Added i18n component and related scripts

.github/workflows/deploy-pages.yml

@@ -30,6 +30,11 @@ jobs:
  sudo apt-get install -y texlive texlive-fonts-extra latexmk
  - name: Fetch Yarn dependencies
  run: yarn install --frozen-lockfile
+ - name: Clone translated_xmls
+ run: |
+ git clone -b translated_xmls https://github.com/source-academy/sicp.git translated_xmls
+ mv translated_xmls/* xml/
+ rm -r translated_xmls
  - name: Build
  run: |
  set -euxo pipefail

.github/workflows/translate-changed.yml

@@ -0,0 +1,55 @@
+on:
+ workflow_dispatch:
+ workflow_call:
+
+concurrency:
+ group: translate-changed
+ cancel-in-progress: true
+
+jobs:
+ translate-changed:
+ runs-on: ubuntu-20.04
+ steps:
+ - uses: actions/checkout@v4
+ - name: cd
+ run: |
+ cd i18n
+ - name: Use Node.js
+ uses: actions/setup-node@v4
+ with:
+ node-version: 20
+ cache: yarn
+ - name: Fetch Yarn Dependencies
+ run: |
+ yarn install
+ - name: Clone translated_xmls
+ run: |
+ git clone -b translated_xmls https://github.com/source-academy/sicp.git translation_output
+ - name: Get Changed Files
+ id: changed-files
+ uses: tj-actions/changed-files@ed68ef82c095e0d48ec87eccea555d944a631a4c # v46
+ with:
+ files: |
+ xml/**.xml
+ - name: Create .env
+ run: |
+ touch .env
+ echo API_MODEL=gpt-4.1-nano >> .env
+ echo API_KEY=${{ secrets.OPENAI_KEY }} >> .env
+ # echo API_KEY=${{ secrets.OPENAI_KEY2 }} >> .env
+ echo MAX_LEN=5000 >> .env
+ echo MAX_TRANSLATION_NO=10 >> .env
+ echo CONTEXT=3 >> .env
+ echo TOKEN_WARNING=7000 >> .env
+ - name: Translate Changed Files
+ if: steps.changed-files.outputs.anychanged == 'true'
+ env:
+ CHANGED_FILES: ${{ steps.changed_files.outputs.all_changed_files }}
+ run: |
+ npx tsx index.ts "${CHANGED_FILES[@]}"
+ - name: Deploy
+ uses: peaceiris/actions-gh-pages@v4
+ with:
+ github_token: ${{ secrets.GITHUB_TOKEN }}
+ publish_dir: ./translation_output
+ force_orphan: false # leave the possibility for direct modification on translated xmls

.github/workflows/translate-everything.yml

@@ -0,0 +1,43 @@
+on:
+ workflow_dispatch:
+ workflow_call:
+
+concurrency:
+ group: translate-everything
+ cancel-in-progress: true
+
+jobs:
+ translate-everything:
+ runs-on: ubuntu-20.04
+ steps:
+ - uses: actions/checkout@v4
+ - name: cd
+ run: |
+ cd i18n
+ - name: Use Node.js
+ uses: actions/setup-node@v4
+ with:
+ node-version: 20
+ cache: yarn
+ - name: Fetch Yarn Dependencies
+ run: |
+ yarn install
+ - name: Create .env
+ run: |
+ touch .env
+ echo API_MODEL=gpt-4.1-nano >> .env
+ echo API_KEY=${{ secrets.OPENAI_KEY }} >> .env
+ # echo API_KEY=${{ secrets.OPENAI_KEY2 }} >> .env
+ echo MAX_LEN=5000 >> .env
+ echo MAX_TRANSLATION_NO=10 >> .env
+ echo CONTEXT=3 >> .env
+ echo TOKEN_WARNING=7000 >> .env
+ - name: Run Translation
+ run: |
+ npx tsx index.ts
+ - name: Deploy
+ uses: peaceiris/actions-gh-pages@v4
+ with:
+ github_token: ${{ secrets.GITHUB_TOKEN }}
+ publish_dir: ./translation_output
+ force_orphan: false # leave the possiblity for direct modification on translated xmls

.gitignore

@@ -10,4 +10,8 @@ test_node_env/expect.txt
 test_node_env/result.txt
 test_node_env/node_modules
 .DS_Store
+.env
 *.icloud
+
+/xml_*
+/logs

i18n/.env.example

@@ -0,0 +1,28 @@
+API_KEY=
+
+# Model to be used for translation
+# ensure that selected assistant supports file search API
+# models that are tried and shown to work includes: o3-mini, 4o and 4o mini
+AI_MODEL=gpt-4.1-nano
+
+# maximum number of characters to be sent for translation per batch, default: 3000
+# smaller number of characters could lead to greater translation quality
+# but worse continuity
+MAX_LEN=5000
+
+#Optional, only specify if calling alternative LLM provider
+# Note: We require the Assistant API which only OpenAI provides as of 13 April 2025
+AI_BASEURL=
+
+# maximum number of concurent translations that are allowed
+# if unspecified default to 5
+MAX_TRANSLATION_NO=10
+
+# Number of previous messages to be referred to in a thread for LLM Translation, if unspecified, default to 3
+# Higher context size might lead to greater continuity in translation but higher cost and lower speed.
+CONTEXT=3
+
+# Number of token per query which when exceeded will log an error
+# Formula to derive an adequate value: MAX_LEN * CONTEXT * k, where k is a constant
+# Recommended k = 1.5
+TOKEN_WARNING=7000

i18n/.gitignore

@@ -0,0 +1,3 @@
+/node_modules
+/translation_output
+/logs

i18n/ai_files/zh_CN/dictionary.txt

@@ -0,0 +1 @@
+function: 函数