v2.1.5

• Added check to checkMWRresults() for results data to verify all values in the quantitation limit column are numbers, with updated instructions in the results template file
• Parameter list (paramMWR) updated with Phycocyanin, Phycocyanin (probe), RFU as unit for chlorophyll, including appropriate updates to thresholdMWR, results template, and WQX phys-chem template
• Removed prettymapr, raster, added maptile and tidyterra for anlzMWRmap() basemaps, this changes the options that are accepted by the maptype argument

v2.1.4

• Improved warning message for all empty or na entries in checkMWRacc() and checkMWRfrecom()
• Fixed spelling of tabMWRwqx() output in the Result Detection Condition column for "Present Above Quantification Limit"
• Fix to checkMWRacc() if more than two value ranges in DQO accuracy file as ascending, previously resulted in error
• Fix to incorrect parameter paste in error message in checkMWRfrecom() if values less than 0 or greater than 100 are found
• Improved tests to check error or warning messages for checkMWRfrecom() and checkMWRacc()

v2.1.3

• anlzMWRmap() has different basemap options using the maptype argument, these include "cartolight", "cartodark", "osm", or "hotstyle". These changes accommodate recent issues with access to stamen tiles.
• ggmap dependency was removed, added raster and prettymapr dependencies
• Fixed issue in Roxygen documentation for itemized lists

v2.1.2

• Warning is now returned if a column for a QC check includes all na entries on import of the DQO accuracy file with readMWRacc()
• readMWRacc() no longer uses dplyr::na_if() on all columns, only for numeric, to correctly identify columns that have all na entries for a QC check
• Sorting behavior for dplyr::arrange() reverted to .locale = 'en' to ignore case
• utilMWRfre() function added to prep results data for frequency checks, similar to existing utilMWRlimits() function
• Correct number of data records is now reported by tabMWRfre() following value range filtering
• tabMWRfre() and qcMWRfre() now require the DQO accuracy file as input to identify appropriate ranges to check for each parameter using the value range column
• Added utilMWRvaluerange() function to check for na, gaps, and overlap in the value range column on import of the DQO accuracy file with readMWRacc()
• Error is now returned if overlapping value ranges are present in the DQO accuracy file
• Warning now returned if gaps are present in the value ranges for a parameter in the DQO accuracy file
• Fix to checkMWRacc() to convert MDL and UQL columns in DQO accuracy as numeric following import as text
• Fixed bug to evaluate lab spike QC checks as the absolute difference, was previously a relative difference
• New error message if the upper value range in the DQO accuracy file is not a percent value for lab spike QC checks with units as percentage
• Lab spikes entered as a percent measure are now evaluated against the correct data quality objective using the upper value range
• Better error and warning messages for tabMWRacc() for incorrect and required data for individual QC checks

v2.1.1

• Fix to incorrect secondary label for Enterococcus in thresholdMWR
• Corrected column name to Record ID User Supplied in WQX output from tabMWRwqx()
• Fix for tabMWRwqx() output if no entries in QC Reference Value column in the results file, this previously resulted in an error
• formMWRresults() no longer imports dplyr and tidyr in Roxygen documentation
• formMWRresults() now handles time and text inputs for Activity Start Date from Excel, all times are correctly formatted as HH:MM in 24 hour time. This includes more robust testing of inputs in text-formMWRresults.R.
• Added check for data quality objectives for accuracy that returns an error if na entries are included in the Value Range column, should be all
• Fix to date output for the Activity Start Date column returned by tabMWRwqx() to be YYYY-MM-DD character, empty time was previously included as date time object

v2.1.0

• Added option to report mean, median, min, max, or geometric mean in summary plots and maps
• Updated utilMWRtitle() function to accommodate the previous point
• Replaced utilMWRconfint() function with utilMWRsummary() for additional summary functions
• Added option to change the scale bar from km to mi on the map created with anlzMWRmap()
• Added option to reverse the color ramp in the legend created with anlzMWRmap()
• All columns in the results file are now required on input with readMWRresults(), this includes a new column Local Record ID
• Added a new check for depth data in the results file used by checkMWRresults(), must be present with no missing values
• Removed cryptic error messages when confidence intervals cannot be plotted, replaced with informative warning
• Fix to qcMWRacc() if output list elements return a data frame with zero rows
• Fix to output message that was creating duplicates when a file is saved to disk
• Fixed incorrect entry description for E. coli in thresholdMWR
• Data input template files and example files updated
• Created a vignette about the utility functions and added an example using patchwork in the modifying plots vignette

v2.0.2

• utilMWRfilter has fix to date filter and no longer uses lubridate::ymd
• All but one test removed for utilMWRfilter, covered in other tests
• Text for modifying vignette edited for better examples
• Out of state article to vignettes
• Added utilMWRhttpgrace for graceful fails if http requests do not work, applies to anlzMWRmap

v2.0.1

• Removed any use of case_when and replaced with ifelse to fix performance issues with dplyr v1.1.0
• System files for examples replaced with real data
• Slight change to method id and method context logic in tabMWRwqx
• Functions that write files (e.g., qcMWRreview) require explicit entry for the output_dir argument
• anlzMWRmap now imports NHD layers from external source (no user changes)

v2.0.0

• Release for CRAN submission
• Addition of WQX functions
• Numerous fixes and additions from beta testing (see commit log)

v1.0.0

First release prior to beta testing