post_crawler/.gitea/workflows/ai-review.yml

name: AI Review

on:
  pull_request:
    types: [opened, synchronize, reopened]

permissions:
  contents: read
  pull-requests: read
  issues: write

jobs:
  ai-review:
    runs-on: linux
    timeout-minutes: 10
    steps:
      - name: Review pull request diff
        env:
          OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
          OPENAI_API_KEY_VAR: ${{ vars.OPENAI_API_KEY }}
          OPENAI_BASE_URL: ${{ secrets.OPENAI_BASE_URL }}
          OPENAI_BASE_URL_VAR: ${{ vars.OPENAI_BASE_URL }}
          OPENAI_MODEL: ${{ secrets.OPENAI_MODEL }}
          OPENAI_MODEL_VAR: ${{ vars.OPENAI_MODEL }}
          GITEA_TOKEN: ${{ secrets.GITEA_TOKEN }}
          GITEA_API_URL: ${{ github.api_url }}
          GITEA_REPOSITORY: ${{ github.repository }}
          GITEA_SHA: ${{ github.sha }}
          PR_NUMBER: ${{ github.event.pull_request.number }}
          PR_TITLE: ${{ github.event.pull_request.title }}
        shell: bash
        run: |
          set -euo pipefail

          OPENAI_API_KEY="${OPENAI_API_KEY:-${OPENAI_API_KEY_VAR:-}}"
          OPENAI_BASE_URL="${OPENAI_BASE_URL:-${OPENAI_BASE_URL_VAR:-https://api.openai.com/v1}}"
          OPENAI_MODEL="${OPENAI_MODEL:-${OPENAI_MODEL_VAR:-gpt-5}}"
          OPENAI_BASE_URL="${OPENAI_BASE_URL%/}"
          export OPENAI_API_KEY OPENAI_BASE_URL OPENAI_MODEL

          if [ -z "${OPENAI_API_KEY:-}" ]; then
            echo "OPENAI_API_KEY is not configured. Add it in repository Settings -> Actions -> Secrets."
            exit 1
          fi

          if [ -z "${GITEA_TOKEN:-}" ]; then
            echo "GITEA_TOKEN is not available. Check Actions token permissions for this repository."
            exit 1
          fi

          if [ -z "${PR_NUMBER:-}" ] || [ "${PR_NUMBER}" = "null" ]; then
            echo "This workflow only reviews pull requests."
            exit 0
          fi

          curl --fail-with-body -L \
            -H "Authorization: token ${GITEA_TOKEN}" \
            -H "Accept: text/plain" \
            "${GITEA_API_URL}/repos/${GITEA_REPOSITORY}/pulls/${PR_NUMBER}.diff" \
            -o pr.diff

          diff_bytes="$(wc -c < pr.diff | tr -d ' ')"
          if [ "${diff_bytes}" -eq 0 ]; then
            printf 'No diff content found for PR #%s.\n' "${PR_NUMBER}" > review.md
            should_fail=false
          else
            node <<'NODE'
          const fs = require('fs');

          const diff = fs.readFileSync('pr.diff', 'utf8');
          const title = process.env.PR_TITLE || '';
          const maxDiffChars = 120000;
          const truncated = diff.length > maxDiffChars;
          const inputDiff = truncated ? diff.slice(0, maxDiffChars) : diff;

          const payload = {
            model: process.env.OPENAI_MODEL || 'gpt-5',
            input: [
              {
                role: 'system',
                content: [
                  {
                    type: 'input_text',
                    text: [
                      'You are a senior code reviewer for a Gitea pull request.',
                      'Focus on correctness, security, data loss, concurrency, backward compatibility, and missing tests.',
                      'Ignore formatting-only comments unless they hide a real bug.',
                      'Return strict JSON only. Do not use markdown fences.',
                      'Schema: {"summary": string, "findings": [{"severity": "critical"|"high"|"medium"|"low", "file": string, "line": number|null, "title": string, "detail": string}], "should_fail": boolean}.',
                      'Set should_fail true only for critical or high confidence high severity issues.'
                    ].join(' ')
                  }
                ]
              },
              {
                role: 'user',
                content: [
                  {
                    type: 'input_text',
                    text: `PR title: ${title}\nDiff was truncated: ${truncated}\n\n${inputDiff}`
                  }
                ]
              }
            ]
          };

          fs.writeFileSync('openai-request.json', JSON.stringify(payload));
          NODE

            curl --fail-with-body "${OPENAI_BASE_URL}/responses" \
              -H "Authorization: Bearer ${OPENAI_API_KEY}" \
              -H "Content-Type: application/json" \
              -d @openai-request.json \
              -o openai-response.json

            node <<'NODE'
          const fs = require('fs');

          const response = JSON.parse(fs.readFileSync('openai-response.json', 'utf8'));

          function collectText(node, out = []) {
            if (!node) return out;
            if (typeof node === 'string') {
              out.push(node);
            } else if (Array.isArray(node)) {
              for (const item of node) collectText(item, out);
            } else if (typeof node === 'object') {
              if (typeof node.output_text === 'string') out.push(node.output_text);
              if (typeof node.text === 'string') out.push(node.text);
              if (node.content) collectText(node.content, out);
              if (node.output) collectText(node.output, out);
            }
            return out;
          }

          const raw = collectText(response).join('\n').trim();
          let review;
          try {
            review = JSON.parse(raw);
          } catch {
            review = {
              summary: raw || 'AI review returned no readable output.',
              findings: [],
              should_fail: false
            };
          }

          const findings = Array.isArray(review.findings) ? review.findings : [];
          const lines = [];
          lines.push('## AI Review');
          lines.push('');
          lines.push(review.summary || 'Review completed.');
          lines.push('');

          if (findings.length === 0) {
            lines.push('No high-confidence issues found.');
          } else {
            lines.push('Findings:');
            for (const finding of findings.slice(0, 20)) {
              const severity = finding.severity || 'medium';
              const file = finding.file || 'unknown file';
              const line = finding.line ? `:${finding.line}` : '';
              const title = finding.title || 'Potential issue';
              const detail = finding.detail || '';
              lines.push(`- [${severity}] ${file}${line} - ${title}`);
              if (detail) lines.push(`  ${detail}`);
            }
          }

          fs.writeFileSync('review.md', lines.join('\n') + '\n');
          fs.writeFileSync('should-fail.txt', review.should_fail ? 'true' : 'false');
          NODE

            should_fail="$(cat should-fail.txt)"
          fi

          node <<'NODE'
          const fs = require('fs');
          fs.writeFileSync('comment.json', JSON.stringify({
            body: fs.readFileSync('review.md', 'utf8')
          }));
          NODE

          curl --fail-with-body -X POST \
            -H "Authorization: token ${GITEA_TOKEN}" \
            -H "Content-Type: application/json" \
            -d @comment.json \
            "${GITEA_API_URL}/repos/${GITEA_REPOSITORY}/issues/${PR_NUMBER}/comments"

          if [ "${should_fail}" = "true" ]; then
            echo "AI review found high severity issues."
            exit 1
          fi