formatting, mainly to force numpy 2.0 github actions (#202)

Author: zingale

README.md

@@ -56,7 +56,8 @@ https://python-hydro.github.io/pyro2
   - We require `numpy`, `numba`, `matplotlib`, and `h5py` for running pyro
     and `setuptools_scm` for the install.
 
-  - There are several ways to install pyro.  The simplest way is via PyPI/pip:
+  - There are several ways to install pyro.  The simplest way is via
+    PyPI/pip:
 
     ```
     pip install pyro-hydro
@@ -85,7 +86,8 @@ https://python-hydro.github.io/pyro2
       print matplotlib.pyplot.get_backend()
       ```
 
- - If you want to run the unit tests, you need to have `pytest` installed.
+ - If you want to run the unit tests, you need to have `pytest`
+   installed.
 
  - Finally, you can run a quick test of the advection solver:
 
@@ -105,8 +107,8 @@ on the grid are described in a jupyter notebook:
 https://github.com/python-hydro/pyro2/blob/main/pyro/mesh/mesh-examples.ipynb
 
 
-Many of the methods here rely on multigrid.  The basic multigrid solver is
-demonstrated in the juputer notebook:
+Many of the methods here rely on multigrid.  The basic multigrid
+solver is demonstrated in the juputer notebook:
 
 https://github.com/python-hydro/pyro2/blob/main/pyro/multigrid/multigrid-constant-coefficients.ipynb
 
@@ -123,7 +125,8 @@ pyro provides the following solvers (all in 2-d):
   - `advection_fv4`: a fourth-order accurate finite-volume advection
     solver that uses RK4 time integration.
 
-  - `advection_nonuniform`: a solver for advection with a non-uniform velocity field.
+  - `advection_nonuniform`: a solver for advection with a non-uniform
+    velocity field.
 
   - `advection_rk`: a second-order unsplit solver for linear advection
     that uses Runge-Kutta integration instead of characteristic
@@ -132,27 +135,28 @@ pyro provides the following solvers (all in 2-d):
   - `advection_weno`: a method-of-lines WENO solver for linear
     advection.
 
-  - `burgers`: a second-order unsplit solver for invsicid Burgers' equation.
+  - `burgers`: a second-order unsplit solver for invsicid Burgers'
+    equation.
 
-  - `viscous_burgers`: a second-order unsplit solver for viscous Burgers' equation
-    with constant-coefficient diffusion. It uses Crank-Nicolson time-discretized
-    solver for solving diffusion.
+  - `viscous_burgers`: a second-order unsplit solver for viscous
+    Burgers' equation with constant-coefficient diffusion. It uses
+    Crank-Nicolson time-discretized solver for solving diffusion.
 
   - `compressible`: a second-order unsplit solver for the Euler
     equations of compressible hydrodynamics.  This uses characteristic
     tracing and corner coupling for the prediction of the interface
     states and a 2-shock or HLLC approximate Riemann solver.
 
-  - `compressible_fv4`: a fourth-order accurate finite-volume compressible
-     hydro solver that uses RK4 time integration.  This is built from the
-     method of McCourquodale and Colella (2011).
+  - `compressible_fv4`: a fourth-order accurate finite-volume
+    compressible hydro solver that uses RK4 time integration.  This
+    is built from the method of McCourquodale and Colella (2011).
 
   - `compressible_rk`: a second-order unsplit solver for Euler
-     equations that uses Runge-Kutta integration instead of
-     characteristic tracing.
+    equations that uses Runge-Kutta integration instead of
+    characteristic tracing.
 
-  - `compressible_sdc`: a fourth-order compressible solver,
-     using spectral-deferred correction (SDC) for the time integration.
+  - `compressible_sdc`: a fourth-order compressible solver, using
+    spectral-deferred correction (SDC) for the time integration.
 
   - `diffusion`: a Crank-Nicolson time-discretized solver for the
     constant-coefficient diffusion equation.
@@ -188,29 +192,30 @@ of the simulation specifying the analysis routines that can be used
 with their data.
 
   - `pyro/util/compare.py`: this takes two simulation output files as
-    input and compares zone-by-zone for exact agreement. This is used as
-    part of the regression testing.
+    input and compares zone-by-zone for exact agreement. This is used
+    as part of the regression testing.
 
       usage: `./compare.py file1 file2`
 
-  - `pyro/plot.py`: this takes the an output file as input and plots the
-    data using the solver's dovis method.
+  - `pyro/plot.py`: this takes the an output file as input and plots
+    the data using the solver's dovis method.
 
       usage: `./plot.py file`
 
   - `pyro/analysis/`
 
-      * `dam_compare.py`: this takes an output file from the
-        shallow water dam break problem and plots a slice through the domain
-        together with the analytic solution (calculated in the script).
+      * `dam_compare.py`: this takes an output file from the shallow
+        water dam break problem and plots a slice through the domain
+        together with the analytic solution (calculated in the
+        script).
 
          usage: `./dam_compare.py file`
 
-      * `gauss_diffusion_compare.py`: this is for the diffusion solver's
-        Gaussian diffusion problem. It takes a sequence of output
-        files as arguments, computes the angle-average, and the plots
-        the resulting points over the analytic solution for comparison
-        with the exact result.
+      * `gauss_diffusion_compare.py`: this is for the diffusion
+        solver's Gaussian diffusion problem. It takes a sequence of
+        output files as arguments, computes the angle-average, and the
+        plots the resulting points over the analytic solution for
+        comparison with the exact result.
 
           usage: `./gauss_diffusion_compare.py file*`
 
@@ -233,9 +238,9 @@ with their data.
 
           usage: `./sedov_compare.py file`
 
-      * `smooth_error.py`: this takes an output file from the advection
-        solver's smooth problem and compares to the analytic solution,
-        outputting the L2 norm of the error.
+      * `smooth_error.py`: this takes an output file from the
+        advection solver's smooth problem and compares to the analytic
+        solution, outputting the L2 norm of the error.
 
           usage: `./smooth_error.py file`
 
@@ -248,8 +253,8 @@ with their data.
 
 ## Understanding the algorithms
 
-  There is a set of notes that describe the background and details of the
-  algorithms that pyro implements:
+  There is a set of notes that describe the background and details of
+  the algorithms that pyro implements:
 
   http://open-astrophysics-bookshelf.github.io/numerical_exercises/
 
@@ -260,10 +265,10 @@ with their data.
 
 ## Regression and unit testing
 
-  The `pyro/test.py` script will run several of the problems (as well as
-  some stand-alone multigrid tests) and compare the solution to stored
-  benchmarks (in each solver's `tests/` subdirectory).  The return value
-  of the script is the number of tests that failed.
+  The `pyro/test.py` script will run several of the problems (as well
+  as some stand-alone multigrid tests) and compare the solution to
+  stored benchmarks (in each solver's `tests/` subdirectory).  The
+  return value of the script is the number of tests that failed.
 
   Unit tests are controlled by pytest and can be run simply via
 
@@ -273,8 +278,8 @@ with their data.
 
 ## Acknowledgements
 
-  If you use pyro in a class or workshop, please e-mail us to let us know
-  (we'd like to start listing these on the website).
+  If you use pyro in a class or workshop, please e-mail us to let us
+  know (we'd like to start listing these on the website).
 
   If pyro was used for a publication, please cite the article found in
   the `CITATION` file.

docs/source/refs.bib

@@ -32,7 +32,6 @@ @article{ZALESAK1979335
 year = "1979",
 issn = "0021-9991",
 doi = "https://doi.org/10.1016/0021-9991(79)90051-2",
-url = "http://www.sciencedirect.com/science/article/pii/0021999179900512",
 author = "Steven T Zalesak",
 abstract = "The theory of flux-corrected transport (FCT) developed by Boris and Book [J. Comput. Phys. 11 (1973) 38; 18 (1975) 248; 20 (1976) 397] is placed in a simple, generalized format, and a new algorithm for implementing the critical flux limiting stage' in multidimensions without resort to time splitting is presented. The new flux limiting algorithm allows the use of FCT techniques in multidimensional fluid problems for which time splitting would produce unacceptable numerical results, such as those involving incompressible or nearly incompressible flow fields. The “clipping” problem associated with the original one dimensional flux limiter is also eliminated or alleviated. Test results and applications to a two dimensional fluid plasma problem are presented."
 }

pyro/compressible/unsplit_fluxes.py

@@ -287,6 +287,7 @@ def unsplit_fluxes(my_data, my_aux, rp, ivars, solid, tc, dt):
 
     riemann = rp.get_param("compressible.riemann")
 
+    riemannFunc = None
     if riemann == "HLLC":
         riemannFunc = ifc.riemann_hllc
     elif riemann == "CGF":

pyro/compressible_rk/fluxes.py

@@ -166,6 +166,7 @@ def fluxes(my_data, rp, ivars, solid, tc):
 
     riemann = rp.get_param("compressible.riemann")
 
+    riemannFunc = None
     if riemann == "HLLC":
         riemannFunc = interface.riemann_hllc
     elif riemann == "CGF":

pyro/compressible_sr/unsplit_fluxes.py

@@ -270,6 +270,7 @@ def unsplit_fluxes(my_data, my_aux, rp, ivars, solid, tc, dt):
 
     riemann = rp.get_param("compressible.riemann")
 
+    riemannFunc = None
     if riemann == "HLLC":
         riemannFunc = ifc.riemann_hllc
     elif riemann == "CGF":

pyro/incompressible/simulation.py

@@ -37,6 +37,7 @@ def initialize(self, other_bc=False, aux_vars=()):
 
         # phi -- used for the projections. Has neumann BC's if v is dirichlet
         # Assuming BC's are either all periodic or all dirichlet
+        phi_bc = None
         if bc.xlb == "periodic":
             phi_bc = bc
         elif bc.xlb == "dirichlet":

pyro/lm_atm/simulation.py

@@ -75,6 +75,7 @@ def initialize(self):
                    self.rp.get_param("mesh.xrboundary"),
                    self.rp.get_param("mesh.ylboundary"),
                    self.rp.get_param("mesh.yrboundary")]:
+            bctype = None
             if bc == "periodic":
                 bctype = "periodic"
             elif bc in ["reflect", "slipwall"]:

pyro/mesh/array_indexer.py

@@ -507,6 +507,7 @@ def pretty_print(self, n=0, fmt=None, show_ghost=True):
 
         # print j descending, so it looks like a grid (y increasing
         # with height)
+        ilo = ihi = jlo = jhi = -1
         if show_ghost:
             if self.idir == 1:
                 ilo = 0
@@ -534,6 +535,8 @@ def pretty_print(self, n=0, fmt=None, show_ghost=True):
         for j in reversed(range(jlo, jhi+1)):
             for i in range(ilo, ihi+1):
 
+                gc = None
+
                 if self.idir == 1:
                     if (j < self.g.jlo or j > self.g.jhi or
                         i < self.g.ilo or i > self.g.ihi+1):

pyro/mesh/mesh-examples.ipynb

@@ -703,11 +703,12 @@ def create(self):
         if self.idir == 1:
             _tmp = np.zeros((self.grid.qx+1, self.grid.qy, self.nvar),
                             dtype=self.dtype)
+            self.data = ArrayIndexerFC(_tmp, idir=self.idir, grid=self.grid)
+
         elif self.idir == 2:
             _tmp = np.zeros((self.grid.qx, self.grid.qy+1, self.nvar),
                             dtype=self.dtype)
-
-        self.data = ArrayIndexerFC(_tmp, idir=self.idir, grid=self.grid)
+            self.data = ArrayIndexerFC(_tmp, idir=self.idir, grid=self.grid)
 
         self.initialized = 1