When creating an automatic trading BOT for virtual currency When using the API of the exchange, errors are frequently detected on the exchange side. Exception handling was implemented as a countermeasure. I will leave it as a memo for myself so that I will not forget it.
It is a simple countermeasure such as server error.
Even if the code you write is correct, you may get an error at runtime.
For example, when you make a request to an external server, the server is busy or
For example, if the server is temporarily under maintenance and the server is not running.
A server error (error code) etc. will be returned as a response result.
Especially for programs that run automatically, such as automatic trading BOT
It would be a problem if it was stopped every time due to a server error. .. .. (No longer an automatic trading BOT)
As a countermeasure, you can prevent unexpected errors by writing a try-except statement (exception handling).
You can check whether the correct response to the squirrel quest was obtained by the HTTP status code. The following is a part of the HTTP status code, and the major classification is like this. I often got errors in the 500s.
--100s: Processing --200s: Success --300s: Redirect --400s: Client error --500s: Server error
sample.py
response = requests.get("https://***********")
print(response.status_code)
You can check the HTTP status code with Response object .status_code like the above code.
sample.py
try:
response = requests.get("https://***********")
response.raise_for_status() #
print(response.status_code)
except requests.exceptions.RequestException as e:
print("error: ",e)
An exception may occur in the try: part, but the process you want to execute
ʻExcept: Describe the processing when an exception occurs in the` part.
In the case of the above code, request communication is made to the external server by GET method and the response result is
Check the HTTP status code with the raise_for_status () method, and if it is in the 200s, only the try: part will be executed.
On the contrary, if the HTTP status code is other than the 200s, it is considered that an exception has occurred.
ʻExcept:` part is executed.
The raise_for_status () method checks if the HTTP status code is in the 200s and
If it is not in the 200s, it raises an exception, that is, it spits out an error and stops.
except is written as ʻexcept exception type as variable name: Exception objects (such as error messages) can be stored and used in variables. The variable name can be any name, but it is common to write ʻe or ʻerr`.
You can check the content of the error by outputting the variable.
For exception types, search for "library name exception handling", "library name error handling", etc.
Most of the time you may want to know.
sample.py
try:
response = requests.get("https://***********")
response.raise_for_status()
except requests.exceptions.RequestException as e:
print("error: ",e)
else:
print("Processed successfully")
In the ʻelse: part, describe the processing to be performed after normal completion without exception. If an exception occurs and the ʻexcept: part is executed, the ʻelse:` part is not executed.
sample.py
try:
response = requests.get("https://***********")
response.raise_for_status()
except requests.exceptions.RequestException as e:
print("error: ",e)
else:
print("Processed successfully")
finally:
print("Always process on exit")
In the finally: part, describe the process you want to do last regardless of the occurrence of an exception.
You can also use ʻelse:andfinally:at the same time. At the time of normal termination, the process offinally: is executed after the process of ʻelse: is executed.
Recommended Posts