init

Author: haozewu

.gitignore

@@ -0,0 +1,84 @@
+# This file is used to ignore files which are generated
+# ----------------------------------------------------------------------------
+.qtcreator
+*.spec
+
+*~
+*.autosave
+*.a
+*.core
+*.moc
+*.o
+*.obj
+*.orig
+*.rej
+*.so
+*.so.*
+*_pch.h.cpp
+*_resource.rc
+*.qm
+.#*
+*.*#
+core
+!core/
+tags
+.DS_Store
+.directory
+*.debug
+Makefile*
+*.prl
+*.app
+moc_*.cpp
+ui_*.h
+qrc_*.cpp
+Thumbs.db
+*.res
+*.rc
+/.qmake.cache
+/.qmake.stash
+
+# qtcreator generated files
+*.pro.user*
+*.qbs.user*
+CMakeLists.txt.user*
+
+# xemacs temporary files
+*.flc
+
+# Vim temporary files
+.*.swp
+
+# Visual Studio generated files
+*.ib_pdb_index
+*.idb
+*.ilk
+*.pdb
+*.sln
+*.suo
+*.vcproj
+*vcproj.*.*.user
+*.ncb
+*.sdf
+*.opensdf
+*.vcxproj
+*vcxproj.*
+
+# MinGW generated files
+*.Debug
+*.Release
+
+# Python byte code
+*.pyc
+
+# Binaries
+# --------
+*.dll
+*.exe
+
+# Directories with generated files
+.moc/
+.obj/
+.pch/
+.rcc/
+.uic/
+/build*/

README.md

@@ -0,0 +1,51 @@
+# Translate File Windows (TFW)
+
+A lightweight Windows desktop tool for translating Word (`.docx`) documents between Chinese, English, French, and Spanish. The application supports optional automatic source-language detection and generates the translated file in the same directory with a language-tag suffix.
+
+## Background
+
+For many non-native English researchers, academic papers are often first written in their native language for clarity and efficiency. Once the manuscript reaches a mature stage, it needs to be translated into English for submission, and in some cases translated back into the local language again for internal reviews, presentations, or institutional reporting.
+
+Performing paragraph-by-paragraph translation manually is time-consuming and error-prone, especially for long technical documents. This tool was developed to streamline that process while offering a basic level of content protection during translation.
+
+To reduce full-document exposure when using online translation endpoints, paragraphs are translated in a randomized order. Combined with a resumable state file, users can manually switch network nodes or proxies during a run, allowing different parts of the document to be processed by different endpoints. This design provides a practical compromise between usability and privacy for completed academic work.
+
+## Features
+
+* Translate Word (`.docx`) documents between **zh / en / fr / es** in any direction
+* Optional automatic detection of the source language
+* Drag-and-drop a `.docx` onto the window to start translation immediately
+* Paragraphs are translated in a randomized order to avoid sequential content exposure
+* Resumable translation via a temporary state file
+* Supports manual switching of translation nodes or network proxies mid-run
+* Background translation thread with a progress bar and live status updates
+* Output file is automatically saved alongside the input file (example: `input.docx` -> `input-en.docx`)
+
+## Download (Windows EXE)
+
+To use the application, download the **prebuilt Windows executable** from the **Release page**. No Python runtime is required.
+
+1. Go to the **Release** section of this repository.
+2. Download the latest release asset: `TranslateFileWindows.exe`.
+3. (Optional) Verify the checksum if provided.
+4. Double-click the executable to launch the application.
+
+You do **not** need to clone the repository or build the project locally.
+
+## Usage
+
+Once launched:
+
+1. Drag a Word (`.docx`) file onto the application window, or click to select a file.
+2. Choose the source language (or enable auto-detection) and the target language.
+3. Click **Start** to begin translation.
+4. The progress bar and status text will update as paragraphs are processed.
+5. After completion, the translated file is saved in the same directory as the original file, with a `-<lang>` suffix (e.g., `-en`, `-cn`, `-fr`, `-es`).
+
+## Resume & Fault Tolerance
+
+If the translation process stops midway, simply relaunch the program and translate the same document again.
+
+* The application automatically resumes using a temporary state file stored alongside the input document.
+* The state file is removed after a successful run.
+* Paragraphs are processed in a shuffled order, so restarting the task or switching network nodes during execution does not invalidate already completed translations.

main_en2cn.py

@@ -0,0 +1,19 @@
+# This Python file uses the following encoding: utf-8
+import sys
+
+from PySide6.QtWidgets import QApplication
+
+from widget import TranslatorWidget
+
+
+def main() -> None:
+    """Launch the GUI translator."""
+    app = QApplication([])
+    window = TranslatorWidget()
+    window.resize(640, 200)
+    window.show()
+    sys.exit(app.exec())
+
+
+if __name__ == "__main__":
+    main()

pyproject.toml

@@ -0,0 +1,5 @@
+[project]
+name = "PySide Project"
+
+[tool.pyside6-project]
+files = ["widget.py", "translator.py"]

requirements.txt

@@ -0,0 +1,3 @@
+PySide6
+python-docx==0.8.11
+googletrans==4.0.2

scripts/build.ps1

@@ -0,0 +1,58 @@
+param(
+    [string]$Entry = "main_en2cn.py",
+    [string]$Name = "TranslateFileWindows"
+)
+
+Set-StrictMode -Version Latest
+$ErrorActionPreference = "Stop"
+
+$root = Resolve-Path (Join-Path $PSScriptRoot "..")
+
+$venvActivate = Join-Path $root ".qtcreator/Python_3_12_10venv/Scripts/Activate.ps1"
+if (-not (Test-Path $venvActivate)) {
+    Write-Error "Virtual env activate script not found at $venvActivate"
+    exit 1
+}
+
+Write-Host "Activating virtual environment ..." -ForegroundColor Cyan
+. $venvActivate
+
+$entryPath = Join-Path $root $Entry
+if (-not (Test-Path $entryPath)) {
+    Write-Error "Entry file not found: $entryPath"
+    exit 1
+}
+
+$icoPath = Join-Path $root "static/favicon.ico"
+
+$iconArg = @("--icon", $icoPath)
+if (-not (Test-Path $icoPath)) {
+    Write-Error "favicon.ico not found at $icoPath. Place your icon there before building."
+    exit 1
+}
+
+$buildDir = Join-Path $root "build"
+$distDir = Join-Path $root "dist"
+
+foreach ($dir in @($buildDir, $distDir)) {
+    if (Test-Path $dir) {
+        Remove-Item -Recurse -Force $dir
+    }
+}
+
+$dataArg = @("--add-data", "$($root)\\static\\favicon.ico;static")
+
+Write-Host "Building $Name from $entryPath ..." -ForegroundColor Cyan
+& pyinstaller --clean --noconfirm --onefile --windowed --name $Name @iconArg @dataArg $entryPath
+
+if ($LASTEXITCODE -ne 0) {
+    Write-Error "PyInstaller build failed with exit code $LASTEXITCODE"
+    exit $LASTEXITCODE
+}
+
+$exePath = Join-Path -Path (Join-Path $root "dist") -ChildPath "$Name.exe"
+if (Test-Path $exePath) {
+    Write-Host "Build succeeded: $exePath" -ForegroundColor Green
+} else {
+    Write-Warning "Build finished, but dist/$Name.exe not found."
+}

static/favicon.ico

@@ -0,0 +1,135 @@
+import asyncio
+import json
+import os
+import random
+import time
+from typing import Generator, Tuple
+
+from docx import Document
+from googletrans import Translator
+
+# Language codes used across the app
+LANG_SOURCE_CODES = ["auto", "zh-cn", "en", "fr", "es"]
+LANG_TARGET_CODES = ["zh-cn", "en", "fr", "es"]
+
+
+def load_or_initialize_state(state_file: str, num_paragraphs: int) -> dict:
+    if os.path.exists(state_file):
+        with open(state_file, "r", encoding="utf-8") as f:
+            return json.load(f)
+    return {
+        "translated_count": 0,
+        "paragraphs": [],
+        "order": [],
+        "translated_paragraphs": [None] * num_paragraphs,
+    }
+
+
+def save_state(state_file: str, state: dict) -> None:
+    with open(state_file, "w", encoding="utf-8") as f:
+        json.dump(state, f, ensure_ascii=False, indent=4)
+
+
+class DocumentTranslator:
+    """Translate a Word document paragraph by paragraph with resumable state."""
+
+    def __init__(
+        self,
+        input_file: str,
+        output_file: str,
+        state_file: str,
+        src_lang: str = "auto",
+        dest_lang: str = "zh-cn",
+    ) -> None:
+        self.input_file = input_file
+        self.output_file = output_file
+        self.state_file = state_file
+        self.src_lang = src_lang
+        self.dest_lang = dest_lang
+
+    def translate(self) -> Generator[Tuple[int, dict], None, None]:
+        """
+        Run the translation and yield (progress_percent, status_payload).
+
+        Progress updates can be consumed by a GUI thread without blocking the UI.
+        """
+        doc = Document(self.input_file)
+        paragraphs = [(i, para.text) for i, para in enumerate(doc.paragraphs)]
+
+        state = load_or_initialize_state(self.state_file, len(paragraphs))
+
+        if not state["paragraphs"]:
+            state["paragraphs"] = paragraphs
+            state["order"] = list(range(len(paragraphs)))
+            random.shuffle(state["order"])
+            save_state(self.state_file, state)
+
+        translator = Translator()
+        loop = asyncio.new_event_loop()
+        asyncio.set_event_loop(loop)
+
+        try:
+            total = len(state["order"])
+            for i in range(state["translated_count"], total):
+                original_index = state["order"][i]
+                index, text = state["paragraphs"][original_index]
+
+                if not text.strip():
+                    state["translated_paragraphs"][index] = text
+                    status = {
+                        "event": "skip_empty",
+                        "index": i + 1,
+                        "total": total,
+                    }
+                else:
+                    result = translator.translate(
+                        text, src=self.src_lang, dest=self.dest_lang
+                    )
+                    # googletrans 3.4.0+ returns a coroutine; handle both sync/async
+                    if asyncio.iscoroutine(result):
+                        result = loop.run_until_complete(result)
+                    translated_text = result.text
+                    state["translated_paragraphs"][index] = translated_text
+                    status = {
+                        "event": "translated",
+                        "index": i + 1,
+                        "total": total,
+                        "src": self.src_lang,
+                        "dest": self.dest_lang,
+                    }
+                    time.sleep(random.uniform(3, 5))
+
+                state["translated_count"] = i + 1
+                save_state(self.state_file, state)
+                progress = int(((i + 1) / total) * 100)
+                yield progress, status
+
+            translated_doc = Document()
+            for para in state["translated_paragraphs"]:
+                if para:
+                    translated_doc.add_paragraph(para)
+
+            translated_doc.save(self.output_file)
+            yield 100, {
+                "event": "completed",
+                "output": self.output_file,
+                "src": self.src_lang,
+                "dest": self.dest_lang,
+            }
+        finally:
+            # clean up async client
+            try:
+                loop.run_until_complete(translator.client.aclose())
+            except Exception:
+                pass
+            loop.close()
+
+            # clean state only when translation completes
+            if (
+                os.path.exists(self.state_file)
+                and state.get("translated_count") == len(state.get("order", []))
+            ):
+                try:
+                    os.remove(self.state_file)
+                except OSError:
+                    pass