simplify_idaifield()
get_field_index()
get_field_index()
more efficient by utilizing vectors instead of lapply()
. get_field_index()
and simplify_idaifield()
is now an option. This way, in get_field_index()
the "shortDescription" can be added without the more time intensive recursive layer search that comes with the now available option find_layers = TRUE
. For info on what this does, see documentation of find_layer()
.simplify_idaifield()
is now slightly faster as it uses find_layer()
differently.find_layer()
can now handle a vector of multiple ids.project
/projectname
-argument will be removed soon, and project needs to be specified in connect_idaifield()
. Relevant functions currently issue a warning to that effect. idf_last_changed(connection, n = n)
: Returns a vector with the last n changed resources in the database. idf_get_changes(connection, ids = c(...))
: Returns a data.frame in which each change to one of the resources listed in ids (can be either their 'identifier' or 'UUID') corresponds to one row separated by creation or modification.remove_config_names()
issues a message about duplicate field or category names if it is appropriate. Functions may override data when multiple columns with the same name would otherwise exist, as e.g. 'diameter' and 'projectName:diameter'. Attaches an attribute that lists the duplicate field/category names.check_idaifieldr_version()
).reformat_geometry()
(MultiPolygons have to be unnested before processing). (Imported Polygons may be formatted improperly, unnest if necessary (reformat_geometry()
). Geometry is still a work in progress.)get_field_index()
, where it would return an empty data.frame if there was no configuration-resource.idf_json_query()
which lets users construct their own queries to the CouchDB-API find_layer()
get_language_lookup()
idf_query()
-function where it would not get values from fields with possibility of multiple entriesselect_by()
to idf_select_by()
show_categories()
to idf_show_categories()
idf_select_by()
and idf_show_categories()
idf_query()
/idf_index_query()
: rename "type" to "category" in resource listsidaifield_as_matrix()
simplify_idaifield()
convert_to_onehot()
unnest_docs()
& find_resource()
)get_field_index()
- getting the uidlist/index directly from the databasefind_layer()
significantlyget_uid_list()
idf_query()
with type & categorysimplify_idaifield()
, category is currently switched to type,
but this will be changed later -> #TODO: rename type everywhere to
category to reflect actual db structureget_language_lookup()
) for custom config fieldsdownload_language_list()
to get current translations from GitHubget_idaifield_docs()
will now remove the Configuration as a doc and attach it as attribute to all custom classes, thus simplify_idaifield()
will use the config-attribute directly and works without a connection. simplify_idaifield()
now allows to keep checkbox-fields etc. as-isidf_query()
and idf_index_query()
to return docs instead of simplified list to allow choiceidaifield_...
-lists are now named (with the identifier of each resource)data("idaifieldr_demodata")
)get_idaifield_docs()
should still work.find_layer()
, replace_uid()
and check_if_uid()
find_layer()
(internal), seems to work as intended now.check_if_uid()
now handles vectorsget_uid_list()
, replace_uid()
, convert_to_onehot()
, and fix_relations()
(a lot)get_idaifield_docs()
and processing to be more usable with simplify_idaifield()
, with advice to use the latter only on subsets so the time is more manageable. check_and_unnest()
to exportget_uid_list()
using lapply()
idf_query()
to specifically query the db for groups without first downloading everythingidf_index_query()
to specifically query the db for things that are only available in the uidlist (needs a uidlist to do that)simplified = TRUE
(simplify_idaifield()
) will now convert checkbox field to multiple columns. I should probably make this an option in the future, but I need to clean up that function first (TODO).idaifield_as_matrix()
now returns a matrix with character values, not listsI stated that I don't intend any structural changes, but that - apparently - was impossible. Old scripts will not work without interference with the new version.
First release.
I will try to limit future changes so that they will not affect the structural outcome of the core functions as they are now.
TODOs: Find a way to handle the remaining lists better, esp in combination with the non-list columns. Try to assign column type automatically (numeric for numbers etc.) * How to deal with the dimension...-fields (measurements)? Currently ignored, can be flattened with dplyr and friends.
NEWS.md
file to track changes to the package.Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.