parsechain 0.0.3

Making parsing concise

A way to parse html fluently and as declarative as possible. This is in alpha stage now, things will change.

Installation

pip install parsechain

Usage

import requests
from parsechain import C, Response

# Fetch html and cast it
response = Response.cast(requests.get(...))

# Get a movie title and rating
title = response.css('h1 .title').text
rating = response.css('.left-box').inner_text.re(r'IMDb: ([\d.]+)').float

# Or both
movie = response.root.multi({
    'title': C.css('h1 .title').text,
    'rating': C.css('.left-box').inner_text.re(r'IMDb: ([\d.]+)').float,
})

The last example could be extended to show chains reuse:

def by_label(label):
    return C.css('.left-box').inner_text.re(fr'{label}: ([\w.]+)')

parse_movie = C.multi({
    'title': C.css('h1 .title').text,
    'rating': by_label('IMDb').float,
    'status': by_label('Status').trim,
})

movie = parse_movie(response.root)  # Pass a root of a tree

The complete list of available ops could be seen in parsechain.chains.Ops class. Proper documentation to follow, some day ;)