Merge branch 'main' into ParSNSTOP

Author: marcomangano

.github/build_real.sh

@@ -7,4 +7,11 @@ if [[ $IMAGE == "private" ]]; then
     cp -r $HOME/SNOPT/* pyoptsparse/pySNOPT/source
 fi
 
+# temporarily disable pip constraints file for build-time dependencies
+mv ~/.config/pip/constraints.txt ~/.config/pip/constraints.txt.bkup
+touch  ~/.config/pip/constraints.txt
+
 pip install .[optview,testing] -v
+
+# move pip constraints file back
+mv ~/.config/pip/constraints.txt.bkup ~/.config/pip/constraints.txt

.github/environment.yml

@@ -1,19 +1,22 @@
 dependencies:
 # build
   - python >=3.9
-  - numpy >=1.21,<2
-  - ipopt
+  - numpy >=2.0
   - swig
   - meson >=1.3.2
   - compilers
   - pkg-config
   - pip
   - setuptools
   - build
+  - flang 5*
+  - libpgmath
+# runtime
   - packaging
+  - mdolab-baseclasses >=1.3.1
+  - scipy >=1.7
+  - sqlitedict >=1.6
+  - cyipopt
 # testing
   - parameterized
   - testflo
-  - scipy >=1.7
-  - mdolab-baseclasses >=1.3.1
-  - sqlitedict >=1.6

.github/test_real.sh

@@ -11,4 +11,5 @@ cd tests
 # we have to copy over the coveragerc file to make sure it's in the
 # same directory where codecov is run
 cp ../.coveragerc .
-testflo --pre_announce --disallow_deprecations -v --coverage --coverpkg pyoptsparse $EXTRA_FLAGS --timeout 60
+
+testflo -i --pre_announce --disallow_deprecations -v --coverage --coverpkg pyoptsparse $EXTRA_FLAGS --timeout 60

.github/windows.yaml

@@ -27,7 +27,6 @@ jobs:
       displayName: Install mamba and update environment
 
     - script: |
-        set IPOPT_DIR=%CONDA_PREFIX%\Library
         set CC=cl
         set FC=flang
         set CC_LD=link

.github/workflows/linux-build.yml

@@ -0,0 +1,34 @@
+name: PyOptSparse Linux Actions
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+    branches:
+      - main
+
+jobs:
+  build-linux:
+    runs-on: ubuntu-latest
+    timeout-minutes: 15
+    strategy:
+      fail-fast: false
+      matrix:
+        versions: [{python: "3.10", numpy: "1.22"}, {python: "3.11", numpy: "1.25"}, {python: "3.12", numpy: "2.1"}]
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-python@v5 
+        with:
+          python-version: ${{ matrix.versions.python }}
+      - name: Install runtime numpy ${{ matrix.versions.numpy }}
+        run: |
+          pip install -U pip wheel
+          pip install numpy~=${{ matrix.versions.numpy }}
+      - name: Build and install pyoptsparse
+        run: |
+          pip install .['testing']
+      - name: Run tests
+        run: |
+          cd tests
+          testflo -i --pre_announce -v -n 1 .

.github/workflows/windows-build.yml

@@ -11,35 +11,45 @@ on:
 jobs:
   build-windows:
     runs-on: windows-latest
-
+    timeout-minutes: 15
+    strategy:
+      fail-fast: false
+      matrix:
+        numpy_version: ["1.21.6", "1.25.2", "2.1.3"]
     steps:
-      - uses: actions/checkout@v2
-      - uses: conda-incubator/setup-miniconda@v2
+      - uses: actions/checkout@v4
+      - uses: conda-incubator/setup-miniconda@v3
         with:
-          python-version: 3.9
-          miniforge-variant: Mambaforge
+          python-version: "3.10"
           channels: conda-forge,defaults
           channel-priority: strict
           activate-environment: pyos-build
           environment-file: .github/environment.yml
-      - name: Install libpgmath
-        shell: bash -l {0}
-        run: |
-          conda activate pyos-build
-          mamba install libpgmath
       - name: Build and install pyoptsparse
         shell: cmd /C CALL {0}
         run: |
           call conda activate pyos-build
-          set IPOPT_DIR=%CONDA_PREFIX%\Library
-          set CC=cl
-          set FC=flang
-          set CC_LD=link
+
+          :: debug - list conda environment
+          call conda list
+
+          :: debug - list environment vars
+          set
+
+          :: set fortran compiler, flang 5 activation doesn't seem to do it
+          set FC=flang.exe
+
           python -m build -n -x .
+
           pip install --no-deps --no-index --find-links dist pyoptsparse
+      - name: Install runtime numpy ${{ matrix.numpy_version }}
+        shell: bash -l {0}
+        run: |
+          conda activate pyos-build
+          conda install numpy==${{ matrix.numpy_version }}
       - name: Run tests
         shell: bash -l {0}
         run: |
           conda activate pyos-build
           cd tests
-          testflo --pre_announce -v -n 1 --timeout 60 .
+          testflo -i --pre_announce -v -n 1 --timeout 60 .

README.md

@@ -23,7 +23,7 @@ ALPSO, CONMIN, IPOPT, NLPQLP, NSGA2, PSQP, SLSQP, ParOpt and SNOPT are currently
 
 We do not provide the source code for SNOPT and NLPQLP, due to their restrictive license requirements.
 Please contact the authors of the respective optimizers if you wish to obtain them.
-Furthermore, ParOpt and IPOPT are available as a open source package but must be installed separately.
+Furthermore, ParOpt and IPOPT are available as open source packages but must be installed separately.
 Please see the documentation page of each optimizer for purchase and installation instructions.
 
 ## Integration into other frameworks

doc/install.rst

@@ -34,7 +34,6 @@ pyOptSparse has the following dependencies:
   We recommend ``gcc`` and ``gfortran`` which can be installed via the package manager for your operating system.
 
 Please make sure these are installed and available for use.
-In order to use NSGA2, SWIG (v1.3+) is also required, which can be installed via the package manager.
 Python dependencies are automatically handled by ``pip``, so they do not need to be installed separately.
 
 .. note::
@@ -68,6 +67,12 @@ If you plan to modify pyOptSparse, installing with the developer option, i.e. wi
   Some optimizers are proprietary, and their sources are not distributed with pyOptSparse.
   To use them, please follow the instructions on specific optimizer pages.
 
+To see the list of installed optimizers, use the following:
+
+.. prompt:: bash
+
+  python -c "import pyoptsparse; print(pyoptsparse.list_optimizers())"
+
 Specifying compilers
 ~~~~~~~~~~~~~~~~~~~~
 To specify a non-default compiler (e.g. something other than ``/usr/bin/gcc``), meson recognizes certain `special environment variables <https://mesonbuild.com/Reference-tables.html#compiler-and-linker-selection-variables>`__.
@@ -184,19 +189,6 @@ An ``environment.yml`` file is provided in the ``pyoptsparse`` repo:
     conda config --env --set channel_priority strict
 
     conda env update -f .github\environment.yml
-    conda install libpgmath
-
-Next, we need to tell the compiler where to find IPOPT:
-
-.. tabs::
-
-  .. code-tab:: bash Linux/OSX
-
-    export IPOPT_DIR="$CONDA_PREFIX"
-
-  .. code-tab:: powershell Windows
-
-    set IPOPT_DIR=%CONDA_PREFIX%\Library
 
 Finally, build the wheel and install it using pip:
 

doc/optimizers/IPOPT.rst

@@ -4,71 +4,14 @@ IPOPT
 =====
 IPOPT (Interior Point OPTimizer) is an open source interior point optimizer, designed for large-scale nonlinear optimization.
 The source code can be found `here <https://www.coin-or.org/download/source/Ipopt/>`_.
-The latest version we support is 3.13.2.
+The latest version we support is 3.14.17.
 
 Installation
 ------------
-IPOPT must be installed separately, then linked to pyOptSparse when building.
-For the full installation instructions, please see `their documentation <https://coin-or.github.io/Ipopt/INSTALL.html>`_.
-OpenMDAO also has a very helpful `script <https://github.com/OpenMDAO/build_pyoptsparse/>`_ which can be used to install IPOPT with other linear solvers.
-Here we explain a basic setup using MUMPS as the linear solver, together with METIS adapted from the OpenMDAO script.
-
-#. Download the tarball and extract it to ``$IPOPT_DIR`` which could be set to for example ``$HOME/packages/Ipopt``.
-
-#. Install METIS, which can be used to improve the performance of the MUMPS linear solver.
-
-   .. code-block:: bash
-
-      # build METIS
-      cd $IPOPT_DIR
-      git clone https://github.com/coin-or-tools/ThirdParty-Metis.git
-      cd ThirdParty-Metis
-      ./get.Metis
-      ./configure --prefix=$IPOPT_DIR
-      make
-      make install
-
-#. Install MUMPS
-
-   .. code-block:: bash
-
-      # build MUMPS
-      cd $IPOPT_DIR
-      git clone https://github.com/coin-or-tools/ThirdParty-Mumps.git
-      cd ThirdParty-Mumps
-      ./get.Mumps
-      ./configure --with-metis --with-metis-lflags="-L${IPOPT_DIR}/lib -lcoinmetis" \
-           --with-metis-cflags="-I${IPOPT_DIR}/include -I${IPOPT_DIR}/include/coin-or -I${IPOPT_DIR}/include/coin-or/metis" \
-           --prefix=$IPOPT_DIR CFLAGS="-I${IPOPT_DIR}/include -I${IPOPT_DIR}/include/coin-or -I${IPOPT_DIR}/include/coin-or/metis" \
-           FCFLAGS="-I${IPOPT_DIR}/include -I${IPOPT_DIR}/include/coin-or -I${IPOPT_DIR}/include/coin-or/metis"
-      make
-      make install
-
-#. Build IPOPT
-
-   .. code-block:: bash
-
-      # build IPOPT
-      cd $IPOPT_DIR
-      mkdir build
-      cd build
-      ../configure --prefix=${IPOPT_DIR} --disable-java --with-mumps --with-mumps-lflags="-L${IPOPT_DIR}/lib -lcoinmumps" \
-           --with-mumps-cflags="-I${IPOPT_DIR}/include/coin-or/mumps"
-      make
-      make install     
-
-#. You must add the IPOPT library path to the ``LD_LIBRARY_PATH`` variable for things to work right.
-   This could be done for example by adding the following to your ``.bashrc``:
-
-   .. code-block:: bash
-
-     export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$IPOPT_DIR/lib
-
-   Furthermore, the environment variable ``$IPOPT_DIR`` must be set correctly in order to link to pyOptSparse.
-   Alternatively, you can manually define the variables ``$IPOPT_LIB`` and ``$IPOPT_INC`` for the lib and include paths separately.
-
-
-#. Now clean build pyOptSparse. Verify that IPOPT works by running the relevant tests.
+IPOPT and its Python interface `cyipopt <https://github.com/mechmotum/cyipopt>` must be installed separately.
+Follow the instructions `here <https://cyipopt.readthedocs.io/en/stable/install.html>`_.
+OpenMDAO also has a very helpful `script <https://github.com/OpenMDAO/build_pyoptsparse/>`_ which can be used to install IPOPT with other linear solvers,
+but it does not install ``cyipopt`` for you.
 
 Options
 -------

meson_options.txt

@@ -1,6 +1,3 @@
-option('ipopt_dir', type: 'string', value: '',
-        description: 'Top-level dir for ipopt')
-
 option('incdir_numpy', type: 'string', value: '',
     description: 'Include directory for numpy. If left empty Meson will try to find it on its own.')