Boost logo

Boost :

Subject: Re: [boost] Library list generated from meta/libraries.json files
From: Daniel James (dnljms_at_[hidden])
Date: 2015-01-05 12:14:14


On 5 January 2015 at 14:10, Peter Dimov <lists_at_[hidden]> wrote:
> I've put a library list generated from the meta/libraries.json files, at:
>
> http://www.pdimov.com/tmp/boost/index.html

We're already dealing with this on the website, this is how the
documentation list is generated. The full data is at:

http://www.boost.org/doc/libraries.xml

Which was set up by Rene years ago. It includes libraries with no
existing metadata file. Nowadays it's maintained by a mix of php
scripts and manual editing. The way it stores historical data is
awkward because I shoehorned it into existing code, so it'll probably
be easier to handle if I write a little web service to get a release's
details in json format.

> Library authors may wish to take a look at their library entry and see if
> they're happy with it. What I've noticed so far is that many libraries are
> missing the boost-version (First Release) entry, and that Config is missing
> authors for some reason.

Historical data is handled centrally, so it would be redundant and
error prone to store it in the library repos (e.g. if a library was
pulled from a release, but wasn't updated accordingly). But some of
the first metadata files still have it.

> The other thing is that the category names in the json files are the
> internal identifiers, which I've left as-is, as I didn't want to hardcode a
> mapping. We'll have to decide whether to use the proper names in the json
> files, or whether to have a separate categories.json with the mapping
> somewhere. I tend towards the first option.

As you've probably noticed by now, the categories are in the xml file.


Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk