Related
I'm trying to implement a simple app for Google Assistant.
All works fine, but now I have a problem with the "permission" helper :
https://developers.google.com/actions/assistant/helpers#helper_intents
I have an intent connected with webhook to my java application. When an user types a sentence similar to "near to me", I want to ask to him his location and then use lat/lon to perform a search.
es: Brazilian restaurant near to me
my intent "searchRestaurant" is fired
I receive the webhook request and I parse it
if I find a parameter that is connected to a sentence like "near to me", so instead to response with a "Card" or a "List" I return a JSON that represent the helper request :
{
"conversationToken": "[]",
"expectUserResponse": true,
"expectedInputs": [
{
"inputPrompt": {
"initialPrompts": [
{
"textToSpeech": "PLACEHOLDER_FOR_PERMISSION"
}
],
"noInputPrompts": []
},
"possibileIntents": [
{
"intent": "actions.intent.PERMISSION",
"inputValueData": {
"#type": "type.googleapis.com/google.actions.v2.PermissionValueSpec",
"optContext": "Posso accedere alla tua posizione?",
"permission": [
"NAME",
"DEVICE_PRECISE_LOCATION"
]
}
}
]
}
]
}
but something seems to be wrong, and I receive an error:
"{\n \"responseMetadata\": {\n \"status\": {\n \"code\": 10,\n \"message\": \"Failed to parse Dialogflow response into AppResponse because of empty speech response\",\n \"details\": [{\n \"#type\": \"type.googleapis.com/google.protobuf.Value\",\n \"value\": \"{\\"id\\":\\"1cc45c5e-c398-4ea7-98a5-408f31ce142d\\",\\"timestamp\\":\\"2018-08-02T14:45:05.752Z\\",\\"lang\\":\\"it\\",\\"result\\":{},\\"alternateResult\\":{},\\"status\\":{\\"code\\":206,\\"errorType\\":\\"partial_content\\",\\"errorDetails\\":\\"Webhook call failed. Error: Failed to parse webhook JSON response: Cannot find field: conversationToken in message google.cloud.dialogflow.v2.WebhookResponse.\\"},\\"sessionId\\":\\"1533221100163\\"}\"\n }]\n }\n }\n}"
The "conversationToken" is filled, so I don't understand the error message.
Maybe I'm trying to perform the operation in a wrong way.
So, which is the correct way to call this helper?
--> I've created a second intent "askGeolocation" that have "actions_intent_PERMISSION" as "Event", and ... if I understand correctly the documentation, should be trigger if the request for helper is correct.
How can I get this working?
UPDATE :
I find some example of the json response for ask permission and seems that it should be different from the one above that i'm using :
https://github.com/dialogflow/fulfillment-webhook-json/blob/master/responses/v2/ActionsOnGoogle/AskForPermission.json
{
"payload": {
"google": {
"expectUserResponse": true,
"systemIntent": {
"intent": "actions.intent.PERMISSION",
"data": {
"#type": "type.googleapis.com/google.actions.v2.PermissionValueSpec",
"optContext": "To deliver your order",
"permissions": [
"NAME",
"DEVICE_PRECISE_LOCATION"
]
}
}
}
}
}
so, i've implemented it and now the response seems to be good (no more error on parsing it), but i still receive an error on it validation :
UnparseableJsonResponse
API Version 2: Failed to parse JSON response string with 'INVALID_ARGUMENT' error: "permission: Cannot find field."
so, a problem still persist.
Anyone know the cause?
Thanks
After some tests i found the problem.
I was returning a wrong json repsonse with "permission" instead of "permissions":
"permission**s**": [
"NAME",
"DEVICE_PRECISE_LOCATION"
]
So the steps to ask for location are correct. I report them here as a little tutorial in order to help who is facing on it for the first time:
1) In DialogFlow, add some logic to your Intent, in order to understand when is ok to ask to user his location. In my case, i've added a "parameter" that identify sentences like "nearby" and so on.
2) When my Intent is fired i receive to my java application a request like this :
...
"queryResult": {
"queryText": "ristorante argentino qui vicino",
"action": "bycategory",
"parameters": {
"askgeolocation": "qui vicino",
"TipoRistorante": ["ristorante", "argentino"]
},
...
3) If "askgeolocation" parameter is filled, instead to return a "simple message" o other type of message, i return a json for ask the permission to geolocation :
{
"payload": {
"google": {
"expectUserResponse": true,
"systemIntent": {
"intent": "actions.intent.PERMISSION",
"data": {
"#type": "type.googleapis.com/google.actions.v2.PermissionValueSpec",
"optContext": "To deliver your order",
"permissions": [
"NAME",
"DEVICE_PRECISE_LOCATION"
]
}
}
}
}
}
4) You MUST have a second Intent that is configured with "actions_intent_PERMISSION " event :
No training phrases
No Action and params
No Responses
But with Fulfillment active :
5) Once your response arrive to Google Assistant this is the message that appear :
6) Now, if user answer "ok" you receive this json on your webhook :
{
"responseId": "32cf46cf-80d8-xxxxxxxxxxxxxxxxxxxxx",
"queryResult": {
"queryText": "actions_intent_PERMISSION",
"action": "geoposition",
"parameters": {
},
"allRequiredParamsPresent": true,
"fulfillmentMessages": [{
"text": {
"text": [""]
}
}],
"outputContexts": [{
"name": "projects/xxxxxxxxxxxxxxxxxxxxx"
}, {
"name": "projects/xxxxxxxxxxxxxxxxxxxxx",
"parameters": {
"PERMISSION": true
}
}, {
"name": "projects/xxxxxxxxxxxxxxxxxxxxx"
}, {
"name": "projects/xxxxxxxxxxxxxxxxxxxxx"
}, {
"name": "projects/xxxxxxxxxxxxxxxxxxxxx"
}, {
"name": "projects/xxxxxxxxxxxxxxxxxxxxx"
}],
"intent": {
"name": "projects/xxxxxxxxxxxxxxxxxxxxx",
"displayName": "geoposition"
},
"intentDetectionConfidence": 1.0,
"languageCode": "it"
},
"originalDetectIntentRequest": {
"source": "google",
"version": "2",
"payload": {
"isInSandbox": true,
"surface": {
"capabilities": [{
"name": "actions.capability.MEDIA_RESPONSE_AUDIO"
}, {
"name": "actions.capability.SCREEN_OUTPUT"
}, {
"name": "actions.capability.AUDIO_OUTPUT"
}, {
"name": "actions.capability.WEB_BROWSER"
}]
},
"requestType": "SIMULATOR",
"inputs": [{
"rawInputs": [{
"inputType": "KEYBOARD"
}],
"arguments": [{
"textValue": "true",
"name": "PERMISSION",
"boolValue": true
}],
"intent": "actions.intent.PERMISSION"
}],
"user": {
"lastSeen": "2018-08-03T08:55:20Z",
"permissions": ["NAME", "DEVICE_PRECISE_LOCATION"],
"profile": {
"displayName": ".... full name of the user ...",
"givenName": "... name ...",
"familyName": "... surname ..."
},
"locale": "it-IT",
"userId": "xxxxxxxxxxxxxxxxxxxxx"
},
"device": {
"location": {
"coordinates": {
"latitude": 45.xxxxxx,
"longitude": 9.xxxxxx
}
}
},
"conversation": {
"conversationId": "xxxxxxxxxxxxxxxxxxxxx",
"type": "ACTIVE",
"conversationToken": "[]"
},
"availableSurfaces": [{
"capabilities": [{
"name": "actions.capability.SCREEN_OUTPUT"
}, {
"name": "actions.capability.AUDIO_OUTPUT"
}, {
"name": "actions.capability.WEB_BROWSER"
}]
}]
}
},
"session": "projects/xxxxxxxxxxxxxxxxxxxxx"
}
that contains, name/surname and latitude/longitude. This information can be saved in your application, in order to not perform again this steps.
I hope this helps.
Davide
In your intent, you can ask for a parameter with a custom Entity. This you can do like this:
entity you can define as "near"
put all the synonyms for near for which you want to trigger location permission in this entity
do not mark this parameter as "required"
do not put any prompt
in the training phrases, add some phrases with this parameter
in your webhook, keep a check on the parameter, if present ask for permission if not continue.
add a permission event to another intent
do your post permission processing in that intent
Entity
Intent
I hope you get it.
There are samples on this topic specifically that guide you through exactly what's needed for requesting permissions in Node and Java.
Note: There are helper intents samples available in Node and Java as well.
I have some documents in the "company" collection structured this way :
[
{
"company_name": "Company 1",
"contacts": {
"main": {
"email": "main#company1.com",
"name": "Mainuser"
},
"store1": {
"email": "store1#company1.com",
"name": "Store1 user"
},
"store2": {
"email": "store2#company1.com",
"name": "Store2 user"
}
}
},
{
"company_name": "Company 2",
"contacts": {
"main": {
"email": "main#company2.com",
"name": "Mainuser"
},
"store1": {
"email": "store1#company2.com",
"name": "Store1 user"
},
"store2": {
"email": "store2#company2.com",
"name": "Store2 user"
}
}
}
]
I'm trying to retrieve the doc that have store1#company2.com as a contact but cannot find how to query a specific value of a specific propertie of an "indexed" list of objects.
My feeling is that the contacts lists should not not be indexed resulting in the following structure :
{
"company_name": "Company 1",
"contacts": [
{
"email": "main#company1.com",
"name": "Mainuser",
"label": "main"
},
{
"email": "store1#company1.com",
"name": "Store1 user",
"label": "store1"
},
{
"email": "store2#company1.com",
"name": "Store2 user",
"label": "store2"
}
]
}
This way I can retrieve matching documents through the following request :
db.company.find({"contacts.email":"main#company1.com"})
But is there anyway to do a similar request on document using the previous structure ?
Thanks a lot for your answers!
P.S. : same question for documents structured this way :
{
"company_name": "Company 1",
"contacts": {
"0": {
"email": "main#company1.com",
"name": "Mainuser"
},
"4": {
"email": "store1#company1.com",
"name": "Store1 user"
},
"1": {
"email": "store2#company1.com",
"name": "Store2 user"
}
}
}
Short answer: yes, they can be queried but it's probably not what you want and it's not going to be really efficient.
The document structure in the first and third block is basically the same - you have an embedded document. The only difference between are the name of the keys in the contacts object.
To query document with that kind of structure you will have to do a query like this:
db.company.find({ $or : [
{"contacts.main.email":"main#company1.com"},
{"contacts.store1.email":"main#company1.com"},
{"contacts.store2.email":"main#company1.com"}
]});
This query will not be efficient, especially if you have a lot of keys in the contacts object. Also, creating a query will be unnecessarily difficult and error prone.
The second document structure, with an array of embedded objects, is optimal. You can create a multikey index on the contacts array which will make your query faster. The bonus is that you can use a short and simple query.
I think the easiest is really to shape your document using the structure describe in your 2nd example : (I have not fixed the JSON)
{
"company_name": "Company 1",
"contacts":{[
{"email":"main#company1.com","name":"Mainuser", "label": "main", ...}
{"email":"store1#company1.com","name":"Store1 user", "label": "store1",...}
{"email":"store2#company1.com","name":"Store2 user", "label": "store2",...}
]}
}
like that you can easily query on email independently of the "label".
So if you really want to use the other structure, (but you need to fix the JSON too) you will have to write more complex code/aggregation pipeline, since we do not know the name and number of attributes when querying the system. Theses structures are also probably hard to use by the developers independently of MongoDB queries.
Since it was not clear let me show what I have in mind
db.company.save(
{
"company_name": "Company 1",
"contacts":[
{"email":"main#company1.com","name":"Mainuser", "label": "main"},
{"email":"store1#company1.com","name":"Store1 user", "label": "store1"},
{"email":"store2#company1.com","name":"Store2 user", "label": "store2"}
]
}
);
db.company.save(
{
"company_name": "Company 2",
"contacts":[
{"email":"main#company2.com","name":"Mainuser", "label": "main"},
{"email":"store1#company2.com","name":"Store1 user", "label": "store1"},
{"email":"store2#company2.com","name":"Store2 user", "label": "store2"}
]
}
);
db.company.ensureIndex( { "contacts.email" : 1 } );
db.company.find( { "contacts.email" : "store1#company2.com" } );
This allows you to store many emails, and query with an index.
I have a working REST request that returns a large results collection. (trimmed here)
The original URL is:
http://intranet.domain.com//_api/SP.UserProfiles.PeopleManager/GetPropertiesFor(accountName=#v)?#v='domain\kens'&$select=AccountName,DisplayName,Email,Title,UserProfileProperties
The response is:
{
"d": {
"__metadata": {
"id": "stuff",
"uri": "morestuff",
"type": "SP.UserProfiles.PersonProperties"
},
"AccountName": "domain\\KenS",
"DisplayName": "Ken Sanchez",
"Email": "KenS#domain.com",
"Title": "Research Assistant",
"UserProfileProperties": {
"results": [
{
"__metadata": {
"type": "SP.KeyValue"
},
"Key": "UserProfile_GUID",
"Value": "1c419284-604e-41a8-906f-ac34fd4068ab",
"ValueType": "Edm.String"
},
{
"__metadata": {
"type": "SP.KeyValue"
},
"Key": "SID",
"Value": "S-1-5-21-2740942301-4273591597-3258045437-1132",
"ValueType": "Edm.String"
},
{
"__metadata": {
"type": "SP.KeyValue"
},
"Key": "ADGuid",
"Value": "",
"ValueType": "Edm.String"
},
{
"__metadata": {
"type": "SP.KeyValue"
},
"Key": "AccountName",
"Value": "domain\\KenS",
"ValueType": "Edm.String"
}...
Is it possible to change the REST request with a $filter that only returns the Key Values from the results collection where Key=SID OR Key= other values?
I only need about 3 values from the results collection by name.
In OData, you can't filter an inner feed.
Instead you could try to query the entity set that UserProfileProperties comes from and expand the associated SP.UserProfiles.PersonProperties entity.
The syntax will need to be adjusted for your scenario, but I'm thinking something along these lines:
service.svc/UserProfileProperties?$filter=Key eq 'SID' and RelatedPersonProperties/AccountName eq 'domain\kens'&$expand=RelatedPersonProperties
That assumes you have a top-level entity set of UserProfileProperties and each is tied back to a single SP.UserProfiles.PersonProperties entity via a navigation property called (in my example) RelatedPersonProperties.
Facebook realtime docs specify that the callback data is of the following format
{
"object": "user",
"entry": [
{
"uid": 1335845740,
"changed_fields": [
"name",
"picture"
],
"time": 232323
}, {
"uid": 1234,
"changed_fields": [
"friends"
],
"time": 232325
}]}
If two different objects, say 'user' and 'page' are to be sent back, does facebook
Batch them together?
Send them separately?
Different objects are sent in different requests. But if there are many user requests then it will batch those user requests and send it. And also you can have different end points for user and page. The format of page request is also different.
It looks something like this
{
"object": "page",
"entry": [
{
"id": "408518775908252",
"time": 1360643280,
"changes": [
{
"field": "feed",
"value": {
"item": "like",
"verb": "add",
"user_id": 5900878
}
}
]
}
]
}
I'm writing an app that gathers info from a users Facebook notifications. Using the Graph Explorer, I request:
me/notifications?include_read=true
and this returns a bunch of data. Each item is a notification item you would see when logging into Facebook. For example (ID/names changed slightly):
{
"id": "notif_630262196_168132987",
"from": {
"name": "John Bloggs",
"id": "822724665"
},
"to": {
"name": "Dermot Bloggs",
"id": "680265196"
},
"created_time": "2013-01-23T22:58:28+0000",
"updated_time": "2013-01-23T22:58:28+0000",
"title": "John Bloggs commented on your link: \"Goodbye Great Barrier Reef. Goodbye...\"",
"link": "http://www.facebook.com/<removed>/posts/330788937030559?comment_id=1702155",
"application": {
"name": "Links",
"id": "2309869772"
},
"unread": 0,
"object": null
}
The notification is in relation to a comment, but the "title:" field gets truncated if it is too long.
Is there a clean way I can programmatically access the comment directly, so I can get all the text, even if it is a 1000 character comment?
Thanks! :-)
You should be able to get the comment by parsing the link field and querying:
/330788937030559_1702155
or
/POSTID_COMMENTID
When I tested this on my account, I am getting a comments object in the returned data, with the full text of the comments. I've got most permissions enabled in my Graph API. I suspect adding read_stream to the permissions is what will give you this data.
You can avoid parsing! I've been doing a lot of experimentation with notifications and getting what I've called the 'source' object (the source object being the Facebook Graph Object from which the notification originates). Unfortunately, I can't find any documentation on how exactly this mechanism works, but if your app has been granted a lot of permissions, the "object" field which in the JSON you posted is null will actually be populated with the 'source object'. Instead of just "object" : null it will look like this:
"object": {
"id": "587140489_588632201147717",
"from": {
"name": "Dave Rodríguez",
"id": "587140489"
},
"message": ":D ",
"picture": "https://fbcdn-vthumb-a.akamaihd.net/hvthumb-ak-prn1/632550_478955408819349_478954732152750_60483_1892_t.jpg",
"link": "https://www.facebook.com/photo.php?v=478954732152750",
"source": "http://video.ak.fbcdn.net/hvideo-ak-prn1/v/754708_478955348819355_1792301950_n.mp4?oh=c7295ccfc3773f24de931e4c29f512ce&oe=513F1728&__gda__=1363127340_9302c108824176369427a0b17491b800",
"name": "¡¡¡EL SECRETO DE LA CHANCLA!!!!",
"description": "EL SECRETO DE LAS MADRES, EL MÁS UTILIZADO EN TODAS LAS GENERACIONES, VÉALO USTED MISMO Y ((COMPARTA))!",
"properties": [
{
"name": "Length",
"text": "1:13"
}
],
"icon": "https://fbstatic-a.akamaihd.net/rsrc.php/v2/yD/r/aS8ecmYRys0.gif",
"actions": [
{
"name": "Comment",
"link": "https://www.facebook.com/587140489/posts/588632201147717"
},
{
"name": "Like",
"link": "https://www.facebook.com/587140489/posts/588632201147717"
}
],
"privacy": {
"value": ""
},
"type": "video",
"status_type": "shared_story",
"object_id": "478954732152750",
"application": {
"name": "Video",
"namespace": "video",
"id": "2392950137"
},
"created_time": "2013-03-10T20:45:20+0000",
"updated_time": "2013-03-10T20:45:20+0000",
"likes": {
"data": [
{
"name": "Adrian Guerra Cuenta Verificada",
"id": "674364748"
}
],
"count": 1
},
"comments": {
"count": 0
}
}
I hope somebody more knowledgeable than me can tell us exactly what permissions we need to see this field populated. I'd figure it out myself but there are so many permissions that brute-forcing the SUM(nCk(77,k),k,1,77)=151115727451828646838271 possible combinations of permissions would take a very long time.