glyrepr 0.13.0

Breaking changes

• Remove the .parallel argument from smap(), smap2(), spmap(), and their variants, and drop the unused furrr and future dependencies. (#62)

New features

• Add structure_nodes(), structure_edges(), and structure_from_tibbles() for converting glycan structures to and from graph-table tibbles. (#60)
• Rename get_anomer_pos() to infer_anomer_pos() to better describe that the position is inferred from a monosaccharide name; get_anomer_pos() remains available as a backward-compatible alias. (#61)
• get_anomer_pos() now accepts generic monosaccharide names such as "Hex". (#57)
• Add NGc and Gc to the supported substituent list for N-glycolyl and glycolyl substituents. (#54)
• Add as.list() support for glyrepr_composition and glyrepr_structure vectors. (#52)

Minor improvements and bug fixes

• glyrepr_structure objects no longer return TRUE from is.character(). Use as.character() for explicit IUPAC-condensed string conversion. (#59)
• Fix as_glycan_structure() so character vectors containing NA reject mixed concrete and generic structures consistently. (#58)
• Fix as_glycan_structure() parsing for monosaccharide names that start with digits, such as 6dGul and 4eLeg. (#55)
• Fix as_glycan_structure() parsing for substituent names such as Pyr, PC, PPEtn, and PEtn that share the P prefix. (#53)
• Ambiguous linkages like a2-3/6 are also regarded as unknown for has_linkage() and get_structure_level(). (#51)