Liking cljdoc? Tell your friends :D

common-crawl-utils

Various Common Crawl utilities in Clojure.

Utilities:

  • Selected content fetcher from the Common Crawl Archive
  • Clojure Wrapper on the Common Crawl Index API
  • Reader for the raw Common Crawl WARC archives

Selected Content Fetcher

(common-crawl-utils.fetcher/fetch-content 
  {:cdx-api "https://index.commoncrawl.org/CC-MAIN-2019-09-index" 
   :url "tokenmill.lt" 
   :filter ["status:200"]})
=>
({:offset "272838009",
  :content {:warc "WARC/1.0\r
                   WARC-Type: response\r
                   WARC-Date: 2019-02-17T12:51:41Z\r
                   WARC-Record-ID: <urn:uuid:114c36c3-b278-49bf-b4c0-0cf2a0eaac7c>\r
                   Content-Length: 33486\r
                   Content-Type: application/http; msgtype=response\r
                   WARC-Warcinfo-ID: <urn:uuid:64ce1a6d-7cdc-44ec-a802-f4feac213f0c>\r
                   WARC-Concurrent-To: <urn:uuid:3143fe9c-ec79-4f53-8665-86c668cb46d8>\r
                   WARC-IP-Address: 79.98.31.5\r
                   WARC-Target-URI: http://tokenmill.lt/\r
                   WARC-Payload-Digest: sha1:U3FWVBI7XZ2KVBD72MRR7TCHHXSX2FJS\r
                   WARC-Block-Digest: sha1:MPJRHAU5WNOIDVX2AZTDOEP2YPXXLC66\r
                   WARC-Identified-Payload-Type: text/html",
            :header "HTTP/1.1 200 OK\r
                     Date: Sun, 17 Feb 2019 12:51:41 GMT\r
                     Content-Type: text/html;charset=UTF-8\r
                     X-Crawler-Transfer-Encoding: chunked\r
                     Server: Jetty(7.x.y-SNAPSHOT)",
            :html "THE ACTUAL HTML"},
  :digest "U3FWVBI7XZ2KVBD72MRR7TCHHXSX2FJS",
  :mime "text/html",
  :charset "UTF-8",
  :mime-detected "text/html",
  :filename "crawl-data/CC-MAIN-2019-09/segments/1550247481992.39/warc/CC-MAIN-20190217111746-20190217133746-00381.warc.gz",
  :status "200",
  :urlkey "lt,tokenmill)/",
  :url "http://tokenmill.lt/",
  :length "8548",
  :languages "eng",
  :timestamp "20190217125141"})

Uses Common Crawl Index API to get coordinates to content which is stored on AWS.

Common Crawl Index API Wrapper

Common Crawl Index API wrapper allows to query Common Crawl Index API for coordinates into the Common Crawl data.

Constructing Queries

All queries must contain "url" key. To return all existing coordinates that match specified host, we append "/*" to it or set key "matchType" to "host".

{:url "tokenmill.lt/*"}

{:url "tokenmill.lt" :matchType "host"}

Additionally, results can be filtered. Below is an example, where coordinates that have "status" containing "200" and "mime" containing "html" are returned.

{:url "tokenmill.lt/*" :filter ["status:200" "mime:html"]}

All available fields: urlkey, timestamp, url, mime, status, digest, length, offset, filename.

Full reference can be found at CDX Server API.

Coordinates

Common Crawl is updated on a monthly basis. Each crawl has a specific index API, which we can query like this:

(common-crawl-utils.coordinates/fetch {:cdx-api "https://index.commoncrawl.org/CC-MAIN-2019-09-index" :url "tokenmill.lt" :filter ["status:200"]})
=>
({:offset "272838009",
 :digest "U3FWVBI7XZ2KVBD72MRR7TCHHXSX2FJS",
 :mime "text/html",
 :charset "UTF-8",
 :mime-detected "text/html",
 :filename "crawl-data/CC-MAIN-2019-09/segments/1550247481992.39/warc/CC-MAIN-20190217111746-20190217133746-00381.warc.gz",
 :status "200",
 :urlkey "lt,tokenmill)/",
 :url "http://tokenmill.lt/",
 :length "8548",
 :languages "eng",
 :timestamp "20190217125141"})

When "cdx-api" keyword is not specified, most recent one is used. Currently available index collections can be accessed with "common-crawl-utils.utils/get-crawls" or can be found at: https://index.commoncrawl.org/collinfo.json

Reader

Can directly read Common Crawl .warc files containing content, as well as .cdx files containing coordinates.

(common-crawl-utils.reader/read-warc)

(common-crawl-utils.reader/read-coordinates)

If no arguments are specified, reads from latest crawl. Otherwise, we can specify crawl "id" which can be found at https://index.commoncrawl.org/collinfo.json.

(first (common-crawl-utils.reader/read-warc "CC-MAIN-2019-09"))
=>
{:content-length 61992,
  :content-type #object[org.jwat.common.ContentType 0x3eb133dd "application/http; msgtype=response"],
  :date #inst"2019-02-15T19:26:02.000-00:00",
  :filename nil,
  :target-uri #object[org.jwat.common.Uri 0x3efffa13 "http://0204mm.com/?PUT=a_show&AID=68666&FID=1361239&R2=&CHANNEL="],
  :target-uri-str "http://0204mm.com/?PUT=a_show&AID=68666&FID=1361239&R2=&CHANNEL=",
  :warc-type "response",
  :payload-stream #object[org.jwat.common.ByteCountingPushBackInputStream
                          0x1b9da111
                          "org.jwat.common.ByteCountingPushBackInputStream@1b9da111"]}

License

Copyright © 2019 TokenMill UAB.

Distributed under the The Apache License, Version 2.0.

Can you improve this documentation? These fine people already did:
Dainius Jocas, Rokas Ramanauskas, Rokas & Žygimantas Medelis
Edit on GitHub

cljdoc is a website building & hosting documentation for Clojure/Script libraries

× close