datacleanr 1.0.4

improved plotting performance of large, "narrowly distributed" distributed data, such as a 1:1 scatter or a high resolution time series (i.e., little variation in either x or y dimension). The low performance seems to come from changes made in how hover events are handled in plotly in newer versions (2.x.x). See plotly/plotly.js#5790 (comment), which is still unresolved at 06 June, 2022. The fix uses the default hovermode = "closest", but sets hoverdistance = 200, which should ensure a better responsiveness in most cases. The increased hoverdistance called for a disabling of hover labels to prevent confusion. Note, the increased hoverdistance may cause some unintended consequences when clicking into plotting 'white space', as the click snaps to the closest point within 200 pixels.

datacleanr 1.0.3 - Publication version

Release for Zenodo Archive of published version of datacleanr.

datacleanr 1.0.3

• fixed filter_scoped() test to conform with dplyr 1.0.8 release (dropped dplyr error class)
• linked with Zenodo

datacleanr 1.0.2

• removed navbarPageWithInputs, replaced with shiny::navbarpage() and bslib::nav_item(), as recommended
• adjusted button placement (close/exit) with css styles
• plotly no longer serves Plotly.d3, which was used to select individual traces on the plot in the viz tab; it now requires additional import. This is currently done via a D3 script dependency from a CDN, which requires an internet connection (checked by can_internet())

datacleanr 1.0.0

• First major version with features for exploration, filtering, visualizing and cleaning/annotating data in a reproducible manner.
• Added a NEWS.md file to track changes to the package.