This commit is contained in:
fuzhongyun 2025-12-29 16:36:36 +08:00
commit 80d3ffa0e7
15 changed files with 560 additions and 0 deletions

12
.gitignore vendored Executable file
View File

@ -0,0 +1,12 @@
# Python-generated files
__pycache__/
*.py[oc]
.pytest*
build/
dist/
wheels/
*.egg-info
*.png
# Virtual environments
.venv

24
Dockerfile Normal file
View File

@ -0,0 +1,24 @@
# 使用官方轻量级 Python 镜像
FROM python:3.12-slim
# 设置工作目录
WORKDIR /app
# 安装依赖
COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt
# 复制源码和字体文件
# 注意:我们将当前目录的所有内容复制到 /app
COPY . .
# 确保 assets 目录存在 (以防本地没有创建但有字体文件散落在根目录的情况,虽然我们在 deploy 前已经整理了)
# 设置字体路径环境变量(可选,如果在代码中使用了默认值)
ENV FONT_PATH_REGULAR=/app/assets/simsun.ttc
ENV FONT_PATH_BOLD=/app/assets/simhei.ttf
# 暴露端口
EXPOSE 8000
# 启动命令
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "8000"]

97
README.md Executable file
View File

@ -0,0 +1,97 @@
# Excel2Pic API Service
这是一个轻量级的 Excel 转图片 API 服务,专为插件化部署设计。
它使用 FastAPI 构建,底层基于 `openpyxl``Pillow` (PIL) 实现高效的表格渲染,无需依赖任何 Office 组件或浏览器环境。
## 特性
* 🚀 **高性能**:基于 FastAPI原生异步支持。
* 📦 **极轻量**Docker 镜像体积 < 200MB
* 🎨 **纯 Python 实现**:无需 LibreOffice 或 Headless Chrome。
* 🔠 **中文支持**:内置宋体和黑体支持,解决中文乱码问题。
* 🐳 **一键部署**:提供 Dockerfile 和一键部署脚本。
## 目录结构
```
excel2pic/
├── app.py # FastAPI 应用入口
├── core/ # 核心逻辑模块
│ ├── renderer.py # 渲染引擎 (ExcelRenderer)
│ └── __init__.py
├── assets/ # 静态资源 (字体文件)
│ ├── simsun.ttc # 宋体
│ └── simhei.ttf # 黑体
├── tests/ # 单元测试
├── deploy.sh # 一键部署脚本
├── Dockerfile # Docker 构建文件
└── requirements.txt # 项目依赖
```
## 快速开始
### 1. 本地运行
**前置条件**: Python 3.12+
```bash
# 1. 创建并激活虚拟环境
python -m venv .venv
source .venv/bin/activate # Linux/Mac
# .venv\Scripts\activate # Windows
# 2. 安装依赖
pip install -r requirements.txt
# 3. 启动服务
uvicorn app:app --reload
```
### 2. Docker 部署
```bash
# 赋予执行权限
chmod +x deploy.sh
# 一键构建并启动
./deploy.sh
```
## API 文档
启动服务后,访问 `http://localhost:8000/docs` 查看交互式 Swagger 文档。
### 核心接口
**POST** `/api/v1/convert`
将上传的 Excel 文件转换为 PNG 图片。
* **参数**:
* `file`: (Required) Excel 文件 (.xlsx / .xls)
* `sheet_name`: (Optional) 指定要转换的工作表名称,默认为活动工作表。
* **响应**:
* `Content-Type`: `image/png` (直接返回图片二进制流)
**示例调用 (cURL)**:
```bash
curl -X POST "http://localhost:8000/api/v1/convert" \
-F "file=@/path/to/data.xlsx" \
-F "sheet_name=Sheet1" \
--output result.png
```
## 维护指南
### 添加新字体
1. 将 `.ttf``.ttc` 文件放入 `assets/` 目录。
2. 修改 `core/renderer.py` 中的 `__init__` 默认参数,或在初始化 `ExcelRenderer` 时传入路径。
3. 重新构建 Docker 镜像。
### 运行测试
```bash
export PYTHONPATH=$PYTHONPATH:.
pytest tests/
```

62
app.py Normal file
View File

@ -0,0 +1,62 @@
from fastapi import FastAPI, UploadFile, File, Form, HTTPException
from fastapi.responses import StreamingResponse, JSONResponse
from core.renderer import ExcelRenderer
import io
import logging
# Configure logging
logging.basicConfig(level=logging.INFO)
logger = logging.getLogger(__name__)
app = FastAPI(
title="Excel2Pic API",
description="A lightweight service to convert Excel sheets to images.",
version="0.1.0"
)
@app.post("/api/v1/convert", summary="Convert Excel to Image")
async def convert_excel(
file: UploadFile = File(..., description="The Excel file to convert"),
sheet_name: str = Form(None, description="Name of the sheet to convert (optional, defaults to active sheet)"),
):
"""
Convert an uploaded Excel file to a PNG image.
"""
# Validation
if not file.filename.endswith(('.xlsx', '.xls')):
raise HTTPException(status_code=400, detail="Invalid file format. Please upload .xlsx or .xls file.")
try:
# Read file content
contents = await file.read()
# Initialize Renderer
# Note: In a real deployment, font paths might come from env vars
renderer = ExcelRenderer(contents)
# Render
image_bytes = renderer.render_to_bytes(sheet_name=sheet_name)
# Return as streaming response
# Handle Chinese filenames in Content-Disposition
from urllib.parse import quote
filename = file.filename.split('.')[0] + ".png"
encoded_filename = quote(filename)
return StreamingResponse(
io.BytesIO(image_bytes),
media_type="image/png",
headers={"Content-Disposition": f"inline; filename*=utf-8''{encoded_filename}"}
)
except ValueError as ve:
# Often raised when sheet name is not found
logger.warning(f"Value Error: {str(ve)}")
raise HTTPException(status_code=400, detail=str(ve))
except Exception as e:
logger.error(f"Internal Server Error: {str(e)}", exc_info=True)
raise HTTPException(status_code=500, detail=f"An error occurred during conversion: {str(e)}")
@app.get("/health", summary="Health Check")
def health_check():
return {"status": "ok"}

BIN
assets/simhei.ttf Executable file

Binary file not shown.

BIN
assets/simsun.ttc Executable file

Binary file not shown.

BIN
assets/simsun.ttf Executable file

Binary file not shown.

0
core/__init__.py Normal file
View File

210
core/renderer.py Normal file
View File

@ -0,0 +1,210 @@
import io
import warnings
from typing import Optional, Union, Tuple, BinaryIO
from openpyxl import load_workbook
from openpyxl.worksheet.worksheet import Worksheet
from PIL import Image, ImageDraw, ImageFont
# Suppress warnings
warnings.filterwarnings('ignore')
class ExcelRenderer:
def __init__(self, file_content: Union[bytes, BinaryIO], font_path_regular: str = "assets/simsun.ttc", font_path_bold: str = "assets/simhei.ttf"):
"""
Initialize the renderer with Excel file content and font paths.
"""
self.file_content = file_content if isinstance(file_content, io.BytesIO) else io.BytesIO(file_content)
self.font_path_regular = font_path_regular
self.font_path_bold = font_path_bold
self._load_fonts()
def _load_fonts(self):
"""
Load fonts with fallback mechanisms.
"""
try:
self.font_regular = ImageFont.truetype(self.font_path_regular, 12)
except OSError:
# Fallback to default if custom font not found
try:
self.font_regular = ImageFont.truetype("arial.ttf", 12)
except OSError:
self.font_regular = ImageFont.load_default()
try:
self.font_bold = ImageFont.truetype(self.font_path_bold, 12)
except OSError:
try:
self.font_bold = ImageFont.truetype("arialbd.ttf", 12)
except OSError:
self.font_bold = ImageFont.load_default()
def render_to_bytes(self, sheet_name: Optional[str] = None, dpi: int = 200, padding: int = 20) -> bytes:
"""
Render the specified sheet to a PNG image and return bytes.
"""
img = self._render_image(sheet_name, padding)
output = io.BytesIO()
img.save(output, format='PNG', dpi=(dpi, dpi))
output.seek(0)
return output.getvalue()
def _render_image(self, sheet_name: Optional[str], padding: int) -> Image.Image:
"""
Internal method to draw the Excel sheet onto a PIL Image.
"""
wb = load_workbook(self.file_content, data_only=True)
if sheet_name is None:
sheet = wb.active
else:
if sheet_name in wb.sheetnames:
sheet = wb[sheet_name]
else:
raise ValueError(f"Sheet '{sheet_name}' not found. Available sheets: {wb.sheetnames}")
return self._draw_sheet(sheet, padding)
def _draw_sheet(self, sheet: Worksheet, padding: int) -> Image.Image:
cell_height = 40 # Default cell height
max_row = sheet.max_row
max_col = sheet.max_column
# Calculate column widths and image dimensions
col_widths_pixels = []
img_width = 2 * padding
for col in range(1, max_col + 1):
col_letter = sheet.cell(row=1, column=col).column_letter
# Get column width (approximate conversion)
col_dim = sheet.column_dimensions[col_letter]
col_width_excel = col_dim.width if col_dim.width else 10
# Excel width to pixels (approximate factor ~7 + padding)
width_px = int(col_width_excel * 7) + 5
col_widths_pixels.append(width_px)
img_width += width_px
img_height = max_row * cell_height + 2 * padding
# Create image
img = Image.new('RGB', (img_width, img_height), color='white')
draw = ImageDraw.Draw(img)
# Pre-calculate column x-positions
col_x_positions = [padding]
current_x = padding
for width in col_widths_pixels:
current_x += width
col_x_positions.append(current_x)
# Draw cells
for row in range(1, max_row + 1):
for col in range(1, max_col + 1):
cell = sheet.cell(row=row, column=col)
x1 = col_x_positions[col - 1]
y1 = padding + (row - 1) * cell_height
x2 = col_x_positions[col]
y2 = y1 + cell_height
self._draw_cell(draw, cell, x1, y1, x2, y2)
return img
def _draw_cell(self, draw: ImageDraw.ImageDraw, cell, x1, y1, x2, y2):
# Background color
fill_color = cell.fill.start_color.rgb
bg_color = self._parse_color(fill_color, default=(255, 255, 255))
# Draw background and border
draw.rectangle([x1, y1, x2, y2], fill=bg_color, outline=(200, 200, 200))
# Content
cell_value = cell.value
if cell_value is None:
return
text = self._format_cell_value(cell, cell_value)
if not text:
return
# Font handling
is_bold = cell.font and cell.font.bold
current_font = self.font_bold if is_bold else self.font_regular
# Font color
font_color_hex = cell.font.color.rgb if (cell.font and cell.font.color) else None
text_color = self._parse_color(font_color_hex, default=(0, 0, 0))
# Alignment
h_align = cell.alignment.horizontal if (cell.alignment and cell.alignment.horizontal) else 'left'
v_align = cell.alignment.vertical if (cell.alignment and cell.alignment.vertical) else 'center'
# Text rendering with simple truncation
self._draw_text(draw, text, x1, y1, x2, y2, current_font, text_color, h_align, v_align)
def _parse_color(self, color_code, default=(0, 0, 0)) -> Tuple[int, int, int]:
if not color_code or color_code == '00000000' or not isinstance(color_code, str):
return default
# Handle ARGB (Excel often uses this)
if len(color_code) > 6:
# Strip alpha if present (usually first 2 chars for ARGB)
# Example: FF000000 -> 000000 (Black), FFFFFFFF -> FFFFFF (White)
# Note: This is a simplification.
if color_code.startswith('FF') or len(color_code) == 8:
color_code = color_code[2:]
try:
return tuple(int(color_code[i:i + 2], 16) for i in (0, 2, 4))
except:
return default
def _format_cell_value(self, cell, value) -> str:
if isinstance(value, (int, float)):
# Simple number formatting
if cell.number_format:
if '0.00' in str(cell.number_format):
return format(value, '.2f')
elif '0.0' in str(cell.number_format):
return format(value, '.1f')
return str(value)
return str(value)
def _draw_text(self, draw, text, x1, y1, x2, y2, font, color, h_align, v_align):
# Calculate available width
max_width = x2 - x1 - 10
text_width = draw.textlength(text, font=font)
# Simple truncation if too long
if text_width > max_width and len(text) > 3:
# Estimate chars that fit
char_ratio = max_width / text_width
keep_chars = int(len(text) * char_ratio) - 2
if keep_chars > 0:
text = text[:keep_chars] + "..."
text_width = draw.textlength(text, font=font) # Re-measure
# Horizontal Position
if h_align == 'center':
text_x = x1 + (x2 - x1 - text_width) / 2
elif h_align == 'right':
text_x = x2 - text_width - 5
else: # left
text_x = x1 + 5
# Vertical Position (Approximate, using fixed height)
# Assuming font size 12 approx height 12-15 pixels
font_height = 12
if v_align == 'top':
text_y = y1 + 5
elif v_align == 'bottom':
text_y = y2 - font_height - 5
else: # center
text_y = y1 + (y2 - y1 - font_height) / 2
draw.text((text_x, text_y), text, fill=color, font=font)

40
deploy.sh Executable file
View File

@ -0,0 +1,40 @@
#!/bin/bash
set -e
# 定义变量
IMAGE_NAME="excel2pic"
CONTAINER_NAME="excel2pic_container"
PORT=8000
echo "开始部署 $IMAGE_NAME ..."
# 1. 构建 Docker 镜像
echo "构建 Docker 镜像..."
docker build -t $IMAGE_NAME:latest .
# 2. 检查并停止旧容器
if [ "$(docker ps -aq -f name=$CONTAINER_NAME)" ]; then
echo "停止并移除旧容器..."
docker stop $CONTAINER_NAME || true
docker rm $CONTAINER_NAME || true
fi
# 3. 启动新容器
echo "启动新容器..."
docker run -d \
--name $CONTAINER_NAME \
-p $PORT:8000 \
$IMAGE_NAME:latest
# 4. 验证部署
echo "等待服务启动..."
sleep 2
if docker ps | grep -q $CONTAINER_NAME; then
echo "部署成功!"
echo "API 文档地址: http://localhost:$PORT/docs"
echo "测试转换接口: POST http://localhost:$PORT/api/v1/convert"
else
echo "部署失败,容器未运行。"
docker logs $CONTAINER_NAME
exit 1
fi

8
requirements.txt Normal file
View File

@ -0,0 +1,8 @@
fastapi
uvicorn
python-multipart
pandas
openpyxl
pillow
pytest
httpx

BIN
tests/data.xlsx Executable file

Binary file not shown.

BIN
tests/data1.xlsx Executable file

Binary file not shown.

60
tests/test_api.py Normal file
View File

@ -0,0 +1,60 @@
import pytest
from fastapi.testclient import TestClient
from app import app
from openpyxl import Workbook
import io
client = TestClient(app)
@pytest.fixture
def sample_excel_file():
wb = Workbook()
ws = wb.active
ws.title = "APITest"
ws['A1'] = "API"
out = io.BytesIO()
wb.save(out)
out.seek(0)
return out
def test_health_check():
response = client.get("/health")
assert response.status_code == 200
assert response.json() == {"status": "ok"}
def test_convert_endpoint(sample_excel_file):
files = {'file': ('test.xlsx', sample_excel_file, 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet')}
response = client.post("/api/v1/convert", files=files)
assert response.status_code == 200
assert response.headers["content-type"] == "image/png"
assert len(response.content) > 0
def test_convert_invalid_file_type():
files = {'file': ('test.txt', io.BytesIO(b"dummy"), 'text/plain')}
response = client.post("/api/v1/convert", files=files)
assert response.status_code == 400
assert "Invalid file format" in response.json()["detail"]
def test_convert_specific_sheet(sample_excel_file):
# Re-create file because previous read might have consumed it if not handled carefully (TestClient usually handles this)
# But let's be safe and use the fixture which returns a new BytesIO if we construct it that way.
# Actually the fixture returns the same object, let's seek 0 just in case.
sample_excel_file.seek(0)
files = {'file': ('test.xlsx', sample_excel_file, 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet')}
data = {'sheet_name': 'APITest'}
response = client.post("/api/v1/convert", files=files, data=data)
assert response.status_code == 200
def test_convert_missing_sheet(sample_excel_file):
sample_excel_file.seek(0)
files = {'file': ('test.xlsx', sample_excel_file, 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet')}
data = {'sheet_name': 'MissingSheet'}
response = client.post("/api/v1/convert", files=files, data=data)
assert response.status_code == 400
assert "Sheet 'MissingSheet' not found" in response.json()["detail"]

47
tests/test_renderer.py Normal file
View File

@ -0,0 +1,47 @@
import pytest
from openpyxl import Workbook
import io
from core.renderer import ExcelRenderer
from PIL import Image
@pytest.fixture
def sample_excel_bytes():
wb = Workbook()
ws = wb.active
ws.title = "TestSheet"
ws['A1'] = "Hello"
ws['B1'] = "World"
ws['A2'] = 123
ws['B2'] = 456.78
# Add some color
from openpyxl.styles import PatternFill
fill = PatternFill(start_color="FFFF0000", end_color="FFFF0000", fill_type="solid")
ws['A1'].fill = fill
out = io.BytesIO()
wb.save(out)
out.seek(0)
return out.getvalue()
def test_renderer_initialization(sample_excel_bytes):
renderer = ExcelRenderer(sample_excel_bytes)
assert renderer is not None
def test_render_to_bytes(sample_excel_bytes):
renderer = ExcelRenderer(sample_excel_bytes)
img_bytes = renderer.render_to_bytes(sheet_name="TestSheet")
assert isinstance(img_bytes, bytes)
assert len(img_bytes) > 0
# Verify it's a valid image
img = Image.open(io.BytesIO(img_bytes))
assert img.format == "PNG"
assert img.width > 0
assert img.height > 0
def test_render_invalid_sheet(sample_excel_bytes):
renderer = ExcelRenderer(sample_excel_bytes)
with pytest.raises(ValueError, match="Sheet 'NonExistent' not found"):
renderer.render_to_bytes(sheet_name="NonExistent")