# AVAPI
### Get data from Alpha Vantage into python.
Authors:
- Ole Olaussen: ole.olaussen@ekkobit.com
- Xuan Ling: xuan.ling@ekkobit.com
## Documentation
For more complete documentation, please go to [ReadTheDocs](https://avapi.readthedocs.io).
There are other python wrappers for the alpha vantage API, such as [alphavantage](https://pypi.org/project/alphavantage/), [alpha-vantage-downloader](https://pypi.org/project/alpha-vantage-downloader/), [alpha_vantage](https://pypi.org/project/alpha_vantage/) and [alphavantage-wrapper](https://pypi.org/project/alphavantage-wrapper/). They all have in common that they have different functions for each of the data types that Alpha Vantage offers. This may be less practical when downloading multiple data types at multiple resolutions in one session. Avapi has a single function that can download any data type from Alpha Vantage: avapi.get_data() takes \*\*kwargs as input and outputs a dictionary, unless datatype="csv" is given, in which case a csv file is saved. If the data type is time series, avapi.to_df() converts it to a Pandas data frame. The intention is a simple and flexible way to get data from Alpha Vantage into python.
### Installation
```python
pip install avapi
```
### Before use
Follow instructions on the [Alpha vantage](https://www.alphavantage.co) website to get a free API-key. The [docs](https://www.alphavantage.co/documentation/) provide all necessary info for à la carte downloading of historical data and indicators. Please checkout Alpha Vantage [support](https://www.alphavantage.co/support/#support) and read through the [Frequently Asked Questions](https://www.alphavantage.co/support/#support).
#### Examples
The "demo" api-key only works for these specific examples. If you modify the code you also need to provide your own api-key.
##### Single download
```python
import avapi as aa
import pandas as pd
data = aa.get_data(function='VWAP', symbol='MSFT',
interval='15min', apikey='demo')
df = aa.to_df(data)
df.head()
```
##### Multiple files download
There are limits for free accounts: up to 5 API requests per minute and 500 requests per day. In loops, if you don't incorporate time.sleep(), you might get errors.
```python
import avapi as aa
import pandas as pd
from time import sleep
data_path = 'data/'
data_calls = [
{
'function': 'TIME_SERIES_DAILY',
'symbol': 'MSFT',
'apikey': 'demo',
'datatype': 'csv',
},
{
'function':'TIME_SERIES_INTRADAY',
'symbol': 'MSFT',
'interval': '5min',
'apikey': 'demo',
'datatype': 'csv',
}
]
for call in data_calls:
save_to = data_path + call['function'] + '_' + call['symbol'] + '.csv'
data = aa.get_data(save_to=save_to, **call)
sleep(15)
```
##### Unexpected results
If you get unexpected results, you may check the latest, original response from the Alpha Vantage server in the following way:
```python
response = aa.response()
print(response)
```
## Contribute
- Issue Tracker: https://github.com/ekkobit/avapi/issues
- Source Code: https://github.com/ekkobit/avapi
- Documentation: https://avapi.readthedocs.io
