invest: GEOS and PROJ bump causing Wave Energy / Coastal Vulnerability tests to fail
I noticed tests failing not related to PR #475 . When looking at the conda env diff, I saw some important updates.
When updating to Proj=7.2.0 in Conda locally I could reproduce the test failures and I also got the dependent updates:
The following packages will be UPDATED:
cairo 1.16.0-hd28d34b_1006 --> 1.16.0-hba8bd2f_1007
expat 2.2.9-h33f27b4_2 --> 2.2.10-h39d44d4_0
gdal 3.2.1-py38hb9e1232_0 --> 3.2.1-py38hb9e1232_4
geos 3.8.1-he025d50_0 --> 3.9.0-h39d44d4_0
geotiff 1.6.0-h8884d1a_3 --> 1.6.0-h5d775bf_4
icu 67.1-h33f27b4_0 --> 68.1-h0e60522_0
libclang 10.0.1-default_hf44288c_1 --> 11.0.1-default_h5c34c98_1
libgdal 3.2.1-h0d43894_0 --> 3.2.1-hd376e63_4
libspatialite 5.0.0-hf693123_0 --> 5.0.1-h6966176_2
openjpeg 2.3.1-h48faf41_3 --> 2.4.0-h48faf41_0
poppler 0.89.0-h5d62644_1 --> 0.89.0-h3772339_5
proj 7.1.1-h7d85306_3 --> 7.2.0-h3e70539_1
qt 5.12.9-hb2cf2c5_0 --> 5.12.9-h5909a2a_3
shapely 1.7.1-py38hc96c142_1 --> 1.7.1-py38hc55ee08_2
tiledb 2.1.6-hf84e3da_1 --> 2.2.3-hf84e3da_0
About this issue
- Original URL
- State: closed
- Created 3 years ago
- Comments: 18 (18 by maintainers)
Commits related to this issue
- I. #476 bound geos version — committed to dcdenu4/invest by dcdenu4 3 years ago
- I. #476 Update Wave Energy coords — committed to dcdenu4/invest by dcdenu4 3 years ago
- I. #476 need to set PROJ lower bound — committed to dcdenu4/invest by dcdenu4 3 years ago
- I. #476 bump gdal to bump PROJ — committed to dcdenu4/invest by dcdenu4 3 years ago
- I. #476 skip tests for geos/proj updates — committed to dcdenu4/invest by dcdenu4 3 years ago
- updated test data revision. #476. — committed to davemfish/invest by davemfish 3 years ago
- stop skipping tests that were failing due to a geos bug - it's been patched. #476. — committed to davemfish/invest by davemfish 3 years ago
- I. #476 make assert vector more robust — committed to dcdenu4/invest by dcdenu4 3 years ago
- I. #476 fix typo — committed to dcdenu4/invest by dcdenu4 3 years ago
One step forward and two steps back! Glad the conda env diffing has been helping!
Wave energy was not touched in 505. I tried unskipping them with the new GEOS version, but that didn’t seem to be the problem so I left them skipped.
hooray for the conda env diff actions step!