In June 2017 del.icio.us was bought by Maciej Ceglowski, the founder of Pinboard. He restored the export functionality in del.icio.us, so you don't need this script anymore. Please use the built-in export instead.
A script for scraping and exporting your bookmarks from del.icio.us.
del.icio.us is a bookmarking service that went through a series of increasingly more unfortunate acquisitions. As of January 1st, 2017, the website is frequently down, yet doesn't allow users to export their bookmarks. At the same time, it no longer offers an API, and prohibits anyone from content scraping.
That's deeply unsettling, since many users have used del.icio.us to store thousands of bookmarks, capturing years of internet browsing history.
This script allows users to export their content to a HTML bookmarks file, in the format accepted by major browsers and other bookmark services. It will preserve tags and dates. An additional option enables validation and skipping of dead URLs.
- Ruby
- nokogiri gem. Check nokogiri documentation for installation instructions.
- curl
./export.rb -u USERNAME -o bookmarks.html
USERNAME the name of del.icio.us user whose bookmarks will be exported. bookmarks.html is the output file, where the bookmarks will be saved.
The above command will export all public links from selected del.icio.us account, including dead ones. This may take a few minutes, depending on the number of links.
./export.rb -u USERNAME -p 'PASSWORD' -o bookmarks.html
If you provide a password, the script will attempt to login on your behalf and export all links, including those marked as private. To make it easier to identify private links later, the script adds a special tag to them: ___private.
Don't forget to enclose the password in single quotes if it contains any characters that may confuse your shell.
./export.rb -u USERNAME -o bookmarks.html --validate
When you provide the --validate option, the script will try to fix or skip dead links:
- If the server doesn't respond within 5 seconds, the link will be skipped.
- If the server responds with a transient error (timeout or 5xx error) the script will retry up to 3 times, then skip the link.
- If the server sends a redirect header, the script will follow it and save the target URL.
This mode is much slower and can take about 2 minutes for every hundred links.
If you are behind a proxy and the script doesn't work correctly, set up following environmental variables before running the script:
export http_proxy http://proxy.server:3128
export https_proxy=https://proxy.server:6443
Replace proxy.server and the port number if information about your proxy server.
If you don't use the --validate flag, only https_proxy variable is necessary.
The script is available under provisions of the public domain license. You are free to copy and modify it without asking for author's permission. The author doesn't provide any warranty or support.
Please be aware that content scraping is currently not allowed by del.icio.us' terms and conditions. The author doesn't take any responsibility for your usage of this script.