v2.4.3

Fixes issue where certain links (e.g. https://example.com/) would not be downloaded even though all others would.

2.4.2

Did more work on our previous sanitization efforts in 2.4.1. This should get rid of most remaining issues when downloading URLs with differently encoded characters in it.

Full Changelog: v2.4.1...v2.4.2

v2.4.1

Fixes an issue where links with HTML tags on them (e.g. https://example.com/foo <br> bar) failed to download.

Full Changelog: v2.4.0...v2.4.1

2.4.0

What's Changed

• Initial implementation of a 'composite snapshot' feature (see #22). It's still experimental and not listed in the documentation just yet, but it more or less attempts to mimic the WMD feature of creating a 'snapshot' of a site, combining snapshots near the dates you selected for a more or less complete site. This was added in 2.3.11, I just forgot to mention it here. You can test it using --composite-at [date].
• A refactoring of tidy_bytes.rb. It's not very neat, but it likely fixes issues with downloading links containing non-UTF-8 characters (see #25)
• Refactored the url rewrite modal into a separate file (thanks to adampweb!)

Full Changelog: v2.3.12...v2.4.0

v2.3.12

Fixes issue where 2.3.11 was not loading. That's really it.

v2.3.11

• Added experimental --recursive-subdomains command, which will (per title) recursively download content from subdomains. By default, subdomain recursion is set at a value of 1, which you can change using --subdomain-depth DEPTH (by replacing DEPTH with the number of your choice)
• Prevent fetching off non RFC3968-compliant URLs (by underarchiver - cherry-picked to this repo)
• Added note for users of the old WMD
• Fixed issue where some regex filters were not being processed (by AlfonsoCorrado - cherry-picked to this repo)

Full Changelog: v2.3.10...v2.3.11

v2.3.10

Fixes an issue where Windows users got an Error processing file: no implicit conversion of Proc into String message when attempting to download files.

Full Changelog: v2.3.9...v2.3.10

v2.3.9

What's Changed

• Don’t append /* when using —exact-url, which broke it in 2.3.8 (thanks to elidickinson!)
• Uses correct file path and sanitization on Windows. Our previous method was too aggressive and led to a few issues with selected paths and also file names (see #16)

Full Changelog: v2.3.8...v2.3.9

v2.3.8

A couple of updates since last time!

What's Changed

• Workaround for API only showing html files for some domains (thanks to elidickinson!). Fixed #6
• Fix bug with archive urls containing square brackets (thanks again to elidickinson)
• Downloading data from sitename/* will now list the results in a sitename folder, and not * as previously
• Adds some nil checks when fetching a list of snapshots, fixing one minor bug

Full Changelog: v2.3.7...v2.3.8

v2.3.7

Fixes the --all command. At some point after 2.3.3 it broke; I'm not sure exactly how, or if it was broken in the first place and I just didn't notice, but a basic implementation of it is here. It should download both redirect (30x) and error (40x/50x) pages if used.

Full Changelog: v2.3.6...v2.3.7