Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update Dependency List with dependencies.yaml [skip gpuci] #803

Merged
merged 20 commits into from
Nov 21, 2022
Merged
Show file tree
Hide file tree
Changes from 2 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -59,7 +59,7 @@ git clone https://github.com/rapidsai/cuspatial.git $CUSPATIAL_HOME
2. clone the cuSpatial repo

```shell
conda env update --file conda/environments/cuspatial_dev_cuda11.5.yml
conda env update --file conda/environments/all_cuda-115_arch-x86_64.yaml
```

### Build and install cuSpatial
Expand Down
31 changes: 31 additions & 0 deletions conda/environments/all_cuda-115_arch-x86_64.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,31 @@
# This file is generated by `rapids-dependency-file-generator`.
# To make changes, edit dependencies.yaml and run `rapids-dependency-file-generator`.
channels:
- rapidsai
- rapidsai-nightly
- conda-forge
- nvidia
dependencies:
- black=22.3.0
- clang-tools=11.1.0
- clang=11.1.0
- cmake>=3.23.1
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Suggested change
- cmake>=3.23.1
- cmake>=3.23.1,!=3.25.0

- cmakelang=0.6.13
- cudatoolkit=11.5
- cudf=22.12.*
- cython>=0.29,<0.30
- flake8=3.8.3
- gcovr>=5.0
- gdal>=3.0.2
isVoid marked this conversation as resolved.
Show resolved Hide resolved
- geopandas>=0.11.0
- isort=5.10.1
- myst-parser
- numba>=0.49
- numpy>=1.19
- pydata-sphinx-theme
- pytest
- pytest-cov
- python>=3.8,<3.10
- scikit-build>=0.13.1
name: all_cuda-115_arch-x86_64

20 changes: 0 additions & 20 deletions conda/environments/cuspatial_dev_cuda11.5.yml

This file was deleted.

100 changes: 100 additions & 0 deletions dependencies.yaml
Original file line number Diff line number Diff line change
@@ -0,0 +1,100 @@
# Dependency list for https://github.com/rapidsai/dependency-file-generator
files:
all:
output: conda
matrix:
cuda: ["11.5"]
arch: [x86_64]
includes:
- build
- cudatoolkit
- develop
- run
- test_python
test_python:
output: none
includes:
- cudatoolkit
- py_version
- test_python
test_cpp:
output: none
includes:
- cudatoolkit
- test_cpp
checks:
output: none
includes:
- build
- develop
- py_version
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Style checks should not be dependent on a specific Python version being specified in the matrix (if no value is specified, the matrix match will fail). The Python requirement in the build list should be sufficient. However, we may want a default (empty matrix in the last entry) of py_version defining >=3.8,<3.10 instead of putting it in build.

Suggested change
- py_version

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Any reason doing it just for py_version? Why not also cython?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

And are you suggesting removing python entry from build and use py_version in all?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think you did what I wanted -- remove py_version from the style check dependency list.

As for removing python from build -- is there a C++ library that can be built without Python? If so, I would remove python, cython, ... from build and put them in a separate list like build_python. We're trying to achieve a level of modularity/granularity that is useful for minimizing dependencies at each stage of a build.

channels:
- rapidsai
- rapidsai-nightly
- conda-forge
- nvidia
dependencies:
build:
common:
- output_types: [conda, requirements]
packages:
- cudf=22.12.*
bdice marked this conversation as resolved.
Show resolved Hide resolved
- cmake>=3.23.1
ajschmidt8 marked this conversation as resolved.
Show resolved Hide resolved
- cython>=0.29,<0.30
- python>=3.8,<3.10
- scikit-build>=0.13.1
- gdal>=3.0.2
- geopandas>=0.11.0
cudatoolkit:
specific:
- output_types: conda
matrices:
- matrix:
cuda: "11.5"
packages:
- cudatoolkit=11.5
develop:
bdice marked this conversation as resolved.
Show resolved Hide resolved
common:
- output_types: [conda, requirements]
packages:
- black=22.3.0
- cmakelang=0.6.13
- flake8=3.8.3
- gcovr>=5.0
isVoid marked this conversation as resolved.
Show resolved Hide resolved
- isort=5.10.1
- pydata-sphinx-theme
- myst-parser
- output_types: conda
isVoid marked this conversation as resolved.
Show resolved Hide resolved
packages:
- clang=11.1.0
- clang-tools=11.1.0
py_version:
specific:
- output_types: conda
matrices:
- matrix:
py: "3.8"
packages:
- python=3.8
- matrix:
py: "3.9"
packages:
- python=3.9
run:
common:
- output_types: [conda, requirements]
packages:
- numba>=0.49
isVoid marked this conversation as resolved.
Show resolved Hide resolved
- numpy>=1.19
test_cpp:
common:
- output_types: [conda]
isVoid marked this conversation as resolved.
Show resolved Hide resolved
packages:
- gtest=1.10.0
isVoid marked this conversation as resolved.
Show resolved Hide resolved
- gmock=1.10.0
test_python:
common:
- output_types: [conda, requirements]
packages:
- pytest
- pytest-cov