✅build: add features to check links status and workflow to generate l…

…ink status page
xxrjun · Apr 26, 2024 · 927c2cd · 927c2cd
1 parent 432ac6f
commit 927c2cd
Show file tree

Hide file tree

Showing 4 changed files with 193 additions and 0 deletions.
diff --git a/.github/workflows/link-status-report.yml b/.github/workflows/link-status-report.yml
@@ -0,0 +1,36 @@
+name: Generate Link Status Report
+
+on:
+  push:
+    branches:
+      - main
+
+jobs:
+  generate_report:
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout repository
+      uses: actions/checkout@v3
+
+    - name: Set up Python
+      uses: actions/setup-python@v4
+      with:
+        python-version: '3.x'
+
+    - name: Install dependencies
+      run: |
+        python -m pip install --upgrade pip
+        pip install aiohttp
+
+    - name: Generate link status report
+      run: |
+        python utils/link_status_report.py
+
+    - name: Deploy report
+      uses: peaceiris/actions-gh-pages@v3
+      with:
+        github_token: ${{ secrets.GH_TOKEN }}
+        publish_dir: ./
+        destination_dir: link-status-report
+        keep_files: true
diff --git a/docs/index.md b/docs/index.md
@@ -7,3 +7,6 @@
 > 作為自己學習 Computer Science 的紀錄與未來規劃，為自己在這茫茫碼海中留下一點痕跡
 
 > 期許自己的血液裡都流著程式碼的那天
+
+
+> [**Link Status Report**](https://xxrjun.github.io/cs-resources/link-status-report/links_report.html)
diff --git a/utils/README.md b/utils/README.md
@@ -0,0 +1,17 @@
+> [!TIP]
+> Use [./link_statuc_repot.py](./link_statuc_repot.py) to check all links status in this folder.
+
+
+## Usage
+
+Install the required packages:
+
+```bash
+pip install -r requirements.txt
+```
+
+Run the script:
+
+```bash
+python link_statuc_repot.py
+```
diff --git a/utils/link_status_report.py b/utils/link_status_report.py
@@ -0,0 +1,137 @@
+import asyncio
+import aiohttp
+import re
+import webbrowser
+import os
+
+# Regular expression to find Markdown links
+markdown_link_re = r"\[([^\]]+)\]\((http[s]?://[^\s]+)\)"
+
+async def is_link_active(url, session, semaphore):
+    async with semaphore:  # Limit the number of concurrent requests
+        try:
+            async with session.head(url, allow_redirects=True, timeout=5) as response:
+                return ("✓", "green") if response.status == 200 else (f"{response.status}", "red")
+        except asyncio.TimeoutError:
+            return ("✕", "orange")
+        except aiohttp.ClientError as e:
+            return ("✕", "red")
+
+async def check_links_in_markdown(file_path):
+    results = []
+
+    with open(file_path, 'r', encoding='utf-8') as file:
+        content = file.read()
+
+    links = re.findall(markdown_link_re, content)
+
+    semaphore = asyncio.Semaphore(30)  # Adjust the concurrency level as needed
+
+    async with aiohttp.ClientSession() as session:
+        tasks = [is_link_active(url, session, semaphore) for _, url in links]
+        statuses = await asyncio.gather(*tasks)
+
+    for (text, url), (status, color) in zip(links, statuses):
+        # remove '**' from text
+        text = text.replace('**', '')
+        results.append((text, status, url, color))
+
+    return results
+
+async def generate_report(folder_path, report_path):
+    markdown_files = []
+    for root, dirs, files in os.walk(folder_path):
+        for file in files:
+            if file.endswith('.md'):
+                markdown_files.append(os.path.join(root, file))
+
+    html_content = """
+    <html>
+    <head>
+        <title>Link Status Report</title>
+        <style>
+            body {
+                font-family: Arial, sans-serif;
+                margin: 0;
+                padding: 20px;
+                background-color: #1c1c1c;
+                color: #f2f2f2;
+            }
+            h1 {
+                color: #f2f2f2;
+                text-align: center;
+                margin-bottom: 30px;
+            }
+            h2 {
+                color: #f2f2f2;
+                margin-top: 40px;
+            }
+            table {
+                border-collapse: collapse;
+                width: 100%;
+                margin-bottom: 30px;
+            }
+            th, td {
+                border: 1px solid #444;
+                padding: 12px;
+                text-align: left;
+            }
+            th {
+                background-color: #333;
+                font-weight: bold;
+            }
+            tr:nth-child(even) {
+                background-color: #282828;
+            }
+            a {
+                color: #6fb3d2;
+                text-decoration: none;
+            }
+            a:hover {
+                text-decoration: underline;
+            }
+            .status {
+                font-weight: bold;
+                text-align: center;
+                width: 60px;
+            }
+            .green {
+                color: #8bc34a;
+            }
+            .red {
+                color: #f44336;
+            }
+            .orange {
+                color: #ff9800;
+            }
+        </style>
+    </head>
+    <body>
+        <h1>Link Status Report</h1>
+    """
+
+    for file in markdown_files:
+        results = await check_links_in_markdown(file)
+
+        html_content += f"<h2>{file}</h2>"
+        html_content += "<table><tr><th>Link Text</th><th class='status'>Status</th><th>URL</th></tr>"
+
+        for text, status, url, color in results:
+            html_content += f"<tr><td>{text}</td><td class='status {color}'>{status}</td><td><a href='{url}' target='_blank'>{url}</a></td></tr>"
+
+        html_content += "</table>"
+
+    html_content += "</body></html>"
+
+    with open(report_path, 'w', encoding='utf-8') as report_file:
+        report_file.write(html_content)
+
+    report_path = 'file://' + os.path.realpath(report_path)
+    webbrowser.open(report_path)
+    return report_path
+
+if __name__ == "__main__":
+    markdown_folder_path = '../docs'
+    report_path = '../links_status_report.html'
+    report_path = asyncio.run(generate_report(markdown_folder_path, report_path))
+    print(f"Link status report generated for markdown files in {markdown_folder_path} at {report_path}")
Original file line number	Diff line number	Diff line change
Expand Up		@@ -7,3 +7,6 @@
		> 作為自己學習 Computer Science 的紀錄與未來規劃，為自己在這茫茫碼海中留下一點痕跡

		> 期許自己的血液裡都流著程式碼的那天


		> [Link Status Report](https://xxrjun.github.io/cs-resources/link-status-report/links_report.html)