Unfortunately MetaCPAN has some problems and seems to serve stale data.
MetaCPAN's recent page is usually many hours behind (currently it's 17h)
and possibly also other MetaCPAN APIs have the same problem. I opened
https://github.com/metacpan/metacpan-api/issues/1388 some days ago.
Regards,
Slaven
23. 10. 2025. u 23:14, Doug Bell piše:
This is happening because the CPAN Testers database is missing some
PAUSE data. I'm running a re-sync from MetaCPAN which should correct
that.
Doug Bell
[email protected]
On Oct 3, 2024, at 6:59 AM, James E Keenan <[email protected]> wrote:
Many CPANtesters reports are being entered into the metabase without
their [AUTHOR/Distribution-Name] data being properly recorded. As a
result, the summary pages for those distributions are not being
updated properly at fast-matrix.cpantesters.org and at
matrix.cpantesters.org.
Just now I went to http://metabase.cpantesters.org/tail/log.txt and
searched for the string '[/]'. I got 33 matches -- reports whose
uploader and distribution name were not recorded. Now, it happens
that all these reports were from '[Chris Williams (BINGOS)]', but
*that* string was found 383 times on the page.
Moreover, I noticed this over the weekend for reports I myself had
generated. Type-Tiny version 2.006000 was released 3 days ago
(https://metacpan.org/pod/Type::Tiny) and I generated some reports
that appeared in the log file with '[/]'. At
http://fast-matrix.cpantesters.org/?dist=Type-Tiny, you will be told
"NOTE: no report for latest version 2.006000" and the most recent
version with reports presented will be 2.005_002. On the other hand,
when I went just now to
http://matrix.cpantesters.org/?dist=Type-Tiny, a matrix for reports
for 2.006000 was shown but was told "Reusing old cached file, 2.0
days old".
Can this be investigated? We need this information to determine
whether certain CPAN distributions have worked around "breakage" in
recent development versions of the core.
Thank you very much.
