`urlskip` not working as expected?

[ravilov]

Prerequisites

• I verified that this is not a filter list issue. Report any issues with filter lists or broken website functionality in the uAssets issue tracker.
• This is NOT a YouTube, Facebook or Twitch report. These sites MUST be reported by clicking their respective links.
• This is not a support issue or a question. For support, questions, or help, visit /r/uBlockOrigin.
• I performed a cursory search of the issue tracker to avoid opening a duplicate issue.
• The issue is not present after disabling uBO in the browser.
• I checked the documentation to understand that the issue I am reporting is not normal behavior.

I tried to reproduce the issue when...

• uBO is the only extension.
• uBO uses default lists and settings.
• using a new, unmodified browser profile.

Description

I added the following rules to my local list:

||click.convertkit-mail4.com/*
||click.convertkit-mail4.com/*$urlskip=-blocked /\/[^\/]+\/[^\/]+\/(.+)/ -safebase64 +https

The idea is the offending URL is initially blocked, and stays blocked unless the urlskip chain succeeds and redirects to a clean URL. However no matter how I put it I'm only ever getting the standard uBO "request blocked" page.

A specific URL where the issue occurs.

https://c19cfde3.click.convertkit-mail4.com/v8u4p808noimun8xzmaghvk8o834c9hkx9g/3ohphdurer8855hrhn/aHR0cHM6Ly9tYWlsc2Nvb3AuaW8=

Steps to Reproduce

1. Add the rules above.
2. Visit the sample URL.
3. Observe the unexpected behavior.

Expected behavior

I get redirected to mailscoop.io, the result of urlskip operations on the URL above.

Troubleshooting information

uBlock Origin: 1.71.0
Chromium Mobile: 137
filterset (summary):
 network: 375155
 cosmetic: 282490
 scriptlet: 74685
 html: 0
listset (total-discarded, last-updated):
 added:
  https://127.0.0.1:4096/ublock/filters.txt: 3155-2, 28m
  adguard-generic: 96500-7205, 1d.3h.35m
  adguard-spyware-url: 2534-15, 6d.5h.1m
  block-lan: 91-0, 27d.12h.50m
  curben-phishing: 164633-57, 1h.4m
  adguard-cookies: 38831-3915, 1d.16h.1m
  ublock-cookies-adguard: 5733-36, 6d.5h
  fanboy-cookiemonster: 46895-173, 4d.11h.42m
  ublock-cookies-easylist: 5733-5733, 6d.5h
  adguard-social: 26486-998, 1d.16h.1m
  fanboy-social: 17677-22, 6d.5h
  fanboy-thirdparty_social: 69-0, 3h.49m
  [11 lists not shown]: [too many]
 default:
  user-filters: 10-0, never
  ublock-filters: 51819-417, 1h.4m Δ
  ublock-badware: 10384-28, 1h.4m Δ
  ublock-privacy: 4053-125, 1h.4m Δ
  ublock-unbreak: 2765-3, 1h.4m Δ
  ublock-quick-fixes: 441-0, 1h.4m Δ
  adguard-mobile: 12476-127, 1d.16h.1m
  easylist: 88756-645, 1h.4m Δ
  easyprivacy: 55838-916, 1h.4m Δ
  urlhaus-1: 24165-12, 1h.4m
  plowe-0: 3504-2, 1d.12h.13m
filterset (user): [array of 6 redacted]
trustedset:
 added: [array of 13 redacted]
switchRuleset:
 added: [array of 7 redacted]
hostRuleset:
 added: [array of 3 redacted]
userSettings:
 advancedUserEnabled: true
 contextMenuEnabled: false
 largeMediaSize: 1000000
 suspendUntilListsAreLoaded: true
 tooltipsDisabled: true
 userFiltersTrusted: true
hiddenSettings:
 allowGenericProceduralFilters: true
 dnsResolveEnabled: false
 filterAuthorMode: true
 trustedListPrefixes: ublock- user- file: https://127.0.0.1:4096/ http://127.0.0.1:4096/
 uiStyles: font-family: sans-serif; font-size: 80%;
 updateAssetBypassBrowserCache: true
 userResourcesLocation: [redacted]
supportStats:
 allReadyAfter: 9745 ms
 maxAssetCacheWait: 4262 ms
 cacheBackend: indexedDB

[gorhill]

I don't see valid base64 component in your URL. Which part is supposed to be a valid base64 string?

[ravilov]

The part after the last slash - aHR0cHM6Ly9tYWlsc2Nvb3AuaW8=. It's the part I'm trying to capture with the regex in the rule. I decoded it manually and it yields https://mailscoop.io as expected.

[uBlock-user]

||click.convertkit-mail4.com^
||click.convertkit-mail4.com/*/*/$urlskip=/[^\/]+\/(aHR0c[^\/]+)/ -base64 -blocked

Try that.

[gorhill]

Try:

||click.convertkit-mail4.com/$urlskip=/([^/]+$)/ -base64 -blocked

[ravilov]

||click.convertkit-mail4.com^
||click.convertkit-mail4.com/*/*/$urlskip=/[^\/]+\/(aHR0c[^\/]+)/ -base64 -blocked

Try that.

Thanks, that works. I'm honestly not too happy about "hardcoding" htt there though. 🙂

Try:

||click.convertkit-mail4.com/$urlskip=/([^/]+$)/ -base64 -blocked

That works too. In fact, after playing with it all a bit, I realized my original rule(s) work just fine if I replace .+ with [^\/]+. Not sure why "not a slash" is valid but "any character" isn't. Should be equivalent in a string that doesn't have slashes. 🤔

My guess is that with \/[^\/]+\/(.+) the first portion matches too little and the second portion then captures too much, including the slash and probably some chars before, which ends up being invalid base64 and then an invalid URL.

Either way, my rules now work, so thank you!