Is your feature request related to a problem? Please describe.
Pretty selfish ask here, but we'd like to consider allowing a wider range of allowed python-dateutil, specifically allowing something below 2.8.1 in the generated client TOML.
Some of our internal tooling which uses both our generated openapi client and aws-sam-cli is incompatible due to a dependency conflict. Specifically, aws-sam-cli requires python-dateutil < 2.8.1. We asked about expanding the acceptable range in aws-sam-cli, since the .1 patch version in particular seems arbitrary and were basically told to pound sand.
I know this is pretty esoteric to our individual use case, but thought I'd ask.
Describe the solution you'd like
Support a more permissive range of python-dateutil under 2.8.1.
Describe alternatives you've considered
We can either make this change in our fork directly - which I was hoping to avoid because I'd like to be able to soon ignore our fork and just use the dependency as is - or continue an internal workaround we have around not using the tooling that causes the dependency conflict.
Is your feature request related to a problem? Please describe.
Pretty selfish ask here, but we'd like to consider allowing a wider range of allowed
python-dateutil, specifically allowing something below2.8.1in the generated client TOML.Some of our internal tooling which uses both our generated openapi client and aws-sam-cli is incompatible due to a dependency conflict. Specifically, aws-sam-cli requires python-dateutil < 2.8.1. We asked about expanding the acceptable range in aws-sam-cli, since the
.1patch version in particular seems arbitrary and were basically told to pound sand.I know this is pretty esoteric to our individual use case, but thought I'd ask.
Describe the solution you'd like
Support a more permissive range of python-dateutil under 2.8.1.
Describe alternatives you've considered
We can either make this change in our fork directly - which I was hoping to avoid because I'd like to be able to soon ignore our fork and just use the dependency as is - or continue an internal workaround we have around not using the tooling that causes the dependency conflict.