Releases: fastscape-lem/fastscapelib
Releases · fastscape-lem/fastscapelib
Release v0.3.0
What's Changed
- Create a container API for structured grids by @adriendelsalle in #155
- Add flow kernels API and parallel execution by @adriendelsalle in #157
- Bump googletest by @benbovy in #164
- Bump Python / Numpy supported versions by @benbovy in #166
- Fix fastscapelib cmake config file by @benbovy in #171
- Setup pixi by @benbovy in #170
- MNT: pin xtensor, fix RTD, update pixi conf by @benbovy in #176
- Provide hints where CMake should find xtensor. by @bangerth in #173
- Avoid warnings about unnecessary semicolons. by @bangerth in #175
- Add a set of missing braces. by @bangerth in #179
- Remove a variable that is only ever set but not read. by @bangerth in #178
- CI: test with xtensor master branch by @benbovy in #180
- CI: test xtensor master (tweaks) by @benbovy in #181
- Update to xtensor 0.26 by @benbovy in #182
- Add Healpix grid by @benbovy in #162
- CI: python wheels, bump windows version by @benbovy in #183
- Rename trimesh_xt -> trimesh by @benbovy in #184
- Flow kernel improvements by @benbovy in #185
- Refactor FlowKernelEroder by @benbovy in #186
- DOC: flow kernels API reference and user guide by @benbovy in #187
New Contributors
Full Changelog: v0.2.2...v0.3.0
Release v0.2.2
Release v0.2.1
Bug fixes
- Fixed
pflood_sink_resolver
that was "flooding too much" (#145, #146). - Fixed
RuntimeWarning
(invalid value in cast) issued when calling
FlowGraph.basins()
in Python (#147). - Python bindings: fixed Python interpreter crash (segfault) when accessing data
members of the implementation of a graph snapshot viaFlowGraph.impl()
(#148).
Release v0.2.0
A complete re-write of Fastscapelib with brand new features and API (note: the
API of the previous version has been mostly removed). Some highlights:
- A flexible grid system, including 1D profile grid, 2D raster grid (with
support of diagonals vs. non-diagonals connectivity) and 2D triangular mesh - Full support of looped (periodic) boundary conditions for uniform rectangular
grids - Flexible flow routing using a "flow graph" and "flow operators"
- Support for both single direction and multiple direction flow (grid-agnostic
implementation) - Efficient resolution of closed depressions in the topography while routing the
flow paths, based either on explicit computation of a graph of basins or on
the priority flood algorithm (grid-agnostic implementation) - The current flow graph implementation graph based on fixed-size arrays is
extensible to alternative representations (e.g., sparse matrix, linked-lists) - First-class C++ and Python APIs
- Detailed documentation (examples, user-guide, API reference, etc.)
Thanks to the contributors to this release: Benoît Bovy, Adrien Delsalle,
Guillaume Cordonnier and Hannah Sophia Davies. Thanks also to Johan Mabille and
QuantStack (https://quantstack.net/) for their contribution and advice on the
design of the library.
v0.1.3
v0.1.2
v0.1.1
Initial release
Initial release with efficient implementations of basic features such as flow routing, sink filling, bedrock channel erosion using the Stream Power Law and hillslope erosion by linear diffusion.