caching/vendor/knative.dev/test-infra/scripts/library.sh

676 lines
24 KiB
Bash

#!/usr/bin/env bash
# Copyright 2018 The Knative Authors
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# This is a collection of useful bash functions and constants, intended
# to be used in test scripts and the like. It doesn't do anything when
# called from command line.
# GCP project where all tests related resources live
readonly KNATIVE_TESTS_PROJECT=knative-tests
# Default GKE version to be used with Knative Serving
readonly SERVING_GKE_VERSION=gke-latest
readonly SERVING_GKE_IMAGE=cos
# Conveniently set GOPATH if unset
if [[ -z "${GOPATH:-}" ]]; then
export GOPATH="$(go env GOPATH)"
if [[ -z "${GOPATH}" ]]; then
echo "WARNING: GOPATH not set and go binary unable to provide it"
fi
fi
# Useful environment variables
[[ -n "${PROW_JOB_ID:-}" ]] && IS_PROW=1 || IS_PROW=0
readonly IS_PROW
[[ -z "${REPO_ROOT_DIR:-}" ]] && REPO_ROOT_DIR="$(git rev-parse --show-toplevel)"
readonly REPO_ROOT_DIR
readonly REPO_NAME="$(basename ${REPO_ROOT_DIR})"
# Useful flags about the current OS
IS_LINUX=0
IS_OSX=0
IS_WINDOWS=0
case "${OSTYPE}" in
darwin*) IS_OSX=1 ;;
linux*) IS_LINUX=1 ;;
msys*) IS_WINDOWS=1 ;;
*) echo "** Internal error in library.sh, unknown OS '${OSTYPE}'" ; exit 1 ;;
esac
readonly IS_LINUX
readonly IS_OSX
readonly IS_WINDOWS
# Set ARTIFACTS to an empty temp dir if unset
if [[ -z "${ARTIFACTS:-}" ]]; then
export ARTIFACTS="$(mktemp -d)"
fi
# On a Prow job, redirect stderr to stdout so it's synchronously added to log
(( IS_PROW )) && exec 2>&1
# Print error message and exit 1
# Parameters: $1..$n - error message to be displayed
function abort() {
echo "error: $@"
exit 1
}
# Display a box banner.
# Parameters: $1 - character to use for the box.
# $2 - banner message.
function make_banner() {
local msg="$1$1$1$1 $2 $1$1$1$1"
local border="${msg//[-0-9A-Za-z _.,\/()\']/$1}"
echo -e "${border}\n${msg}\n${border}"
# TODO(adrcunha): Remove once logs have timestamps on Prow
# For details, see https://github.com/kubernetes/test-infra/issues/10100
echo -e "$1$1$1$1 $(TZ='America/Los_Angeles' date)\n${border}"
}
# Simple header for logging purposes.
function header() {
local upper="$(echo $1 | tr a-z A-Z)"
make_banner "=" "${upper}"
}
# Simple subheader for logging purposes.
function subheader() {
make_banner "-" "$1"
}
# Simple warning banner for logging purposes.
function warning() {
make_banner "!" "$1"
}
# Checks whether the given function exists.
function function_exists() {
[[ "$(type -t $1)" == "function" ]]
}
# Waits until the given object doesn't exist.
# Parameters: $1 - the kind of the object.
# $2 - object's name.
# $3 - namespace (optional).
function wait_until_object_does_not_exist() {
local KUBECTL_ARGS="get $1 $2"
local DESCRIPTION="$1 $2"
if [[ -n $3 ]]; then
KUBECTL_ARGS="get -n $3 $1 $2"
DESCRIPTION="$1 $3/$2"
fi
echo -n "Waiting until ${DESCRIPTION} does not exist"
for i in {1..150}; do # timeout after 5 minutes
if ! kubectl ${KUBECTL_ARGS} > /dev/null 2>&1; then
echo -e "\n${DESCRIPTION} does not exist"
return 0
fi
echo -n "."
sleep 2
done
echo -e "\n\nERROR: timeout waiting for ${DESCRIPTION} not to exist"
kubectl ${KUBECTL_ARGS}
return 1
}
# Waits until all pods are running in the given namespace.
# Parameters: $1 - namespace.
function wait_until_pods_running() {
echo -n "Waiting until all pods in namespace $1 are up"
local failed_pod=""
for i in {1..150}; do # timeout after 5 minutes
local pods="$(kubectl get pods --no-headers -n $1 2>/dev/null)"
# All pods must be running (ignore ImagePull error to allow the pod to retry)
local not_running_pods=$(echo "${pods}" | grep -v Running | grep -v Completed | grep -v ErrImagePull | grep -v ImagePullBackOff)
if [[ -n "${pods}" ]] && [[ -z "${not_running_pods}" ]]; then
# All Pods are running or completed. Verify the containers on each Pod.
local all_ready=1
while read pod ; do
local status=(`echo -n ${pod} | cut -f2 -d' ' | tr '/' ' '`)
# Set this Pod as the failed_pod. If nothing is wrong with it, then after the checks, set
# failed_pod to the empty string.
failed_pod=$(echo -n "${pod}" | cut -f1 -d' ')
# All containers must be ready
[[ -z ${status[0]} ]] && all_ready=0 && break
[[ -z ${status[1]} ]] && all_ready=0 && break
[[ ${status[0]} -lt 1 ]] && all_ready=0 && break
[[ ${status[1]} -lt 1 ]] && all_ready=0 && break
[[ ${status[0]} -ne ${status[1]} ]] && all_ready=0 && break
# All the tests passed, this is not a failed pod.
failed_pod=""
done <<< "$(echo "${pods}" | grep -v Completed)"
if (( all_ready )); then
echo -e "\nAll pods are up:\n${pods}"
return 0
fi
elif [[ -n "${not_running_pods}" ]]; then
# At least one Pod is not running, just save the first one's name as the failed_pod.
failed_pod="$(echo "${not_running_pods}" | head -n 1 | cut -f1 -d' ')"
fi
echo -n "."
sleep 2
done
echo -e "\n\nERROR: timeout waiting for pods to come up\n${pods}"
if [[ -n "${failed_pod}" ]]; then
echo -e "\n\nFailed Pod (data in YAML format) - ${failed_pod}\n"
kubectl -n $1 get pods "${failed_pod}" -oyaml
echo -e "\n\nPod Logs\n"
kubectl -n $1 logs "${failed_pod}" --all-containers
fi
return 1
}
# Waits until all batch jobs complete in the given namespace.
# Parameters: $1 - namespace.
function wait_until_batch_job_complete() {
echo -n "Waiting until all batch jobs in namespace $1 run to completion."
for i in {1..150}; do # timeout after 5 minutes
local jobs=$(kubectl get jobs -n $1 --no-headers \
-ocustom-columns='n:{.metadata.name},c:{.spec.completions},s:{.status.succeeded}')
# All jobs must be complete
local not_complete=$(echo "${jobs}" | awk '{if ($2!=$3) print $0}' | wc -l)
if [[ ${not_complete} -eq 0 ]]; then
echo -e "\nAll jobs are complete:\n${jobs}"
return 0
fi
echo -n "."
sleep 2
done
echo -e "\n\nERROR: timeout waiting for jobs to complete\n${jobs}"
return 1
}
# Waits until the given service has an external address (IP/hostname).
# Parameters: $1 - namespace.
# $2 - service name.
function wait_until_service_has_external_ip() {
echo -n "Waiting until service $2 in namespace $1 has an external address (IP/hostname)"
for i in {1..150}; do # timeout after 15 minutes
local ip=$(kubectl get svc -n $1 $2 -o jsonpath="{.status.loadBalancer.ingress[0].ip}")
if [[ -n "${ip}" ]]; then
echo -e "\nService $2.$1 has IP $ip"
return 0
fi
local hostname=$(kubectl get svc -n $1 $2 -o jsonpath="{.status.loadBalancer.ingress[0].hostname}")
if [[ -n "${hostname}" ]]; then
echo -e "\nService $2.$1 has hostname $hostname"
return 0
fi
echo -n "."
sleep 6
done
echo -e "\n\nERROR: timeout waiting for service $2.$1 to have an external address"
kubectl get pods -n $1
return 1
}
# Waits for the endpoint to be routable.
# Parameters: $1 - External ingress IP address.
# $2 - cluster hostname.
function wait_until_routable() {
echo -n "Waiting until cluster $2 at $1 has a routable endpoint"
for i in {1..150}; do # timeout after 5 minutes
local val=$(curl -H "Host: $2" "http://$1" 2>/dev/null)
if [[ -n "$val" ]]; then
echo -e "\nEndpoint is now routable"
return 0
fi
echo -n "."
sleep 2
done
echo -e "\n\nERROR: Timed out waiting for endpoint to be routable"
return 1
}
# Returns the name of the first pod of the given app.
# Parameters: $1 - app name.
# $2 - namespace (optional).
function get_app_pod() {
local pods=($(get_app_pods $1 $2))
echo "${pods[0]}"
}
# Returns the name of all pods of the given app.
# Parameters: $1 - app name.
# $2 - namespace (optional).
function get_app_pods() {
local namespace=""
[[ -n $2 ]] && namespace="-n $2"
kubectl get pods ${namespace} --selector=app=$1 --output=jsonpath="{.items[*].metadata.name}"
}
# Capitalize the first letter of each word.
# Parameters: $1..$n - words to capitalize.
function capitalize() {
local capitalized=()
for word in $@; do
local initial="$(echo ${word:0:1}| tr 'a-z' 'A-Z')"
capitalized+=("${initial}${word:1}")
done
echo "${capitalized[@]}"
}
# Dumps pod logs for the given app.
# Parameters: $1 - app name.
# $2 - namespace.
function dump_app_logs() {
echo ">>> ${REPO_NAME_FORMATTED} $1 logs:"
for pod in $(get_app_pods "$1" "$2")
do
echo ">>> Pod: $pod"
kubectl -n "$2" logs "$pod" --all-containers
done
}
# Sets the given user as cluster admin.
# Parameters: $1 - user
# $2 - cluster name
# $3 - cluster region
# $4 - cluster zone, optional
function acquire_cluster_admin_role() {
echo "Acquiring cluster-admin role for user '$1'"
local geoflag="--region=$3"
[[ -n $4 ]] && geoflag="--zone=$3-$4"
# Get the password of the admin and use it, as the service account (or the user)
# might not have the necessary permission.
local password=$(gcloud --format="value(masterAuth.password)" \
container clusters describe $2 ${geoflag})
if [[ -n "${password}" ]]; then
# Cluster created with basic authentication
kubectl config set-credentials cluster-admin \
--username=admin --password=${password}
else
local cert=$(mktemp)
local key=$(mktemp)
echo "Certificate in ${cert}, key in ${key}"
gcloud --format="value(masterAuth.clientCertificate)" \
container clusters describe $2 ${geoflag} | base64 --decode > ${cert}
gcloud --format="value(masterAuth.clientKey)" \
container clusters describe $2 ${geoflag} | base64 --decode > ${key}
kubectl config set-credentials cluster-admin \
--client-certificate=${cert} --client-key=${key}
fi
kubectl config set-context $(kubectl config current-context) \
--user=cluster-admin
kubectl create clusterrolebinding cluster-admin-binding \
--clusterrole=cluster-admin \
--user=$1
# Reset back to the default account
gcloud container clusters get-credentials \
$2 ${geoflag} --project $(gcloud config get-value project)
}
# Run a command through tee and capture its output.
# Parameters: $1 - file where the output will be stored.
# $2... - command to run.
function capture_output() {
local report="$1"
shift
"$@" 2>&1 | tee "${report}"
local failed=( ${PIPESTATUS[@]} )
[[ ${failed[0]} -eq 0 ]] && failed=${failed[1]} || failed=${failed[0]}
return ${failed}
}
# Print failed step, which could be highlighted by spyglass.
# Parameters: $1...n - description of step that failed
function step_failed() {
local spyglass_token="Step failed:"
echo "${spyglass_token} $@"
}
# Create a temporary file with the given extension in a way that works on both Linux and macOS.
# Parameters: $1 - file name without extension (e.g. 'myfile_XXXX')
# $2 - file extension (e.g. 'xml')
function mktemp_with_extension() {
local nameprefix
local fullname
nameprefix="$(mktemp $1)"
fullname="${nameprefix}.$2"
mv ${nameprefix} ${fullname}
echo ${fullname}
}
# Create a JUnit XML for a test.
# Parameters: $1 - check class name as an identifier (e.g. BuildTests)
# $2 - check name as an identifier (e.g., GoBuild)
# $3 - failure message (can contain newlines), optional (means success)
function create_junit_xml() {
local xml="$(mktemp_with_extension ${ARTIFACTS}/junit_XXXXXXXX xml)"
local failure=""
if [[ "$3" != "" ]]; then
# Transform newlines into HTML code.
# Also escape `<` and `>` as here: https://github.com/golang/go/blob/50bd1c4d4eb4fac8ddeb5f063c099daccfb71b26/src/encoding/json/encode.go#L48,
# this is temporary solution for fixing https://github.com/knative/test-infra/issues/1204,
# which should be obsolete once Test-infra 2.0 is in place
local msg="$(echo -n "$3" | sed 's/$/\&#xA;/g' | sed 's/</\\u003c/' | sed 's/>/\\u003e/' | sed 's/&/\\u0026/' | tr -d '\n')"
failure="<failure message=\"Failed\" type=\"\">${msg}</failure>"
fi
cat << EOF > "${xml}"
<testsuites>
<testsuite tests="1" failures="1" time="0.000" name="$1">
<testcase classname="" name="$2" time="0.0">
${failure}
</testcase>
</testsuite>
</testsuites>
EOF
}
# Runs a go test and generate a junit summary.
# Parameters: $1... - parameters to go test
function report_go_test() {
# Run tests in verbose mode to capture details.
# go doesn't like repeating -v, so remove if passed.
local args=" $@ "
local go_test="go test -v ${args/ -v / }"
# Just run regular go tests if not on Prow.
echo "Running tests with '${go_test}'"
local report="$(mktemp)"
capture_output "${report}" ${go_test}
local failed=$?
echo "Finished run, return code is ${failed}"
# Install go-junit-report if necessary.
run_go_tool github.com/jstemmer/go-junit-report go-junit-report --help > /dev/null 2>&1
local xml="$(mktemp_with_extension ${ARTIFACTS}/junit_XXXXXXXX xml)"
cat ${report} \
| go-junit-report \
| sed -e "s#\"\(github\.com/knative\|knative\.dev\)/${REPO_NAME}/#\"#g" \
> ${xml}
echo "XML report written to ${xml}"
if [[ -n "$(grep '<testsuites></testsuites>' ${xml})" ]]; then
# XML report is empty, something's wrong; use the output as failure reason
create_junit_xml _go_tests "GoTests" "$(cat ${report})"
fi
# Capture and report any race condition errors
local race_errors="$(sed -n '/^WARNING: DATA RACE$/,/^==================$/p' ${report})"
create_junit_xml _go_tests "DataRaceAnalysis" "${race_errors}"
if (( ! IS_PROW )); then
# Keep the suffix, so files are related.
local logfile=${xml/junit_/go_test_}
logfile=${logfile/.xml/.log}
cp ${report} ${logfile}
echo "Test log written to ${logfile}"
fi
return ${failed}
}
# Install Knative Serving in the current cluster.
# Parameters: $1 - Knative Serving manifest.
function start_knative_serving() {
header "Starting Knative Serving"
subheader "Installing Knative Serving"
echo "Installing Serving CRDs from $1"
kubectl apply --selector knative.dev/crd-install=true -f "$1"
echo "Installing the rest of serving components from $1"
kubectl apply -f "$1"
wait_until_pods_running knative-serving || return 1
}
# Install Knative Monitoring in the current cluster.
# Parameters: $1 - Knative Monitoring manifest.
function start_knative_monitoring() {
header "Starting Knative Monitoring"
subheader "Installing Knative Monitoring"
# namespace istio-system needs to be created first, due to the comment
# mentioned in
# https://github.com/knative/serving/blob/4202efc0dc12052edc0630515b101cbf8068a609/config/monitoring/tracing/zipkin/100-zipkin.yaml#L21
kubectl create namespace istio-system 2>/dev/null
echo "Installing Monitoring from $1"
kubectl apply -f "$1" || return 1
wait_until_pods_running knative-monitoring || return 1
wait_until_pods_running istio-system || return 1
}
# Install the stable release Knative/serving in the current cluster.
# Parameters: $1 - Knative Serving version number, e.g. 0.6.0.
function start_release_knative_serving() {
start_knative_serving "https://storage.googleapis.com/knative-releases/serving/previous/v$1/serving.yaml"
}
# Install the latest stable Knative Serving in the current cluster.
function start_latest_knative_serving() {
start_knative_serving "${KNATIVE_SERVING_RELEASE}"
}
# Install Knative Eventing in the current cluster.
# Parameters: $1 - Knative Eventing manifest.
function start_knative_eventing() {
header "Starting Knative Eventing"
subheader "Installing Knative Eventing"
echo "Installing Eventing CRDs from $1"
kubectl apply --selector knative.dev/crd-install=true -f "$1"
echo "Installing the rest of eventing components from $1"
kubectl apply -f "$1"
wait_until_pods_running knative-eventing || return 1
}
# Install the stable release Knative/eventing in the current cluster.
# Parameters: $1 - Knative Eventing version number, e.g. 0.6.0.
function start_release_knative_eventing() {
start_knative_eventing "https://storage.googleapis.com/knative-releases/eventing/previous/v$1/eventing.yaml"
}
# Install the latest stable Knative Eventing in the current cluster.
function start_latest_knative_eventing() {
start_knative_eventing "${KNATIVE_EVENTING_RELEASE}"
}
# Run a go tool, installing it first if necessary.
# Parameters: $1 - tool package/dir for go get/install.
# $2 - tool to run.
# $3..$n - parameters passed to the tool.
function run_go_tool() {
local tool=$2
local action=get
[[ $1 =~ ^[\./].* ]] && action=install
# Avoid running `go get` from root dir of the repository, as it can change go.sum and go.mod files.
# See discussions in https://github.com/golang/go/issues/27643.
if [[ -z "$(which ${tool})" ]]; then
local action=get
[[ $1 =~ ^[\./].* ]] && action=install
# Avoid running `go get` from root dir of the repository, as it can change go.sum and go.mod files.
# See discussions in https://github.com/golang/go/issues/27643.
if [[ ${action} == "get" && $(pwd) == "${REPO_ROOT_DIR}" ]]; then
local temp_dir="$(mktemp -d)"
local install_failed=0
# Swallow the output as we are returning the stdout in the end.
pushd "${temp_dir}" > /dev/null 2>&1
GOFLAGS="" go ${action} $1 || install_failed=1
popd > /dev/null 2>&1
(( install_failed )) && return ${install_failed}
else
GOFLAGS="" go ${action} $1
fi
fi
shift 2
${tool} "$@"
}
# Run go-licenses to update licenses.
# Parameters: $1 - output file, relative to repo root dir.
# $2 - directory to inspect.
function update_licenses() {
cd "${REPO_ROOT_DIR}" || return 1
local dst=$1
local dir=$2
shift
run_go_tool github.com/google/go-licenses go-licenses save "${dir}" --save_path="${dst}" --force || return 1
# Hack to make sure directories retain write permissions after save. This
# can happen if the directory being copied is a Go module.
# See https://github.com/google/go-licenses/issues/11
chmod -R +w "${dst}"
}
# Run go-licenses to check for forbidden licenses.
function check_licenses() {
# Check that we don't have any forbidden licenses.
run_go_tool github.com/google/go-licenses go-licenses check "${REPO_ROOT_DIR}/..." || return 1
}
# Run the given linter on the given files, checking it exists first.
# Parameters: $1 - tool
# $2 - tool purpose (for error message if tool not installed)
# $3 - tool parameters (quote if multiple parameters used)
# $4..$n - files to run linter on
function run_lint_tool() {
local checker=$1
local params=$3
if ! hash ${checker} 2>/dev/null; then
warning "${checker} not installed, not $2"
return 127
fi
shift 3
local failed=0
for file in $@; do
${checker} ${params} ${file} || failed=1
done
return ${failed}
}
# Check links in the given markdown files.
# Parameters: $1...$n - files to inspect
function check_links_in_markdown() {
# https://github.com/raviqqe/liche
local config="${REPO_ROOT_DIR}/test/markdown-link-check-config.rc"
[[ ! -e ${config} ]] && config="${_TEST_INFRA_SCRIPTS_DIR}/markdown-link-check-config.rc"
local options="$(grep '^-' ${config} | tr \"\n\" ' ')"
run_lint_tool liche "checking links in markdown files" "-d ${REPO_ROOT_DIR} ${options}" $@
}
# Check format of the given markdown files.
# Parameters: $1..$n - files to inspect
function lint_markdown() {
# https://github.com/markdownlint/markdownlint
local config="${REPO_ROOT_DIR}/test/markdown-lint-config.rc"
[[ ! -e ${config} ]] && config="${_TEST_INFRA_SCRIPTS_DIR}/markdown-lint-config.rc"
run_lint_tool mdl "linting markdown files" "-c ${config}" $@
}
# Return whether the given parameter is an integer.
# Parameters: $1 - integer to check
function is_int() {
[[ -n $1 && $1 =~ ^[0-9]+$ ]]
}
# Return whether the given parameter is the knative release/nightly GCF.
# Parameters: $1 - full GCR name, e.g. gcr.io/knative-foo-bar
function is_protected_gcr() {
[[ -n $1 && $1 =~ ^gcr.io/knative-(releases|nightly)/?$ ]]
}
# Return whether the given parameter is any cluster under ${KNATIVE_TESTS_PROJECT}.
# Parameters: $1 - Kubernetes cluster context (output of kubectl config current-context)
function is_protected_cluster() {
# Example: gke_knative-tests_us-central1-f_prow
[[ -n $1 && $1 =~ ^gke_${KNATIVE_TESTS_PROJECT}_us\-[a-zA-Z0-9]+\-[a-z]+_[a-z0-9\-]+$ ]]
}
# Return whether the given parameter is ${KNATIVE_TESTS_PROJECT}.
# Parameters: $1 - project name
function is_protected_project() {
[[ -n $1 && "$1" == "${KNATIVE_TESTS_PROJECT}" ]]
}
# Remove symlinks in a path that are broken or lead outside the repo.
# Parameters: $1 - path name, e.g. vendor
function remove_broken_symlinks() {
for link in $(find $1 -type l); do
# Remove broken symlinks
if [[ ! -e ${link} ]]; then
unlink ${link}
continue
fi
# Get canonical path to target, remove if outside the repo
local target="$(ls -l ${link})"
target="${target##* -> }"
[[ ${target} == /* ]] || target="./${target}"
target="$(cd `dirname ${link}` && cd ${target%/*} && echo $PWD/${target##*/})"
if [[ ${target} != *github.com/knative/* && ${target} != *knative.dev/* ]]; then
unlink ${link}
continue
fi
done
}
# Returns the canonical path of a filesystem object.
# Parameters: $1 - path to return in canonical form
# $2 - base dir for relative links; optional, defaults to current
function get_canonical_path() {
# We don't use readlink because it's not available on every platform.
local path=$1
local pwd=${2:-.}
[[ ${path} == /* ]] || path="${pwd}/${path}"
echo "$(cd ${path%/*} && echo $PWD/${path##*/})"
}
# Returns the current branch.
function current_branch() {
local branch_name=""
# Get the branch name from Prow's env var, see https://github.com/kubernetes/test-infra/blob/master/prow/jobs.md.
# Otherwise, try getting the current branch from git.
(( IS_PROW )) && branch_name="${PULL_BASE_REF:-}"
[[ -z "${branch_name}" ]] && branch_name="$(git rev-parse --abbrev-ref HEAD)"
echo "${branch_name}"
}
# Returns whether the current branch is a release branch.
function is_release_branch() {
[[ $(current_branch) =~ ^release-[0-9\.]+$ ]]
}
# Returns the URL to the latest manifest for the given Knative project.
# Parameters: $1 - repository name of the given project
# $2 - name of the yaml file, without extension
function get_latest_knative_yaml_source() {
local repo_name="$1"
local yaml_name="$2"
# If it's a release branch, the yaml source URL should point to a specific version.
if is_release_branch; then
# Extract the release major&minor version from the branch name.
local branch_name="$(current_branch)"
local major_minor="${branch_name##release-}"
# Find the latest release manifest with the same major&minor version.
local yaml_source_path="$(
gsutil ls gs://knative-releases/${repo_name}/previous/v${major_minor}.*/${yaml_name}.yaml 2> /dev/null \
| sort \
| tail -n 1 \
| cut -b6-)"
# The version does exist, return it.
if [[ -n "${yaml_source_path}" ]]; then
echo "https://storage.googleapis.com/${yaml_source_path}"
return
fi
# Otherwise, fall back to nightly.
fi
echo "https://storage.googleapis.com/knative-nightly/${repo_name}/latest/${yaml_name}.yaml"
}
# Initializations that depend on previous functions.
# These MUST come last.
readonly _TEST_INFRA_SCRIPTS_DIR="$(dirname $(get_canonical_path ${BASH_SOURCE[0]}))"
readonly REPO_NAME_FORMATTED="Knative $(capitalize ${REPO_NAME//-/ })"
# Public latest nightly or release yaml files.
readonly KNATIVE_SERVING_RELEASE="$(get_latest_knative_yaml_source "serving" "serving")"
readonly KNATIVE_EVENTING_RELEASE="$(get_latest_knative_yaml_source "eventing" "eventing")"
readonly KNATIVE_MONITORING_RELEASE="$(get_latest_knative_yaml_source "serving" "monitoring")"