Bump version to 1.20.0~dev post-release

[cdesiniotis]

No description provided.

[elezar]

As a note. This is what fills up our Shipit view with versions that we don't want to release. Does it make sense to use something like rc.0 here instead so that ordering at least remains consistent?

[cdesiniotis]

Could you clarify how adding rc.0 here would help with the ordering?

[cdesiniotis]

Updated to rc.0 as you suggested.

[elezar]

To be clear, this could be solved elsewhere in the stack (e.g. where we construct the kitmaker build version), but see the following example:

Note that the 1.19.0~dev releases ALWAYS come before the 1.19.0 releases.

Looking at the how we construct the kitmaker version, we could use dev here and then change the following line https://github.com/elezar/nvidia-container-toolkit/blob/5eee5ce7b8ab19012f5aabdafd525f21b3a54496/scripts/release-kitmaker-artifactory.sh#L196 too:

diff --git a/scripts/release-kitmaker-artifactory.sh b/scripts/release-kitmaker-artifactory.sh
index 58ea8a61..0f4780fe 100755
--- a/scripts/release-kitmaker-artifactory.sh
+++ b/scripts/release-kitmaker-artifactory.sh
@@ -193,7 +193,7 @@ function upload_archive() {
 component="nvidia_container_toolkit"
 version="${VERSION%~rc.*}"
 version_suffix=$(date -r "${IMAGE_EPOCH}" '+%Y.%m.%d.%s' || date -d @"${IMAGE_EPOCH}" '+%Y.%m.%d.%s')
-kitmaker_version="${VERSION%~rc.*}.${version_suffix}"
+kitmaker_version="${VERSION%~*}.${version_suffix}"
 kitmaker_os="linux"
 
 # create_and_upload creates a kitmaker archive for the specified component, os, and arch and uploads it.

That might be a cleaner long-term change.

[cdesiniotis]

Got it. Your proposal makes sense. I have changed LIB_TAG back to dev to align with what was done previously. I have opened #1764 to hopefully clean up our future builds in ShipIt.

[coveralls]

Coverage Report for CI Build 24370293315

Coverage remained the same at 43.4%

Details

• Coverage remained the same as the base build.
• Patch coverage: No coverable lines changed in this PR.
• No coverage regressions found.

Uncovered Changes

No uncovered changes found.

Coverage Regressions

No coverage regressions found.

---

Coverage Stats

Relevant Lines:	14825
Covered Lines:	6434
Line Coverage:	43.4%
Coverage Strength:	0.48 hits per line

---

💛 - Coveralls