Improve SKIPS handling in regression tests

This patch changes the regression test schedule generation to only
completely rebuild the schedule when TESTS is specified. In all
other cases the existing schedule is amended. This means any
parallelity present in the cmake generated schedule will be kept.

test/pg_regress.sh

@@ -1,5 +1,8 @@
 #!/usr/bin/env bash
 
+# shellcheck disable=SC2053
+# shellcheck disable=SC2235
+
 # Wrapper around pg_regress and pg_isolation_regress to be able to control the schedule with environment variables
 #
 # The following control variables are supported:
@@ -35,8 +38,6 @@ PSQL="${PSQL} -X" # Prevent any .psqlrc files from being executed during the tes
 # all bgw tests would be "*bgw*" while with =~ it would be ".*bgw.*"
 matches() {
   for pattern in $1; do
-    # shellcheck disable=SC2053
-    # We do want to match globs in $pattern here.
     if [[ $2 == $pattern ]]; then
       return 0
     fi
@@ -59,11 +60,11 @@ if [[ -z ${TESTS} ]] && [[ -z ${SKIPS} ]] && [[ -z ${IGNORES} ]]; then
 
   SCHEDULE=${TEST_SCHEDULE}
 
-elif [[ -z ${TESTS} ]] && [[ -z ${SKIPS} ]] && [[ -n ${IGNORES} ]]; then
-  # if we only have IGNORES we can use the cmake created schedule
+elif [[ -z ${TESTS} ]] && ( [[ -n ${SKIPS} ]] || [[ -n ${IGNORES} ]] ); then
+  # If we only have IGNORES or SKIPS we can use the cmake created schedule
   # and just prepend ignore lines for the tests whose result should be
-  # ignored. This will allow us to retain the parallel groupings from
-  # the supplied schedule
+  # ignored and strip out the skipped tests. This will allow us to retain
+  # the parallel groupings from the supplied schedule.
 
   echo > ${TEMP_SCHEDULE}
 
@@ -71,23 +72,35 @@ elif [[ -z ${TESTS} ]] && [[ -z ${SKIPS} ]] && [[ -n ${IGNORES} ]]; then
 
   # to support wildcards in IGNORES we match the IGNORES
   # list against the actual list of tests
-  for test_pattern in ${IGNORES}; do
-    for test_name in ${ALL_TESTS}; do
-      # shellcheck disable=SC2053
-      # We do want to match globs in $test_pattern here.
-      if [[ $test_name == $test_pattern ]]; then
-        echo "ignore: ${test_name}" >> ${TEMP_SCHEDULE}
-      fi
+  if [[ -n ${IGNORES} ]]; then
+    for test_pattern in ${IGNORES}; do
+      for test_name in ${ALL_TESTS}; do
+        if [[ $test_name == $test_pattern ]]; then
+          echo "ignore: ${test_name}" >> ${TEMP_SCHEDULE}
+        fi
+      done
     done
-  done
+  fi
 
   cat ${TEST_SCHEDULE} >> ${TEMP_SCHEDULE}
 
+  # to support wildcards in SKIPS we match the SKIPS
+  # list against the actual list of tests
+  if [[ -n ${SKIPS} ]]; then
+    for test_pattern in ${SKIPS}; do
+      for test_name in ${ALL_TESTS}; do
+        if [[ $test_name == $test_pattern ]]; then
+          sed -i ${TEMP_SCHEDULE} -e "s!^test:\s*${test_name}\s*\$!!"
+          sed -i ${TEMP_SCHEDULE} -e "s!\b${test_name}\b!!"
+        fi
+      done
+    done
+  fi
+
   SCHEDULE=${TEMP_SCHEDULE}
 
 else
-  # either TESTS or SKIPS was specified so we need to create a new schedule
-  # based on those
+  # TESTS was specified so we need to create a new schedule based on that
 
   ALL_TESTS=$(grep '^test: ' ${TEST_SCHEDULE} | sed -e 's!^test: !!' |tr '\n' ' ')
 
@@ -102,8 +115,6 @@ else
   for test_pattern in ${TESTS}; do
     for test_name in ${ALL_TESTS}; do
       if ! matches "${SKIPS}" "${test_name}"; then
-        # shellcheck disable=SC2053
-        # We do want to match globs in $test_pattern here.
         if [[ $test_name == $test_pattern ]]; then
           current_tests="${current_tests} ${test_name}"
         fi
@@ -127,8 +138,6 @@ else
   for test_pattern in ${IGNORES}; do
     for test_name in ${ALL_TESTS}; do
       if ! matches "${SKIPS}" "${test_name}"; then
-        # shellcheck disable=SC2053
-        # We do want to match globs in $test_pattern here.
         if [[ $test_name == $test_pattern ]]; then
           echo "ignore: ${test_name}" >> ${TEMP_SCHEDULE}
         fi