NEWS.md

daiquiri (development version)

New features

• New field_types_advanced() function. Allows just a subset of the columns in the source df to be named explicitly in the specification, with the remaining columns set to the .default_field_type parameter. (#16)

Bug fixes and minor improvements

• Improved scaling of heatmaps when there are lots of fields. (#17)

daiquiri 1.1.1 (2023-07-18)

New features

• New ft_strata() field type, which when specified, produces a report where the aggregated data and individual data fields plots are stratified by the ft_strata() column's distinct values

• Column-specific strings for missing values can now be set in the field_types() specification (#13)

Bug fixes and minor improvements

• Categorical data fields now show a heatmap plot as well as the individual time series plots when aggregate_by_each_category option is set to TRUE

• Categorical data fields now retain special characters in labels when aggregate_by_each_category option is set to TRUE

• Print method of daiquiri_object now displays the location the report was saved to

• When a data field contains all missing values, this now shows correctly in the various tabs (#12)

• When running package from within rmarkdown/quarto (rmd/qmd) files, the parent file can now contain a chunk labelled setup without causing an error. (#7)

• Hex logo now appears on reports, adding dependency to xfun

daiquiri 1.0.3 (2022-12-06)

Bug fixes and minor improvements

• Validation warnings now match column names correctly when field_types are specified in a different order to the supplied data frame columns

• Passing in a data frame containing integer columns no longer causes an aggregation error (#9)

• Calling functions with package prefix no longer causes an error (#10)

• Fixed (some) errors about duplicate chunk labels when running package from within rmarkdown/quarto (rmd/qmd) files (bug introduced in previous release 1.0.2). This now allows chunks in the parent file to be unlabelled but unfortunately still errors when there is a chunk labelled setup. (#7)

daiquiri 1.0.2 (2022-11-21)

• When rendering reports, intermediate files are now written to tempdir() instead of to the directory of the report_htmldoc.Rmd file (the default behaviour of rmarkdown::render()). This fixes errors caused when the library location is read-only.

daiquiri 1.0.1 (2022-11-11)

First release to CRAN

• Replaced calls to deprecated function aes_string() in ggplot2

daiquiri 1.0.0 (2022-11-01)

This release incorporates changes requested for acceptance into https://ropensci.org/. There are many breaking changes as objects have been renamed for better consistency and style.

Breaking changes

• daiquiri_report() replaces create_report() and some parameters have been renamed.

• field_types() replaces fieldtypes().

• prepare_data(), aggregate_data(), and report_data() parameters have been renamed.

• initialise_log() replaces log_initialise().

• close_log() replaces log_close().

• template_field_types() replaces fieldtypes_template()

Bug fixes and minor improvements

• Fixed error when user passes in a data.table (to daiquiri_report() or prepare_data()) that contains non-character columns.

• daiquiri_report (formerly create_report()) and report_data() accept a new parameter report_title.

• report_data() now accepts ... parameter to be passed through to rmarkdown::render().

• close_log() now returns the path to the closed log file (if any).

• example_prescriptions.csv replaces example_dataset.csv as the example dataset supplied with the package.

daiquiri 0.7.0 (2022-04-20)

This release moves the reading of csv files out into a separate function in order to make it more configurable and to handle the parsing of all fields as character data for the user.

Breaking changes

• create_report() now only accepts a dataframe as the first parameter. The textfile_contains_columnnames parameter has been removed.

• load_data() has been replaced with read_data() and prepare_data().

• log_initialise() function: dirpath parameter renamed to log_directory.

New features

• New function read_data() reads data from a delimited file, with all columns read in as character type.

• New function prepare_data() validates a dataframe against a fieldtypes specification, and prepares it for aggregation.

• create_report() accepts a new parameter dataset_shortdesc for the user to specify a dataset description to appear on the report.

• export_aggregated_data() function accepts new save_fileprefix parameter.

• New function fieldtypes_template() generates template code for creating a fieldtypes specification based on an existing dataframe, and outputs it to the console.

Bug fixes and minor improvements

• Fixed ALL_FIELDS_COMBINED calculated field rowsumming NAs incorrectly.

• Fixed plots failing when all values are missing.

• Fixed log_message() trying to write to different log file when called from Rmd folder (and relative path used).

• Made '[DUPLICATES]' and '[ALL_FIELDS_COMBINED]' reserved names for data fields.

• Allow column names in supplied dataframe to contain special characters.

• Reduced real estate at top of report.

• Removed datatype column and fixed validation warnings total from Source data tab in report.

• Updated example data.

• Added further validation checks for user-supplied params.

• Added CITATION file.

daiquiri 0.6.1 (2022-02-23)

Beta release. Complete list of functions exported:

• aggregate_data()
• create_report() accepts either a dataframe or csv filename as the first parameter. This may change in future.
• export_aggregated_data()
• field_types()
• ft_categorical()
• ft_datetime()
• ft_freetext()
• ft_ignore()
• ft_numeric()
• ft_simple()
• ft_timepoint()
• ft_uniqueidentifier()
• load_data() accepts either a dataframe or csv filename as the first parameter. This may change in future.
• log_close()
• log_initialise()
• report_data()