No tweepy

[lebowskichill]

I would like to start using twitter apis without tweepy. Does anyone have a code example to get
twitter api and for example a function getting the last 20 tweets from a user?
I am new and i dont understand de twitter apis documentation. I have been using tweepy very well but sometimes i have problems

[IgorBrigadir]

As an alternative to tweepy, GitHub - geduldig/TwitterAPI: Minimal python wrapper for Twitter's REST and Streaming APIs is a good minimal library to use.

[jessicagarson]

Thanks for reaching out. Do you have a code sample you’ve been working with that you are having trouble with?

[lebowskichill]

Sometimes i request for 5000 tweets and i get 3800 or i get two or three tweets with the same datetime.
Also, i dont why the apis documentation says that i can only retrieve tweets with 30-day endpoint and with tweepy i can get tweets from more than 30 days?

def get_tweets(user, n):
    import tweepy
    from tweepy import OAuthHandler
    import json
    '''Devuelve los ultimos n favoritos del usuario indicado.
    Info Favorito:
        -texto
        -fecha y horario del favorito
        -Usuario del tweet al cual se le puso "me gusta"
        -Hashtags presentes en el tweet
    
    Pre: user = type str. Example: 'lebowskichill', WITHOUT "@"
         n = type int.
    Pos: Devuelve una lista de diccionarios
    '''
    
    tweets_extraidos = []
    j = 0 #Cantidad de tweets extraidos
    api = create_twitter_api() # Twitter Api
    try:
        for data in tweepy.Cursor(api.user_timeline, user, tweet_mode= "extended").items(n):
            record = {} #Contenedor de favorito
            data=json.dumps(data._json,indent=2) #Paso de tweepy.models.Status a una String de JSON
            js=json.loads(data)  #Paso del String de json a un dict de json
            
            #Modifico tiempo del tweet a timezone Argentina
            from datetime import datetime, timedelta
            delta_t = timedelta(hours = 3) #Diferencia de horario #UTC/GMT -3 hours
            cadena_horario_fecha = js['created_at']
            date_object = datetime.strptime(cadena_horario_fecha, '%a %b %d %H:%M:%S %z %Y') #Transformo a date object
            date_object = date_object - delta_t #Tranformo a huso horario Bs As
            
            #Recoleccion de fecha y horario de favorito
            record['created_at'] = date_object 

            

[lebowskichill]

   if 'quoted_status' in js.keys():
                #Obtengo el texto del usuario estudiado y el texto del quote
                record['text'] = js['full_text']
                record['quoted_text'] = js['quoted_status']['full_text']
                
                #Datos del usuario estudiado
                record['arroba_name'] = js['user']['screen_name']
                record['user_name'] = js['user']['name']
                record['user_id'] = js['user']['id']
                
                #Datos del usuario quoted
                record['quoted_arroba_name'] = js['quoted_status']['user']['screen_name']
                record['quoted_user_name'] = js['quoted_status']['user']['name']
                record['quoted_user_id'] = js['quoted_status']['user']['id']
                
                
                #Informo que es un quoted_tweet
                record['type'] = 'quoted'
                
            else: #Tweet unico (Retweet o tweet)
                
        
                
                #En caso de ser un Tweet:
                if 'RT @' not in js['full_text']:
                    
                    record['text'] = js['full_text'] #Extraigo texto del "tweet"
                    #Datos del usuario del tweet
                    record['arroba_name'] = js['user']['screen_name']
                    record['user_name'] = js['user']['name']
                    record['user_id'] = js['user']['id']
                    
                    #Si es una respuesta o un tweet unitario:
                    if js['in_reply_to_screen_name'] == None:
                        record['type'] = 'tweet'
                    else:    
                        record['type'] = 'reply'
                        
                #En caso de ser un Retweet:
                else:
                    record['text'] = js['retweeted_status']['full_text'] #Extraigo texto de "retweet"
                    #Datos del usuario retweetiado
                    record['arroba_name'] = js['retweeted_status']['user']['screen_name']
                    record['user_name'] = js['retweeted_status']['user']['name']
                    record['user_id'] = js['retweeted_status']['user']['id']      
                    
                    record['type'] = 'retweet'
                
                #El tweet o retweet al no ser quoted_tweets:
                
                record['quoted_arroba_name'] = False
                record['quoted_user_name'] = False
                record['quoted_user_id'] = False
                record['quoted_text'] = False
                
                    
                
                    
            #Recolección de Hashtags:
            hashtags = []    
            for d in js['entities']['hashtags']:
                hashtags.append(d['text'])
            record['hashtags'] = hashtags            
                
            #Meto tweet en la lista
            tweets_extraidos.append(record) 
            j+=1  #Se extrajo un nuevo tweet
    finally:  #Si me paso de solicitudes
        
        #Requests disponibles luego de utilizar api.favorites  
        data2=api.rate_limit_status()
        print('Api.favorites solicitudes restantes:', data2["resources"]['statuses']['/statuses/user_timeline']['remaining'])
        print(f'Se extrajeron {j} tweets')
            
            
        return tweets_extraidos

[andypiper]

Just to be clear here:

• the user timeline v1.1 API (Twitter API) provides up to 3200 Tweets, which could be more than 30 days … this is based on the code sample you’ve provided.
• you mention “30 day endpoint”, this is a premium search feature that Tweepy does not (to my knowledge?) support.
• in the future, as we move to v2 API, Tweepy may not support the current functionality, you may want to seek out a library that works with the modern API.

Tweepy itself is not provided or supported by Twitter, so for some of these questions you would be best checking with the community around that library.

As @IgorBrigadir said there are indeed other Python libraries, including TwitterAPI.

[lebowskichill]

Thank you!

[lebowskichill]

Thanks!!