MTA contains a Lucene index of sha1 hash values for known software artifacts.
It currently contains about 27 million entries. The size of the index is causing the MTA distributions, that have the index bundled within them, to get very large.

In the medium term we need to think about alternative ways off identifying known software artifacts.

But in the short term, and the purpose of this Jira, is to trim the index to remove superfluous entries (artifacts that are Release candidates, alpha, beta, snapshots, etc.).

artifacts that contain the following (case insensitive) in their name should be excluded
alpha, beta, .pre, -pre, .rc, -rc, snap

Other suffixes that we may want to consider using to eliminate

• android
• artifacts are those that follow this pattern (e.g. Jetty server)
  x.y.z.vCCYYMMDD