https://github.com/marks/streamlit-airtable-connection

Airtable connection for Streamlit, includes AI ✨ example to ask questions from records
https://github.com/marks/streamlit-airtable-connection

airtable airtable-api pyairtable streamlit streamlit-component streamlit-components

Last synced: 8 days ago
JSON representation

Airtable connection for Streamlit, includes AI ✨ example to ask questions from records

• Host: GitHub
• URL: https://github.com/marks/streamlit-airtable-connection
• Owner: marks
• Created: 2023-07-30T18:45:19.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-08-27T14:37:11.000Z (3 months ago)
• Last Synced: 2024-10-29T23:44:37.506Z (21 days ago)
• Topics: airtable, airtable-api, pyairtable, streamlit, streamlit-component, streamlit-components
• Language: Python
• Homepage: https://marks-airtable-connection-demo.streamlit.app
• Size: 32.2 KB
• Stars: 16
• Watchers: 1
• Forks: 2
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Airtable ⚡ Streamlit Connection

This repo includes a `st.connection` for Airtable which wraps the

popular community created and maintained

[pyAirtable](https://github.com/gtalarico/pyairtable) library.

The initial focus of this connector is on read-only operations: specifically

[listing records](https://airtable.com/developers/web/api/list-records) and

[retrieving the schema](https://airtable.com/developers/web/api/get-base-schema)

of an Airtable base. Write-back support may be added in the future.

### 👀 Live demo w/ AI ✨

A live demo of the code example in  [`examples/explore.py`](./examples/explore.py) can be accessed on Streamlit Cloud at [http://marks-airtable-connection-demo.streamlit.app](http://marks-airtable-connection-demo.streamlit.app) [[demo video](https://share.support.airtable.com/RBuJRnyL)]. Provide your Open AI API key to be able to ask questions about the data in the selected Airtable table. The underlying data can be viewed and cloned [here](https://airtable.com/appdbRXgibDiQkNQN/shrIO0m8oyeQJTf9T).

---

### Minimal example: retrieve all records from base's first table

_See `examples/` for additional examples_

#### [`.streamlit/secrets.toml`](./examples/.streamlit/secrets.toml.example)

```

[connections.your_connection_name]

personal_access_token = "patXXX" # REQUIRED

base_id = "appXXX" # optional

table_id = "tblXXX" # optional

```

ℹ️ The Airtable

[personal access token](https://airtable.com/developers/web/guides/personal-access-tokens)

you use should have both

[`data.records:read`](https://airtable.com/developers/web/api/scopes#data-records-read)

and

[`schema.bases:read`](https://airtable.com/developers/web/api/scopes#schema-bases-read)

scopes to use all functionality of this connector.

#### [`minimal_example.py`](./examples/minimal_example.py)

```python

import streamlit as st

from streamlit_airtable import AirtableConnection

# Create connection

conn = st.connection("your_connection_name", type=AirtableConnection)

# Retrieve base schema

base_schema = conn.get_base_schema()

with st.expander("Base schema"):

    st.json(base_schema)

# Get the first table's ID

first_table = base_schema["tables"][0]

st.markdown(f"First table ID: `{first_table['id']}` (named `{first_table['name']}`)")

# Retrieve all records for the first table (pyAirtable paginates automatically)

# (Note you can also pass in parameters supported by pyAirtable

# (https://pyairtable.readthedocs.io/en/stable/api.html#parameters) such as as

# max_records, view, sort, and formula into conn.query() like so:

# table_records = conn.query(first_table["id"], max_records=25, view='viwXXX')

table_records = conn.query(table_id=first_table["id"])

st.markdown(f"{len(table_records)} records retrieved")

st.dataframe(table_records)

```

#### Steps to replicate the minimal example

1. Clone/download this repo

2. Install the connector (`pip install -e .`)

3. Move into the `examples/` dir (`cd examples/`)

4. Copy `.streamlit/secrets.toml.example` to `.streamlit/secrets.toml`

   and provide your own values

5. Run `streamlit run minimal_example.py`

### Acknowledgements & thank yous

- Streamlit connection examples:

  [Google Sheets](https://github.com/streamlit/gsheets-connection) &

  [DuckDB](https://github.com/streamlit/release-demos/blob/master/1.22/st-experimental-connection/duckdb_connection/connection.py)

- [pyAirtable](https://github.com/gtalarico/pyairtable)