- Improved ping_edi() function to gracefully handle network errors and disabled network-dependent tests to align with CRAN policies and prevent CRAN check issues. See issue #154.
- Resolved rate limiting errors in ping_edi() triggered by recent updates to the rate limiting mechanism in the EDI data repository and due to the lack of rate limiting procedures in the ecocomDP R package.
- Fixed and cleaned up NEON data mapping functions to work with latest NEON data release, return abundance values that are finite and non-negative.
- Added code to resolve duplicate taxon counts within sample/event IDs for NEON data.
- Updated plotting functions.
- Fix NEON mosquito mapping error in internal function
map_neon.ecocomdp.10043.001.001
that resulted from an update to the NEON "Mosquitoes sampled from CO2 traps" (DP1.10043.001) data model that affects the 2022 data release (and later). The fix is backwards compatible to older NEON data releases and also fixes a problem that resulted in pulling duplicate records from the mos_archivepooling table (archive records are no longer accessed using ecocomDP). - Fix vulnerability to absence of retiring r-spatial packages. See issue #147.
- Relax constraints on minimum version requirements. See commit f2bdb91a4e1a3b633ebed3bffbf1773ac8565e80.
- Fix
plot_sites()
to work with version 0.6.0 of the usmap package. See issue #136.
- Release new plotting functions for taxonomic and geographic exploration.
- Allow multiple input types to the new
data
parameter of "use" functions (supercedesdataset
), thereby simplifying and accelerating exploratory manipulation and plotting. - Reformat dataset object returned by
read_data()
. Generally, this simplifies dataset indexing and usage within iteration contexts. It removes a layer of nesting and places the dataset id at the same level as metadata, tables, and validation_issues. - Release new
annotation_dictionary()
function. This feature will allow ecocomDP creators to view the semantic annotations for existing ecocomDP datasets. Annotation dictionary is published as an EDI data package to apply occasional updates. - Release new
validate_mapped_id()
function. Ensures values listed under /variable_mapping/mapped_id resolve without error - Release new Model Overview vignette. The model overview vignette consists of the authoritative set of table descriptions, relationships, and requirements for the ecocomDP model and a visualization of table contents and relationships.
- Release new Shared Practices Vignette. This document delineates the practices that all creators of ecocomDP formatted datasets should adopt in order for the community to build a cohesive and interoperable collection. It contains detailed descriptions of practices, definitions of concepts, and solutions to many common issues.
- Prevent L1 EML from inheriting L0 provenance and additional metadata, these should not be inherited by the L1 in
create_eml()
- Discourage use of the
observation
parameter of plotting functions and replace it with the more flexibledata
parameter (see above). These changes are backwards compatible with previous releases. - Discourage use of
plot_taxa_sample_time()
in favor ofplot_sample_space_time()
. These changes are backwards compatible with previous releases. - Discourage use of the previous
dataset
object returned byread_data()
in favor of a new one. See details above. These changes are backwards compatible with previous releases.
- Link to ecocomDP article.
- Enable full return of L0 location columns in
flatten_data()
. - Implement a human readable row sorting based on location_id values in
create_location()
. - Check create_ecocomDP.R for expected function and associated arguments in
create_eml()
.
- Fix alignment of categorical variable names and definitions in
create_eml()
. - Allow only one basisOfRecord in
create_eml()
. - Fix numeric type detection in
create_eml()
. - Fix taxonomic hierarchy expander in
create_eml()
. - Fix empty annotation defaults in
create_eml()
. - Remove all L0 data entities from the L1 EML, these should not be inherited by the L1 in
create_eml()
. - Handle both present/absent L0 taxonomic coverage in
create_eml()
. - Fix assignment of self referencing ids in
create_location()
. - Only return unique locations in
create_location()
. - Fix methods in
calc_number_of_years_sampled()
andcalc_length_of_survey_years()
. - Fix handling of datetimes with YYYY format in
read_data()
. - Include ancillary table datetime in join operations of
flatten_data()
. - Remove XML attributes to prevent id clashing and schema invalidation when constructing provenance nodes in
create_eml()
andconvert_to_dwca()
. - Fix basisOfRecord reference in
convert_to_dwca()
. - Incorporate L0 methods markdown blocks in
convert_to_dwca()
. - Prevent namespace clash with the taxonomyCleanr package.
- The ecocomDP package is now available on CRAN