Updates Build Instructions to Adjust for Devcontainer Instructions and Remove Stale Information

docs/source/developer_guide/build.md

@@ -18,30 +18,49 @@ git clone https://github.com/rapidsai/cuspatial.git $CUSPATIAL_HOME
 2. clone the cuSpatial repo
 
 ```shell
-conda env update --file conda/environments/all_cuda-118_arch-x86_64.yaml
+conda env create -n cuspatial --file conda/environments/all_cuda-118_arch-x86_64.yaml
 ```
 
-## Build and install cuSpatial
+## Build cuSpatial
 
-1. Compile and install
-   ```shell
-   cd $CUSPATIAL_HOME && \
-   chmod +x ./build.sh && \
-   ./build.sh
-   ```
+### From the cuSpatial Dev Container:
 
-2. Run C++/Python test code
+Execute `build-cuspatial-cpp to build `libcuspatial`. The following options may be added.
+ - `-DBUILD_TESTS=ON`: build `libcuspatial` unit tests.
+ - `-DBUILD_BENCHMARKS=ON`: build `libcuspatial` benchmarks.
+ - `-DCMAKE_BUILD_TYPE=Debug`: Create a Debug build of `libcuspatial` (default is Release).
+In addition, `build-cuspatial-python` to build cuspatial cython components.
 
-   Some tests using inline data can be run directly, e.g.:
+### From Bare Metal:
 
-   ```shell
-   $CUSPATIAL_HOME/cpp/build/gtests/LEGACY_HAUSDORFF_TEST
-   $CUSPATIAL_HOME/cpp/build/gtests/POINT_IN_POLYGON_TEST
-   python python/cuspatial/cuspatial/tests/legacy/test_hausdorff_distance.py
-   python python/cuspatial/cuspatial/tests/test_pip.py
-   ```
+Compile libcuspatial (C++), cuspatial (cython) and C++ tests:
+```shell
+cd $CUSPATIAL_HOME && \
+chmod +x ./build.sh && \
+./build.sh libcuspatial cuspatial tests
+```
+Additionally, the following options are also commonly used:
+- `benchmarks`: build libcuspatial benchmarks
+- `clean`: remove all existing build artifacts and configuration
+Execute `./build.sh -h` for full list of available options.
+
+## Validate Installation with C++ and Python Tests
+
+- C++ tests are located within the `$CUSPATIAL_HOME/cpp/build/gtests` directory.
+- Python tests are located within the `$CUSPATIAL_HOME/python/cuspatial/cuspatial/tests` directory.
+
+```note
+Dev Container users:
+To manage difference between branches and build types, the build directories are located at
+`$CUSPATIAL_HOME/cpp/build/[release|debug]` depending on build type, and  `$CUSPATIAL_HOME/cpp/build/latest`.
+is a symbolic link to the most recent build directory.
 
-   Some other tests involve I/O from data files under `$CUSPATIAL_HOME/test_fixtures`.
-   For example, `$CUSPATIAL_HOME/cpp/build/gtests/SHAPEFILE_READER_TEST` requires three
-   pre-generated polygon shapefiles that contain 0, 1 and 2 polygons, respectively. They are available at
-   `$CUSPATIAL_HOME/test_fixtures/shapefiles` <br>
+Execute C++ tests:
+```shell
+ninja -C $CUSPATIAL_HOME/cpp/build/ test
+```
+
+Execute Python tests:
+```
+pytest $CUSPATIAL_HOME/python/cuspatial/cuspatial/tests/
+```

docs/source/developer_guide/development_environment.md

@@ -1,15 +1,25 @@
 # Creating a Development Environment
 
-cuSpatial follows the RAPIDS release schedule, so developers are encouraged to develop 
-using the latest development branch of RAPIDS libraries that cuspatial depends on. Other
-cuspatial dependencies can be found in `conda/environments/`.
-
-Maintaining a local development environment can be an arduous task, especially after each
-RAPIDS release. Most cuspatial developers today use
-[rapids-compose](https://github.com/trxcllnt/rapids-compose) to setup their development environment.
-It contains helpful scripts to build a RAPIDS development container image with the required
-dependencies and RAPIDS libraries automatically fetched and correctly versioned. It also provides
-script commands for simple building and testing of all RAPIDS libraries, including cuSpatial.
-`rapids-compose` is the recommended way to set up your environment to develop for cuspatial.
-
-For developers who would like to build from conda or from source, see the [build page](https://docs.rapids.ai/api/cuspatial/stable/developer_guide/build.html).
+cuSpatial recommends using [Dev Containers](https://containers.dev/) to setup the development environment.
+To setup Dev Containers for cuspatial, please refer to [documentation](https://github.com/rapidsai/cuspatial/tree/main/.devcontainer).
+
+## From Bare Metal
+
+RAPIDS keeps a single source of truth for library dependencies in `dependencies.yaml`. This file divides
+the dependencies into several dimensions: building, testing, documentations, notebooks etc. As a developer,
+you generally want to generate an environment recipe that includes everything that the library *may* use.
+
+To do so, install the rapids-dependency-file-generator via pip:
+```shell
+pip install rapids-dependency-file-generator
+```
+
+And run under the repo root:
+```shell
+rapids-dependency-file-generator --clean
+```
+
+The environment recipe is generated within the `conda/environments` directory. To continue the next step of building,
+see the [build page](https://docs.rapids.ai/api/cuspatial/stable/developer_guide/build.html).
+
+For more information about how RAPIDS manages dependencies, see [README of rapids-dependency-file-generator repo](https://github.com/rapidsai/dependency-file-generator).