As part of my work for Perl 5 Porters in analyzing "Blead Breaks CPAN" (BBC) bug tickets, I frequently have occasion to build perl at the HEAD of the blead (main development) branch, install it and 'cpanm' locally, install Task::CPAN::Reporter against that perl, then try to install CPAN distributions which have been cited in BBC.

Yesterday, I built and installed perl at this commit:

#####
$ ./bin/perl -Ilib -v | head -2 | tail -1
This is perl 5, version 37, subversion 3 (v5.37.3 (v5.37.2-33-gff5bbe677e)) built for amd64-freebsd-thread-multi
#####

I then installed Task::CPAN::Reporter (via 'cpanm') and then used 'cpan' to install Test::Trap and Sub::Metadata, generating CPANtesters reports along the way.

#####
...
cpan[1]> install Sub::Metadata Test::Trap
...
CPAN::Reporter: Test result is 'pass', All tests successful.
CPAN::Reporter: preparing a CPAN Testers report for Sub-Metadata-0.002
CPAN::Reporter: sending test report with 'pass' via Metabase
...
CPAN::Reporter: Test result is 'pass', All tests successful.
CPAN::Reporter: preparing a CPAN Testers report for Test-Trap-v0.3.5
CPAN::Reporter: sending test report with 'pass' via Metabase
...
$ ./bin/perl -Ilib -MTask::CPAN::Reporter -MTest::Trap -MSub::Metadata -E 'say q|hello world|;'
hello world
#####

These two CPAN distros (and their prerequisites) were duly logged as having had CPANtesters reports generated:

#####
$ tail -n 4 ~/.cpanreporter/reports-sent.db
test PASS Module-Build-0.4231 (perl-5.37.3) amd64-freebsd-thread-multi 12.3-release-p1 test PASS Sub-Metadata-0.002 (perl-5.37.3) amd64-freebsd-thread-multi 12.3-release-p1 test PASS Data-Dump-1.25 (perl-5.37.3) amd64-freebsd-thread-multi 12.3-release-p1 test PASS Test-Trap-v0.3.5 (perl-5.37.3) amd64-freebsd-thread-multi 12.3-release-p1
#####

However, when I checked matrix.cpantesters.org today for those reports, I found reports for only 2 of those 4 distros:

#####
Distribution|Report URL|Timestamp
Data-Dump|http://www.cpantesters.org/cpan/report/a8c864ee-1050-11ed-84a8-20dc05b3d53a|2022-07-30 21:43 Test-Trap|http://www.cpantesters.org/cpan/report/ad848512-1050-11ed-a8e1-724c06b3d53a|2022-07-30 21:43
#####

#####
Distribution|No Report URL
Module-Build|no report from me on Jul 30 2022 at http://matrix.cpantesters.org/?dist=Module-Build;os=freebsd;reports=1#sl=4,1 Sub-Metadata|no report from me on Jul 30 2022 at http://matrix.cpantesters.org/?dist=Sub-Metadata;os=freebsd;reports=1#sl=4,1
#####

For reference, as I write this message it is now:

#####
$ date
Sun Jul 31 13:39:36 UTC 2022
#####

... approximately 16 hours after the tests were generated.

While I understand that there are delays (hours?) between the time a report is logged at http://metabase.cpantesters.org/tail/log.txt and the time it can be located via http://matrix.cpantesters.org/, my expectation has been that *all* reports generated in a given batch should become available via the web interface simultaneously.

Is that expectation incorrect?

If not, then why have the reports for Module-Build and Sub-Metadata not yet appeared on http://matrix.cpantesters.org?

Reply via email to