Changelog • movepub

movepub (development version)

movepub now relies on frictionless >= 2.0.1. When using create_package() or add_resource() it will follow the Data Package v2 specification (#136). The main changes in a datapackage.json will be:
1. $schema is added to package, resources and schemas.
2. type is used instead of profile for resources.
3. primaryKey and foreignKeys.fields in resources are always arrays.

write_dwc() now requires animal-taxon, animal-id and tag-id in the source data. It also gracefully handles any non-required missing fields (#120).
write_dwc() now adds georeferenceSources (set to "GPS" for GPS sensor data) and dwc:identificationVerificationStatus (set to "verified by expert" for all records, since the taxon is assumed to be well-known before the tag was attached).
write_dwc() now parses location-error-text (in addition to location-error-numerical) for dwc:coordinateUncertaintyInMeters (#139).
write_dwc() now always uses the canonical URI (with http) for dwc:measurementTypeID (#128).

get_mvb_term() is now defunct and returns an error. Use move2::movebank_get_vocabulary() instead (#143).
movepub now reexports add_properties() from frictionless (#136).
write_eml() now handles multiple Issued dates in the DataCite metadata (#144).
add_resource() now automatically sets type to "datetime" if a field is defined with the format "yyyy-MM-dd HH:mm:ss.SSS" in the Movebank Attribute Dictionary. Before it only did this for format, which led to validation errors for empty datetime fields (#142).

write_eml() is now a separate function from write_dwc() (#57).
write_eml() now formats the derived_paragraph as DocBook (rather than HTML), a format supported by EML and the GBIF IPT (#102). The paragraph is also shorter and added at the end of the abstract (#76).
New html_to_docbook() allows to convert a string or character vector from HTML to DocBook. You can use this to convert descriptions in order to have valid EML (#101).
write_eml() and write_dwc() no longer add "[subsampled representation]" to the dataset title (#76).

write_dwc() can now be used for an unpublished dataset (i.e. without metadata on DataCite, which was required for the previously build-in write_eml() functionality). Some record-level terms (e.g. dwc:datasetName) can be provided as arguments (#57, #72).
write_dwc() (and write_eml()) no longer writes to "." by default, since this is not allowed by CRAN policies. The user needs to explicitly define a directory (#70).
write_dwc() now writes the output file as occurrence.csv (previously dwc_occurrence.csv) and adds a meta.xml file. The sex and life stage of the animal are - in addition to dwc:sex and dwc:lifeStage in occurrence.csv - expressed in an Extended Measurement Or Facts extension file (emof.csv), for better support with OBIS (#71, #77, #78).
write_dwc() provides a message regarding the matching of scientific names with WoRMS Aphia IDs. These IDs are now clickable URLs, making it easier to verify the match (#58).
write_dwc() is now more modular, facilitating extension for non-GPS tracking data (#66).

movepub now relies on R >= 4.1.0 (because of move2 dependency) and uses base pipes (|> rather than %>%) (#98).
Many functions of frictionless are now reexported by movepub, so you no longer have to load that package to create Data Packages (#54).
get_mvb_term() is deprecated in favour of move2::movebank_get_vocabulary() (#60).
Sanne Govaert is added as author.

write_dwc() now makes use of dplyr rather than SQL for its transformation. This reduces the number of dependencies (#61).

As per OBIS recommendations, write_dwc() now adds a scientificNameID to all occurrences, with the WoRMS LSID for that taxon. It does so using the new get_aphia_id().
The cli package is now used for all messages.