I'm pleased to announce v0.14 of the haskell chart library. This is a library for drawing 2D data plots. It's features include + Use of the cairo graphics engine, supporting a variety of output types: ps, pdf, png, and gtk windows. + A variety of plot types, including: points, lines, error bars, candlesticks, bar charts and pie charts. + Strong typing. Charts are parameterized by the types of their coordinates. One benefit of this is that labels etc are automatically generated appropriately for the type of the data. Currently axis types include: Double, Int, Log, LocalTime, and Indexed. + (some) support for interactivity. Charts support dynamic resizing, and mapping from device coordinates back to source values. The library is available on hackage. Additional information including details of the mailing list can be found here: http://dockerz.net/software/chart.html Thanks to Malcolm Wallace, Eugene Kirpichov, and Matt Brown for their contributions to this release. Tim Docker ---------------------------------------------------------------------- New features in v0.14 --------------------- * Plot Type: AreaSpots4D Spots with varying area and colour * Plot Type: CandleStick charts A specific type of plot often used in financial markets for stock price series. * Multiline text in Annotation Plots with flexible control over the anchoring. * Picking. The library now contains logic required to map device coordinates back to input elements. This facilitates interactivity. See tests/TestPicking.hs. * Multiple Layers of axis labels This is currently used by the LocalTime axis to show an additional set of context labels - these show the next sensible granularity of time enclosing the more detailed labels and ticks. * Avoid overlapping axis labels The axis rendering code now skips labels where there would be overlaps. This is done dynamically as the chart is rendered. * Code refactor Plots and axes have been split into separate file hierarchies, with one file per type. This makes it easier to understand, and (hopefully) makes it clearer how to add new plot types.