Re: CPAN Statistics API for mswin32 build failures

> On Feb 14, 2020, at 6:05 PM, David Cantrell <[email protected]> wrote:
> 
> On 2020-02-14 18:55, Robert Ryley wrote:
> 
>> I'm looking to see if anyone had previously looked for patterns of errors across modules on a particular platform -- especially mswin32. 
> 
> Then you might want to look at Andreas's analysis site http://analysis.cpantesters.org/

There's also the "CPAN River", which rates distributions based on how many descendants depend on them. To my knowledge, those two statistics (platform and river position) have not been combined for analysis.

Another option you have to do analysis on large sections of the database is to build your own database and populate it from the primary database. The CPAN::Testers::Schema project can do that (using the "cpantesters-schema fetch" command, which could be enhanced to pull different sets of data based on criteria other than distribution+version), and there's docker tools available to spin up a local CPAN Testers infrastructure in https://github.com/cpan-testers/cpantesters-deploy.

Once you have a local database, you can use SQL tools to do your analysis, which will likely be a lot easier than using the API :)

As for the documentation: If you've got specific questions, feel free to ask them here, on IRC (https://kiwiirc.com/nextclient/#irc://irc.perl.org/%23cpantesters-discuss?nick=mc-guest-?), or in the Github issue tracker. If you see anything that's out-of-date or not well-covered, please submit patches (especially to the OpenAPI docs, which could use some).