Open SPlanzer opened 1 month ago
I can't even get your workaround to work.
What I have been using is a conda environment file (which still uses pip to install vector2dggs):
name: vector2dggs
channels:
- conda-forge
- defaults
dependencies:
- python=3.11
- pip
- gdal=3.9.1
- pip:
- vector2dggs>=0.5.4
If you save this to env.yml
and run conda env create --name v2d --file env.yml
... well, it still fails. (However, it definitely has worked in the past.)
However here's my new workaround:
conda create --name v2d python=3.11 gdal=3.9.1 -c conda-forge
pip install vector2dggs==0.5.4
Installing 0.6.0 (current release) does not work even with this method. This is probably due to an upgrade in the gdal dependency introduced in https://github.com/manaakiwhenua/vector2dggs/commit/c61b7dbf0a3ab9bfc8ae0e80c391e79263dbbc65
Since it seems to matter, my system has libgeos-dev
on Ubuntu 22.04 LTS at 3.12.1-1.
name: vector2dggs
channels:
- conda-forge
- defaults
dependencies:
- python=3.11
- pip
- gdal=3.9.1
- pip:
- vector2dggs==0.5.4
Works. So this is ultimately associated with upgrading the python GDAL dependency; perhaps some conflict with system-installed GEOS. I'll see if we even need the upgrade to GDAL for v0.6; I don't think it's necesssary.
With latest release (0.6.1) I can now do a bare pip install vector2dggs
successfully.
@SPlanzer if you can do this too, please close the issue.
Pip and Poetry install failure
pip install
is failing (Ubuntu with a new Conda environment)The below collapsible section has the details of the failed install. Looks to be related to compiling of the Cython code with pygeos
Installation Log: pip install vector2dggs
```plaintext pip install vector2dggs Collecting vector2dggs Using cached vector2dggs-0.6.0-py3-none-any.whl.metadata (9.8 kB) Collecting click<9.0.0,>=8.1.3 (from vector2dggs) Using cached click-8.1.7-py3-none-any.whl.metadata (3.0 kB) Collecting click-log<0.5.0,>=0.4.0 (from vector2dggs) Using cached click_log-0.4.0-py2.py3-none-any.whl.metadata (1.2 kB) Collecting dask<2024.0.0,>=2023.3.0 (from vector2dggs) Using cached dask-2023.12.1-py3-none-any.whl.metadata (3.7 kB) Collecting dask-geopandas<0.4.0,>=0.3.0 (from vector2dggs) Using cached dask_geopandas-0.3.1-py3-none-any.whl.metadata (3.6 kB) Collecting gdal<4.0.0,>=3.8.0 (from vector2dggs) Using cached gdal-3.9.1.tar.gz (840 kB) Installing build dependencies ... done Getting requirements to build wheel ... done Preparing metadata (pyproject.toml) ... done Collecting geopandas<0.13.0,>=0.12.2 (from vector2dggs) Using cached geopandas-0.12.2-py3-none-any.whl.metadata (1.0 kB) Collecting h3pandas<0.3.0,>=0.2.6 (from vector2dggs) Using cached h3pandas-0.2.6-py3-none-any.whl Collecting psycopg2<3.0.0,>=2.9.6 (from vector2dggs) Using cached psycopg2-2.9.9-cp312-cp312-linux_x86_64.whl Collecting pyarrow<15.0.0,>=14.0.1 (from vector2dggs) Using cached pyarrow-14.0.2-cp312-cp312-manylinux_2_28_x86_64.whl.metadata (3.0 kB) Collecting pygeos<0.14,>=0.13 (from vector2dggs) Using cached pygeos-0.13.tar.gz (141 kB) Installing build dependencies ... done Getting requirements to build wheel ... error error: subprocess-exited-with-error × Getting requirements to build wheel did not run successfully. │ exit code: 1 ╰─> [59 lines of output]poetry install
returns similar resultsInstallation Log: poetry install vector2dggs
```plaintext - Installing pygeos (0.14): Failed ChefBuildError Backend subprocess exited when trying to invoke get_requires_for_build_wheelFor Now
I was successful in install this with the below