A python library for caching data while scraping using the famous requests-module

Often when scraping, something goes wrong. In order to minimize the burden on the website you are scraping, its a good idea to cache the data. This library does all this transparently.

With this module you can create an instance of crequests.Session (which inherits from requests.Session). With this instance you can use all the standard methods of requests.Session. In the background all the magic caching happens, while you use all the well known methods of the requests library

Normal use of requests

A normal way to do a get request with the requests-module goes like this:

import requests
html = requests.get("http://httpbin.org/html").content

If you need to run your program several times in a test or development phase, the content will be fetched from the website every time, putting a strain on the website. It can also be slow.

In this small example it's probably not a big problem, but if your program is traversing hundreds of pages, it's not nice to start all over.

Crequests to the rescue

First install crequests:

pip install crequests

You can now achieve the same request like this:

import crequests
crs = crequests.Session("cachedata")
html = crs.get("http://httpbin.org/html").content

Running this program will create a cache folder in the current working directory called "cachedata". The raw html will be extracted and returned in the same way as before. But besides this, a local cache copy is stored.

Next time you run the code, the exact same data will be retrieved from disk and returned instead.

You can delete the cache by deleting the folder

Technical info

The crequests.Session class extends the requests.Session class.

All methods should be exposed like the original class. The most useful are .get / .put / .post... But have a look in their documentation for more details.

All cached methods returns an requests.models.Response object.

Full Example

import crequests
import logging
from bs4 import BeautifulSoup

logging.basicConfig(
    level=logging.INFO, format="%(asctime)s [%(name)s] %(levelname)7s: %(message)s", datefmt="%H:%M:%S",
)

url = "http://httpbin.org/html"

crs = crequests.Session("cachedata")
for _ in range(3):  # Do the same over and over... Check that we get cache hits - this should be fast
    rawHtml = crs.get(url).content
    if rawHtml:
        soup = BeautifulSoup(rawHtml, "html.parser")
        print(soup.body.h1)

By running this example, the log output shows that the cache is working like expected:

07:01:45 [crequests.Session]    INFO: CACHE-MISS. 'http://httpbin.org/html' not in cache.
07:01:45 [crequests.Session]    INFO: Getting data directly from: http://httpbin.org/html
07:01:45 [crequests.Session]    INFO: Writing cachefile 'cachedata/httpbin.org/da/daff1d3c15c93dda35b7b95ca41f7e06d70b9551' with content of 'http://httpbin.org/html'
<h1>Herman Melville - Moby-Dick</h1>
07:01:47 [crequests.Session]    INFO: CACHE-HIT. 'http://httpbin.org/html' got cacheinfo in 'cachedata/httpbin.org/da/daff1d3c15c93dda35b7b95ca41f7e06d70b9551'
<h1>Herman Melville - Moby-Dick</h1>
07:01:47 [crequests.Session]    INFO: CACHE-HIT. 'http://httpbin.org/html' got cacheinfo in 'cachedata/httpbin.org/da/daff1d3c15c93dda35b7b95ca41f7e06d70b9551'
<h1>Herman Melville - Moby-Dick</h1>

Have fun...

Alex Skov Jensen