toplevel location code can now deal with excessively large regions and shrinks them down to size if possible or rejects them.

load_XXXX() methods now return the number of adaptors which have appeared in the registry due to that method's invocation. This can be used to report success or failure to load databases.

Improving the error handling of the DBI connection in load_registry_from_db(). We now report an error if it cannot connect to the MySQL DB and if it was not pingable. Hopefully this catches both erronious scenarios.

similarity events.  Require similarity events to be within 1.5% (not 2%)
of top score.

Reformat routine while here.

Bigger penalty for exonerate matches on other seq_regions.

Rescore penalty altered to 0.9.

gene rescore penalty from 0.8 to 0.9.

in the ananlysis table, return a reference to an empty list.  Without
this change, all features would have been returned which would quickly
make the client run out of memory.

added fetch_all_by_name which gets all the xrefs that have the accession, display_label or synonym that is passed to it. external database source is optional

warning if species alias is not found.

In get_DBAdaptor(): Thow exception if species alias not found (excactly
as in get_adaptor()).

merge in changes from ensembl-64 for retrieving non-reference slices. Don't need to supply a value for logappend, verbose, interactive and dry_run parameters

Functionality suggested by Ensembl Genomes, used when building a variation DB.
New flag: start_equals_end which controls the simplified sql for 1bp slice.

(defaults to 200).

Test will fail if run on non-deterministic data storage engines as we have 2 features with the queried for ditag_id

Introduced Utils::IO namespace along with serializer code for GFF3 and human-readable formats. Pre-release, not intended for public use.
Also added Utils::BiotypeMapper that performs ontology mappings.

	formatting

	safe use of @aliases (don't expect registry to make copy of the list)

	quote argument names and values