search_answers.py

"""
Accept the error from another function.
Functions:
- Get questions based on the error_message
- Get top 5 answers from the site based on the question ids
"""

from dotenv import load_dotenv
load_dotenv()
import os
import requests
import urllib.request
import json
from bs4 import BeautifulSoup
from get_error import main
import click
import csv
import io


class ask():

    def __init__(self,error_list):
        self.key=os.environ.get('key')
        self.client_id=os.environ.get('client_id')
        self.answer_filter='!bN4iJfRmwXV)w1'
        #error_message is a list of errors
        self.error_list=error_list
        self.error_message=' '.join(error_list)
        #levy's hardcoded line
        #self.error_message='unsupported operand type(s) for int and str'
    
    def get_question_id(self):
        """
        returns top 5 question id's
    
        """

        url=f'https://api.stackexchange.com/2.2/search/advanced?site=stackoverflow.com&q={self.error_message}' 
        jData=requests.get(url)
        data=jData.json()
        question_ids=[]
        for i in data['items']:
            ques_id=i.get('question_id')
            question_ids.append(str(ques_id))
        #query accepts ids only delimited by semicolon
        return ';'.join(question_ids)
        
    def get_answer(self,count=5):
        """
        returns answers based on question id's given by get_function_id()
        """
        ids=self.get_question_id()

        if self.key:
            query=f'https://api.stackexchange.com/2.2/questions/{ids}/answers?client_id={self.client_id}&site=stackoverflow&key={self.key}&filter={self.answer_filter}'
        else:
            query=f'https://api.stackexchange.com/2.2/questions/{ids}/answers?&site=stackoverflow&filter={self.answer_filter}'


        jData=requests.get(query)
        data=jData.json()
        answers=[]
        c=0


        for i in data['items']:   
                ans=i.get('body')
                soup = BeautifulSoup(ans,features="html.parser")

                #levys
                answers.append(soup.get_text())
                '''
                #moved to cli.py
                print('\n')
                click.echo(click.style(f"        ************** ", fg=colors[0]))
                click.echo(click.style(soup.get_text(), fg=colors[1]))
                '''
                #arlyns
                #answers.append(soup.get_text().replace('\n',' '))


                c+=1
                if c==count:
                    break


        # opening the csv file in 'w' mode 
        file = io.open('g4g.csv', 'w', newline ='', encoding="utf-8") 
          
        with file: 
            # identifying header   
            header = ['Number', 'Answer', 'Current'] 
            writer = csv.DictWriter(file, fieldnames = header) 
            writer.writeheader() 
            # writing data row-wise into the csv file
            count = 1
            for answer in answers:
                if count==1:
                    writer.writerow({'Number' : count, 'Answer': answer, 'Current': True})
                else:
                    writer.writerow({'Number' : count, 'Answer': answer, 'Current': False})
                count += 1
        
        return answers