API.ai stopped sending to webhook? - dialogflow-es

The _actions_on_google_ context stopped appearing in the contexts section, and I think this may be a clue to what's gone wrong.
It's just stopped doing anything, it'll get the required parameter then the JSON says it's sent it to webhook but there's only the 'secure' context in that (I remember seeing an actions on google one by default). When I remove my custom context it doesn't help, and I've updated firebase.
How do I get this back?

I think the issue was that I had deleted something in the code, maybe the }); at the end.
If you have the same issue, go OCD through your code and match everything together to check it's all good.

Related

Receiving “(my app name) isn't responding right now. Please try again soon” on Actions on google

Keep receiving “(my app name) isn't responding right now. Please try again soon” on my actions on google simulator, I’ve tried almost all the solution that I found on the internet and ended up with this, I decided to copy the code from google https://developers.google.com/assistant/df-asdk/helpers#calling_the_helper it works just the permission one , for the permission handler it doesn’t work it seems like if I do not put the default response in dialogflow It wouldn’t work so I don’t want that default response because it keeps saying that response, and when I removed it , it gave me an error ,but in the debug tap it shows everything(display name, location etc.)

Gmail API Watch Request Not Filtering Label IDs

Issue
Trying to get the Gmail API to send only new messages to Cloud Pub/Sub by using the Gmail APIs Watch function but when setting the labelIds in order to filter to just INBOX, i am still notified of all changes, files sent, drafts etc.
I've looked online and can see that there are people that are experiencing this issue and there is even a bug out for it https://issuetracker.google.com/issues/36759803.
But i find it kind of odd that the entire feature of filtering just doesn't work and its been an issue for 5 years according to the bug report. Makes me think its just me that is doing something wrong. Would be cool to hear from anyone that has got this to work.
Where im at
I'm using the Gmail API, trying to get my Gmail account to publish messages to Cloud Pub/Sub. I've managed to do this however im trying to filter what i get notifications for to just new messages. This is where i have a problem.
According to this Gmail API documentation, it appears all i have to call is the code below and that should be that. I will add however, that im calling the Kotlin equivalent, but i don't think this should matter.
request = {
'labelIds': ['INBOX'],
'topicName': 'projects/myproject/topics/mytopic'
}
gmail.users().watch(userId='me', body=request).execute()
What im calling:
val request = WatchRequest().apply {
labelIds = listOf("INBOX")
topicName = "projects/myproject/topics/mytopic"
}
gmail.users().watch("me", request).execute()
Other things ive tried
I have tried adding labelFilterAction = Include and other variations but they all seem to be ignored.
After trying the ideas listed by ziganotschka, I'm more confident that the issue is not related to specific client libraries.
I managed to speak to someone at Google who in turn contacted the G-mail API team. They said they were aware that this is an issue, that it has been an issue since 2015 and was most likely caused by a change they made back then.
A ticket is still open, but doesn't look like it will be fixed.
Perform the following trouble-shooting actions
Provide a request body including labelIds, labelFilterAction AND topicName as specified here
Try the label "UNREAD" instead of "INBOX"
To discart Kotlin implementation related issues, test your request in a programming language that allows you to pass the request as a JSON body in exactly the same way as specified in the documentation.

Slack #here mention in channel via Incoming webhook

So I have an alert channel that collects messages from slack via incoming webhook. It's one of many tools we use to monitor these couple of apps that normally run without issue, but if there is a problem we want to know right away so we can do something about it. I'm getting emails, etc. but I also want the channel wide slack messages.
What I'm trying to do seems simple enough, but it's not working. I want to trigger an #channel or #here message to alert all the users in the channel at the time, so if the error happens in an environment that matters everyone gets a notification. I'm following examples in the documentation using the <#ABC123> channel id but it's not translating into an #here directive, it's just plain text.
Either what I'm doing can't be done, or more likely I'm missing something somewhere or approaching this incorrectly. I'm guessing it's the latter, but at this point confirmation that I can't do what I'm trying to do is fine, I'll come up with something else.
Found it
Apparently the convention for the channel alerts are <!channel>, not <#C123> as is with users or channel ids, or <#channel> with channel mentions (though that does work, just not as a channel wide alert). Hope this helps someone eventually.

User account email verification links not working

I am having a lot of troubles getting verification emails to work, when a user signs up with parse-server (using MailGun & Heroku).
The verification link sent to the user has this form:
https://myapp.herokuapp.com/apps/.............
but it does not work/
I get this, when clicking it:
Cannot GET /apps/.......
And even if I type only this in the web browser:
https://myapp.herokuapp.com/apps/
I still get this:
Cannot GET /apps/
At one point I was suspecting I could have made some confusion between parse-server-mailgun and parse-server-simple-mailgun-adapter (I am in fact not fully aware of the difference between the two), so I tried both but I still got the same kind of error.
Since, I can reach the point where a user account is actually created and a verification email is sent, I suppose I must have only made some detail mistake in my settings; but I can't see where. Any help or relevant hint from some experienced person would be highly apreciated.
In case someone think I should share some code because the explanation above is not enough, please tell me what you need to know to clarifly my point. And I will provide the necessary information.
You should see your publicServerURL to https://myapp.heroku.com/parse

Mozpay hitting postbackURL 6 times

I have set up mozpay in my app. When I send the signed jwt with the post-back URL to the app, and then the app uses it with mozpay, during simulation, the post-back URL gets hit 6 times. I wanted to know if that's intended. If so i can work around it, but if its not intended i would rather fix that problem.
Is it possible that the 6 post-backs only occur during simulation?
Also is this any different in the charge-back URL?
Is there a setting that I can set in the signed jwt to change this?
If there is some documentation on this, I haven't been able to find it, and I would greatly appreciate a link.
What you describe is expected behavior. Since your postback URL was not responding correctly, Mozilla's server retried the post 5 additional times before giving up. As you mention, when you fixed the transaction response you no longer saw so many requests.
There was an answer here, but I dont see it anymore. The answer was fairly correct.
We were sending back "webpay: {transactionID}" what we didnt realize is that {transactionID} includes the text "webpay:" So all we had to do was send back "{transactionID}" instead.
I would set as the answer the person who posted, but i no longer see the post. Thanks.

Resources