develooper Front page | perl.qa | Postings from March 2017

Re: BBC testing from upriver downwards

Thread Previous | Thread Next
From:
Neil Bowers
Date:
March 21, 2017 09:26
Subject:
Re: BBC testing from upriver downwards
Message ID:
C2D14107-A3E8-4DCF-8B3D-E39C5F00B69B@cogendo.com
> ​> ​1. Compose a list of CPAN distros starting with those farthest up river, i.e., distros that only depend on the perl 5 core.  Within that set of distros I'd like to order them from most reverse dependencies to fewest.  Then go down river from there.
> 
> ​David Golden has a script for generating the CPAN river heirarchy -- the last run of this that I know of (at least, it's the data I've been working off of) is at https://gist.github.com/xdg/af7a32c5b21d45a6b255 <https://gist.github.com/xdg/af7a32c5b21d45a6b255>. It lists the distributions in their position on the CPAN river, and their top 5 users (reverse dependencies).  This would be a natural place to start testing, and determining which branches of the river have blockages (the first instance of a no-dot-in-INC issue).  David, can you generate a fresh copy?​

I am generating the river data roughly weekly. You can get it from here:

	http://neilb.org/river-of-cpan.json.gz <http://neilb.org/river-of-cpan.json.gz>

The file is a big array of hashes, one hash per distribution. Here’s an example, for Try-Tiny:

	{
	    "dist" : "Try-Tiny”,
	    "total" : 13045,
	    "bucket" : 5,
	    "immediate" : 1077
	},

The keys are:
“immediate” is the number of directly dependent CPAN distributions.
“total” is the total number of dists dependent on it.
“bucket” chunks the river: bucket 5 is dists with 10k or more total dependents. Bucket 4 is 1k up to 9999, and so on until bucket 0, which is dists that have no dependents.

At the moment there are 56 dists in bucket 5 (“the head of the river”).

This file is being imported into MetaCPAN. For example, for Try-Tiny, have a look at:

	https://fastapi.metacpan.org/distribution/Try-Tiny <https://fastapi.metacpan.org/distribution/Try-Tiny>

At the moment this is being generated as a by-product of my scripts that run the PRC, generate the adoption list, etc. I’m currently working on a stand-alone generator for this data, which I’m hoping to have complete by PTS, so I can work on productising it with Leo Lapworth.

Neil



Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About