'' instead of NA for missing values at the suggestion of Davis Vaughan (issue #212).vctrs update.Minor update
taxonomy printing errordefault.stringsAsFactors since it is depreciated.LARGE CHANGES:
The beginning of a complete rewrite of the taxa package to make the more basic component classes more like base R vectors. The taxmap class is not yet reimplemented, but will be similar to the class in the previous versions of taxa. The old version of taxa has been incorporated into the metacoder package until this version of taxa is mature, at which time metacoder will also use this version.
n_obs that would cause an error when used on an object with tables with columns named by numbers.taxmap are now supportedtaxonomy_table option add_id_col now works (issue #191).parse_tax_data option class_col now accepts negative column indexes, meaning “all other columns”.taxonomy_table function that converts the information in a taxmap or taxonomy object into a table with taxa as rows and ranks as columns.print_tree function that prints text-based trees of taxmap or taxonomy objects (issue #173).get_dataset function to get a single data set from taxmap objects. Useful for piping with %>%.filter_taxa and filter_obs can now subset anything that has names, length, and can be subset, not just tables, lists, and vectors. For example, DNAbin objects from the ape package can now be used in taxmap objects (issue #178).taxmap and taxonomy parsers now treat taxa with the same name and same place in the taxonomy, but different ranks, database IDs, or authorities, as different taxa.filter_obs can now filter multiple datasets at once if they are the same length (issue #179).select_obs and arrange_obs can now work on multiple datasets at once."taxon_rank" value for the class_key options work with extract_tax_data.taxmap print method when printing tables with only a taxon ID column (issue #181).target in many functions renamed to data to make it more intuitive.parse_tax_data can now incorporate rank information which can be accessed by result$taxon_ranks() (issue #113).taxmap print methods now have more information and color (issue #124).leaves_apply function that works like subtaxa_apply, but on leaves (issue #126).value option now return named taxon indexes by default, instead of unnamed taxon indexes (issue #128).lookup_tax_data and extract_tax_data can now use “fuzzy” matching when looking up taxon names, so taxon names can be misspelled and still be founds.lookup_tax_data and extract_tax_data now only look up unique sequence IDs, improving download speed.filter_obs now can filter out observations in non-target data sets that are associated with taxa that are removed when drop_taxa = TRUE (issue #143). This is done using filter_taxa, so the supertaxa, subtaxa, and reassign_obs options are now available to filter_obs to control how taxon removal is done.lookup_tax_data and extract_tax_data now have progress bars instead of printing lots of text when downloading information.mutate_obs now creates new vector/tables if the data set specified does not exist (issue #121).filter_taxa option keep_order that preserves input taxon order. It is TRUE by default, which changes how it used to work. Set to FALSE for old behavior.simplify option in many functions is now always handled the same way: If all vectors in a list are names, then unique key-value pairs are returned. Otherwise, names are ignored and unique values are returned.leaves option now behaves like subtaxa, returning all leaves for each taxon. The old behavior can be replicated by setting the new simplify option to TRUE (issue #127).filter_taxa now has better error messages for invalid inputs (issue #117).filter_taxa when no taxa pass filter (issue #116).parse_tax_data when class_key was not named (issue #131).hierarchy print method with taxon_id class was not used (issue #138).parse_tax_data when all classification data is NA.taxmap print method when printing zero-length lists and vectors (issue #148).subset (issue #88, issue #89)lookup_tax_data and extract_tax_data that caused an error when one of the queries failed too download.obs_apply to not work when passed as a variable (issue #97)map_data_ for mapping without using NSE.n_obs and n_obs_1 and make them available for NSE (issue #91parse_tax_data/extract_tax_data can now parse things like phylum;Nitrosopumilales;order;Nitrosopumilaceae;family; and split out the rank and taxon names by using multiple matches to the class_regex when class_sep is NULL.extract_tax_data now gives warnings if a regex does not match.n_supertaxa_1 function to get number of immediate supertaxa (always 1 or 0).branches function to go with roots, leaves, and stems. (issue #56)internodes and is_internode functions to go with roots, leaves, branches, and stems. Useful for removing uninformative taxonomic ranks/taxa.taxon, taxon_name, taxon_id, taxon_rank, and taxa to handle NULL inputs as first class citizens to handle cases when you have essentially a blank taxon (use case comes from taxize package) #95 #107classification as an abstraction to get either hierarchy of taxon indexes, names, or ids (issue #57)get_data_frame for both Taxonomy and Taxmap objects that wraps around get_data to coerce into a data.frame. (issue #58) (PR #105)parse_tax_data, I moved “taxon_id” and “input_index” columns to front and “input” to rear. Also “tax_data” now comes before “class_data”.