gws_biota

Product features

• New task to zip/export the BIOTA database (BiotaDbZipper)

Technical features

• All 10 DB creator tasks now validate that no critical column is entirely NULL after creation — raises an error if a column has zero non-NULL values, catching parsing/format compatibility issues early
• DbService.ensure_pigz_installed(): pigz is now automatically installed at runtime if missing (required for .tar.gz decompression in Taxonomy and Enzyme creators)
• Compound creator: removed global transaction wrapper to prevent MySQL "server has gone away" timeout during long parsing operations
• Taxonomy and Enzyme creators: .tar.gz files are now extracted using Python's built-in tarfile module as fallback, eliminating the hard dependency on pigz

Bug Fixes

• Fixed ChEBI v252 format compatibility: formula, inchi, inchikey, smiles, mass, monoisotopic_mass, charge columns were all NULL because ChEBI migrated property namespaces from http://purl.obolibrary.org/obo/chebi/* to chemrof:* — both formats are now supported
• Fixed kegg_id and metacyc_id xref parsing: ChEBI v252 changed KEGG: → kegg.compound: and MetaCyc: → metacyc.compound: — both formats now handled
• Fixed Docker: corrected BIOTA DB container name

Breaking Changes

• None