CRAN Release 2.6.3

soilDB 2.6.3 (2021-07-23)

• SDA_query() and all functions that call SDA_query() get proper column class handling (related to #190), however:
  • be careful with the use of CAST(): unknown datatypes may not be correctly interpreted
  • previous column classes that were incorrectly guessed by type.convert() may have changed (e.g. component.wei)
• SDA_spatialQuery() can now be used to return soil survey area symbols or geometry using what="areasymbol" or what="sapolygon", respectively
• Added new columns to soil classification ("SC") table result of get_soilseries_from_NASIS(); now including taxonomic mineralogy class which may contain multiple parts for series with strongly contrasting control sections
• Updates to get_SDA_*() methods
  • Extends get_SDA_property(property = ...) and get_SDA_interpretation(rulename = ...) vectorization over property/rulename to work with any aggregation method.
    • Now supports: Dominant Condition, Min, Max, Dominant Component, Weighted Average
  • Add query_string argument (default: FALSE). Set as TRUE to skip submitting query to SDA, instead returning a string of the query that would have been sent instead of data.frame result
  • get_SDA_property: better handling of NULL, miscellaneous areas, and property-specific weighting
    • Remove ISNULL(x, 0) logic that affected weighted averages in presence of missing data
    • Conditional calculation of horizon weights considering NULL values for requested properties (unique weights for each property)
    • New default argument include_minors=FALSE includes only components where majcompflag = 'Yes' in result
    • New default argument miscellaneous_areas=FALSE removes miscellaneous land types compkind values from result
    • Organic and bedrock layers are no longer removed from "Weighted Average", "MIN" or "MAX" aggregations
  • get_SDA_interpretation: added argument not_rated_value with default value of NA to set "not rated"" values across methods/queries. For backwards compatibility with original SQL use not_rated_value = 99.0
  • Standardizing MUKEY column name (and other keys) as lowercase in results
  • More informative error messages for bad input / arguments inconsistent with specified method
• Thanks to @hammerly who pointed out weighted averaging of NASIS phlabresults wasn't working as expected and for highlighting some more improvements (#192)
• get_OSD() TXT and HTML formats now supported (in addition to JSON) through a common function interface
• Added get_NASIS_table_key_by_name() get_NASIS_fkey_by_name(), get_NASIS_pkeyref_by_name(), get_NASIS_pkey_by_name(), get_NASIS_table_name_by_purpose() methods for helping get information on primary/foreign keys and thematic groups of NASIS tables (useful for creating SQLite/external snapshots of NASIS tables)
• get_mapunit_from_NASIS(), get_legend_from_NASIS() and get_lmuaoverlap_from_NASIS() now works for "MLRA Survey Area" areatypename and no longer is limited by constraints on legendsuituse or mustatus