DataFrameup

 $ pip install frameup
 $ frameup <path-to-csvfile>

Frameup is the easiest way to get your Pandas DataFrame up into a Python-based web application. Simply import frameup and your DataFrames will become URL query parameter, and pagination aware.

Supports:

Most of the Pandas DataFrame to_html parameters as URL query parameters
Pagination query parameters: page, limit, offset
Filtering of the DataFrame using the parameter query and the Pandas DataFrame query syntax

Zero dependencies, except Pandas of course.

Quick look

Serve a csv as a frameup dataframe on localhost

 $ python -m frameup.serve <path-to-csv-file>

Then navigate to http://localhost:8000/. Use the Pandas DataFrame query syntax in the query box.

... or, get a JSON payload:

 $ curl 'http://localhost:8000/?query=&limit=10&page=1' | python -m json.tool

Use it in your web application

Flask example

Given a template similar to example.j2.html

from flask import Flask, jsonify, render_template, request, url_for
import pandas as pd
import frameup

app = Flask(__name__)

df = pd.read_csv(YOUR_CSV_FILE)

@app.route('/mydataframe')
def main():
    data = df.frameup.data(path=url_for('main'), **request.args)
    return render_template('example.j2.html', **data)

For something ajaxy, just replace the return with:

return jsonify(**data)

On query parameter objects

Be sure the query parameter object you pass frameup does not return lists for values. classes is the only multi-valued parameter accepted, and should be passed as a comma-delimited string rather than multiple classes keys.

Python web frameworks all have their own way of dealing with the vagaries of GET parameter specification hell. Most implement some concept of a MultiDict, but the APIs for these vary from one framework to the next. Thus, the requirement of only single-valued GET params greatly simplifies things here.

Other projects