An unofficial X SDK for Python

Installation

corvx works with Python +3.10

You can install the corvx package directly with pip:

pip install -e .

or

pip install corvx

Usage

To work with the X Scraper module you have to import the corresponding module first:

from corvx import Corvx

# You can also set the tokens with environment variables X_AUTH_TOKEN and X_CSRF_TOKEN
corvx = Corvx(
    auth_token=os.getenv('X_AUTH_TOKEN'),
    csrf_token=os.getenv('X_CSRF_TOKEN'),
)

The available methods and its usage are described below.

Query Formats

corvx supports two query formats:

1. Simple String Format

You can directly pass strings as queries:

queries = [
    'python programming',
    'javascript development',
]

for tweet in corvx.search(queries=queries):
    print(tweet)

2. Advanced Query Language

For more complex searches, a JSON-based query language is available. The query must be specified as a Python dictionary containing a list of fields and global options.

Global options

---

lang

This option will force the tweets to match a given language. The language must be specified with its ISO 639-1 two-letter code (e.g., es for Spanish).

since

This parameter refers to the minimum allowed date. It has to be specified in the YYYY-MM-DD format.

until

This parameter refers to the maximum allowed date. It has to be specified in the YYYY-MM-DD format.

near

It has to be specified with a tuple object composed of a text location and a range in miles (e.g., ('Santiago de Compostela', 15)).

Fields

---

A query can specify multiple fields which are Python dictionaries with one or more keys and values:

items

This is a list of strings, either terms or phrases.

exact

If True, the specified terms or phrases must match exactly as they were written on the tweets (case/latin insensitive). If this flag is set, the target parameter will be ignored.

match

If not specified, the tweets will match every item.

• 'any' (the tweets must match at least one of the items)
• 'none' (the tweets won't match any item)

target

If not specified, the tweets will match ordinary keywords.

• 'hashtag' (tweets containing #item)
• 'mention' (tweets mentioning @item)
• 'from' (tweets written by @item)
• 'to' (tweets that are replies to @item)

Examples

---

Simple search with multiple queries:

# Simple string queries
queries = ['python', 'javascript']
for tweet in corvx.search(queries=queries):
    print(tweet)

Advanced search with a single query:

# Advanced query format
query = {
    'fields': [
        {'items': ['Santiago']},
        {'items': ['Chile'], 'match': 'none'},
    ],
    'lang': 'es'
}

for tweet in corvx.search(query=query):
    print(tweet)

Advanced search with multiple queries:

# Multiple advanced queries
queries = [
    {
        'fields': [{'items': ['python'], 'match': 'any'}],
        'lang': 'en'
    },
    {
        'fields': [{'items': ['javascript'], 'match': 'any'}],
        'lang': 'en'
    }
]

for tweet in corvx.search(queries=queries):
    print(tweet)

Search Options

Deep Search

Search for all available results by going back in time:

for tweet in corvx.search(queries=queries, deep=True):
    print(tweet)

Limit Results

Limit the number of results:

for tweet in corvx.search(queries=queries, limit=100):
    print(tweet)

Rate Limiting

Control the time between API calls:

for tweet in corvx.search(queries=queries, sleep_time=30):
    print(tweet)

Stream

Search constantly for new results:

# Stream with multiple queries
queries = ['python', 'javascript']
for tweet in corvx.stream(queries=queries):
    print(tweet)

Debug Logging

Enable debug logging to see detailed information:

import logging

# Configure logging at the start of your script
logging.basicConfig(
    level=logging.DEBUG,
    format='[%(asctime)s] %(levelname)s - %(message)s',
    datefmt='%Y-%m-%d %H:%M:%S',
)

# Configure corvx logger
logger = logging.getLogger('corvx')
logger.setLevel(logging.DEBUG)
logger.propagate = True