- fix NA in plot method of
onto
.
- include information about previous matches in the "description" column in matching tables, to assist in finding the correct matches, based on previous information.
- change automatically set matches to "exact" in
edit_matches()
. - fix a bug when mapping external concepts that had already previously been mapped.
- simplify
get_concept()
. Search parameters are now not provided via a table anymore, but as a combination of column name and value to search for in that column. They can also be provided as a string that represents a query that would be used indplyr::filter()
, where several search conditions can be combined arbitrarily.
- also allow building trees of the parents, not only the children in
make_tree()
- fix acknowledgement and adapt to update of dplyr
- implement better filtering on parent concepts to simplify editing matches.
- include matches with 0 difference already as close match, for convenience.
- fix a little bug located in
new_mapping()
.
- when matching a new dataseries with old concepts, give a message and ignore non-close matches.
- fuzzy matching implemented. When concepts that are new for mapping do not match exactly, fuzzy matching is carried out and three new columns are offered for assigning the new concepts to harmonised concepts
- small fix that would not allow to match with an empty set of values.
- introduce function where a table with columns per classes in the ontology can be used to retrieve concepts. This table is matched with a flattened ontology. Currently, no mappings are attached to this table.
- reconstruct regex functionality and enable extracting of external concepts in
get_concept()
- reconstruct extracting mappings in
get_concept()
- write the getters
get_class()
andget_source()
- reconstruct sorting in
get_concept()
so that it corresponds to the order of the input table
- describe an ontology either by version, or by date
- revision of vignettes
- include functions
export_as_rdf()
for being able to explort any ontology in other, semantic-web based, formats.
- various fixes of convenience and to make ontologics work smoothly together with
arealDB
- allow it that also classes are handled with the
new_mapping()
function, which also means thatnew_class()
can now only set harmonised classes.
- adapt the onto-class structure according to issue #6; this entails that concepts and classes are summarised into their respective slot, with sub-tables "harmonised" and "external", featuring specific columns.
- change all functions to support this new structure
- this changes the behaviour of the function
new_concept()
insofar that this only creates harmonised concepts (so no source needs to be provided here anymore) and of the functionnew_mapping()
insofar that this now creates external concepts and their mapping to harmonised concepts (and thus writes into the sub-table "external" and into the "mapping-columns" of the sub-table "harmonised")
- adapt column names according to issue #5
- revise how to extract concepts via
get_concept()
. Now you have to provide a table that contains the column name on which to subset and the values in that column that should be filtered from the ontology.
- adapt
new_concept()
so that also concepts can be defined that have no class (for various reasons). This now gives a warning, and inserts the concepts with an "undefined" class.
- revise all functions that access an ontology to work either in the
.GlobalEnv
, or at the given path. The argumentpath =
was therefore changed toontology =
. - Instead of taking merely the terms, functions
new_concept()
andnew_mapping()
now take the output ofget_concept()
of already existing concepts, or at least a table with columns 'code', 'label_en' and 'class'.
- revise
get_concept()
so that it's faster.
- Initial commit