action.yml

name: 'GitHub Traffic Archiver'
description: 'Stop losing your GitHub traffic data! Automatically archives clones and views into a private long-term CSV database.'
author: 'groda'
branding:
  icon: 'database' 
  color: 'green'
  
inputs:
  observed-repo:
    description: "Repository to fetch metrics from (OWNER/REPO)"
    required: false
    default: ${{ github.repository }}
  metrics-repo:
    description: "Target repo for storing metrics (OWNER/REPO)"
    required: true
  metrics-pat:
    description: "Personal Access Token (PAT) with Contents Write permissions"
    required: true


    
# This section is required for Marketplace listing
runs:
  using: "composite"
  steps:
    - name: Install jq
      shell: bash
      run: sudo apt-get update && sudo apt-get install -y jq

    - name: Fetch Traffic Stats (Clones and Views)
      shell: bash
      env:
        GH_TOKEN: ${{ inputs.metrics-pat }}
        OBSERVED_REPO: ${{ inputs.observed-repo }}
      run: |
        # Fetch Clones
        curl -L -H "Authorization: Bearer $GH_TOKEN" \
          -H "Accept: application/vnd.github+json" \
          -H "X-GitHub-Api-Version: 2022-11-28" \
          -o clones.json \
          https://api.github.com/repos/$OBSERVED_REPO/traffic/clones

        # Fetch Views
        curl -L -H "Authorization: Bearer $GH_TOKEN" \
          -H "Accept: application/vnd.github+json" \
          -H "X-GitHub-Api-Version: 2022-11-28" \
          -o views.json \
          https://api.github.com/repos/$OBSERVED_REPO/traffic/views

    - name: Append and Deduplicate metrics CSV
      shell: bash
      run: |
        set -e

        # 1. Setup Repo
        git clone https://x-access-token:${{ inputs.metrics-pat }}@github.com/${{ inputs.metrics-repo }}.git metrics-temp
        cd metrics-temp
        git config user.name "metrics-bot"
        git config user.email "metrics-bot@users.noreply.github.com"

        # 2. Setup File Logic
        # We use the observed-repo input to define the filename
        FILE="data/$(echo '${{ inputs.observed-repo }}' | tr '/' '__').csv"
        HEADER="date,repository,type,count,uniques"
        mkdir -p data
        
        # Check for existence and Migration (4 cols to 5)
        if [ -f "$FILE" ]; then
          COL_COUNT=$(head -n 1 "$FILE" | grep -o "," | wc -l)
          if [ "$COL_COUNT" -eq 3 ]; then
            echo "Migrating old 4-column CSV..."
            echo "$HEADER" > "$FILE.migrated"
            tail -n +2 "$FILE" | awk -F, '{print $1","$2",clones,"$3","$4}' >> "$FILE.migrated"
            mv "$FILE.migrated" "$FILE"
          fi
        else
          echo "$HEADER" > "$FILE"
        fi

        REPO="${{ inputs.observed-repo }}"

        # 3. Process Clones
        LATEST_CLONE_DATE=$(jq -r '.clones[-1].timestamp[0:10] // empty' ../clones.json)
        [ -z "$LATEST_CLONE_DATE" ] && LATEST_CLONE_DATE=$(date +%F)
        jq -r --arg REPO "$REPO" '.clones[] | [.timestamp[0:10], $REPO, "clones", .count, .uniques] | @csv' ../clones.json >> "$FILE"
        jq -r --arg REPO "$REPO" --arg DATE "$LATEST_CLONE_DATE" \
          '[ "\($DATE)~ 14-day total", $REPO, "clones_total", .count, .uniques ] | @csv' ../clones.json >> "$FILE"

        # 4. Process Views
        LATEST_VIEW_DATE=$(jq -r '.views[-1].timestamp[0:10] // empty' ../views.json)
        [ -z "$LATEST_VIEW_DATE" ] && LATEST_VIEW_DATE=$(date +%F)
        jq -r --arg REPO "$REPO" '.views[] | [.timestamp[0:10], $REPO, "views", .count, .uniques] | @csv' ../views.json >> "$FILE"
        jq -r --arg REPO "$REPO" --arg DATE "$LATEST_VIEW_DATE" \
          '[ "\($DATE)~ 14-day total", $REPO, "views_total", .count, .uniques ] | @csv' ../views.json >> "$FILE"

        # 5. DEDUPLICATE AND SORT
        echo "$HEADER" > "$FILE.tmp"
        tail -n +2 "$FILE" | awk -F, '$3!="" {a[$1,$2,$3]=$0} END {for (i in a) print a[i]}' | sort -t, -k1,1r >> "$FILE.tmp"
        mv "$FILE.tmp" "$FILE"

        # 6. Commit and Push
        git add "$FILE"
        git commit -m "metrics: record clones and views for $REPO" || true
        git push