Awesome

WriteVTK.jl

This package allows to write VTK XML files for visualisation of multidimensional datasets using tools such as ParaView. A wide range of VTK formats is supported, including different kinds of structured and unstructured grids, as well as metadata files for describing time series or multi-block domains.

Installation

WriteVTK can be installed using the Julia package manager:

julia> ] add WriteVTK

Quick start

The vtk_grid function is the entry point for creating different kinds of VTK files. In the simplest cases, one just passes coordinate information to this function. WriteVTK.jl then determines the VTK format that is more adapted for the provided data.

For instance, it is natural in Julia to describe a uniform three-dimensional grid (with regularly-spaced increments) as a set of ranges:

x = 0:0.1:1
y = 0:0.2:1
z = -1:0.05:1

This specific way of specifying coordinates is compatible with the image data VTK format, which has the .vti extension. The following creates such a file, with some scalar data attached to each point:

vtk_grid("fields", x, y, z) do vtk
    vtk["temperature"] = rand(length(x), length(y), length(z))
end

This will create a fields.vti file with the data. Note that the file extension should not be included in the filename, as it will be attached automatically according to the dataset type.

By changing the coordinate specifications, the above can be naturally generalised to non-uniform grid spacings and to curvilinear and unstructured grids. In each case, the correct kind of VTK file will be generated.

Supported VTK formats

This package attempts to implement writers for all VTK XML formats described in the VTK specification. Note that legacy (non-XML) files are not supported.

Supported dataset formats include:

image data (.vti),
rectilinear grid (.vtr),
structured (or curvilinear) grid (.vts),
unstructured grid (.vtu),
polydata (.vtp, a specific type of unstructured grid).

Moreover, the following metadata formats are supported:

multiblock files (.vtm),
ParaView collections (.pvd, typically used for time series),
parallel files (.pvt*).

Authors