On-demand crawling

[tomzx]

Hello,

Here's my current use case: I use versioneye to track the dependencies of internal projects (mostly PHP-related). Obviously we do not use all the packages available on packagist. We would also like not to depend on VersionEye public API to obtain packages information. However, crawling all of packagist while we're probably interested in 500-1000 packages seems a bit excessive.

Thus, it would probably make sense to have tracked projects dictate which packages should be crawled. Then, the crawler/producer would only have to walk through these packages in order to fetch a minimal list of metadata.

[reiz]

Hi @tomzx. Are you using your own VersionEye instance? Like described here: https://github.com/versioneye/ops_contrib?

[tomzx]

@reiz Yes. Sorry if that wasn't clear from my post.