From 5a1dbc04e196c416342b70ff8731a4257721701d Mon Sep 17 00:00:00 2001 From: fluzzi Date: Fri, 12 May 2023 12:05:25 -0300 Subject: [PATCH] add confirmation and fixes --- README.md | 2 + connpy/__init__.py | 2 + connpy/_version.py | 2 +- connpy/ai.py | 115 ++++++++++++++++++++-- connpy/api.py | 13 ++- docs/connpy/index.html | 211 +++++++++++++++++++++++++++++++++++++---- 6 files changed, 320 insertions(+), 25 deletions(-) diff --git a/README.md b/README.md index 719ccb3..fffec94 100644 --- a/README.md +++ b/README.md @@ -246,6 +246,8 @@ With the Connpy API you can run commands on devices using http requests - A JSON object with the results of the executed commands on the nodes. +--- + ### 4. Ask AI **Endpoint**: `/ask_ai` diff --git a/connpy/__init__.py b/connpy/__init__.py index 266b1c0..0031366 100644 --- a/connpy/__init__.py +++ b/connpy/__init__.py @@ -152,6 +152,8 @@ With the Connpy API you can run commands on devices using http requests - A JSON object with the results of the executed commands on the nodes. +--- + ### 4. Ask AI **Endpoint**: `/ask_ai` diff --git a/connpy/_version.py b/connpy/_version.py index c2cc2aa..05103bc 100644 --- a/connpy/_version.py +++ b/connpy/_version.py @@ -1,2 +1,2 @@ -__version__ = "3.2.2" +__version__ = "3.2.3" diff --git a/connpy/ai.py b/connpy/ai.py index 5acbc42..7eb9513 100755 --- a/connpy/ai.py +++ b/connpy/ai.py @@ -1,9 +1,11 @@ import openai +import time import json import re import ast from textwrap import dedent from .core import nodes +from copy import deepcopy class ai: ''' This class generates a ai object. Containts all the information and methods to make requests to openAI chatGPT to run actions on the application. @@ -116,9 +118,34 @@ class ai: self.__prompt["command_assistant"]= """ router1: ['show running-config'] """ + self.__prompt["confirmation_system"] = """ + Please analyze the user's input and categorize it as either an affirmation or negation. Based on this analysis, respond with: + + 'True' if the input is an affirmation like 'do it', 'go ahead', 'sure', etc. + 'False' if the input is a negation. + If the input does not fit into either of these categories, kindly express that you didn't understand and request the user to rephrase their response. + """ + self.__prompt["confirmation_user"] = "Yes go ahead!" + self.__prompt["confirmation_assistant"] = "True" self.model = model self.temp = temp + def _retry_function(self, function, max_retries, backoff_num, *args): + #Retry openai requests + retries = 0 + while retries < max_retries: + try: + myfunction = function(*args) + break + except (openai.error.APIConnectionError, openai.error.RateLimitError): + wait_time = backoff_num * (2 ** retries) + time.sleep(wait_time) + retries += 1 + continue + if retries == max_retries: + myfunction = False + return myfunction + def _clean_original_response(self, raw_response): #Parse response for first request to openAI GPT. info_dict = {} @@ -178,6 +205,14 @@ class ai: pass return info_dict + def _clean_confirmation_response(self, raw_response): + #Parse response for confirmation request to openAI GPT. + value = raw_response.strip() + if value.strip(".").lower() == "true": + value = True + elif value.strip(".").lower() == "false": + value = False + return value def _get_commands(self, user_input, nodes): #Send the request for commands for each device to openAI GPT. @@ -233,7 +268,51 @@ class ai: output["response"] = self._clean_original_response(output["raw_response"]) return output - def ask(self, user_input, dryrun = False, chat_history = None): + def _get_confirmation(self, user_input): + #Send the request to identify if user is confirming or denying the task + message = [] + message.append({"role": "system", "content": dedent(self.__prompt["confirmation_system"])}) + message.append({"role": "user", "content": dedent(self.__prompt["confirmation_user"])}) + message.append({"role": "assistant", "content": dedent(self.__prompt["confirmation_assistant"])}) + message.append({"role": "user", "content": user_input}) + response = openai.ChatCompletion.create( + model=self.model, + messages=message, + temperature=self.temp, + top_p=1 + ) + output = {} + output["dict_response"] = response + output["raw_response"] = response["choices"][0]["message"]["content"] + output["response"] = self._clean_confirmation_response(output["raw_response"]) + return output + + def confirm(self, user_input, max_retries=3, backoff_num=1): + ''' + Send the user input to openAI GPT and verify if response is afirmative or negative. + + ### Parameters: + + - user_input (str): User response confirming or denying. + + ### Optional Parameters: + + - max_retries (int): Maximum number of retries for gpt api. + - backoff_num (int): Backoff factor for exponential wait time + between retries. + + ### Returns: + + bool or str: True, False or str if AI coudn't understand the response + ''' + result = self._retry_function(self._get_confirmation, max_retries, backoff_num, user_input) + if result: + output = result["response"] + else: + output = f"{self.model} api is not responding right now, please try again later." + return output + + def ask(self, user_input, dryrun = False, chat_history = None, max_retries=3, backoff_num=1): ''' Send the user input to openAI GPT and parse the response to run an action in the application. @@ -250,9 +329,13 @@ class ai: ### Optional Parameters: - - dryrun (bool): Set to true to get the arguments to use to run - in the app. Default is false and it will run - the actions directly. + - dryrun (bool): Set to true to get the arguments to use to + run in the app. Default is false and it + will run the actions directly. + - chat_history (list): List in gpt api format for the chat history. + - max_retries (int): Maximum number of retries for gpt api. + - backoff_num (int): Backoff factor for exponential wait time + between retries. ### Returns: @@ -279,7 +362,11 @@ class ai: ''' output = {} - original = self._get_filter(user_input, chat_history) + original = self._retry_function(self._get_filter, max_retries, backoff_num, user_input, chat_history) + if not original: + output["app_related"] = False + output["response"] = f"{self.model} api is not responding right now, please try again later." + return output output["input"] = user_input output["app_related"] = original["response"]["app_related"] output["dryrun"] = dryrun @@ -301,7 +388,12 @@ class ai: if not type == "command": output["action"] = "list_nodes" else: - commands = self._get_commands(user_input, thisnodes) + commands = self._retry_function(self._get_commands, max_retries, backoff_num, user_input, thisnodes) + if not commands: + output = [] + output["app_related"] = False + output["response"] = f"{self.model} api is not responding right now, please try again later." + return output output["args"] = {} output["args"]["commands"] = commands["response"]["commands"] output["args"]["vars"] = commands["response"]["variables"] @@ -310,10 +402,21 @@ class ai: output["action"] = "test" else: output["action"] = "run" + if dryrun: + output["task"] = [] + if output["action"] == "test": + output["task"].append({"Task": "Verify if expected value is in command(s) output"}) + output["task"].append({"Expected value to verify": output["args"]["expected"]}) + elif output["action"] == "run": + output["task"].append({"Task": "Run command(s) on devices and return output"}) + varstocommands = deepcopy(output["args"]["vars"]) + del varstocommands["__global__"] + output["task"].append({"Devices": varstocommands}) if not dryrun: mynodes = nodes(self.config.getitems(output["nodes"]),config=self.config) if output["action"] == "test": output["result"] = mynodes.test(**output["args"]) + output["logs"] = mynodes.output elif output["action"] == "run": output["result"] = mynodes.run(**output["args"]) return output diff --git a/connpy/api.py b/connpy/api.py index 6d4b0e7..9bcffbb 100755 --- a/connpy/api.py +++ b/connpy/api.py @@ -70,6 +70,13 @@ def ask_ai(): ai = myai(conf) return ai.ask(input, dryrun, chat_history) +@app.route("/confirm", methods=["POST"]) +def confirm(): + conf = app.custom_config + data = request.get_json() + input = data["input"] + ai = myai(conf) + return str(ai.confirm(input)) @app.route("/run_commands", methods=["POST"]) def run_commands(): @@ -99,7 +106,7 @@ def run_commands(): mynodes = nodes(mynodes, config=conf) try: - args["vars"] = data["variables"] + args["vars"] = data["vars"] except: pass try: @@ -111,7 +118,9 @@ def run_commands(): if action == "run": output = mynodes.run(**args) elif action == "test": - output = mynodes.test(**args) + output = {} + output["result"] = mynodes.test(**args) + output["output"] = mynodes.output else: error = "Wrong action '{}'".format(action) return({"DataError": error}) diff --git a/docs/connpy/index.html b/docs/connpy/index.html index 28b3bf7..d400042 100644 --- a/docs/connpy/index.html +++ b/docs/connpy/index.html @@ -147,6 +147,7 @@ options: +

4. Ask AI

Endpoint: /ask_ai

Method: POST

@@ -401,6 +402,8 @@ With the Connpy API you can run commands on devices using http requests - A JSON object with the results of the executed commands on the nodes. +--- + ### 4. Ask AI **Endpoint**: `/ask_ai` @@ -682,9 +685,34 @@ __pdoc__ = { self.__prompt["command_assistant"]= """ router1: ['show running-config'] """ + self.__prompt["confirmation_system"] = """ + Please analyze the user's input and categorize it as either an affirmation or negation. Based on this analysis, respond with: + + 'True' if the input is an affirmation like 'do it', 'go ahead', 'sure', etc. + 'False' if the input is a negation. + If the input does not fit into either of these categories, kindly express that you didn't understand and request the user to rephrase their response. + """ + self.__prompt["confirmation_user"] = "Yes go ahead!" + self.__prompt["confirmation_assistant"] = "True" self.model = model self.temp = temp + def _retry_function(self, function, max_retries, backoff_num, *args): + #Retry openai requests + retries = 0 + while retries < max_retries: + try: + myfunction = function(*args) + break + except (openai.error.APIConnectionError, openai.error.RateLimitError): + wait_time = backoff_num * (2 ** retries) + time.sleep(wait_time) + retries += 1 + continue + if retries == max_retries: + myfunction = False + return myfunction + def _clean_original_response(self, raw_response): #Parse response for first request to openAI GPT. info_dict = {} @@ -744,6 +772,14 @@ __pdoc__ = { pass return info_dict + def _clean_confirmation_response(self, raw_response): + #Parse response for confirmation request to openAI GPT. + value = raw_response.strip() + if value.strip(".").lower() == "true": + value = True + elif value.strip(".").lower() == "false": + value = False + return value def _get_commands(self, user_input, nodes): #Send the request for commands for each device to openAI GPT. @@ -799,7 +835,51 @@ __pdoc__ = { output["response"] = self._clean_original_response(output["raw_response"]) return output - def ask(self, user_input, dryrun = False, chat_history = None): + def _get_confirmation(self, user_input): + #Send the request to identify if user is confirming or denying the task + message = [] + message.append({"role": "system", "content": dedent(self.__prompt["confirmation_system"])}) + message.append({"role": "user", "content": dedent(self.__prompt["confirmation_user"])}) + message.append({"role": "assistant", "content": dedent(self.__prompt["confirmation_assistant"])}) + message.append({"role": "user", "content": user_input}) + response = openai.ChatCompletion.create( + model=self.model, + messages=message, + temperature=self.temp, + top_p=1 + ) + output = {} + output["dict_response"] = response + output["raw_response"] = response["choices"][0]["message"]["content"] + output["response"] = self._clean_confirmation_response(output["raw_response"]) + return output + + def confirm(self, user_input, max_retries=3, backoff_num=1): + ''' + Send the user input to openAI GPT and verify if response is afirmative or negative. + + ### Parameters: + + - user_input (str): User response confirming or denying. + + ### Optional Parameters: + + - max_retries (int): Maximum number of retries for gpt api. + - backoff_num (int): Backoff factor for exponential wait time + between retries. + + ### Returns: + + bool or str: True, False or str if AI coudn't understand the response + ''' + result = self._retry_function(self._get_confirmation, max_retries, backoff_num, user_input) + if result: + output = result["response"] + else: + output = f"{self.model} api is not responding right now, please try again later." + return output + + def ask(self, user_input, dryrun = False, chat_history = None, max_retries=3, backoff_num=1): ''' Send the user input to openAI GPT and parse the response to run an action in the application. @@ -816,9 +896,13 @@ __pdoc__ = { ### Optional Parameters: - - dryrun (bool): Set to true to get the arguments to use to run - in the app. Default is false and it will run - the actions directly. + - dryrun (bool): Set to true to get the arguments to use to + run in the app. Default is false and it + will run the actions directly. + - chat_history (list): List in gpt api format for the chat history. + - max_retries (int): Maximum number of retries for gpt api. + - backoff_num (int): Backoff factor for exponential wait time + between retries. ### Returns: @@ -845,7 +929,11 @@ __pdoc__ = { ''' output = {} - original = self._get_filter(user_input, chat_history) + original = self._retry_function(self._get_filter, max_retries, backoff_num, user_input, chat_history) + if not original: + output["app_related"] = False + output["response"] = f"{self.model} api is not responding right now, please try again later." + return output output["input"] = user_input output["app_related"] = original["response"]["app_related"] output["dryrun"] = dryrun @@ -867,7 +955,12 @@ __pdoc__ = { if not type == "command": output["action"] = "list_nodes" else: - commands = self._get_commands(user_input, thisnodes) + commands = self._retry_function(self._get_commands, max_retries, backoff_num, user_input, thisnodes) + if not commands: + output = [] + output["app_related"] = False + output["response"] = f"{self.model} api is not responding right now, please try again later." + return output output["args"] = {} output["args"]["commands"] = commands["response"]["commands"] output["args"]["vars"] = commands["response"]["variables"] @@ -876,10 +969,21 @@ __pdoc__ = { output["action"] = "test" else: output["action"] = "run" + if dryrun: + output["task"] = [] + if output["action"] == "test": + output["task"].append({"Task": "Verify if expected value is in command(s) output"}) + output["task"].append({"Expected value to verify": output["args"]["expected"]}) + elif output["action"] == "run": + output["task"].append({"Task": "Run command(s) on devices and return output"}) + varstocommands = deepcopy(output["args"]["vars"]) + del varstocommands["__global__"] + output["task"].append({"Devices": varstocommands}) if not dryrun: mynodes = nodes(self.config.getitems(output["nodes"]),config=self.config) if output["action"] == "test": output["result"] = mynodes.test(**output["args"]) + output["logs"] = mynodes.output elif output["action"] == "run": output["result"] = mynodes.run(**output["args"]) return output @@ -887,7 +991,7 @@ __pdoc__ = {

Methods

-def ask(self, user_input, dryrun=False, chat_history=None) +def ask(self, user_input, dryrun=False, chat_history=None, max_retries=3, backoff_num=1)

Send the user input to openAI GPT and parse the response to run an action in the application.

@@ -902,9 +1006,13 @@ __pdoc__ = { expected value.

Optional Parameters:

-
- dryrun  (bool): Set to true to get the arguments to use to run
-                  in the app. Default is false and it will run 
-                  the actions directly.
+
- dryrun       (bool): Set to true to get the arguments to use to
+                       run in the app. Default is false and it
+                       will run the actions directly.
+- chat_history (list): List in gpt api format for the chat history.
+- max_retries   (int): Maximum number of retries for gpt api.
+- backoff_num   (int): Backoff factor for exponential wait time
+                       between retries.
 

 
Returns:

 
dict: Dictionary formed with the following keys:
@@ -930,7 +1038,7 @@ __pdoc__ = {
 
 Expand source code
 
-
def ask(self, user_input, dryrun = False, chat_history = None):
+
def ask(self, user_input, dryrun = False, chat_history = None,  max_retries=3, backoff_num=1):
     '''
     Send the user input to openAI GPT and parse the response to run an action in the application.
 
@@ -947,9 +1055,13 @@ __pdoc__ = {
 
     ### Optional Parameters:  
 
-        - dryrun  (bool): Set to true to get the arguments to use to run
-                          in the app. Default is false and it will run 
-                          the actions directly.
+        - dryrun       (bool): Set to true to get the arguments to use to
+                               run in the app. Default is false and it
+                               will run the actions directly.
+        - chat_history (list): List in gpt api format for the chat history.
+        - max_retries   (int): Maximum number of retries for gpt api.
+        - backoff_num   (int): Backoff factor for exponential wait time
+                               between retries.
 
     ### Returns:  
 
@@ -976,7 +1088,11 @@ __pdoc__ = {
 
     '''
     output = {}
-    original = self._get_filter(user_input, chat_history)
+    original = self._retry_function(self._get_filter, max_retries, backoff_num, user_input, chat_history)
+    if not original:
+        output["app_related"] = False
+        output["response"] = f"{self.model} api is not responding right now, please try again later."
+        return output
     output["input"] = user_input
     output["app_related"] = original["response"]["app_related"]
     output["dryrun"] = dryrun
@@ -998,7 +1114,12 @@ __pdoc__ = {
         if not type == "command":
             output["action"] = "list_nodes"
         else:
-            commands = self._get_commands(user_input, thisnodes)
+            commands = self._retry_function(self._get_commands, max_retries, backoff_num, user_input, thisnodes)
+            if not commands:
+                output = []
+                output["app_related"] = False
+                output["response"] = f"{self.model} api is not responding right now, please try again later."
+                return output
             output["args"] = {}
             output["args"]["commands"] = commands["response"]["commands"]
             output["args"]["vars"] = commands["response"]["variables"]
@@ -1007,15 +1128,72 @@ __pdoc__ = {
                 output["action"] = "test"
             else:
                 output["action"] = "run"
+            if dryrun:
+                output["task"] = []
+                if output["action"] == "test":
+                    output["task"].append({"Task": "Verify if expected value is in command(s) output"})
+                    output["task"].append({"Expected value to verify": output["args"]["expected"]})
+                elif output["action"] == "run":
+                    output["task"].append({"Task": "Run command(s) on devices and return output"})
+                varstocommands = deepcopy(output["args"]["vars"])
+                del varstocommands["__global__"]
+                output["task"].append({"Devices": varstocommands})
             if not dryrun:
                 mynodes = nodes(self.config.getitems(output["nodes"]),config=self.config)
                 if output["action"] == "test":
                     output["result"] = mynodes.test(**output["args"])
+                    output["logs"] = mynodes.output
                 elif output["action"] == "run":
                     output["result"] = mynodes.run(**output["args"])
     return output
+
+def confirm(self, user_input, max_retries=3, backoff_num=1) +
+
+

Send the user input to openAI GPT and verify if response is afirmative or negative.

+

Parameters:

+
- user_input (str): User response confirming or denying.
+
+

Optional Parameters:

+
- max_retries (int): Maximum number of retries for gpt api.
+- backoff_num (int): Backoff factor for exponential wait time
+                     between retries.
+
+

Returns:

+
bool or str: True, False or str if AI coudn't understand the response
+
+
+ +Expand source code + +
def confirm(self, user_input, max_retries=3, backoff_num=1):
+    '''
+    Send the user input to openAI GPT and verify if response is afirmative or negative.
+
+    ### Parameters:  
+
+        - user_input (str): User response confirming or denying.
+
+    ### Optional Parameters:  
+
+        - max_retries (int): Maximum number of retries for gpt api.
+        - backoff_num (int): Backoff factor for exponential wait time
+                             between retries.
+
+    ### Returns:  
+
+        bool or str: True, False or str if AI coudn't understand the response
+    '''
+    result = self._retry_function(self._get_confirmation, max_retries, backoff_num, user_input)
+    if result:
+        output = result["response"]
+    else:
+        output = f"{self.model} api is not responding right now, please try again later."
+    return output
+
+
@@ -4171,6 +4349,7 @@ tasks:

ai