GitLab

make sure the cache stores the same case as the original sequence else it will look wierd (as if softmasked) if you get it again using the cache

fixed bug where the dbID of an xref that had already been stored was not set in the object when store was called

for origin overlapping genes take into account exons with negative start and with start greater then end

it looks like the features overlapping origin can be attached to a normal slice as well as to an overlapping origin

make sure the sequence is retrieved for both ends of the region overlapping circular chromosome origin

specified.  Patch from Dan S.

not needed and move declaration of a statment handle to aviod having to
declare another one.

Slice::dbID().

for specifying inheritance.

conditional operator.

In fetch_all_by_Transcript():  Be absolutely sure not to add the
exon_transcript table twice.  Should solve a bug reported by Jan.

From Michael S.

external DB type strings contains '%' and use "LIKE" or "=" accordingly.
Document this behaviour in the POD of the method.  Also, safely quote
strings inserted into SQL.

In _type_by_external_db_id(): Fix documentation for method.

Both pointed out by Michael S.

  Fix misspelt "Exmaple".

  In AUTOLOAD(): warning() before throw(), otherwise not much joy.

  Fix documentation for register_assembled(), register_component(),
  _register_chained_special().

  Remove unused variable in _build_combined_mapper().

  Initialise $mid_cs_id in _register_chained_special() properly.

  Use splice() to remove a list element in register_all_chained() rather
  than setting the list length (say what?!).

for the canonical transcript in the canonical_transcript() method of the
Gene object, and also use it in the store() method in GeneAdaptor to
properly set the canonical transcript of genes.

Thanks Bronwen for the heads up.

to zero and then update it when we know what the new internal ID of the
transcript is.

<ensembl-dev@ebi.ac.uk> to <dev@ensembl.org>.

_fetch_by_Slice_start_end_strand_circular().