Drop logical model mapping tests (#320)

* CONTRIBUTING: Get rid of remote-mlmd recommendations

Signed-off-by: Isabella do Amaral <[email protected]>

* robot: drop logical model mapping tests

Since [1] the Python client doesn't access MLMD directly, instead
routing requests through the MR server, there's no need to check for
matching implementations.

[1]: 9b9be50 "Switch Python client backend to use REST (#152)"

Signed-off-by: Isabella do Amaral <[email protected]>

---------

Signed-off-by: Isabella do Amaral <[email protected]>

.github/workflows/run-robot-tests.yaml

@@ -6,20 +6,20 @@ on:
   push:
     # To any branch
     branches:
-      - '*'
+      - "*"
   # For every pull request
   pull_request:
     # But ignore this paths
     paths-ignore:
-      - 'LICENSE*'
-      - 'DOCKERFILE*'
-      - '**.gitignore'
-      - '**.md'
-      - '**.txt'
-      - '.github/ISSUE_TEMPLATE/**'
-      - '.github/dependabot.yml'
-      - 'docs/**'
-      - 'scripts/**'
+      - "LICENSE*"
+      - "DOCKERFILE*"
+      - "**.gitignore"
+      - "**.md"
+      - "**.txt"
+      - ".github/ISSUE_TEMPLATE/**"
+      - ".github/dependabot.yml"
+      - "docs/**"
+      - "scripts/**"
 # Define workflow jobs
 jobs:
   # Job runs Robot Framework tests against locally build image from current code
@@ -36,9 +36,9 @@ jobs:
         uses: actions/setup-python@v5
         with:
           # Set Python version to install
-          python-version: '3.9'
+          python-version: "3.9"
           # Set architecture of Python to install
-          architecture: 'x64'
+          architecture: "x64"
       # Install required Python packages for running Robot Framework tests
       - name: Install required Python packages
         # Install required Python packages using pip
@@ -55,10 +55,6 @@ jobs:
       - name: Run Robot Framework tests (REST mode)
         # Run Robot Framework tests in REST mode from test/robot directory
         run: robot test/robot
-      # Run Robot Framework tests in Python mode against running docker compose
-      - name: Run Robot Framework tests (Python mode)
-        # Run Robot Framework tests in Python mode from test/robot directory
-        run: TEST_MODE=Python robot test/robot/MRandLogicalModel.robot
       # Shutdown docker compose with locally build image from current code
       - name: Shutdown docker compose with local image
         # Shutdown docker compose running in the background

CONTRIBUTING.md

@@ -108,17 +108,6 @@ This colima setups allows to:
 - launch Integration tests in Go (used in Core go layer) with Testcontainers for Go
 - launch DevContainer to be able to install MLMD python wheel dependency (which is x86 specific)
 
-## Remote-only MLMD Python library
-
-The Model Registry Python client wraps `ml-metadata` python dependency (which is [x86 specific](https://pypi.org/project/ml-metadata/#files)).
-
-When developing and contributing to the Model Registry Python client using Apple-silicon/ARM-based computers,
-it can be helpful to evaluate using locally this soft-fork of the upstream MLMD library supporting only remote gRPC connections:
-https://github.com/opendatahub-io/ml-metadata/releases/tag/v1.14.0%2Bremote.1
-
-This dependency can be substituted in the [clients/python/pyproject.toml](clients/python/pyproject.toml) and installed locally with `poetry lock`.
-It is recommended not to check-in this substitution, only helpful for local development while using Apple-silicon/ARM-based computers.
-
 ## DevContainer
 
 Using a [DevContainer](https://containers.dev) is helpful to develop with the Model Registry Python client, since it needs to wrap MLMD python dependency (which is [x86 specific](https://pypi.org/project/ml-metadata/#files)).

test/robot/requirements.txt

@@ -1,4 +1,3 @@
-ml-metadata==1.14.0
 robotframework
 robotframework-requests
 pyyaml