Skip to content

fix(filters): Stage B2 sweep round 5 — 20 recurring off-topic domains - #195

Merged
shaypal5 merged 1 commit into
mainfrom
codex/b2-spam-domain-sweep-5
Jun 10, 2026
Merged

shaypal5 merged 1 commit into
mainfrom
codex/b2-spam-domain-sweep-5

Conversation

@shaypal5

Copy link
Copy Markdown
Member

With the domain-frequency gate killing the single-shot tail, the remaining batch noise is recurring off-topic domains the gate can't catch (car/real-estate/gadget/franchising SEO, traffic-ticket & legal-admin SEO, foreign academic/Danish outlets, Torah essays, reality-TV, gossip, custody advocacy, film/therapy/sex-myth SEO). Blocklisted per docs/batch_scraping_protocol.md.

🤖 Generated with Claude Code

…ic domains

With the domain-frequency gate now killing the single-shot tail, the remaining
batch noise is recurring (>=2 appearances) off-topic domains that the gate
cannot catch: car-deal/real-estate listings (cartube, dh-nadlan), gadget/expo/
franchising/ESG SEO, traffic-ticket and legal-admin SEO (eranwaisman, what2do,
dabullaw), foreign academic/Danish outlets (discovery.researcher.life, fyens.dk),
Torah essays (etzion), reality-TV (boss-il), gossip (ednakarnaval), custody
advocacy (father4justice), film/therapy/sex-myth SEO (cinema, dvorih, eytanim),
municipal tenders (gadish).

Existing rows suppressed operationally per the batch scraping protocol.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Copilot AI review requested due to automatic review settings June 10, 2026 20:07
@shaypal5
shaypal5 merged commit 10ceb63 into main Jun 10, 2026
@shaypal5
shaypal5 deleted the codex/b2-spam-domain-sweep-5 branch June 10, 2026 20:07

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Updates the discovery candidate filters to block additional recurring off-topic domains that are slipping past the domain-frequency gate, reducing wasted scrape/classifier budget during Stage B2 batch operations.

Changes:

  • Added 20 newly identified recurring off-topic domains to _IRRELEVANT_CONTENT_DOMAINS under the “Stage B2 sweep 2026-06 (round 5)” section.
  • Expanded coverage for several SEO/commerce/foreign/religious/entertainment sources that repeatedly surface in discovery results but are not relevant to enforcement-event discovery.

@codecov

codecov Bot commented Jun 10, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.03%. Comparing base (34e72c3) to head (d34a393).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files
@@           Coverage Diff           @@
##             main     #195   +/-   ##
=======================================
  Coverage   93.03%   93.03%           
=======================================
  Files          80       80           
  Lines       11947    11947           
=======================================
  Hits        11115    11115           
  Misses        832      832           
Files with missing lines Coverage Δ
src/denbust/discovery/candidate_filters.py 98.24% <ø> (ø)
🚀 New features to boost your workflow:
  • ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

@github-actions

Copy link
Copy Markdown

pr-agent-context report:

No unresolved review comments, failing checks, or actionable patch coverage gaps were found on PR
#195. Treat this PR as all clear unless new signals appear.

Run metadata:

Tool ref: v4.0.19
Tool version: 4.0.19
Trigger: pull request opened
Workflow run: 27303035304 attempt 1
Comment timestamp: 2026-06-10T20:10:25.468180+00:00
PR head commit: d34a393871355e49a4a5bf8753812137ca7d4302

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants