microdot - request.body

[matthimue]

I wonder about the following .....

• I send a POST with 'content-type text/plain' and only 1 letter 'D' in the body (verified in browser)
• microdot 'request.body' is set to a byte object - b'"D"' - of length 3, counting and including double quotes
• I would expect 'D' only of length 1 as a string, don't counting the doublequotes. Why is that?
• What is the 'pythonic way to get the 'D' back'' here. Setting a = b'"D"' and a.decode() in turn does not help, doing eval(a) instead helps. is there another/better way of getting the plain text back, whatever it is?

Kind regards
Matthias

[miguelgrinberg]

My guess is that your client is converting your D to a JSON string, which would be "D". Any chance of that? I can look into this further if you think this is definitely a problem on the Microdot side, but I don't think it is.

[miguelgrinberg]

Actually, I just tested this and I do not see the behavior you reported. Here is the endpoint that I used:

@app.post('/test')
async def test(request):
    print(request.body, len(request.body))
    return '', 200

The client request:

curl -X POST http://localhost:5000/test -d 'D' -H 'Content-Type: text/plain'

And the output:

Starting async server on 0.0.0.0:5000...
b'D' 1
POST /test 200

[matthimue]

Yes you're right - of course - thanks for help

I found the line in the javascript code/fetch API, where the data indeed has been modified to be represented with double quotes. The devtools tab showed content-length 1, so I didn't assume the quotes to be sent, as I understand string literals can be defined with both single/double quotes both in javascript and python.

So yes, I get request.body as b'D' and with request.body.decode() I got the string back