timescaledb/test/pg_regress.sh
Fabrízio de Royes Mello fbc1fbb64b PG16: Enable CI
This commit enables PG16 in the following workflows:
* Regression Linux (including 32bit)
* ABI test
* Memory tests
* Sanitizer tests
* Coverity
2023-11-04 13:29:02 -03:00

216 lines
7.0 KiB
Bash
Executable File

#!/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:
#
# TESTS only run tests from this list
# IGNORES failure of tests in this list will not lead to test failure
# SKIPS tests from this list are not run
#
# In TESTS you may use wildcards to match multiple test names
# TESTS="compression*" will match all tests whose name starts with compression
# TESTS="*compression*" will match all tests that have compression anywhere in the name
# Wildcard matching also applies to version specific tests so compression-13
# would also be matched by those patterns.
CURRENT_DIR=$(dirname $0)
EXE_DIR=${EXE_DIR:-${CURRENT_DIR}}
PG_REGRESS=${PG_REGRESS:-pg_regress}
PG_REGRESS_DIFF_OPTS=-u
TEST_SCHEDULE=${TEST_SCHEDULE:-}
TEMP_SCHEDULE=${CURRENT_DIR}/temp_schedule
SCHEDULE=
TESTS=${TESTS:-}
IGNORES=${IGNORES:-}
SKIPS=${SKIPS:-}
PSQL=${PSQL:-psql}
PSQL="${PSQL} -X" # Prevent any .psqlrc files from being executed during the tests
PG_VERSION_MAJOR=$(${PSQL} --version | awk '{split($3,v,"."); print v[1]}')
# check if test matches any of the patterns in a list
# $1 list of patterns or test names
# $2 test name
# we use == intentionally and not =~ because the pattern syntax differs between
# those two and == allows for simpler patterns. With == the pattern to match
# all bgw tests would be "*bgw*" while with =~ it would be ".*bgw.*"
matches() {
for pattern in $1; do
if [[ $2 == $pattern ]]; then
return 0
fi
done
return 1
}
if [[ -z ${TEST_SCHEDULE} ]]; then
echo "No test schedule supplied please set TEST_SCHEDULE"
exit 1;
fi
# PG16 removed the `ignore` feature from `pg_regress`
# so as an wraparound if we have any IGNORES entry then
# we merge it together with SKIPS and cleanup the IGNORES
# https://github.com/postgres/postgres/commit/bd8d453e9b5f8b632a400a9e796fc041aed76d82
if [[ ${PG_VERSION_MAJOR} -ge 16 ]]; then
if [[ -n ${IGNORES} ]]; then
if [[ -n ${SKIPS} ]]; then
SKIPS="${SKIPS} ${IGNORES}"
else
SKIPS="${IGNORES}"
fi
IGNORES=""
fi
fi
echo "TESTS ${TESTS}"
if [[ ${PG_VERSION_MAJOR} -lt 16 ]]; then
echo "IGNORES ${IGNORES}"
fi
echo "SKIPS ${SKIPS}"
if [[ -z ${TESTS} ]] && [[ -z ${SKIPS} ]] && [[ -z ${IGNORES} ]]; then
# no filter variables set
# nothing to do here and we can use the cmake generated schedule
SCHEDULE=${TEST_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 and strip out the skipped tests. This will allow us to retain
# the parallel groupings from the supplied schedule.
echo > ${TEMP_SCHEDULE}
ALL_TESTS=$(grep '^test: ' ${TEST_SCHEDULE} | sed -e 's!^test: !!' |tr '\n' ' ')
# to support wildcards in IGNORES we match the IGNORES
# list against the actual list of tests
if [[ -n ${IGNORES} ]]; then
for test_pattern in ${IGNORES}; do
for test_name in ${ALL_TESTS}; do
if [[ -n ${test_name} ]] && [[ $test_name == $test_pattern ]]; then
echo "ignore: ${test_name}" >> ${TEMP_SCHEDULE}
fi
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 == $test_pattern ]]; then
sed -e "s!^test:\s*${test_name}\s*\$!!" -i.backup ${TEMP_SCHEDULE}
sed -e "s!\b${test_name}\b!!" -i.backup ${TEMP_SCHEDULE}
fi
done
done
fi
SCHEDULE=${TEMP_SCHEDULE}
else
# TESTS was specified so we need to create a new schedule based on that
ALL_TESTS=$(grep '^test: ' ${TEST_SCHEDULE} | sed -e 's!^test: !!' |tr '\n' ' ')
if [[ -z "${TESTS}" ]]; then
TESTS=${ALL_TESTS}
fi
# build new test list in current_tests removing entries in SKIPS and
# validating against schedule as TESTS might contain tests from
# multiple suites and not apply to current run
current_tests=""
for test_pattern in ${TESTS}; do
for test_name in ${ALL_TESTS}; do
if ! matches "${SKIPS}" "${test_name}"; then
if [[ $test_name == $test_pattern ]]; then
current_tests="${current_tests} ${test_name}"
fi
fi
done
done
# if none of the tests survived filtering we can exit early
if [[ -z "${current_tests}" ]]; then
exit 0
fi
current_tests=$(echo "${current_tests}" | tr ' ' '\n' | sort)
TESTS=${current_tests}
echo > ${TEMP_SCHEDULE}
# 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 ${ALL_TESTS}; do
if ! matches "${SKIPS}" "${test_name}"; then
if [[ $test_name == $test_pattern ]]; then
echo "ignore: ${test_name}" >> ${TEMP_SCHEDULE}
fi
fi
done
done
for t in ${TESTS}; do
echo "test: ${t}" >> ${TEMP_SCHEDULE}
done
SCHEDULE=${TEMP_SCHEDULE}
fi
function cleanup() {
rm -rf ${EXE_DIR}/sql/dump
rm -rf ${TEST_TABLESPACE1_PREFIX}
rm -rf ${TEST_TABLESPACE2_PREFIX}
rm -rf ${TEST_TABLESPACE3_PREFIX}
rm -f ${TEMP_SCHEDULE}
cat <<EOF | ${PSQL} -U ${USER} -h ${TEST_PGHOST} -p ${TEST_PGPORT} -d template1 >/dev/null 2>&1
DROP TABLESPACE IF EXISTS tablespace1;
DROP TABLESPACE IF EXISTS tablespace2;
DROP TABLESPACE IF EXISTS tablespace3;
EOF
rm -rf ${TEST_OUTPUT_DIR}/.pg_init
}
trap cleanup EXIT
# Generating a prefix directory for all test tablespaces. This should
# be used to build a full path for the tablespace. Note that we
# terminate the prefix with the directory separator so that we can
# easily generate paths independent of the OS.
#
# This mktemp line will work on both OSX and GNU systems
TEST_TABLESPACE1_PREFIX=${TEST_TABLESPACE1_PREFIX:-$(mktemp -d 2>/dev/null || mktemp -d -t 'timescaledb_regress')/}
TEST_TABLESPACE2_PREFIX=${TEST_TABLESPACE2_PREFIX:-$(mktemp -d 2>/dev/null || mktemp -d -t 'timescaledb_regress')/}
TEST_TABLESPACE3_PREFIX=${TEST_TABLESPACE3_PREFIX:-$(mktemp -d 2>/dev/null || mktemp -d -t 'timescaledb_regress')/}
# Creating some defaults for transitioning tests to use the prefix.
TEST_TABLESPACE1_PATH=${TEST_TABLESPACE1_PATH:-${TEST_TABLESPACE1_PREFIX}_default}
TEST_TABLESPACE2_PATH=${TEST_TABLESPACE2_PATH:-${TEST_TABLESPACE2_PREFIX}_default}
TEST_TABLESPACE3_PATH=${TEST_TABLESPACE3_PATH:-${TEST_TABLESPACE3_PREFIX}_default}
mkdir -p $TEST_TABLESPACE1_PATH $TEST_TABLESPACE2_PATH $TEST_TABLESPACE3_PATH
export TEST_TABLESPACE1_PREFIX TEST_TABLESPACE2_PREFIX TEST_TABLESPACE3_PREFIX
export TEST_TABLESPACE1_PATH TEST_TABLESPACE2_PATH TEST_TABLESPACE3_PATH
rm -rf ${TEST_OUTPUT_DIR}/.pg_init
mkdir -p ${EXE_DIR}/sql/dump
export PG_REGRESS_DIFF_OPTS
PG_REGRESS_OPTS="${PG_REGRESS_OPTS} --schedule=${SCHEDULE}"
${PG_REGRESS} "$@" ${PG_REGRESS_OPTS}