feat(website): i18n ja

.github/workflows/build-check.yml

@@ -75,5 +75,6 @@ jobs:
                   npm install -g yarn
                   yarn cache clean
                   export NODE_OPTIONS=--max-old-space-size=8192
-                  yarn && yarn build
+                  yarn
+                  PWA_SERVICE_WORKER_URL=https://doris.apache.org/sw.js yarn docusaurus build --locale en --locale zh-CN
                   rm -rf build

.github/workflows/cron-deploy-website.yml

@@ -39,7 +39,14 @@ jobs:
                   export NODE_OPTIONS=--max-old-space-size=8192
                   yarn
                   node ./scripts/update_github_info.js
-                  yarn build
+                  PWA_SERVICE_WORKER_URL=https://doris.apache.org/sw.js yarn docusaurus build --locale en --locale zh-CN
+                  if [ ! -d "./ja-build" ]; then
+                      echo "ja-build directory not found, aborting to avoid publishing incorrect ja content."
+                      exit 1
+                  fi
+                  rm -rf ./build/ja
+                  mkdir -p ./build/ja
+                  cp -a ./ja-build/. ./build/ja/
                   touch build/.dummy
                   cp .asf-site.yaml ./build/.asf.yaml
                   cp versions.json ./build/

.github/workflows/manual-deploy-website.yml

@@ -41,7 +41,15 @@ jobs:
                   npm install -g yarn
                   yarn cache clean
                   export NODE_OPTIONS=--max-old-space-size=8192
-                  yarn && yarn build
+                  yarn
+                  PWA_SERVICE_WORKER_URL=https://doris.apache.org/sw.js yarn docusaurus build --locale en --locale zh-CN
+                  if [ ! -d "./ja-build" ]; then
+                      echo "ja-build directory not found, aborting to avoid publishing incorrect ja content."
+                      exit 1
+                  fi
+                  rm -rf ./build/ja
+                  mkdir -p ./build/ja
+                  cp -a ./ja-build/. ./build/ja/
                   touch build/.dummy
                   cp .asf-site.yaml ./build/.asf.yaml
                   cp versions.json ./build/

.github/workflows/manual-i18n-translate-workflow.yaml

@@ -0,0 +1,173 @@
+name: Manual Translate Docs (EN -> JA)
+
+on:
+  workflow_dispatch:
+    inputs:
+      source_dir:
+        description: "Directory containing English docs to translate"
+        required: true
+        default: "docs"
+      target_dir:
+        description: "Target i18n root directory for Japanese docs"
+        required: true
+        default: "i18n/ja"
+      file_glob:
+        description: "Glob pattern for files to translate"
+        required: false
+        default: "**/*.{md,mdx}"
+
+
+jobs:
+  prepare_translation_input:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Use Node.js
+        uses: actions/setup-node@v3
+        with:
+            node-version: 20     
+
+      - name: Test AWS API Access
+        env:
+          AWS_API_KEY: ${{ secrets.AWS_API_KEY }}
+        run: |
+          echo "Testing AWS API Access..."
+          node scripts/i18n/test-aws-access.js
+
+      - name: Install dependencies
+        run: | 
+          npm install -g yarn 
+          yarn install --frozen-lockfile
+
+      - name: Prepare translation input
+        run: |
+          mkdir -p .i18n/input
+
+          node scripts/i18n/collect-files.js \
+            "${{ inputs.source_dir }}" \
+            "${{ inputs.file_glob }}" \
+            files.json
+
+          echo "Collected files:"
+          cat files.json
+
+          if [ "$(jq length files.json)" -eq 0 ]; then
+            echo "❌ No markdown files found under ${{ inputs.source_dir }}"
+            exit 1
+          fi
+
+          node scripts/i18n/prepare.js \
+            files.json \
+            .i18n/input \
+            manifest.json \
+            "${{ inputs.target_dir }}"
+
+      - name: Debug:list .i18n/input
+        run: |
+          echo "Contents of .i18n/input:"
+          ls -la .i18n/input || true
+          echo "File count: $(find .i18n/input -type f | wc -l)"
+
+      - name: Verify intermediate files exist
+        run: |
+          COUNT=$(find .i18n/input -type f | wc -l)
+          if [ "$COUNT" -eq 0 ]; then
+            echo "❌ No intermediate files generated in .i18n/input"
+            exit 1
+          fi
+
+      - uses: actions/upload-artifact@v4
+        with:
+          name: i18n-input
+          path: .i18n/input
+          include-hidden-files: true
+
+  translate_docs:
+    runs-on: ubuntu-latest
+    needs: prepare_translation_input
+    steps:
+      - uses: actions/checkout@v4
+
+      - uses: actions/download-artifact@v4
+        with:
+          name: i18n-input
+          path: .i18n/input
+          include-hidden-files: true
+
+      - name: Translate via Claude API
+        env:
+          AWS_API_KEY: ${{ secrets.AWS_API_KEY }}
+        run: |
+          mkdir -p .i18n/output
+          node scripts/i18n/translate-claude.js \
+            .i18n/input \
+            .i18n/output
+
+      - uses: actions/upload-artifact@v4
+        with:
+          name: i18n-output
+          path: .i18n/output
+
+  post_process:
+    runs-on: ubuntu-latest
+    needs: translate_docs
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Use Node.js
+        uses: actions/setup-node@v3
+        with:
+            node-version: 20     
+
+      - name: Install dependencies
+        run: | 
+          npm install -g yarn 
+          yarn install --frozen-lockfile
+
+      - uses: actions/download-artifact@v4
+        with:
+          name: i18n-output
+          path: .i18n/output
+          include-hidden-files: true
+
+      - name: Restore markdown and write ja docs
+        run: |
+          node scripts/i18n/post-process.js \
+            .i18n/output \
+            "${{ inputs.target_dir }}"
+
+      - uses: actions/upload-artifact@v4
+        with:
+          name: i18n-ja-docs
+          path: ${{ inputs.target_dir }}
+          include-hidden-files: true
+
+  create_pr:
+    runs-on: ubuntu-latest
+    needs: post_process
+    permissions:
+      contents: write
+      pull-requests: write
+    steps:
+      # 1. checkout 仓库
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - uses: actions/download-artifact@v4
+        with:
+          name: i18n-ja-docs
+          path: ${{ inputs.target_dir }}
+          include-hidden-files: true
+
+      - uses: peter-evans/create-pull-request@v6
+        with:
+          base: master
+          branch: auto/i18n-ja-${{ github.run_id }}
+          commit-message: chore(i18n):auto translate docs to ja
+          title: "[auto-i18n] Translate docs to Japanese"
+          body: |
+            This PR contains automatically translated Japanese docs.
+          add-paths: |
+            ${{ inputs.target_dir }}/**

crowdin.yml

@@ -0,0 +1,3 @@
+files:
+  - source: /docs/**/*.md
+    translation: i18n/ja

docs/ai/vector-search/hnsw.md

@@ -43,7 +43,7 @@ However, this family of algorithms has two main issues:
 1. As the dataset grows, the number of iterations in the routing phase increases roughly following a power‑law.
 2. It is difficult to build a high‑quality proximity graph; local clusters and poor global connectivity are very common.
 
-![low_quality_pgraph](/images/vector-search/low_quality_pgraph.png)
+![low quality pgraph](/images/vector-search/low-quality-pgraph.png)
 
 The figure above intuitively shows the shape of a problematic proximity graph. Darker points represent nodes with poorer connectivity; some nodes barely have any neighbors at all, which makes them very hard to reach during search.
 

docs/ai/vector-search/performance.md

@@ -33,7 +33,7 @@ For larger-scale results (10M/100M-level datasets) in single-node and distribute
 All machines used in the tests have 16 CPU cores and 64 GB memory, with CPUs of model Intel(R) Xeon(R) Platinum 8369B CPU @ 2.70GHz. FE and BE are co-located on a single 16C64GB machine (note: this is not a recommended production deployment; in production, FE and BE should be deployed separately). The tested version is Apache Doris 4.0.2. The dataset used is VectorDBBench Performance768D1M, which contains 1 million vectors of 768 dimensions.
 
 ## Results
-![performance](/images/vector-search/ann_index_performance_0.jpg)
+![performance](/images/vector-search/ann-index-performance-0.jpg)
 
 ## Analysis
 ```

docusaurus.config.js

@@ -4,15 +4,15 @@ const VERSIONS = require('./versions.json');
 const { markdownBoldPlugin } = require('./config/markdown-bold-plugin');
 const { DEFAULT_VERSION } = require('./src/constant/version');
 
-const lightCodeTheme = themes.dracula;
+	const lightCodeTheme = themes.dracula;
 
-const logoImg = '/images/logo-doris.svg';
+	const logoImg = '/images/logo-doris.svg';
 
-function getDocsVersions() {
-    const result = {};
-    VERSIONS.map(version => {
-        if (version === 'current') {
-            result[version] = {
+	function getDocsVersions() {
+	    const result = {};
+	    VERSIONS.map(version => {
+	        if (version === 'current') {
+	            result[version] = {
                 label: 'Dev',
                 path: 'dev',
                 banner: 'unreleased',
@@ -54,7 +54,7 @@ const config = {
     trailingSlash: true,
     i18n: {
         defaultLocale: 'en',
-        locales: ['en', 'zh-CN'],
+        locales: ['en', 'zh-CN', 'ja'],
         localeConfigs: {
             en: {
                 label: 'English',
@@ -64,13 +64,17 @@ const config = {
                 label: '中文',
                 htmlLang: 'zh-Hans-CN',
             },
+            ja: {
+                label: '日本語',
+                htmlLang: 'ja-JP',
+            },
         },
     },
     scripts: ['/js/custom-script.js',
         {
             async: true,
             src: 'https://widget.kapa.ai/kapa-widget.bundle.js',
-            'data-website-id': 'a5fb90df-217a-4097-95c0-80490220314b',
+            'data-website-id': '10c44fc4-7375-4b37-9554-b39132bcb855',
             'data-modal-title': 'Apache Doris AI',
             'data-project-name': 'Apache Doris Website',
             'data-button-hide': "true",
@@ -181,14 +185,14 @@ const config = {
                     //     // if (versionDocsDirPath === 'versioned_docs/version-dev') {
                     //     //     return `https://github.com/apache/doris-website/edit/master/docs/${locale}/docs/${docPath}`;
                     //     // }
-                    // },
-                    showLastUpdateAuthor: false,
-                    showLastUpdateTime: true,
-                    remarkPlugins: [markdownBoldPlugin, require('remark-math')],
-                    rehypePlugins: [
-                        [
-                            require('rehype-katex'),
-                            {
+	                    // },
+	                    showLastUpdateAuthor: false,
+	                    showLastUpdateTime: false,
+	                    remarkPlugins: [markdownBoldPlugin, require('remark-math')],
+	                    rehypePlugins: [
+	                        [
+	                            require('rehype-katex'),
+	                            {
                                 strict: process.env.CI === 'true' || process.env.GITHUB_ACTIONS === 'true' ? false : 'warn',
                             }
                         ]
@@ -214,7 +218,11 @@ const config = {
                     createSitemapItems: async params => {
                         const { defaultCreateSitemapItems, ...rest } = params;
                         const items = await defaultCreateSitemapItems(rest);
-                        for (let item of items) {
+                        const filteredItems = items.filter(item => {
+                            const pathname = new URL(item.url).pathname.replace(/\/+$/, '');
+                            return !['/search', '/ja/search', '/zh-CN/search'].includes(pathname);
+                        });
+                        for (let item of filteredItems) {
                             if (item.url.includes('docs')) {
                                 item.changefreq = 'daily';
                                 item.priority = 0.8;
@@ -223,7 +231,7 @@ const config = {
                                 item.priority = 0.2;
                             }
                         }
-                        return items;
+                        return filteredItems;
                     },
                 },
             }),
@@ -234,11 +242,11 @@ const config = {
             '@yang1666204/docusaurus-search-local',
             {
                 hashed: true,
-                language: ['en', 'zh'],
+                language: ['en', 'zh', 'ja'],
                 highlightSearchTermsOnTargetPage: true,
                 // indexPages: true,
                 indexDocs: true,
-                docsRouteBasePath: ['/docs/2.1', '/docs/3.x', '/docs/4.x', '/docs/dev'],
+                docsRouteBasePath: ['docs','ja/docs','zh-CN/docs'],
                 indexBlog: false,
                 explicitSearchResultPath: true,
                 searchBarShortcut: true,
@@ -416,6 +424,34 @@ const config = {
                     },
                 ],
             },
+            docNavbarJA: {
+                title: '',
+                logo: {
+                    alt: 'Apache Doris',
+                    src: logoImg,
+                },
+                items: [
+                    {
+                        type: 'search',
+                        position: 'left',
+                        className: 'docs-search',
+                    },
+                    {
+                        type: 'localeDropdown',
+                        position: 'right',
+                    },
+                    {
+                        type: 'docsVersionDropdown',
+                        position: 'right',
+                        dropdownItemsAfter: [
+                            {
+                                label: 'Archived',
+                                to: `/ja/archive-docs`,
+                            },
+                        ],
+                    },
+                ],
+            },
             communityNavbar: {
                 title: '',
                 logo: {

i18n/zh-CN/docusaurus-plugin-content-docs/current/ai/vector-search/hnsw.md

@@ -39,7 +39,7 @@ HNSW (Hierarchical Navigable Small World) 算法是在 [Efficient and robust app
 1. 随着数据量的增加，查询路由阶段的迭代次数会有幂律级别的增加。
 2. 构建一个高质量的 proximity graph 很难，很容易出现局部聚集导致图不具备全局联通的问题。
 
-![low_quality_pgraph](/images/vector-search/low_quality_pgraph.png)
+![low quality pgraph](/images/vector-search/low-quality-pgraph.png)
 
 这里有一张示意图可以直观展示一个不理想的 proximity graph 的形状。图中颜色越深的点表示该点的连通性越差，可以看到有个别点几乎没有找到自己的邻居，那么在搜索阶段这些点就很难被路由到。