2.0.1

• Fix issue where NULL values passed to interpolate() triggering an error

2.0.0

Breaking changes

• major code rework for all data-wrangling and variable manipulation functions with tidyverse libraries and development conventions
• renamed most data-wrangling and variable manipulation functions such that their functionality are more name-intuitive (see below)

Minor changes

• Default code of generated data from all data-wrangling and variable manipulation functions now uses native R pipe |>. An option is added to use the magrittr pipe %>% instead.
• The code() accessor function now returns a string with class inzcode that has a print method which formats the code nicely.

New features

• The following functions are reworked and added new features:
  • renamed aggregatedt() to aggregate_dt() and aggregate_dt() is now a wrapper of aggregate_data() with date component extraction and aggregation functionality built in, and added an option to allow aggregation with additional variables
  • combined unite() and combineCatVars() into combine_vars() and added options to remove empty factor combinations and turn missing values into an explicit factor level
  • combined reshape_data() and stackVars() into reshape_data() and reworked the arguments
  • renamed rankVars() to rank_vars() and included more ranking methods (integer ranking by density and percentage ranking)
  • renamed reorderLevels() to reorder_levels and included more auto-reordering methods (by first appearance and numeric order)
  • added an option to specify time zone in convert_to_datetime()
  • renamed joindata() to join_data(), included right-join method, and reworked its arguments
• pull %||% operator from 'iNZight'

Other changes

• The following functions have changed names and with renamed/reworked arguments (see code documentation for more details)
  • renamed aggregateData() to aggregate_data()
  • renamed appendrows() to append_rows()
  • renamed collapseLevels() to collapse_cat()
  • renamed convertToCat() to convert_to_cat()
  • renamed createNewVar() to create_vars()
  • renamed deleteVars() to delete_vars()
  • renamed extract_part() to extract_dt_comp()
  • renamed filterLevels() to filter_cat()
  • renamed filterNumeric() to filter_num()
  • renamed filterRandom() to random_sample()
  • renamed filterRows() to remove_rows()
  • renamed missingToCat() to missing_to_cat()
  • renamed renameLevels() to rename_levels()
  • renamed renameVars() to rename_vars()
  • renamed selectVars() to select_vars()
  • renamed separate() to separate_var()
  • renamed sortVars() to sort_vars()
  • renamed standardizeVars() to standardize_vars()
  • renamed transformVar() to transform_vars()
• removed countMissing()

1.13.4

• read_text() function handles all invalid characters (as smart_read() does)
• read_text() applies string to factor conversion

1.13.3

• read_text() function replaces spaces with underscores in column names

1.13.2

• add global options to set/override default comment character

1.13.1

• add global options to set/override default comment character (this will allow Lite to change the default without changing the package's default behaviour; default set as # at load time)

1.13.0

New features

Data dictionaries
Users can now import a data dictionary and apply it to a dataset. This will apply text labels to numerically coded variables and set labels for variables with human-friendly names.
Linked data
New functions to load linked datasets from a .inzlnk file, where the file linkage is specified. Optionally users can include a data dictionary in this file, which will also be applied to the data.

Other changes

• relocate survey specification reading files to new package surveyspec
• add survival analysis as an option for fitModel via family = "cox"
• initial implementation of (basic) database connectivity - this is in early alpha stage
• various bug fixes/changes
• fix some issues occuring when column names have spaces (in the CSV) (#200)

1.12.3

• allow 'readr' to use automatic delimiter guessing (csv files) instead of forcing , (#187)
• fix Windows bug in url_to_temp() adding extra lines, causing importing files with more than one line of comments to fail delimiter-guessing
• fix bug where extension guess failed if file extension was not lower case (e.g., file.CSV)

1.12.2

• add new function print_code() to extract, tidy, and print code attached to an object
• allow more than one character as 'sep' argument to combineCatVars()

1.12.1

• fix bug in smart_read() where parsing column types (numeric -> categorical) failed if NAs in column
• fix test failing in new version of survey (2.4)