Prepare a test for multiple test contacts

Author: seberm

tests/report/reportportal/data/test.fmf

@@ -1,11 +1,13 @@
 /bad:
     summary: Failing test
-    contact: [email protected]
+    contact:
+      - [email protected]
+      - [email protected]
     test: echo "Something bad happened!"; false
 
 /good:
     summary: Passing test
-    contact: tester_2@redhat.com
+    contact: tester3@redhat.com
     test: echo "Everything's fine!"
     id: 63f26fb7-69c4-4781-a06e-098e2b58129f
 

tests/report/reportportal/test.sh

@@ -191,6 +191,14 @@ rlJournalStart
                     key="contact"
                     value="$(yq -r ".\"$test_name\".$key" test.fmf)"
                     if [[ $value != null ]]; then
+                        # TODO: Fix the test here! The output of yq is list of
+                        # contact in the case of multiple contacts (test==/bad)
+                        # and a string in a case when one contact is provided
+                        # for the test (test==/good). We must somehow handle
+                        # both situations.
+                        #
+                        # If yq returns list, we must check both contacts in the response JSON
+                        # If yq returns just a string value, we must check only one contact in the response JSON
                         rlAssertGrep "$key" tmp_attributes.json -A1 > tmp_attributes_selection
                         rlAssertGrep "$value" tmp_attributes_selection
                     else
@@ -201,6 +209,10 @@ rlJournalStart
                     rlAssertNotGrep "$key" tmp_attributes.json
                 fi
 
+                # REMOVE ME: just for testing
+                #echo "EXITING OMG"
+                #exit 99
+
                 rm tmp_attributes*
             done