ci: remove stale submodules after checkout

When ‘fetch’ strategy is used, Gitlab removes untracked files before
checking out new revision. However if the new revision doesn’t
include some of the submodules which were present in the old
revision, such submodule directories would not be removed by the
checkout. This extra step ensures that these stale submodules are
removed.

.gitlab-ci.yml

@@ -31,11 +31,25 @@ variables:
   APPLY_BOT_FILTER_SCRIPT: "$CI_PROJECT_DIR/tools/ci/apply_bot_filter.py"
   CHECKOUT_REF_SCRIPT: "$CI_PROJECT_DIR/tools/ci/checkout_project_ref.py"
 
+# When 'fetch' strategy is used, Gitlab removes untracked files before checking out
+# new revision. However if the new revision doesn't include some of the submodules
+# which were present in the old revision, such submodule directories would not be
+# removed by the checkout. This extra step ensures that these stale submodules
+# are removed.
+.git_clean_stale_submodules:  &git_clean_stale_submodules >
+  find . -name '.git' -not -path './.git' -printf '%P\n'
+  | sed 's|/.git||' 
+  | xargs -I {} sh -c '
+  grep -q {} .gitmodules 
+  ||  (echo "Removing {}, has .git directory but not in .gitmodules file" 
+  && rm -rf {});'
+
 # before each job, we need to check if this job is filtered by bot stage/job filter
 .apply_bot_filter: &apply_bot_filter
   python $APPLY_BOT_FILTER_SCRIPT || exit 0
 
 before_script:
+  - *git_clean_stale_submodules
   # apply bot filter in before script
   - *apply_bot_filter
   # add gitlab ssh key
@@ -56,6 +70,7 @@ before_script:
 
 .do_nothing_before:
   before_script: &do_nothing_before
+    - *git_clean_stale_submodules
     # apply bot filter in before script
     - *apply_bot_filter
     - echo "Not setting up GitLab key, not fetching submodules"
@@ -63,6 +78,7 @@ before_script:
 
 .add_gitlab_key_before:
   before_script: &add_gitlab_key_before
+    - *git_clean_stale_submodules
     # apply bot filter in before script
     - *apply_bot_filter
     - echo "Not fetching submodules"