services/device/shell_test.sh - release.go.x.ref - Git at Google

 #!/bin/bash
 # Copyright 2015 The Vanadium Authors. All rights reserved.
 # Use of this source code is governed by a BSD-style
 # license that can be found in the LICENSE file.

 # The shell_test library is used to execute shell tests.

 # IMPORTANT: If your script registers its own "trap" handler, that
 # handler must call shell_test::at_exit to clean up all temporary
 # files and directories created by this library.

 source "$(go list -f {{.Dir}} v.io/x/ref/cmd/mgmt)/shell.sh"

 trap shell_test::at_exit INT TERM EXIT

 # ensure that we print a 'FAIL' message unless shell_test::pass is
 # explicitly called.
 shell_test_EXIT_MESSAGE="FAIL_UNEXPECTED_EXIT"

 # default timeout values.
 shell_test_DEFAULT_SERVER_TIMEOUT="10"
 shell_test_DEFAULT_MESSAGE_TIMEOUT="30"

 # default binary dir and working dir.
 shell_test_BIN_DIR="${shell_test_BIN_DIR:-$(shell::tmp_dir)}"
 shell_test_WORK_DIR="$(shell::tmp_dir)"

 # shell_test::assert_eq GOT WANT LINENO checks that GOT == WANT and if
 # not, fails the test, outputting the offending line number.
 shell_test::assert_eq() {
   local -r GOT="$1"
   local -r WANT="$2"
   local -r LINENO="$3"
   if [[ "${GOT}" != "${WANT}" ]]; then
     shell_test::fail "line ${LINENO}: expected '${GOT}' == '${WANT}'"
   fi
 }

 # shell_test::assert_ge GOT WANT LINENO checks that GOT >= WANT and if
 # not, fails the test, outputting the given line number and error
 # message.
 shell_test::assert_ge() {
   local -r GOT="$1"
   local -r WANT="$2"
   local -r LINENO="$3"
   if [[ "${GOT}" -lt "${WANT}" ]]; then
     shell_test::fail "line ${LINENO}: expected ${GOT} >= ${WANT}"
   fi
 }

 # shell_test::assert_gt GOT WANT LINENO checks that GOT > WANT and if
 # not, fails the test, outputting the given line number and error
 # message.
 shell_test::assert_gt() {
   local -r GOT="$1"
   local -r WANT="$2"
   local -r LINENO="$3"
   if [[ "${GOT}" -le "${WANT}" ]]; then
     shell_test::fail "line ${LINENO}: expected ${GOT} > ${WANT}"
   fi
 }

 # shell_test::assert_le GOT WANT LINENO checks that GOT <= WANT and if
 # not, fails the test, outputting the given line number and error
 # message.
 shell_test::assert_le() {
   local -r GOT="$1"
   local -r WANT="$2"
   local -r LINENO="$3"
   if [[ "${GOT}" -gt "${WANT}" ]]; then
     shell_test::fail "line ${LINENO}: expected ${GOT} <= ${WANT}"
   fi
 }

 # shell_test::assert_lt GOT WANT LINENO checks that GOT < WANT and if
 # not, fails the test, outputting the given line number and error
 # message.
 shell_test::assert_lt() {
   local -r GOT="$1"
   local -r WANT="$2"
   local -r LINENO="$3"
   if [[ "${GOT}" -ge "${WANT}" ]]; then
     shell_test::fail "line ${LINENO}: expected ${GOT} < ${WANT}"
   fi
 }

 # shell_test::assert_ne GOT WANT LINENO checks that GOT != WANT and if
 # not, fails the test, outputting the offending line number.
 shell_test::assert_ne() {
   local -r GOT="$1"
   local -r WANT="$2"
   local -r LINENO="$3"
   if [[ "${GOT}" == "${WANT}" ]]; then
     shell_test::fail "line ${LINENO}: expected '${GOT}' != '${WANT}'"
   fi
 }

 # shell_test::assert_contains GOT WANT LINENO checks that GOT contains WANT and
 # if not, fails the test, outputting the offending line number.
 shell_test::assert_contains() {
   local -r GOT="$1"
   local -r WANT="$2"
   local -r LINENO="$3"
   if [[ ! "${GOT}" =~ "${WANT}" ]]; then
     shell_test::fail "line ${LINENO}: expected '${GOT}' contains '${WANT}'"
   fi
 }

 # shell_test::at_exit is executed when the shell script exits.
 shell_test::at_exit() {
   echo "${shell_test_EXIT_MESSAGE}"
   # Note: shell::at_exit unsets our trap, so it won't run again on exit.
   shell::at_exit
   shell::kill_child_processes
 }

 # shell_test::fail is used to identify a failing test.
 #
 # Example:
 #   <command> || shell_test::fail "line ${LINENO}: <error message>"
 shell_test::fail() {
   [[ "$#" -gt 0 ]] && echo "$0 $*"
   shell_test_EXIT_MESSAGE="FAIL"
   exit 1
 }

 # shell_test::pass is used to identify a passing test.
 shell_test::pass() {
   shell_test_EXIT_MESSAGE="PASS"
   exit 0
 }

 # shell_test::build_go_binary builds the binary for the given go package ($1) with
 # the given output ($2) in $shell_test_BIN_DIR. Before building a binary, it will
 # check whether it already exists. If so, it will skip it.
 shell_test::build_go_binary() {
   pushd "${shell_test_BIN_DIR}" > /dev/null
   local -r PACKAGE="$1"
   local OUTPUT="${2:-$(basename ${PACKAGE})}"
   if [[ -f "${OUTPUT}" ]]; then
     echo "${shell_test_BIN_DIR}/${OUTPUT}"
     return
   fi

   go build -o "${OUTPUT}" "${PACKAGE}" 2>/dev/null || shell_test::fail "line ${LINENO}: failed to build ${OUTPUT}"
   echo "${shell_test_BIN_DIR}/${OUTPUT}"
   popd > /dev/null
 }

 # shell_test::setup_server_test is common boilerplate used for testing vanadium
 # servers. In particular, this function sets up an instance of the mount table
 # daemon, and sets the V23_NAMESPACE environment variable accordingly.  It also
 # sets up credentials as needed.
 shell_test::setup_server_test() {
   if [[ -n ${shell_test_RUNNING_UNDER_AGENT+1} ]]; then
     shell_test::setup_server_test_agent
   else
     shell_test::setup_server_test_no_agent
   fi
 }

 # shell_test::setup_mounttable brings up a mounttable and sets the
 # V23_NAMESPACE to point to it.
 shell_test::setup_mounttable() {
   # Build the mount table daemon and related tools.
   MOUNTTABLED_BIN="$(shell_test::build_go_binary 'v.io/x/ref/services/mounttable/mounttabled')"

   # Start the mounttable daemon.
   local -r MT_LOG=$(shell::tmp_file)
   if [[ -n ${shell_test_RUNNING_UNDER_AGENT+1} ]]; then
     shell::run_server "${shell_test_DEFAULT_SERVER_TIMEOUT}" "${MT_LOG}" "${MT_LOG}" "${VRUN}" "${MOUNTTABLED_BIN}" --v23="127.0.0.1:0" &> /dev/null || (cat "${MT_LOG}" && shell_test::fail "line ${LINENO}: failed to start mounttabled")
   else
     shell::run_server "${shell_test_DEFAULT_SERVER_TIMEOUT}" "${MT_LOG}" "${MT_LOG}" "${MOUNTTABLED_BIN}" --v23.tcp.address="127.0.0.1:0" &> /dev/null || (cat "${MT_LOG}" && shell_test::fail "line ${LINENO}: failed to start mounttabled")
   fi
   shell::timed_wait_for "${shell_test_DEFAULT_MESSAGE_TIMEOUT}" "${MT_LOG}" "Mount table service" || shell_test::fail "line ${LINENO}: failed to find expected output"

   export V23_NAMESPACE=$(grep "Mount table service at:" "${MT_LOG}" | sed -e 's/^.*endpoint: //')
   [[ -n "${V23_NAMESPACE}" ]] || shell_test::fail "line ${LINENO}: failed to read endpoint from logfile"
 }

 # shell_test::setup_server_test_no_agent does the work of setup_server_test when
 # not running under enable_agent (using credentials directories).
 shell_test::setup_server_test_no_agent() {
   [[ ! -n ${shell_test_RUNNING_UNDER_AGENT+1} ]] \
     || shell_test::fail "setup_server_test_no_agent called when running under enable_agent"
   shell_test::setup_mounttable

   # Create and use a new credentials directory /after/ the mount table is
   # started so that servers or clients started after this point start with different
   # credentials.
   local -r TMP=${V23_CREDENTIALS:-}
   if [[ -z "${TMP}" ]]; then
     export V23_CREDENTIALS=$(shell::tmp_dir)
   fi
 }

 # shell_test::setup_server_test_agent does the work of setup_server_test when
 # running under enable_agent.
 shell_test::setup_server_test_agent() {
   [[ -n ${shell_test_RUNNING_UNDER_AGENT+1} ]] \
     || shell_test::fail "setup_server_test_agent called when not running under enable_agent"
   shell_test::setup_mounttable
 }

 # shell_test::start_server is used to start a server. The server is
 # considered started when it successfully mounts itself into a mount
 # table; an event that is detected using the shell::wait_for
 # function.
 #
 # Example:
 #   shell_test::start_server <server> <args>
 shell_test::start_server() {
   START_SERVER_LOG_FILE=$(shell::tmp_file)
   START_SERVER_PID=$(shell::run_server "${shell_test_DEFAULT_SERVER_TIMEOUT}" "${START_SERVER_LOG_FILE}" "${START_SERVER_LOG_FILE}" "$@" -logtostderr -vmodule=publisher=2)
   shell::timed_wait_for "${shell_test_DEFAULT_MESSAGE_TIMEOUT}" "${START_SERVER_LOG_FILE}" "ipc pub: mount"
 }

 # shell_test::credentials is used to create a new VeyronCredentials directory.
 # The directory is initialized with a new principal that has a self-signed
 # blessing of the specified name set as default and shareable with all peers.
 #
 # Example:
 #   shell_test::credentials <self blessing name>
 shell_test::credentials() {
   [[ ! -n ${shell_test_RUNNING_UNDER_AGENT+1} ]] \
     || shell_test::fail "credentials called when running under enable_agent"
   local -r PRINCIPAL_BIN="$(shell_test::build_go_binary 'v.io/x/ref/cmd/principal')"
   local -r CRED=$(shell::tmp_dir)
   "${PRINCIPAL_BIN}" create --overwrite=true "${CRED}" "$1" >/dev/null || shell_test::fail "line ${LINENO}: create failed"
   echo "${CRED}"
 }

 # shell_test::forkcredentials is used to create a new VeyronCredentials directory
 # that is initialized with a principal blessed by the provided principal under the
 # provided extension.
 #
 # Example:
 #   shell_test::forkcredentials <blesser principal's VeyronCredentials> <extension>
 shell_test::forkcredentials() {
   [[ ! -n ${shell_test_RUNNING_UNDER_AGENT+1} ]] \
     || shell_test::fail "forkcredentials called when running under enable_agent"

   local -r PRINCIPAL_BIN="$(shell_test::build_go_binary 'v.io/x/ref/cmd/principal')"
   local -r FORKCRED=$(shell::tmp_dir)
   "${PRINCIPAL_BIN}" create --overwrite=true "${FORKCRED}" self >/dev/null || shell_test::fail "line ${LINENO}: create failed"
   "${PRINCIPAL_BIN}" --v23.credentials="$1" bless --require-caveats=false "${FORKCRED}" "$2" >blessing || shell_test::fail "line ${LINENO}: bless failed"
   "${PRINCIPAL_BIN}" --v23.credentials="${FORKCRED}" store setdefault blessing || shell_test::fail "line ${LINENO}: store setdefault failed"
   "${PRINCIPAL_BIN}" --v23.credentials="${FORKCRED}" store set blessing ... || shell_test::fail "line ${LINENO}: store set failed"
   echo "${FORKCRED}"
 }

 # shell_test::enable_agent causes the test to be executed under the security
 # agent.  It sets up the agent and then runs itself under that agent.
 #
 # Example:
 #   shell_test::enable_agent "$@"
 #   build() {
 #     ...
 #   }
 #   main() {
 #     build
 #     ...
 #     shell_test::pass
 #   }
 #   main "$@"
 shell_test::enable_agent() {
   if [[ ! -n ${shell_test_RUNNING_UNDER_AGENT+1} ]]; then
     local -r AGENTD="$(shell_test::build_go_binary 'v.io/x/ref/services/agent/agentd')"
     local -r WORKDIR="${shell_test_WORK_DIR}"
     export shell_test_RUNNING_UNDER_AGENT=1
     V23_CREDENTIALS="${WORKDIR}/credentials" exec ${AGENTD} --no-passphrase --additional-principals="${WORKDIR}/childcredentials" bash -"$-" "$0" "$@"
     shell_test::fail "failed to run test under agent"
   else
     VRUN="$(shell_test::build_go_binary 'v.io/x/ref/cmd/vrun')"
   fi
 }
	#!/bin/bash
	# Copyright 2015 The Vanadium Authors. All rights reserved.
	# Use of this source code is governed by a BSD-style
	# license that can be found in the LICENSE file.

	# The shell_test library is used to execute shell tests.

	# IMPORTANT: If your script registers its own "trap" handler, that
	# handler must call shell_test::at_exit to clean up all temporary
	# files and directories created by this library.

	source "$(go list -f {{.Dir}} v.io/x/ref/cmd/mgmt)/shell.sh"

	trap shell_test::at_exit INT TERM EXIT

	# ensure that we print a 'FAIL' message unless shell_test::pass is
	# explicitly called.
	shell_test_EXIT_MESSAGE="FAIL_UNEXPECTED_EXIT"

	# default timeout values.
	shell_test_DEFAULT_SERVER_TIMEOUT="10"
	shell_test_DEFAULT_MESSAGE_TIMEOUT="30"

	# default binary dir and working dir.
	shell_test_BIN_DIR="${shell_test_BIN_DIR:-$(shell::tmp_dir)}"
	shell_test_WORK_DIR="$(shell::tmp_dir)"

	# shell_test::assert_eq GOT WANT LINENO checks that GOT == WANT and if
	# not, fails the test, outputting the offending line number.
	shell_test::assert_eq() {
	local -r GOT="$1"
	local -r WANT="$2"
	local -r LINENO="$3"
	if [[ "${GOT}" != "${WANT}" ]]; then
	shell_test::fail "line ${LINENO}: expected '${GOT}' == '${WANT}'"
	fi
	}

	# shell_test::assert_ge GOT WANT LINENO checks that GOT >= WANT and if
	# not, fails the test, outputting the given line number and error
	# message.
	shell_test::assert_ge() {
	local -r GOT="$1"
	local -r WANT="$2"
	local -r LINENO="$3"
	if [[ "${GOT}" -lt "${WANT}" ]]; then
	shell_test::fail "line ${LINENO}: expected ${GOT} >= ${WANT}"
	fi
	}

	# shell_test::assert_gt GOT WANT LINENO checks that GOT > WANT and if
	# not, fails the test, outputting the given line number and error
	# message.
	shell_test::assert_gt() {
	local -r GOT="$1"
	local -r WANT="$2"
	local -r LINENO="$3"
	if [[ "${GOT}" -le "${WANT}" ]]; then
	shell_test::fail "line ${LINENO}: expected ${GOT} > ${WANT}"
	fi
	}

	# shell_test::assert_le GOT WANT LINENO checks that GOT <= WANT and if
	# not, fails the test, outputting the given line number and error
	# message.
	shell_test::assert_le() {
	local -r GOT="$1"
	local -r WANT="$2"
	local -r LINENO="$3"
	if [[ "${GOT}" -gt "${WANT}" ]]; then
	shell_test::fail "line ${LINENO}: expected ${GOT} <= ${WANT}"
	fi
	}

	# shell_test::assert_lt GOT WANT LINENO checks that GOT < WANT and if
	# not, fails the test, outputting the given line number and error
	# message.
	shell_test::assert_lt() {
	local -r GOT="$1"
	local -r WANT="$2"
	local -r LINENO="$3"
	if [[ "${GOT}" -ge "${WANT}" ]]; then
	shell_test::fail "line ${LINENO}: expected ${GOT} < ${WANT}"
	fi
	}

	# shell_test::assert_ne GOT WANT LINENO checks that GOT != WANT and if
	# not, fails the test, outputting the offending line number.
	shell_test::assert_ne() {
	local -r GOT="$1"
	local -r WANT="$2"
	local -r LINENO="$3"
	if [[ "${GOT}" == "${WANT}" ]]; then
	shell_test::fail "line ${LINENO}: expected '${GOT}' != '${WANT}'"
	fi
	}

	# shell_test::assert_contains GOT WANT LINENO checks that GOT contains WANT and
	# if not, fails the test, outputting the offending line number.
	shell_test::assert_contains() {
	local -r GOT="$1"
	local -r WANT="$2"
	local -r LINENO="$3"
	if [[ ! "${GOT}" =~ "${WANT}" ]]; then
	shell_test::fail "line ${LINENO}: expected '${GOT}' contains '${WANT}'"
	fi
	}

	# shell_test::at_exit is executed when the shell script exits.
	shell_test::at_exit() {
	echo "${shell_test_EXIT_MESSAGE}"
	# Note: shell::at_exit unsets our trap, so it won't run again on exit.
	shell::at_exit
	shell::kill_child_processes
	}

	# shell_test::fail is used to identify a failing test.
	#
	# Example:
	# <command> \|\| shell_test::fail "line ${LINENO}: <error message>"
	shell_test::fail() {
	[[ "$#" -gt 0 ]] && echo "$0 $*"
	shell_test_EXIT_MESSAGE="FAIL"
	exit 1
	}

	# shell_test::pass is used to identify a passing test.
	shell_test::pass() {
	shell_test_EXIT_MESSAGE="PASS"
	exit 0
	}

	# shell_test::build_go_binary builds the binary for the given go package ($1) with
	# the given output ($2) in $shell_test_BIN_DIR. Before building a binary, it will
	# check whether it already exists. If so, it will skip it.
	shell_test::build_go_binary() {
	pushd "${shell_test_BIN_DIR}" > /dev/null
	local -r PACKAGE="$1"
	local OUTPUT="${2:-$(basename ${PACKAGE})}"
	if [[ -f "${OUTPUT}" ]]; then
	echo "${shell_test_BIN_DIR}/${OUTPUT}"
	return
	fi

	go build -o "${OUTPUT}" "${PACKAGE}" 2>/dev/null \|\| shell_test::fail "line ${LINENO}: failed to build ${OUTPUT}"
	echo "${shell_test_BIN_DIR}/${OUTPUT}"
	popd > /dev/null
	}

	# shell_test::setup_server_test is common boilerplate used for testing vanadium
	# servers. In particular, this function sets up an instance of the mount table
	# daemon, and sets the V23_NAMESPACE environment variable accordingly. It also
	# sets up credentials as needed.
	shell_test::setup_server_test() {
	if [[ -n ${shell_test_RUNNING_UNDER_AGENT+1} ]]; then
	shell_test::setup_server_test_agent
	else
	shell_test::setup_server_test_no_agent
	fi
	}

	# shell_test::setup_mounttable brings up a mounttable and sets the
	# V23_NAMESPACE to point to it.
	shell_test::setup_mounttable() {
	# Build the mount table daemon and related tools.
	MOUNTTABLED_BIN="$(shell_test::build_go_binary 'v.io/x/ref/services/mounttable/mounttabled')"

	# Start the mounttable daemon.
	local -r MT_LOG=$(shell::tmp_file)
	if [[ -n ${shell_test_RUNNING_UNDER_AGENT+1} ]]; then
	shell::run_server "${shell_test_DEFAULT_SERVER_TIMEOUT}" "${MT_LOG}" "${MT_LOG}" "${VRUN}" "${MOUNTTABLED_BIN}" --v23="127.0.0.1:0" &> /dev/null \|\| (cat "${MT_LOG}" && shell_test::fail "line ${LINENO}: failed to start mounttabled")
	else
	shell::run_server "${shell_test_DEFAULT_SERVER_TIMEOUT}" "${MT_LOG}" "${MT_LOG}" "${MOUNTTABLED_BIN}" --v23.tcp.address="127.0.0.1:0" &> /dev/null \|\| (cat "${MT_LOG}" && shell_test::fail "line ${LINENO}: failed to start mounttabled")
	fi
	shell::timed_wait_for "${shell_test_DEFAULT_MESSAGE_TIMEOUT}" "${MT_LOG}" "Mount table service" \|\| shell_test::fail "line ${LINENO}: failed to find expected output"

	export V23_NAMESPACE=$(grep "Mount table service at:" "${MT_LOG}" \| sed -e 's/^.*endpoint: //')
	[[ -n "${V23_NAMESPACE}" ]] \|\| shell_test::fail "line ${LINENO}: failed to read endpoint from logfile"
	}

	# shell_test::setup_server_test_no_agent does the work of setup_server_test when
	# not running under enable_agent (using credentials directories).
	shell_test::setup_server_test_no_agent() {
	[[ ! -n ${shell_test_RUNNING_UNDER_AGENT+1} ]] \
	\|\| shell_test::fail "setup_server_test_no_agent called when running under enable_agent"
	shell_test::setup_mounttable

	# Create and use a new credentials directory /after/ the mount table is
	# started so that servers or clients started after this point start with different
	# credentials.
	local -r TMP=${V23_CREDENTIALS:-}
	if [[ -z "${TMP}" ]]; then
	export V23_CREDENTIALS=$(shell::tmp_dir)
	fi
	}

	# shell_test::setup_server_test_agent does the work of setup_server_test when
	# running under enable_agent.
	shell_test::setup_server_test_agent() {
	[[ -n ${shell_test_RUNNING_UNDER_AGENT+1} ]] \
	\|\| shell_test::fail "setup_server_test_agent called when not running under enable_agent"
	shell_test::setup_mounttable
	}

	# shell_test::start_server is used to start a server. The server is
	# considered started when it successfully mounts itself into a mount
	# table; an event that is detected using the shell::wait_for
	# function.
	#
	# Example:
	# shell_test::start_server <server> <args>
	shell_test::start_server() {
	START_SERVER_LOG_FILE=$(shell::tmp_file)
	START_SERVER_PID=$(shell::run_server "${shell_test_DEFAULT_SERVER_TIMEOUT}" "${START_SERVER_LOG_FILE}" "${START_SERVER_LOG_FILE}" "$@" -logtostderr -vmodule=publisher=2)
	shell::timed_wait_for "${shell_test_DEFAULT_MESSAGE_TIMEOUT}" "${START_SERVER_LOG_FILE}" "ipc pub: mount"
	}

	# shell_test::credentials is used to create a new VeyronCredentials directory.
	# The directory is initialized with a new principal that has a self-signed
	# blessing of the specified name set as default and shareable with all peers.
	#
	# Example:
	# shell_test::credentials <self blessing name>
	shell_test::credentials() {
	[[ ! -n ${shell_test_RUNNING_UNDER_AGENT+1} ]] \
	\|\| shell_test::fail "credentials called when running under enable_agent"
	local -r PRINCIPAL_BIN="$(shell_test::build_go_binary 'v.io/x/ref/cmd/principal')"
	local -r CRED=$(shell::tmp_dir)
	"${PRINCIPAL_BIN}" create --overwrite=true "${CRED}" "$1" >/dev/null \|\| shell_test::fail "line ${LINENO}: create failed"
	echo "${CRED}"
	}

	# shell_test::forkcredentials is used to create a new VeyronCredentials directory
	# that is initialized with a principal blessed by the provided principal under the
	# provided extension.
	#
	# Example:
	# shell_test::forkcredentials <blesser principal's VeyronCredentials> <extension>
	shell_test::forkcredentials() {
	[[ ! -n ${shell_test_RUNNING_UNDER_AGENT+1} ]] \
	\|\| shell_test::fail "forkcredentials called when running under enable_agent"

	local -r PRINCIPAL_BIN="$(shell_test::build_go_binary 'v.io/x/ref/cmd/principal')"
	local -r FORKCRED=$(shell::tmp_dir)
	"${PRINCIPAL_BIN}" create --overwrite=true "${FORKCRED}" self >/dev/null \|\| shell_test::fail "line ${LINENO}: create failed"
	"${PRINCIPAL_BIN}" --v23.credentials="$1" bless --require-caveats=false "${FORKCRED}" "$2" >blessing \|\| shell_test::fail "line ${LINENO}: bless failed"
	"${PRINCIPAL_BIN}" --v23.credentials="${FORKCRED}" store setdefault blessing \|\| shell_test::fail "line ${LINENO}: store setdefault failed"
	"${PRINCIPAL_BIN}" --v23.credentials="${FORKCRED}" store set blessing ... \|\| shell_test::fail "line ${LINENO}: store set failed"
	echo "${FORKCRED}"
	}

	# shell_test::enable_agent causes the test to be executed under the security
	# agent. It sets up the agent and then runs itself under that agent.
	#
	# Example:
	# shell_test::enable_agent "$@"
	# build() {
	# ...
	# }
	# main() {
	# build
	# ...
	# shell_test::pass
	# }
	# main "$@"
	shell_test::enable_agent() {
	if [[ ! -n ${shell_test_RUNNING_UNDER_AGENT+1} ]]; then
	local -r AGENTD="$(shell_test::build_go_binary 'v.io/x/ref/services/agent/agentd')"
	local -r WORKDIR="${shell_test_WORK_DIR}"
	export shell_test_RUNNING_UNDER_AGENT=1
	V23_CREDENTIALS="${WORKDIR}/credentials" exec ${AGENTD} --no-passphrase --additional-principals="${WORKDIR}/childcredentials" bash -"$-" "$0" "$@"
	shell_test::fail "failed to run test under agent"
	else
	VRUN="$(shell_test::build_go_binary 'v.io/x/ref/cmd/vrun')"
	fi
	}