHECK    |
# +-----------+----+----------+
# | ignorance | eq | strength |
# +-----------+----+----------+
# Did not follow plan: expected 2, ran 3.
Dubious, test returned 3 (wstat 768, 0x300)
Failed 2/2 subtests 
[15:55:51]

Test Summary Report
-------------------
foo.t (Wstat: 768 Tests: 3 Failed: 3)
  Failed tests:  1-3
  Non-zero exit status: 3
  Parse errors: Bad plan.  You planned 2 tests but ran 3.
Files=1, Tests=3,  1 wallclock secs ( 0.02 usr  0.01 sys +  0.16 cusr  0.01 csys =  0.20 CPU)
Result: FAIL

[petdance]

Note that the source didn't show up in the 3rd test, probably because it doesn't have a name, as noted in #1.

[petdance]

My mistake. It's not a matter of buffered vs. streamed. It's calling subtest vs. subtest_buffered vs. subtest_streamed.

#!/var/perl/bin/perl

use warnings;
use strict;

use Test2::V0;
use Test2::Plugin::SourceDiag;

plan 3;

Test2::Tools::Subtest::subtest_streamed 'Freedom subtest' => sub {
    is( 'freedom', 'slavery' );
};
Test2::Tools::Subtest::subtest_buffered 'Freedom subtest' => sub {
    is( 'freedom', 'slavery' );
};
subtest 'Freedom subtest' => sub {
    is( 'freedom', 'slavery' );
};

gives

$ prove -v foo.t                                                                                       
[16:47:31] foo.t ..                                                 
# Seeded srand with seed '20170627' from local date.                
1..3                                                                
# Subtest: Freedom subtest                                          
    not ok 1 - is( 'freedom', 'slavery' );
    # Failed test 'is( 'freedom', 'slavery' );'                     
    # at foo.t line 12.                                             
    # +---------+----+---------+                                    
    # | GOT     | OP | CHECK   |                                    
    # +---------+----+---------+                                    
    # | freedom | eq | slavery |                                    
    # +---------+----+---------+
    1..1                                                            
not ok 1 - Subtest: Freedom subtest                                 
# Failure source code:                                              
# ------------                                                      
# 11: Test2::Tools::Subtest::subtest_streamed 'Freedom subtest' => sub {
# 12:     is( 'freedom', 'slavery' );                               
# 13: };                                                                                                                                
# ------------                                                      
# Failed test 'Subtest: Freedom subtest'                            
# at foo.t line 13.
not ok 2 - Freedom subtest {
    not ok 1 - is( 'freedom', 'slavery' );
    1..1
}
    # Failed test 'is( 'freedom', 'slavery' );'
    # at foo.t line 15.
    # +---------+----+---------+
    # | GOT     | OP | CHECK   |
    # +---------+----+---------+
    # | freedom | eq | slavery |
    # +---------+----+---------+
# Failure source code:
# ------------
# 14: Test2::Tools::Subtest::subtest_buffered 'Freedom subtest' => sub {
# 15:     is( 'freedom', 'slavery' );
# 16: };
# ------------
# Failed test 'Freedom subtest'
# at foo.t line 16.
Can't call method "previous_sibling" on an undefined value at /var/perl5.20.3/lib/site_perl/5.20.3/Test2/Plugin/SourceDiag.pm line 136.
A context appears to have been destroyed without first calling release().
Based on $@ it does not look like an exception was thrown (this is not always
a reliable test)

This is a problem because the global error variables ($!, $@, and $?) will
not be restored. In addition some release callbacks will not work properly from
inside a DESTROY method.

Here are the context creation details, just in case a tool forgot to call
release():
  File: /var/perl5.20.3/lib/site_perl/5.20.3/Test2/Tools/Subtest.pm
  Line: 34
  Tool: Test2::API::run_subtest

Cleaning up the CONTEXT stack...
A context appears to have been destroyed without first calling release().
Based on $@ it does not look like an exception was thrown (this is not always
a reliable test)

This is a problem because the global error variables ($!, $@, and $?) will
not be restored. In addition some release callbacks will not work properly from
inside a DESTROY method.

Here are the context creation details, just in case a tool forgot to call
release():
  File: foo.t
  Line: 19
  Tool: Test2::Tools::Subtest::subtest_buffered

Cleaning up the CONTEXT stack...
# Looks like your test exited with 255 after test #2.
# Did not follow plan: expected 3, ran 2.
Dubious, test returned 255 (wstat 65280, 0xff00)
Failed 3/3 subtests 
[16:47:32]

Test Summary Report
-------------------
foo.t (Wstat: 65280 Tests: 2 Failed: 2)
  Failed tests:  1-2
  Non-zero exit status: 255
  Parse errors: Bad plan.  You planned 3 tests but ran 2.
Files=1, Tests=2,  1 wallclock secs ( 0.02 usr  0.00 sys +  0.12 cusr  0.01 csys =  0.15 CPU)
Result: FAIL

The first two going to subtest_buffered or subtest_streamed work. Going to subtest provided by ::V0 is what it can't handle.