LTP Dashboard

A Rust-based dashboard for Linux Test Project (LTP). Automatically fetches, scores, and displays patches by priority in terminal or web UI.

Features

• Intelligent Scoring: Multi-factor priority scoring based on:
  • Patch version (higher versions = more refined)
  • Age (inverted-U curve: peaks at 31-60 days, decays for very old/stale patches)
  • Series size (standalone patches prioritized over large series)
  • State (under-review vs new)
  • Fix keywords (bug, regression, broken)
  • New test detection
  • Library/core changes
• Priority Tiers: P1 (≥80), P2 (≥60), P3 (≥40), P4 (≥20), P5 (<20)
• Parallel Fetching: Three-level parallelism for fast data retrieval
• Review Detection: Scans email replies for Reviewed-by: and Acked-by: tags
• Two Output Modes:
  • Terminal: Colored ANSI table output
  • Web UI: Dark-themed HTML interface with auto-refresh
• Smart Caching: 5-minute TTL with background refresh

Installation

Prerequisites

• Rust 1.70+

Build

cargo build --release

Usage

Terminal Mode (default)

# Show top 100 patches in terminal
cargo run -- --max 100

# Include CI check results (slower)
cargo run -- --max 100 --checks

Web Mode

# Start web server on port 3030
cargo run -- --web

# Custom port
cargo run -- --web --port 8080

Then open http://localhost:3030 in your browser.

Configuration (optional)

Create config.toml in the project directory or ~/.config/ltp-dashboard/config.toml:

port = 3030
max_patches = 500
checks = false

Priority: ~/.config/ltp-dashboard/config.toml > ./config.toml > defaults

Command-Line Options

Options:
      --web         Expose a web interface instead of printing to
                    terminal
      --port <PORT> Port for the web server (only used with --web)
                    [default: 3030]
      --max <MAX>   Maximum number of patches to fetch [default: 500]
      --checks      Fetch CI check results for each patch (parallel)
  -h, --help        Print help

Scoring System

Priority Tiers

Tier	Score	Color	Meaning
P1	≥80	Red	Critical: High-priority fixes/updates
P2	≥60	Orange	Important: Should be reviewed soon
P3	≥40	Yellow	Moderate: Normal review queue
P4	≥20	Blue	Low: Can wait
P5	<20	Gray	Very Low: Stale, RFC, or low-impact

Score Breakdown

• Version: v1=0, v2=15, v3=25, v4=35, v5+=40
• Age (inverted-U curve):
  • <7 days: 0
  • 7-14 days: 10
  • 15-30 days: 20
  • 31-60 days: 45 (peak)
  • 61-90 days: 35
  • 91-180 days: 20
  • 181-365 days: 10
  • 366+ days: 5 (stale)
• Series Size: single=10, 2-5=0, 6-10=-5, 11+=-15
• State: new=5, under-review=0
• Fix Keywords: fix, bug, regression, broken = +10
• New Test: new test, add test, rewrite = +10
• Library/Core: tst_, lib:, lapi, configure.ac = +5
• Penalties:
  • RFC: -30
  • No activity (v1, <7 days): -20
  • Very large series (11+): -15

Web UI Features

• Dark Theme: Easy on the eyes for long review sessions
• Inline Badges:
  • 🔴 Stale: >60 days old
  • 🟠 RFC: Request for comments
  • 🔵 SERIES N: Part of N-patch series
  • 🟣 Under Review: Currently being reviewed
  • 🟢 Reviewed-by: Has reviewer approval
  • 🔵 Acked-by: Has maintainer acknowledgment
• Green Highlight: Patches with Reviewed-by or Acked-by
• Auto-Refresh: Updates every 5 minutes
• Gzip Compression: Faster page loads

Data Flow

Patchwork API
     ↓
Parallel Fetch (states: new, under-review, needs-review-ack)
     ↓
Filter out cover letters
     ↓
Score each patch
     ↓
Sort by score (descending)
     ↓
Group by tier
     ↓
Output (Terminal or Web)

Development

Run Tests

cargo test

Check Code

cargo clippy
cargo fmt --check

Dependencies

• axum: Web framework
• tokio: Async runtime
• reqwest: HTTP client (rustls, HTTP/2)
• serde/serde_json: JSON serialization
• minijinja: HTML templating
• chrono: Date/time handling
• regex: Pattern matching
• clap: CLI parsing
• anyhow: Error handling
• toml: Config parsing

License

This project is licensed under the GNU General Public License v3.0 or later - see the LICENSE file for details.