complete rework of sequence commands

This commit is contained in:
Alex 2021-05-27 22:51:01 +02:00
parent 736349e6cf
commit 6a0f6e96f5

View File

@ -65,6 +65,8 @@
# (thanks to Michael Winkler) # (thanks to Michael Winkler)
# 2021-01-28: v0.17c simplified volume detection using new DeviceVolumes endpoint # 2021-01-28: v0.17c simplified volume detection using new DeviceVolumes endpoint
# (thanks to Ingo Fischer) # (thanks to Ingo Fischer)
# 2021-05-27: v0.18 complete rework of sequence commands especially for TTS
# Announcement feature is no longer required due to inconsistent SSML handling
# #
### ###
# #
@ -118,17 +120,16 @@ SET_SPEAKVOL="0"
# if no current playing volume can be determined, fall back to normal volume # if no current playing volume can be determined, fall back to normal volume
SET_NORMALVOL="10" SET_NORMALVOL="10"
# max. age in minutes before volume is read from API
SET_VOLMAXAGE="1"
# Device specific volumes (overriding the above) # Device specific volumes (overriding the above)
SET_DEVICEVOLNAME="EchoDot2ndGen Echo1stGen" # SET_DEVICEVOLNAME="EchoDot2ndGen Echo1stGen"
SET_DEVICEVOLSPEAK="100 30" # SET_DEVICEVOLSPEAK="100 30"
SET_DEVICEVOLNORMAL="100 20" # SET_DEVICEVOLNORMAL="100 20"
SET_DEVICEVOLNAME=""
SET_DEVICEVOLSPEAK=""
SET_DEVICEVOLNORMAL=""
# Announcements can be made to multiple devices, while regular SPEAK cannot # max. age in minutes before volume is read from API (local cache time)
# but the notification feature has to be turned on for those devices SET_VOLMAXAGE="1"
SET_USE_ANNOUNCEMENT_FOR_SPEAK="0"
########################################### ###########################################
# nothing to configure below here # nothing to configure below here
@ -153,7 +154,6 @@ VOLMAXAGE=${VOLMAXAGE:-$SET_VOLMAXAGE}
DEVICEVOLNAME=${DEVICEVOLNAME:-$SET_DEVICEVOLNAME} DEVICEVOLNAME=${DEVICEVOLNAME:-$SET_DEVICEVOLNAME}
DEVICEVOLSPEAK=${DEVICEVOLSPEAK:-$SET_DEVICEVOLSPEAK} DEVICEVOLSPEAK=${DEVICEVOLSPEAK:-$SET_DEVICEVOLSPEAK}
DEVICEVOLNORMAL=${DEVICEVOLNORMAL:-$SET_DEVICEVOLNORMAL} DEVICEVOLNORMAL=${DEVICEVOLNORMAL:-$SET_DEVICEVOLNORMAL}
USE_ANNOUNCEMENT_FOR_SPEAK=${USE_ANNOUNCEMENT_FOR_SPEAK:-$SET_USE_ANNOUNCEMENT_FOR_SPEAK}
COOKIE="${TMP}/.alexa.cookie" COOKIE="${TMP}/.alexa.cookie"
DEVLIST="${TMP}/.alexa.devicelist.json" DEVLIST="${TMP}/.alexa.devicelist.json"
@ -222,7 +222,7 @@ usage()
while [ "$#" -gt 0 ] ; do while [ "$#" -gt 0 ] ; do
case "$1" in case "$1" in
--version) --version)
echo "v0.17c" echo "v0.18"
exit 0 exit 0
;; ;;
-d) -d)
@ -410,7 +410,6 @@ case "$COMMAND" in
VOL=${COMMAND##*:} VOL=${COMMAND##*:}
# volume as integer! # volume as integer!
if [ $VOL -le 100 -a $VOL -ge 0 ] ; then if [ $VOL -le 100 -a $VOL -ge 0 ] ; then
# COMMAND='{"type":"VolumeLevelCommand","volumeLevel":'${VOL}'}'
SEQUENCECMD='Alexa.DeviceControls.Volume' SEQUENCECMD='Alexa.DeviceControls.Volume'
SEQUENCEVAL=',\"value\":\"'${VOL}'\"' SEQUENCEVAL=',\"value\":\"'${VOL}'\"'
else else
@ -421,22 +420,14 @@ case "$COMMAND" in
;; ;;
textcommand:*) textcommand:*)
SEQUENCECMD='Alexa.TextCommand\",\"skillId\":\"amzn1.ask.1p.tellalexa' SEQUENCECMD='Alexa.TextCommand\",\"skillId\":\"amzn1.ask.1p.tellalexa'
SEQUENCEVAL=$(echo ${COMMAND##textcommand:} | sed -r s/\"/\'/g) SEQUENCEVAL=$(echo ${COMMAND##textcommand:} | sed s/\"/\'/g)
SEQUENCEVAL=',\"text\":\"'${SEQUENCEVAL}'\"' SEQUENCEVAL=',\"text\":\"'${SEQUENCEVAL}'\"'
;; ;;
speak:*) speak:*)
TTS=$(echo ${COMMAND##speak:} | sed -r s/\"/\'/g) TTS=$(echo ${COMMAND##speak:} | sed s/\"/\'/g)
if [ $USE_ANNOUNCEMENT_FOR_SPEAK -gt 0 ] ; then
if echo $TTS | grep -q -v -E '^<speak>.*<\/speak>$' ; then
# always sending SSML, since this skips the announcement "bing"
TTS="<speak>${TTS}</speak>"
fi
SEQUENCECMD='AlexaAnnouncement'
else
TTS=',\"textToSpeak\":\"'${TTS}'\"' TTS=',\"textToSpeak\":\"'${TTS}'\"'
SEQUENCECMD='Alexa.Speak' SEQUENCECMD='Alexa.Speak'
SEQUENCEVAL=$TTS SEQUENCEVAL=$TTS
fi
;; ;;
sound:*) sound:*)
SEQUENCECMD='Alexa.Sound' SEQUENCECMD='Alexa.Sound'
@ -635,9 +626,36 @@ list_devices()
jq -r '.devices[].accountName' ${DEVLIST} jq -r '.devices[].accountName' ${DEVLIST}
} }
#
# build node_to_execute string
# ARG1 - SEQUENCECMD
# ARG2 - SEQUENCEVAL
#
node()
{
if [ -n "$1" -a -n "$2" ] ; then
echo '{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"'${1}'\",\"operationPayload\":{\"deviceType\":\"'${DEVICETYPE}'\",\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"locale\":\"'${TTS_LOCALE}'\"'${2}'}}'
else
echo '{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"'${SEQUENCECMD}'\",\"operationPayload\":{\"deviceType\":\"'${DEVICETYPE}'\",\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"locale\":\"'${TTS_LOCALE}'\"'${SEQUENCEVAL}'}}'
fi
}
#
# create comma separated string
#
add_node()
{
if [ -n "$1" ] ; then
if [ -n "$2" ] ; then
echo ${1}','${2}
else
echo ${1}
fi
fi
}
# #
# execute command # execute command
# (SequenceCommands by Michael Geramb and Ralf Otto)
# #
run_cmd() run_cmd()
{ {
@ -663,95 +681,44 @@ if [ -n "${SEQUENCECMD}" ] ; then
ALEXACMD='{"behaviorId":"'${AUTOMATION}'","sequenceJson":"'${SEQUENCE}'","status":"ENABLED"}' ALEXACMD='{"behaviorId":"'${AUTOMATION}'","sequenceJson":"'${SEQUENCE}'","status":"ENABLED"}'
else else
# SequenceCommands are generally not supported on WHA devices VOLUMEPRENODESTOEXECUTE=''
if echo $COMMAND | grep -q -E "weather|traffic|flashbriefing|goodmorning|singasong|tellstory|sound|textcommand" ; then VOLUMEPOSTNODESTOEXECUTE=''
NODESTOEXECUTE=''
if [ "${DEVICEFAMILY}" = "WHA" ] ; then if [ "${DEVICEFAMILY}" = "WHA" ] ; then
echo "Skipping unsupported command: ${COMMAND} on dev:${DEVICE} type:${DEVICETYPE} serial:${DEVICESERIALNUMBER} family:${DEVICEFAMILY}" MEMBERDEVICESERIALS=$(jq --arg device "${DEVICE}" -r '.devices[] | select(.accountName == $device) | .clusterMembers[]' ${DEVLIST})
return for DEVICESERIALNUMBER in $MEMBERDEVICESERIALS ; do
fi DEVICETYPE=$(jq --arg device "${DEVICESERIALNUMBER}" -r '.devices[] | select(.serialNumber == $device) | .deviceType' ${DEVLIST})
fi NODESTOEXECUTE=$(add_node "$(node)" "${NODESTOEXECUTE}")
# some commands support an array of devices - try to find the clusterMembers if a WHA device was supplied
if echo $COMMAND | grep -q -E "speak" ; then # add volume setting per device - the WHA volume is unrelyable
MEMBER='{\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"deviceTypeId\":\"'${DEVICETYPE}'\"}' if [ $SPEAKVOL -gt 0 -o -n "${DEVICEVOLSPEAK}" ] ; then
if [ "${DEVICEFAMILY}" = "WHA" ] ; then DEVICE=$(jq --arg device "${DEVICESERIALNUMBER}" -r '.devices[] | select(.serialNumber == $device) | .accountName' ${DEVLIST})
if [ $USE_ANNOUNCEMENT_FOR_SPEAK -gt 0 ] ; then get_volumes
MEMBER="" VOLUMEPRENODESTOEXECUTE=$(add_node $(node Alexa.DeviceControls.Volume ',\"value\":\"'${SVOL}'\"') ${VOLUMEPRENODESTOEXECUTE})
MEMBERDEVICES=$(jq --arg device "${DEVICE}" -r '.devices[] | select(.accountName == $device) | .clusterMembers[]' ${DEVLIST}) VOLUMEPOSTNODESTOEXECUTE=$(add_node $(node Alexa.DeviceControls.Volume ',\"value\":\"'${VOL}'\"') ${VOLUMEPOSTNODESTOEXECUTE})
for MEMBERSERIAL in $MEMBERDEVICES ; do
MEMBERTYPE=$(jq --arg device "${MEMBERSERIAL}" -r '.devices[] | select(.serialNumber == $device) | .deviceType' ${DEVLIST})
if [ -n "${MEMBER}" ] ; then
MEMBER='{\"deviceSerialNumber\":\"'${MEMBERSERIAL}'\",\"deviceTypeId\":\"'${MEMBERTYPE}'\"},'${MEMBER}
else
MEMBER='{\"deviceSerialNumber\":\"'${MEMBERSERIAL}'\",\"deviceTypeId\":\"'${MEMBERTYPE}'\"}'
fi fi
done done
if [ -z "${MEMBER}" ] ; then if [ -z "${NODESTOEXECUTE}" ] ; then
echo "No clusterMembers found for command: ${COMMAND} on dev:${DEVICE} type:${DEVICETYPE} serial:${DEVICESERIALNUMBER} family:${DEVICEFAMILY}" echo "No clusterMembers found for command: ${COMMAND} on dev:${DEVICE} type:${DEVICETYPE} serial:${DEVICESERIALNUMBER} family:${DEVICEFAMILY}"
return return
fi fi
else else
echo "Speak to Multiroom is only possible when USE_ANNOUNCEMENT_FOR_SPEAK is set and notifications on the devices are enabled" NODESTOEXECUTE=$(add_node "$(node)" "${NODESTOEXECUTE}")
echo "Skipping unsupported command: ${COMMAND} on dev:${DEVICE} type:${DEVICETYPE} serial:${DEVICESERIALNUMBER} family:${DEVICEFAMILY}" if [ $SPEAKVOL -gt 0 -o -n "${DEVICEVOLSPEAK}" ] ; then
return get_volumes
fi VOLUMEPRENODESTOEXECUTE=$(add_node $(node Alexa.DeviceControls.Volume ',\"value\":\"'${SVOL}'\"') ${VOLUMEPRENODESTOEXECUTE})
fi VOLUMEPOSTNODESTOEXECUTE=$(add_node $(node Alexa.DeviceControls.Volume ',\"value\":\"'${VOL}'\"') ${VOLUMEPOSTNODESTOEXECUTE})
fi
# the speak command is treated differently if $SPEAKVOL is > 0
if [ -n "${TTS}" -a $SPEAKVOL -gt 0 ] || [ "${COMMAND%%:*}" = 'sound' -a $SPEAKVOL -gt 0 ] ; then
SVOL=$SPEAKVOL
# Not using arrays here in order to be compatible with non-Bash
# Get the list position of the current device type
IDX=0
for D in $DEVICEVOLNAME ; do
if [ "${D}" = "${DEVICE}" ] ; then
break;
fi
IDX=$((IDX+1))
done
# get the speak volume at that position
C=0
for D in $DEVICEVOLSPEAK ; do
if [ $C -eq $IDX ] ; then
if [ -n "${D}" ] ; then SVOL=$D ; fi
break
fi
C=$((C+1))
done
# try to retrieve the "currently playing" volume
VOL=$(get_volume)
if [ -z "${VOL}" ] ; then
# get the normal volume of the current device type
C=0
for D in $DEVICEVOLNORMAL; do
if [ $C -eq $IDX ] ; then
VOL=$D
break
fi
C=$((C+1))
done
# if the volume is still undefined, use $NORMALVOL
if [ -z "${VOL}" ] ; then
VOL=$NORMALVOL
fi fi
fi fi
if [ -n "${TTS}" -a $USE_ANNOUNCEMENT_FOR_SPEAK -gt 0 ] ; then if [ -n "${VOLUMEPRENODESTOEXECUTE}" -a -n "${VOLUMEPOSTNODESTOEXECUTE}" ] ; then
ALEXACMD='{"behaviorId":"PREVIEW","sequenceJson":"{\"@type\":\"com.amazon.alexa.behaviors.model.Sequence\",\"startNode\":{\"@type\":\"com.amazon.alexa.behaviors.model.SerialNode\",\"nodesToExecute\":[{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"Alexa.DeviceControls.Volume\",\"operationPayload\":{\"deviceType\":\"'${DEVICETYPE}'\",\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"locale\":\"'${TTS_LOCALE}'\",\"value\":\"'${SVOL}'\"}},{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"'${SEQUENCECMD}'\",\"operationPayload\":{\"expireAfter\":\"PT5S\",\"content\":[{\"display\":{\"title\":\"AlexaRemoteControl\",\"body\":\"Announcement\"},\"speak\":{\"type\":\"ssml\",\"value\":\"'${TTS}'\"},\"locale\":\"'${TTS_LOCALE}'\"}],\"target\":{\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"devices\":['${MEMBER}']},\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\"}},{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"Alexa.DeviceControls.Volume\",\"operationPayload\":{\"deviceType\":\"'${DEVICETYPE}'\",\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"locale\":\"'${TTS_LOCALE}'\",\"value\":\"'${VOL}'\"}}]}}","status":"ENABLED"}'; # execute serially "set_speak_volume" => "sequence_command" => "set_normal_volume"
# (each subtask is executed in parallel)
ALEXACMD='{"behaviorId":"PREVIEW","sequenceJson":"{\"@type\":\"com.amazon.alexa.behaviors.model.Sequence\",\"startNode\":{\"@type\":\"com.amazon.alexa.behaviors.model.SerialNode\",\"nodesToExecute\":[{\"@type\":\"com.amazon.alexa.behaviors.model.ParallelNode\",\"nodesToExecute\":['${VOLUMEPRENODESTOEXECUTE}']},{\"@type\":\"com.amazon.alexa.behaviors.model.ParallelNode\",\"nodesToExecute\":['${NODESTOEXECUTE}']},{\"@type\":\"com.amazon.alexa.behaviors.model.ParallelNode\",\"nodesToExecute\":['${VOLUMEPOSTNODESTOEXECUTE}']}]}}","status":"ENABLED"}'
else else
ALEXACMD='{"behaviorId":"PREVIEW","sequenceJson":"{\"@type\":\"com.amazon.alexa.behaviors.model.Sequence\",\"startNode\":{\"@type\":\"com.amazon.alexa.behaviors.model.SerialNode\",\"nodesToExecute\":[{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"Alexa.DeviceControls.Volume\",\"operationPayload\":{\"deviceType\":\"'${DEVICETYPE}'\",\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"locale\":\"'${TTS_LOCALE}'\",\"value\":\"'${SVOL}'\"}},{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"'${SEQUENCECMD}'\",\"operationPayload\":{\"deviceType\":\"'${DEVICETYPE}'\",\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"locale\":\"'${TTS_LOCALE}'\"'${SEQUENCEVAL}'}},{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"Alexa.DeviceControls.Volume\",\"operationPayload\":{\"deviceType\":\"'${DEVICETYPE}'\",\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"locale\":\"'${TTS_LOCALE}'\",\"value\":\"'${VOL}'\"}}]}}","status":"ENABLED"}' # execute in parallel "sequence_command"
fi ALEXACMD='{"behaviorId":"PREVIEW","sequenceJson":"{\"@type\":\"com.amazon.alexa.behaviors.model.Sequence\",\"startNode\":{\"@type\":\"com.amazon.alexa.behaviors.model.ParallelNode\",\"nodesToExecute\":['${NODESTOEXECUTE}']}}","status":"ENABLED"}'
else
if [ -n "${TTS}" -a $USE_ANNOUNCEMENT_FOR_SPEAK -gt 0 ] ; then
ALEXACMD='{"behaviorId":"PREVIEW","sequenceJson":"{\"@type\":\"com.amazon.alexa.behaviors.model.Sequence\",\"startNode\":{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"'${SEQUENCECMD}'\",\"operationPayload\":{\"expireAfter\":\"PT5S\",\"content\":[{\"display\":{\"title\":\"AlexaRemoteControl\",\"body\":\"Announcement\"},\"speak\":{\"type\":\"ssml\",\"value\":\"'${TTS}'\"},\"locale\":\"'${TTS_LOCALE}'\"}],\"target\":{\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"devices\":['${MEMBER}']},\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\"}}}","status":"ENABLED"}';
else
ALEXACMD='{"behaviorId":"PREVIEW","sequenceJson":"{\"@type\":\"com.amazon.alexa.behaviors.model.Sequence\",\"startNode\":{\"@type\":\"com.amazon.alexa.behaviors.model.OpaquePayloadOperationNode\",\"type\":\"'${SEQUENCECMD}'\",\"operationPayload\":{\"deviceType\":\"'${DEVICETYPE}'\",\"deviceSerialNumber\":\"'${DEVICESERIALNUMBER}'\",\"customerId\":\"'${MEDIAOWNERCUSTOMERID}'\",\"locale\":\"'${TTS_LOCALE}'\"'${SEQUENCEVAL}'}}}","status":"ENABLED"}'
fi
fi fi
fi fi
@ -763,7 +730,7 @@ if [ -n "${SEQUENCECMD}" ] ; then
-H "csrf: $(awk "\$0 ~/.${AMAZON}.*csrf[ \\s\\t]+/ {print \$7}" ${COOKIE})" -X POST -d @"${TMP}/.alexa.cmd" \ -H "csrf: $(awk "\$0 ~/.${AMAZON}.*csrf[ \\s\\t]+/ {print \$7}" ${COOKIE})" -X POST -d @"${TMP}/.alexa.cmd" \
"https://${ALEXA}/api/behaviors/preview" "https://${ALEXA}/api/behaviors/preview"
rm -f "${TMP}/.alexa.cmd" # rm -f "${TMP}/.alexa.cmd"
else else
${CURL} ${OPTS} -s -b ${COOKIE} -A "${BROWSER}" -H "DNT: 1" -H "Connection: keep-alive" -L\ ${CURL} ${OPTS} -s -b ${COOKIE} -A "${BROWSER}" -H "DNT: 1" -H "Connection: keep-alive" -L\
-H "Content-Type: application/json; charset=UTF-8" -H "Referer: https://alexa.${AMAZON}/spa/index.html" -H "Origin: https://alexa.${AMAZON}"\ -H "Content-Type: application/json; charset=UTF-8" -H "Referer: https://alexa.${AMAZON}/spa/index.html" -H "Origin: https://alexa.${AMAZON}"\
@ -930,6 +897,59 @@ show_queue()
"https://${ALEXA}/api/np/queue?deviceSerialNumber=${DEVICESERIALNUMBER}&deviceType=${DEVICETYPE}" | jq '.' "https://${ALEXA}/api/np/queue?deviceSerialNumber=${DEVICESERIALNUMBER}&deviceType=${DEVICETYPE}" | jq '.'
} }
#
# device specific SPEAKVOL/NORMALVOL (sets SVOL/VOL)
#
get_volumes()
{
VOL=""
SVOL=""
# Not using arrays here in order to be compatible with non-Bash
# Get the list position of the current device type
IDX=0
for D in $DEVICEVOLNAME ; do
if [ "${D}" = "${DEVICE}" ] ; then
break;
fi
IDX=$((IDX+1))
done
# get the speak volume at that position
C=0
for D in $DEVICEVOLSPEAK ; do
if [ $C -eq $IDX ] ; then
if [ -n "${D}" ] ; then SVOL=$D ; fi
break
fi
C=$((C+1))
done
if [ -z "${SVOL}" ] ; then
SVOL=$SPEAKVOL
fi
# try to retrieve the "currently playing" volume
VOLMAXAGE=1
VOL=$(get_volume)
if [ -z "${VOL}" ] ; then
# get the normal volume of the current device type
C=0
for D in $DEVICEVOLNORMAL; do
if [ $C -eq $IDX ] ; then
VOL=$D
break
fi
C=$((C+1))
done
# if the volume is still undefined, use $NORMALVOL
if [ -z "${VOL}" ] ; then
VOL=$NORMALVOL
fi
fi
}
# #
# current volume level # current volume level
# #