@zohodesk/codestandard-validator 1.1.4 → 1.2.4-exp-2

package/build/ai/prompts/PROMPT2.md

@@ -0,0 +1,159 @@
+# 🧠 How To Use AI Prompt Files for Code Review (Before Commit)
+
+This document explains **how every developer should use AI prompt files** to self-review code **before committing changes**.
+
+The goal is to:
+- Catch issues early  
+- Reduce PR review back-and-forth  
+- Maintain consistent standards across the team  
+
+---
+
+## 📌 Which Prompt Should I Use?
+
+| Prompt file | When to use | Mandatory |
+|------------|-------------|-----------|
+| AIPrompts/code-review/UATReview.md | Before committing UAT changes | ✅ Yes |
+| AIPrompts/code-review/ComponentHTMLCSSReview.md | HTML / CSS / UI changes | ✅ Yes |
+| AIPrompts/code-review/MergeRequestReview.md | Before raising PR / MR | ✅ Yes |
+
+---
+
+
+## 🤖 CODEOWNERS-Based Prompt Detection
+
+```text
+Decide which review prompt(s) must be used based on file changes.
+
+Reference:
+resources/code_owners/code_owners.yaml
+
+Input:
+- Uncommitted git diff or attached .diff file
+
+Rules:
+
+1. If a **Design** change is detected
+   (HTML / CSS / SCSS / UI files),
+   - Use the rules from: AIPrompts/code-review/ComponentHTMLCSSReview.md
+
+2. If **UAT_StepDefinitions** or **UAT_Features** changes are detected,
+   - Use the rules from: AIPrompts/code-review/UATReview.md
+
+3. For all other changes,
+   decide the required prompt(s) using code_owners.yaml.
+   First check for any attached .diff file; otherwise, use uncommitted changes.
+
+4. If multiple types of files are changed,
+   run multiple prompts.
+
+Output:
+- List the required prompt file(s)
+- Short reason for each
+```
+
+---
+
+
+## Review Before Commit
+
+This ensures:
+- Code follows agreed standards  
+- Naming and structure are correct  
+- CSS / HTML rules are respected  
+- Obvious bugs are caught early  
+
+---
+
+### Step 1: Open GitHub Copilot Chat
+
+In VS Code:
+- **Mac:** `Cmd + Shift + I`
+- **Windows/Linux:** `Ctrl + Shift + I`
+
+---
+
+### Step 3:
+## For UAT Review
+
+```text
+Use the rules from:
+AIPrompts/code-review/UATReview.md
+
+Review the following uncommitted git diff.
+Identify violations with file and line number.
+Suggest fixes.
+
+Here is the git diff:
+Take the attached diff file or review the uncommitted changes.
+
+Output:
+- The reviewed report must be readable and neatly aligned.
+- Push the detailed review output as a separate HTML file under:
+  AIPrompts/outputs/UATReview.md.html
+```
+
+---
+
+## 🎨 For HTML / CSS / UI Changes
+
+```text
+Use the rules from:
+AIPrompts/code-review/ComponentHTMLCSSReview.md
+
+Review the following uncommitted git diff.
+Focus on class naming, CSS rules, and structure.
+
+Here is the git diff:
+First check for any attached .diff file; otherwise, use uncommitted changes.
+
+Output:
+- The reviewed report must be readable and neatly aligned.
+- Push the detailed review output as a separate HTML file under:
+  AIPrompts/outputs/ComponentHTMLCSSReview.md.html
+```
+
+---
+
+## 🔁 Before Creating PR / MR ( Merge request review)
+
+```text
+Use the rules from:
+AIPrompts/code-review/MergeRequestReview.md
+
+Review the overall changes and provide PR-level feedback.
+
+Here is the git diff:
+Use the attached MR .diff file. If none is attached, ignore this step.
+
+Output:
+- The reviewed report must be readable and neatly aligned.
+- Push the detailed review output as a separate HTML file under:
+  AIPrompts/outputs/MergeRequestReview.md.html
+```
+
+---
+
+### Step 3: Fix Issues
+
+- Fix all reported issues  
+- Re-run `git diff`  
+- Re-run the review if needed  
+
+---
+
+## 📌 Important Guidelines
+
+- When modifying prompt files, please post in the team channel or have a discussion. The DRI must review the changes.  
+- Prompt files are the single source of truth.  
+
+---
+
+## 🎯 Why We Do This
+
+- Right prompt for the right change  
+- Less reviewer confusion  
+- Faster approvals  
+- Shared responsibility  
+
+This is **self-verification**, designed to help everyone write better code with less friction.

package/build/ai/prompts/PROMPT3.md

@@ -0,0 +1,64 @@
+# Code Review Prompt for HelpCentre Client Merge Requests
+
+## Overview
+
+This document acts as the **main entry point** for reviewing Merge Requests (MRs).  
+It defines the mandatory pre-review checks and links to **separate, focused review documents** that should be used based on the scope of changes.
+
+Use only the required review files depending on what the MR contains.
+
+---
+
+## Pre-Review Checklist
+
+Before reviewing the code, confirm the following:
+
+- MR description clearly explains the purpose and scope of changes
+- Related ticket or issue is linked
+- MR targets the correct branch
+- No merge conflicts are present
+- UAT report is attached
+  - If UAT is not required, the reason must be clearly explained
+
+---
+
+## Review Scope Selection
+
+### Components & CSS Files Review Checklist
+
+Use this checklist **only when the MR includes**:
+
+- React components
+- HTML / JSX changes
+- CSS / CSS Modules updates
+
+➡️ Refer to:  
+**[`ComponentHTMLCSSReview.md`](./ComponentHTMLCSSReview.md)**
+
+Review all component structure, accessibility, styling, naming conventions, and CSS rules using that file.
+
+---
+
+### UAT Files Review Checklist
+
+Use this checklist **only when the MR includes**:
+
+- UAT automation changes
+- Feature files (`.feature`)
+- Spec files (`.spec.js`)
+- Page Object Models (POM)
+- Assertions, selectors, or test configurations
+
+➡️ Refer to:  
+**[`UATReview.md`](./UATReview.md)**
+
+Review UAT folder structure, Playwright usage, locator strategy, i18n enforcement, assertions, and configuration standards using this document.
+
+---
+
+## Notes
+
+- Do not run all review documents blindly
+- Select the review file based on the **actual changes in the MR**
+- If an MR includes **both UI and UAT changes**, review using **both documents**
+- Any deviation from standards must be clearly documented in the MR

package/build/ai/prompts/PROMPT4.md

@@ -0,0 +1,64 @@
+# Code Review Prompt for HelpCentre Client Merge Requests
+
+## Overview
+
+This document acts as the **main entry point** for reviewing Merge Requests (MRs).  
+It defines the mandatory pre-review checks and links to **separate, focused review documents** that should be used based on the scope of changes.
+
+Use only the required review files depending on what the MR contains.
+
+---
+
+## Pre-Review Checklist
+
+Before reviewing the code, confirm the following:
+
+- MR description clearly explains the purpose and scope of changes
+- Related ticket or issue is linked
+- MR targets the correct branch
+- No merge conflicts are present
+- UAT report is attached
+  - If UAT is not required, the reason must be clearly explained
+
+---
+
+## Review Scope Selection
+
+### Components & CSS Files Review Checklist
+
+Use this checklist **only when the MR includes**:
+
+- React components
+- HTML / JSX changes
+- CSS / CSS Modules updates
+
+➡️ Refer to:  
+**[`ComponentHTMLCSSReview.md`](./ComponentHTMLCSSReview.md)**
+
+Review all component structure, accessibility, styling, naming conventions, and CSS rules using that file.
+
+---
+
+### UAT Files Review Checklist
+
+Use this checklist **only when the MR includes**:
+
+- UAT automation changes
+- Feature files (`.feature`)
+- Spec files (`.spec.js`)
+- Page Object Models (POM)
+- Assertions, selectors, or test configurations
+
+➡️ Refer to:  
+**[`UATReview.md`](./UATReview.md)**
+
+Review UAT folder structure, Playwright usage, locator strategy, i18n enforcement, assertions, and configuration standards using this document.
+
+---
+
+## Notes
+
+- Do not run all review documents blindly
+- Select the review file based on the **actual changes in the MR**
+- If an MR includes **both UI and UAT changes**, review using **both documents**
+- Any deviation from standards must be clearly documented in the MR

package/build/ai/provider-factory.js

@@ -0,0 +1,19 @@
+"use strict";
+
+const {
+  OllamaProvider
+} = require('./providers/OllamaProvider');
+function createProvider(config) {
+  const providerName = (config.provider || 'ollama').toLowerCase();
+  switch (providerName) {
+    case 'ollama':
+      return new OllamaProvider(config);
+    // case 'openai': return new OpenAIProvider(config);
+    // case 'anthropic': return new AnthropicProvider(config);
+    default:
+      throw new Error(`Unsupported provider: ${providerName}`);
+  }
+}
+module.exports = {
+  createProvider
+};

package/build/ai/providers/OllamaProvider.js

@@ -0,0 +1,106 @@
+"use strict";
+
+class AIProvider {
+  constructor(config = {}) {
+    this.config = config;
+  }
+  supportsStreaming() {
+    return false;
+  }
+  async generate(request) {
+    // request: { prompt, fileName?, onToken? }
+    throw new Error('Not implemented');
+  }
+}
+class OllamaProvider extends AIProvider {
+  constructor(config) {
+    super(config);
+  }
+  supportsStreaming() {
+    return !!this.config.stream;
+  }
+  async generate({
+    prompt,
+    fileName,
+    onToken
+  }) {
+    const url = `${this.config.baseUrl}${this.config.endpoint}`;
+    const body = {
+      model: this.config.model,
+      prompt,
+      stream: !!this.config.stream,
+      raw: this.config.model === 'deepseek-r1',
+      format: 'json',
+      options: {
+        temperature: this.config.temperature,
+        num_predict: this.config.maxTokens
+      }
+    };
+    const controller = new AbortController();
+    const timeout = setTimeout(() => controller.abort(), this.config.timeoutMs);
+    try {
+      const response = await fetch(url, {
+        method: 'POST',
+        headers: {
+          'Content-Type': 'application/json'
+        },
+        body: JSON.stringify(body),
+        signal: controller.signal
+      });
+      clearTimeout(timeout);
+      if (!response.ok) {
+        throw new Error(`Ollama request failed: ${response.status} ${response.statusText}`);
+      }
+      if (this.supportsStreaming()) {
+        const reader = response.body.getReader();
+        const decoder = new TextDecoder('utf-8');
+        let fullText = '';
+        let isDone = false;
+        while (!isDone) {
+          const {
+            value,
+            done
+          } = await reader.read();
+          if (done) break;
+          const chunk = decoder.decode(value, {
+            stream: true
+          });
+          for (const line of chunk.split('\n')) {
+            const trimmed = line.trim();
+            if (!trimmed) continue;
+            try {
+              const obj = JSON.parse(trimmed);
+              if (obj.response) {
+                fullText += obj.response;
+                if (typeof onToken === 'function') onToken(obj.response);
+              }
+              if (obj.done) {
+                isDone = true;
+                break;
+              }
+            } catch {
+              // ignore partial lines
+            }
+          }
+        }
+        return {
+          text: fullText,
+          fileName
+        };
+      } else {
+        const data = await response.json();
+        const text = typeof data.response === 'string' ? data.response : '';
+        return {
+          text,
+          fileName
+        };
+      }
+    } catch (err) {
+      clearTimeout(timeout);
+      throw err;
+    }
+  }
+}
+module.exports = {
+  OllamaProvider
+};

package/build/ai/render.js

@@ -0,0 +1,157 @@
+"use strict";
+
+const {
+  marked
+} = require('marked');
+const {
+  markedTerminal
+} = require('marked-terminal');
+marked.use(markedTerminal({
+  reflowText: true,
+  width: 80,
+  showSectionPrefix: false
+}));
+
+/**
+ * Render a markdown string to styled terminal output.
+ * @param {string} md - Markdown text
+ */
+function renderMarkdown(md) {
+  if (!md || typeof md !== 'string') return '';
+  return marked.parse(md);
+}
+
+/**
+ * Print rendered markdown to stdout.
+ * @param {string} md
+ */
+function printMarkdown(md) {
+  process.stdout.write(renderMarkdown(md));
+}
+
+/**
+ * Format a review result object as readable markdown.
+ * @param {Object} result
+ * @param {string} [fileName]
+ * @returns {string}
+ */
+function formatReviewResult(result, fileName) {
+  if (!result) return '';
+  const lines = [];
+  if (fileName) {
+    lines.push(`## 📄 Review: \`${fileName}\``);
+  }
+  if (result.rawResponse) {
+    lines.push('\n> ⚠️ Could not parse structured JSON. Raw response below:\n');
+    // Wrap raw summary in a fenced code block if it looks like code,
+    // otherwise render as-is (it may already be markdown)
+    const summary = result.summary || '(empty)';
+    if (summary.includes('function ') || summary.includes('const ') || summary.includes('import ')) {
+      lines.push('```');
+      lines.push(summary);
+      lines.push('```');
+    } else {
+      lines.push(summary);
+    }
+    return lines.join('\n');
+  }
+  if (result.overallScore != null) {
+    lines.push(`### Overall Score: **${result.overallScore} / 50**\n`);
+  }
+  if (result.categories) {
+    lines.push('### Categories\n');
+    lines.push('| Category | Score |');
+    lines.push('|----------|-------|');
+    for (const [key, val] of Object.entries(result.categories)) {
+      const label = key.replace(/([A-Z])/g, ' $1').replace(/^./, s => s.toUpperCase());
+      lines.push(`| ${label} | ${val.score ?? '-'} / 10 |`);
+    }
+    lines.push('');
+    for (const [key, val] of Object.entries(result.categories)) {
+      if (val.issues && val.issues.length > 0) {
+        const label = key.replace(/([A-Z])/g, ' $1').replace(/^./, s => s.toUpperCase());
+        lines.push(`#### ${label} Issues\n`);
+        val.issues.forEach(issue => lines.push(`- ${issue}`));
+        lines.push('');
+      }
+    }
+  }
+  if (result.criticalIssues && result.criticalIssues.length > 0) {
+    lines.push('### 🚨 Critical Issues\n');
+    result.criticalIssues.forEach(issue => lines.push(`- ${issue}`));
+    lines.push('');
+  }
+  if (result.suggestions && result.suggestions.length > 0) {
+    lines.push('### 💡 Suggestions\n');
+    result.suggestions.forEach(s => lines.push(`- ${s}`));
+    lines.push('');
+  }
+  if (result.summary) {
+    lines.push(`### Summary\n\n${result.summary}\n`);
+  }
+  return lines.join('\n');
+}
+
+/**
+ * A streaming renderer that buffers tokens and renders complete markdown blocks.
+ * Usage:
+ *   const sr = new StreamRenderer();
+ *   onToken: token => sr.write(token);
+ *   // after streaming ends:
+ *   sr.flush();
+ */
+class StreamRenderer {
+  constructor() {
+    this._buffer = '';
+    this._rendered = '';
+  }
+
+  /**
+   * Feed a token chunk. Renders complete paragraphs/lines as they arrive.
+   * @param {string} token
+   */
+  write(token) {
+    this._buffer += token;
+
+    // Render when we have at least one complete line (ends with \n)
+    const lastNewline = this._buffer.lastIndexOf('\n');
+    if (lastNewline !== -1) {
+      const ready = this._buffer.slice(0, lastNewline + 1);
+      this._rendered += ready;
+      this._buffer = this._buffer.slice(lastNewline + 1);
+
+      // Clear previous output area and re-render accumulated markdown
+      // For simplicity, render only the new complete lines
+      const output = renderMarkdown(ready);
+      process.stdout.write(output);
+    }
+  }
+
+  /**
+   * Flush remaining buffer and render final output.
+   * @returns {string} The full raw text that was streamed
+   */
+  flush() {
+    if (this._buffer.length > 0) {
+      this._rendered += this._buffer;
+      const output = renderMarkdown(this._buffer);
+      process.stdout.write(output);
+      this._buffer = '';
+    }
+    return this._rendered;
+  }
+
+  /**
+   * Get the full raw accumulated text.
+   * @returns {string}
+   */
+  getRawText() {
+    return this._rendered + this._buffer;
+  }
+}
+module.exports = {
+  renderMarkdown,
+  printMarkdown,
+  formatReviewResult,
+  StreamRenderer
+};

package/build/ai/run-review.js

@@ -0,0 +1,50 @@
+// const path = require('path');
+// const { reviewFile, reviewFiles, reviewFolder } = require('./ollama-service');
+// const { printMarkdown, formatReviewResult, StreamRenderer } = require('./render');
+
+// const CONFIG = {
+//   provider: 'ollama',
+//   model: 'deepseek-r1:8b',
+//   temperature: 0.2,
+//   maxTokens: 4096,
+//   timeoutMs: 120000,
+//   stream: true
+// };
+
+// const PROMPT_FOLDER = path.resolve(__dirname, 'prompts');
+
+// const REVIEW_FOLDER = '/Users/raja-17710/deskApp/linter_tools/codestandard-validator/src/mutation';
+
+// const FOLDER_EXTENSIONS = ['.js', '.ts'];
+
+// async function runAll() {
+
+//   printMarkdown('\n---\n\n## Folder Review\n');
+//   try {
+//     const results = await reviewFolder(REVIEW_FOLDER, {
+//       config: CONFIG,
+//       promptFolder: PROMPT_FOLDER,
+//       extensions: FOLDER_EXTENSIONS,
+//       concurrency: 1,
+//       streamToConsole: true,
+//       onResult: out => {
+//         // Render each completed file review as formatted markdown
+//         printMarkdown('\n' + formatReviewResult(out, out.fileName) + '\n');
+//       }
+//     });
+//     printMarkdown('\n---\n\n### Final Summary\n');
+//     printMarkdown(`Reviewed **${results.length}** file(s).\n`);
+//   } catch (err) {
+//     printMarkdown(`\n> ❌ **reviewFolder error:** ${err.message}\n`);
+//   }
+// }
+
+// runAll()
+//   .then(() => {
+//     printMarkdown('\n---\n\n✅ **Done.**\n');
+//   })
+//   .catch(err => {
+//     printMarkdown(`\n> ❌ **Fatal error:** ${err.message}\n`);
+//     process.exit(1);
+//   });
+"use strict";