feat: add custom styling and release automation (#7)

- Implement custom CSS injection and URL rewriting to allow users to style PDFs and fix cross-environment links.
- Add release-please GitHub Action to automate the release process and enable secure NPM publishing.

Author: orange-guo

.github/workflows/release-please.yml

@@ -0,0 +1,45 @@
+name: release-please
+
+on:
+  push:
+    branches:
+      - main
+
+permissions:
+  contents: write
+  pull-requests: write
+  id-token: write # Required for Trusted Publishing
+
+jobs:
+  release-please:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: googleapis/release-please-action@v4
+        id: release
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+          release-type: node
+
+      # The following steps only run after a release PR is merged
+      - uses: actions/checkout@v4
+        if: ${{ steps.release.outputs.release_created }}
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        if: ${{ steps.release.outputs.release_created }}
+        with:
+          node-version: '20'
+          registry-url: 'https://registry.npmjs.org'
+
+      - name: Install dependencies
+        if: ${{ steps.release.outputs.release_created }}
+        run: yarn install
+
+      - name: Build project
+        if: ${{ steps.release.outputs.release_created }}
+        run: yarn build
+
+      - name: Publish to NPM (Trusted Publishing)
+        if: ${{ steps.release.outputs.release_created }}
+        run: npm publish --provenance --access public
+        # NODE_AUTH_TOKEN is NO LONGER NEEDED with Trusted Publishing

.gitignore

@@ -0,0 +1,19 @@
+# Dependency directories
+node_modules/
+
+# Build outputs
+dist/
+*.pdf
+
+# IDE and OS files
+.DS_Store
+.idea/
+.vscode/
+
+# Package manager lockfiles (if multiple exist, usually only one is kept)
+# Keeping yarn.lock as it was already in the repo, ignoring others if they are temporary
+package-lock.json
+
+# Environment variables
+.env
+.env.local

README.md

@@ -13,8 +13,10 @@ You can find a sample PDF generated by this tool at [doccusaurus.pdf](./example/
 * **Full Site PDF Export**: Exports all documentation pages from a Docusaurus site into a single, cohesive PDF file.
 * **Concurrent Processing**: Leverages Puppeteer's concurrency capabilities to speed up the page content fetching process.
 * **Custom Cover Page**: Supports adding a custom PDF cover page using either a URL or a local file path.
-* **Automatic Table of Contents (TOC) Generation**: Automatically generates a clickable PDF table of contents based on the Docusaurus sidebar structure.
+* **Automatic Table of Contents (TOC) Generation**: Automatically generates a professional, clickable PDF table of contents with dot leaders based on the Docusaurus sidebar structure.
 * **Navigable Internal Links**: Rewrites all internal links within the documentation so they remain clickable and navigable in a PDF reader.
+* **Custom CSS Injection**: Supports injecting custom CSS styles directly into the generated PDF for fine-tuned layout control.
+* **Link Rewriting**: Allows rewriting site domains/URLs within hyperlinks to ensure cross-references work correctly across environments.
 * **Configurable Margins**: Supports adjusting PDF page margins via command-line arguments.
 
 ## Installation
@@ -39,10 +41,17 @@ npx docusaurus-docs-to-pdf -h
 
 ### Examples
 
-**Generate pdf from Docusaurus website**
+**Generate PDF with custom margins and concurrency**
 
 ```bash
-docusaurus-docs-to-pdf --docs-url https://docusaurus.io/docs --pdf-path docusaurus.pdf --pdf-cover-image https://docusaurus.io/img/docusaurus_keytar.svg
-# or
-npx docusaurus-docs-to-pdf --docs-url https://docusaurus.io/docs --pdf-path docusaurus.pdf --pdf-cover-image https://docusaurus.io/img/docusaurus_keytar.svg
-```
+docusaurus-docs-to-pdf --docs-url http://localhost:3000/docs --pdf-path local-docs.pdf --pdf-margin-mm 20 --page-concurrency 10
+```
+
+**Advanced: Inject custom CSS and rewrite links**
+
+```bash
+docusaurus-docs-to-pdf --docs-url http://localhost:3000/docs \
+  --pdf-path docs.pdf \
+  --css "h1 { color: #2e8555; }" \
+  --site-rewrite "http://localhost:3000=https://docs.example.com"
+```

src/docusaurus.ts

@@ -176,10 +176,6 @@ export async function getElementOuterHtml(selector: string): Promise<string> {
     const element: HTMLElement | null = document.querySelector(selector);
 
     if (element) {
-        // Apply CSS style to force a page break *after* this element in the generated PDF.
-        // This is crucial for ensuring each extracted document content block starts on a new page.
-        element.style.pageBreakAfter = 'always';
-        
         // Return the outer HTML (including the element itself and its content).
         return element.outerHTML;
     } else {
@@ -198,33 +194,57 @@ export async function getElementOuterHtml(selector: string): Promise<string> {
  * as it converts original relative/absolute URLs (e.g., `/docs/my-page`) into
  * anchor links (e.g., `#my-page-id`) that point to elements within the merged document.
  *
+ * It also supports rewriting specific site domains/URLs to new ones, which is useful
+ * for redirecting links to production environments while fetching content from development/local ones.
+ *
  * @param pathToAnchors An array of tuples, where each tuple contains:
  * - `string`: The original path or href of a link (e.g., '/docs/introduction').
  * - `string`: The corresponding target anchor ID within the merged document
  * (e.g., 'id-timestamp-random').
+ * @param siteRewrites Optional. An array of [from, to] tuples for domain/URL rewriting.
  * @returns {Promise<void>} A Promise that resolves when all matching links have been rewritten.
  */
-export async function rewriteLinks(pathToAnchors: [string, string][]): Promise<void> {
+export async function rewriteLinks(pathToAnchors: [string, string][], siteRewrites?: [string, string][]): Promise<void> {
     // Select all <a> elements that have an 'href' attribute.
     // Cast to HTMLAnchorElement for proper TypeScript type inference and access to link-specific properties.
     const allLinks = document.querySelectorAll('a[href]');
-    
+
     for (let i = 0; i < allLinks.length; i++) {
         const element = allLinks[i] as HTMLAnchorElement;
         // Get the value of the 'href' attribute.
-        let linkPath = element.getAttribute('href') || '';
-        
+        let href = element.getAttribute('href') || '';
+
+        // Priority 1: Internal Documentation Anchor Rewriting
         // Iterate through the provided map of original paths to new anchor IDs.
+        let matchedInternal = false;
         for (let j = 0; j < pathToAnchors.length; j++) {
             const [path, anchor] = pathToAnchors[j];
             // If the current link's href matches an original path in the map,
             // rewrite its href to point to the new anchor ID.
-            if (linkPath === path) {
+            if (href === path) {
                 element.setAttribute('href', '#' + anchor);
+                matchedInternal = true;
                 // Break inner loop as we found a match for this link
                 break;
             }
         }
+
+        // Skip site rewrite if it's already an internal documentation link
+        if (matchedInternal) {
+            continue;
+        }
+
+        // Priority 2: External/Site Domain Rewriting
+        // If site rewrite rules are provided, check if the href matches any rule.
+        if (siteRewrites && siteRewrites.length > 0) {
+            for (const [from, to] of siteRewrites) {
+                if (href.includes(from)) {
+                    element.setAttribute('href', href.replace(from, to));
+                    // Break after the first matching rewrite rule
+                    break;
+                }
+            }
+        }
     }
 }