coURLan: Clean, filter, normalize, and sample URLs
“It is important for the crawler to visit "important" pages first, so that the fraction of the Web that is visited (and kept up to date) is more meaningful.” (Cho et al. 1998)
“Given that the bandwidth for conducting crawls is neither infinite nor free, it is becoming essential to crawl the Web in not only a scalable, but efficient way, if some reasonable measure of quality or freshness is to be maintained.” (Edwards et al. 2001)
This library provides an additional “brain” for web crawling, scraping and management of web archives:
- Avoid loosing bandwidth capacity and processing time for webpages which are probably not worth the effort.
- Stay away from pages with little text content or explicitly target synoptic pages to gather links.
Using content and language-focused filters, Courlan helps navigating the Web so as to improve the resulting document collections. Additional functions include straightforward domain name extraction and URL sampling.
Separate the wheat from the chaff and optimize crawls by focusing on non-spam HTML pages containing primarily text.
- Heuristics for triage of links
- Targeting spam and unsuitable content-types
- Language-aware filtering
- Crawl management
- URL handling
- Usable with Python or on the command-line
Let the coURLan fish out juicy bits for you!
This package is compatible with with all common versions of Python, it is tested on Linux, macOS and Windows systems.
Courlan is available on the package repository PyPI and can notably be installed with the Python package manager
$ pip install courlan # pip3 install on systems where both Python 2 and 3 are installed $ pip install --upgrade courlan # to make sure you have the latest version $ pip install git+https://github.com/adbar/courlan.git # latest available code (see build status above)
Most filters revolve around the
All useful operations chained in
>>> from courlan import check_url # returns url and domain name >>> check_url('https://github.com/adbar/courlan') ('https://github.com/adbar/courlan', 'github.com') # noisy query parameters can be removed my_url = 'https://httpbin.org/redirect-to?url=http%3A%2F%2Fexample.org' >>> check_url(my_url, strict=True) ('https://httpbin.org/redirect-to', 'httpbin.org') # Check for redirects (HEAD request) >>> url, domain_name = check_url(my_url, with_redirects=True)
Language-aware heuristics, notably internationalization in URLs, are available in
# optional argument targeting webpages in English or German >>> url = 'https://www.un.org/en/about-us' # success: returns clean URL and domain name >>> check_url(url, language='en') ('https://www.un.org/en/about-us', 'un.org') # failure: doesn't return anything >>> check_url(url, language='de') >>> # optional argument: strict >>> url = 'https://en.wikipedia.org/' >>> check_url(url, language='de', strict=False) ('https://en.wikipedia.org', 'wikipedia.org') >>> check_url(url, language='de', strict=True) >>>
Define stricter restrictions on the expected content type with
strict=True. Also blocks certain platforms and pages types crawlers should stay away from if they don't target them explicitly and other black holes where machines get lost.
# strict filtering: blocked as it is a major platform >>> check_url('https://www.twitch.com/', strict=True) >>>
Sampling by domain name
>>> from courlan import sample_urls >>> my_urls = ['https://example.org/' + str(x) for x in range(100)] >>> my_sample = sample_urls(my_urls, 10) # optional: exclude_min=None, exclude_max=None, strict=False, verbose=False
Web crawling and URL handling
Determine if a link leads to another host:
>>> from courlan import is_external >>> is_external('https://github.com/', 'https://www.microsoft.com/') True # default >>> is_external('https://google.com/', 'https://www.google.co.uk/', ignore_suffix=True) False # taking suffixes into account >>> is_external('https://google.com/', 'https://www.google.co.uk/', ignore_suffix=False) True
Other useful functions dedicated to URL handling:
extract_domain(url, fast=True): find domain and subdomain or just domain with
get_base_url(url): strip the URL of some of its parts
get_host_and_path(url): decompose URLs in two parts: protocol + host/domain and path
get_hostinfo(url): extract domain and host info (protocol + host/domain)
fix_relative_urls(baseurl, url): prepend necessary information to relative links
>>> from courlan import * >>> url = 'https://www.un.org/en/about-us' >>> get_base_url(url) 'https://www.un.org' >>> get_host_and_path(url) ('https://www.un.org', '/en/about-us') >>> get_hostinfo(url) ('un.org', 'https://www.un.org') >>> fix_relative_urls('https://www.un.org', 'en/about-us') 'https://www.un.org/en/about-us'
Other filters dedicated to crawl frontier management:
is_not_crawlable(url): check for deep web or pages generally not usable in a crawling context
is_navigation_page(url): check for navigation and overview pages
>>> from courlan import is_navigation_page, is_not_crawlable >>> is_navigation_page('https://www.randomblog.net/category/myposts') True >>> is_not_crawlable('https://www.randomblog.net/login') True
Helper function, scrub and normalize:
>>> from courlan import clean_url >>> clean_url('HTTPS://WWW.DWDS.DE:80/') 'https://www.dwds.de'
Basic scrubbing only:
>>> from courlan import scrub_url
Basic canonicalization/normalization only, i.e. modifying and standardizing URLs in a consistent manner:
>>> from urllib.parse import urlparse >>> from courlan import normalize_url >>> my_url = normalize_url(urlparse(my_url)) # passing URL strings directly also works >>> my_url = normalize_url(my_url) # remove unnecessary components and re-order query elements >>> normalize_url('http://test.net/foo.html?utm_source=twitter&post=abc&page=2#fragment', strict=True) 'http://test.net/foo.html?page=2&post=abc'
Basic URL validation only:
>>> from courlan import validate_url >>> validate_url('http://1234') (False, None) >>> validate_url('http://www.example.org/') (True, ParseResult(scheme='http', netloc='www.example.org', path='/', params='', query='', fragment=''))
UrlStore class allow for storing and retrieving domain-classified URLs, where a URL like
https://example.org/path/testpage is stored as the path
/path/testpage within the domain
https://example.org. It features the following methods:
- URL management
add_urls(urls=, appendleft=None, visited=False): Add a list of URLs to the (possibly) existing one. Optional: append certain URLs to the left, specify if the URLs have already been visited.
dump_urls(): Return a list of all known URLs.
print_urls(): Print all URLs in store (URL + TAB + visited or not).
get_known_domains(): Return all known domains as a list.
total_url_number(): Find number of all URLs in store.
is_known(url): Check if the given URL has already been stored.
has_been_visited(url): Check if the given URL has already been visited.
filter_unknown_urls(urls): Take a list of URLs and return the currently unknown ones.
filter_unvisited_urls(urls): Take a list of URLs and return the currently unvisited ones.
find_known_urls(domain): Get all already known URLs for the given domain (ex. "https://example.org").
find_unvisited_urls(domain): Get all unvisited URLs for the given domain.
- Crawling and downloads
get_url(domain): Retrieve a single URL and consider it to be visited (with corresponding timestamp).
get_rules(domain): Return the stored crawling rules for the given website.
get_download_urls(timelimit=10): Get a list of immediately downloadable URLs according to the given time limit per domain.
establish_download_schedule(max_urls=100, time_limit=10): Get up to the specified number of URLs along with a suitable backoff schedule (in seconds).
download_threshold_reached(threshold): Find out if the download limit (in seconds) has been reached for one of the websites in store.
unvisited_websites_number(): Return the number of websites for which there are still URLs to visit.
is_exhausted_domain(domain): Tell if all known URLs for the website have been visited.
The main fonctions are also available through a command-line utility.
$ courlan --inputfile url-list.txt --outputfile cleaned-urls.txt $ courlan --help usage: courlan [-h] -i INPUTFILE -o OUTPUTFILE [-d DISCARDEDFILE] [-v] [--strict] [-l LANGUAGE] [-r] [--sample] [--samplesize SAMPLESIZE] [--exclude-max EXCLUDE_MAX] [--exclude-min EXCLUDE_MIN]
- optional arguments:
-h, --help show this help message and exit
Manage input and output
-i INPUTFILE, --inputfile INPUTFILE name of input file (required) -o OUTPUTFILE, --outputfile OUTPUTFILE name of output file (required) -d DISCARDEDFILE, --discardedfile DISCARDEDFILE name of file to store discarded URLs (optional) -v, --verbose increase output verbosity
Configure URL filters
--strict perform more restrictive tests -l LANGUAGE, --language LANGUAGE use language filter (ISO 639-1 code) -r, --redirects check redirects
Use sampling by host, configure sample size
--sample use sampling --samplesize SAMPLESIZE size of sample per domain --exclude-max EXCLUDE_MAX exclude domains with more than n URLs --exclude-min EXCLUDE_MIN exclude domains with less than n URLs
coURLan is distributed under the GNU General Public License v3.0. If you wish to redistribute this library but feel bounded by the license conditions please try interacting at arms length, multi-licensing with compatible licenses, or contacting me.
courlan is optimized for English and German but its generic approach is also usable in other contexts.
Contributions are welcome!
Feel free to file issues on the dedicated page.
This effort is part of methods to derive information from web documents in order to build text databases for research (chiefly linguistic analysis and natural language processing). Extracting and pre-processing web texts to the exacting standards of scientific research presents a substantial challenge for those who conduct such research. Web corpus construction involves numerous design decisions, and this software package can help facilitate text data collection and enhance corpus quality.
- Barbaresi, A. "Trafilatura: A Web Scraping Library and Command-Line Tool for Text Discovery and Extraction." Proceedings of ACL/IJCNLP 2021: System Demonstrations, 2021, pp. 122-131.
- Barbaresi, A. "Generic Web Content Extraction with Open-Source Software." Proceedings of the 15th Conference on Natural Language Processing (KONVENS 2019), 2019, pp. 267-268.
Software ecosystem: see this graphic.
These Python libraries perform similar normalization tasks but do not entail language or content filters. They also do not focus on crawl optimization:
- Cho, J., Garcia-Molina, H., & Page, L. (1998). Efficient crawling through URL ordering. Computer networks and ISDN systems, 30(1-7), 161–172.
- Edwards, J., McCurley, K. S., and Tomlin, J. A. (2001). "An adaptive model for optimizing performance of an incremental web crawler". In Proceedings of the 10th international conference on World Wide Web - WWW '01, pp. 106–113.