Upload 4 files

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+msyh.ttc filter=lfs diff=lfs merge=lfs -text
+msyhbd.ttc filter=lfs diff=lfs merge=lfs -text
+TwitterColorEmoji.ttf filter=lfs diff=lfs merge=lfs -text

TwitterColorEmoji.ttf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0bceb7ee4941569eab695a785332f7287d59062ec1e9921fc2966e3a005b301
+size 14477376

generate_image.py

@@ -0,0 +1,939 @@
+# utf-8
+# image_generator.py
+"""
+Advanced image card generator with markdown and emoji support
+"""
+import math
+import random
+import os
+from PIL import Image, ImageDraw, ImageFont, ImageOps
+import emoji
+import re
+from dataclasses import dataclass
+from typing import List, Tuple, Optional, Dict
+from datetime import datetime
+
+
+@dataclass
+class TextStyle:
+    """文本样式定义"""
+    font_name: str = 'regular'  # regular, bold, emoji
+    font_size: int = 30  # 字体大小
+    indent: int = 0  # 缩进像素
+    line_spacing: int = 15  # 行间距
+    is_title: bool = False  # 是否为标题
+    is_category: bool = False  # 是否为分类标题
+    keep_with_next: bool = False  # 是否与下一行保持在一起
+
+
+@dataclass
+class TextSegment:
+    """文本片段定义"""
+    text: str  # 文本内容
+    style: TextStyle  # 样式
+    original_text: str = ''  # 原始文本（用于调试）
+
+
+@dataclass
+class ProcessedLine:
+    """处理后的行信息"""
+    text: str  # 实际文本内容
+    style: TextStyle  # 样式信息
+    height: int = 0  # 行高
+    line_count: int = 1  # 实际占用行数
+
+
+class FontManager:
+    """字体管理器"""
+
+    def __init__(self, font_paths: Dict[str, str]):
+        self.fonts = {}
+        self.font_paths = font_paths
+        self._initialize_fonts()
+
+    def _initialize_fonts(self):
+        """初始化基础字体"""
+        sizes = [30, 35, 40]  # 基础字号
+        for size in sizes:
+            self.fonts[f'regular_{size}'] = ImageFont.truetype(self.font_paths['regular'], size)
+            self.fonts[f'bold_{size}'] = ImageFont.truetype(self.font_paths['bold'], size)
+        # emoji字体
+        self.fonts['emoji_30'] = ImageFont.truetype(self.font_paths['emoji'], 30)
+
+    def get_font(self, style: TextStyle) -> ImageFont.FreeTypeFont:
+        """获取对应样式的字体"""
+        if style.font_name == 'emoji':
+            return self.fonts['emoji_30']
+
+        base_name = 'bold' if style.font_name == 'bold' or style.is_title or style.is_category else 'regular'
+        font_key = f'{base_name}_{style.font_size}'
+
+        if font_key not in self.fonts:
+            # 动态创建新字号的字体
+            self.fonts[font_key] = ImageFont.truetype(
+                self.font_paths['bold' if base_name == 'bold' else 'regular'],
+                style.font_size
+            )
+
+        return self.fonts[font_key]
+
+
+def get_gradient_styles() -> List[Dict[str, tuple]]:
+    """
+    获取精心设计的背景渐变样式
+    """
+    return [
+
+        # Mac 高级白
+        {
+            "start_color": (246, 246, 248),  # 珍珠白
+            "end_color": (250, 250, 252)  # 云雾白
+        },
+        {
+            "start_color": (245, 245, 247),  # 奶白色
+            "end_color": (248, 248, 250)  # 象牙白
+        },
+        # macOS Monterey 风格
+        {
+            "start_color": (191, 203, 255),  # 淡蓝紫
+            "end_color": (255, 203, 237)  # 浅粉红
+        },
+        {
+            "start_color": (168, 225, 255),  # 天空蓝
+            "end_color": (203, 255, 242)  # 清新薄荷
+        },
+
+        # 优雅渐变系列
+        {
+            "start_color": (255, 209, 209),  # 珊瑚粉
+            "end_color": (243, 209, 255)  # 淡紫色
+        },
+        {
+            "start_color": (255, 230, 209),  # 奶橘色
+            "end_color": (255, 209, 247)  # 粉紫色
+        },
+
+        # 清新通透
+        {
+            "start_color": (213, 255, 219),  # 嫩绿色
+            "end_color": (209, 247, 255)  # 浅蓝色
+        },
+        {
+            "start_color": (255, 236, 209),  # 杏橘色
+            "end_color": (255, 209, 216)  # 浅玫瑰
+        },
+
+        # 高级灰调
+        {
+            "start_color": (237, 240, 245),  # 珍珠灰
+            "end_color": (245, 237, 245)  # 薰衣草灰
+        },
+        {
+            "start_color": (240, 245, 255),  # 云雾蓝
+            "end_color": (245, 240, 245)  # 淡紫灰
+        },
+
+        # 梦幻糖果色
+        {
+            "start_color": (255, 223, 242),  # 棉花糖粉
+            "end_color": (242, 223, 255)  # 淡紫丁香
+        },
+        {
+            "start_color": (223, 255, 247),  # 薄荷绿
+            "end_color": (223, 242, 255)  # 天空蓝
+        },
+
+        # 高饱和度系列
+        {
+            "start_color": (255, 192, 203),  # 粉红色
+            "end_color": (192, 203, 255)  # 淡紫蓝
+        },
+        {
+            "start_color": (192, 255, 238),  # 碧绿色
+            "end_color": (238, 192, 255)  # 淡紫色
+        },
+
+        # 静谧系列
+        {
+            "start_color": (230, 240, 255),  # 宁静蓝
+            "end_color": (255, 240, 245)  # 柔粉色
+        },
+        {
+            "start_color": (245, 240, 255),  # 淡紫色
+            "end_color": (240, 255, 240)  # 清新绿
+        },
+
+        # 温柔渐变
+        {
+            "start_color": (255, 235, 235),  # 温柔粉
+            "end_color": (235, 235, 255)  # 淡雅紫
+        },
+        {
+            "start_color": (235, 255, 235),  # 嫩芽绿
+            "end_color": (255, 235, 245)  # 浅粉红
+        }
+    ]
+
+
+def create_gradient_background(width: int, height: int) -> Image.Image:
+    """创建渐变背景 - 从左上到右下的对角线渐变"""
+    gradient_styles = get_gradient_styles()
+    style = random.choice(gradient_styles)
+    start_color = style["start_color"]
+    end_color = style["end_color"]
+
+    # 创建基础图像
+    base = Image.new('RGB', (width, height))
+    draw = ImageDraw.Draw(base)
+
+    # 计算渐变
+    for y in range(height):
+        for x in range(width):
+            # 计算当前位置到左上角的相对距离 (对角线渐变)
+            # 使用 position 在 0 到 1 之间表示渐变程度
+            position = (x + y) / (width + height)
+
+            # 为每个颜色通道计算渐变值
+            r = int(start_color[0] * (1 - position) + end_color[0] * position)
+            g = int(start_color[1] * (1 - position) + end_color[1] * position)
+            b = int(start_color[2] * (1 - position) + end_color[2] * position)
+
+            # 绘制像素
+            draw.point((x, y), fill=(r, g, b))
+
+    return base
+
+
+def get_theme_colors() -> Tuple[tuple, str, bool]:
+    """获取主题颜色配置"""
+    current_hour = datetime.now().hour
+    current_minute = datetime.now().minute
+
+    if (current_hour == 8 and current_minute >= 30) or (9 <= current_hour < 19):
+        use_dark = random.random() < 0.1
+    else:
+        use_dark = True
+
+    if use_dark:
+        # 深色毛玻璃效果: 深色半透明背景(50%透明度) + 白色文字
+        return ((50, 50, 50, 128), "#FFFFFF", True)  # alpha值调整为128实现50%透明度
+    else:
+        # 浅色毛玻璃效果: 白色半透明背景(50%透明度) + 黑色文字
+        return ((255, 255, 255, 128), "#000000", False)  # alpha值调整为128实现50%透明度
+
+
+def create_rounded_rectangle(image: Image.Image, x: int, y: int, w: int, h: int, radius: int, bg_color: tuple):
+    """创建圆角毛玻璃矩形"""
+    # 创建透明背景的矩形
+    rectangle = Image.new('RGBA', (int(w), int(h)), (0, 0, 0, 0))
+    draw = ImageDraw.Draw(rectangle)
+
+    # 绘制带透明度的圆角矩形
+    draw.rounded_rectangle(
+        [(0, 0), (int(w), int(h))],
+        radius,
+        fill=bg_color  # 使用带透明度的背景色
+    )
+
+    # 使用alpha通道混合方式粘贴到背景上
+    image.paste(rectangle, (int(x), int(y)), rectangle)
+
+
+def round_corner_image(image: Image.Image, radius: int) -> Image.Image:
+    """将图片转为圆角"""
+    # 创建一个带有圆角的蒙版
+    circle = Image.new('L', (radius * 2, radius * 2), 0)
+    draw = ImageDraw.Draw(circle)
+    draw.ellipse((0, 0, radius * 2, radius * 2), fill=255)
+
+    # 创建一个完整的蒙版
+    mask = Image.new('L', image.size, 255)
+
+    # 添加四个圆角
+    mask.paste(circle.crop((0, 0, radius, radius)), (0, 0))  # 左上
+    mask.paste(circle.crop((radius, 0, radius * 2, radius)), (image.width - radius, 0))  # 右上
+    mask.paste(circle.crop((0, radius, radius, radius * 2)), (0, image.height - radius))  # 左下
+    mask.paste(circle.crop((radius, radius, radius * 2, radius * 2)),
+               (image.width - radius, image.height - radius))  # 右下
+
+    # 创建一个空白的透明图像
+    output = Image.new('RGBA', image.size, (0, 0, 0, 0))
+
+    # 将原图和蒙版合并
+    output.paste(image, (0, 0))
+    output.putalpha(mask)
+
+    return output
+
+
+def add_title_image(background: Image.Image, title_image_path: str, rect_x: int, rect_y: int, rect_width: int) -> int:
+    """添加标题图片"""
+    try:
+        with Image.open(title_image_path) as title_img:
+            # 如果图片不是RGBA模式，转换为RGBA
+            if title_img.mode != 'RGBA':
+                title_img = title_img.convert('RGBA')
+
+            # 设置图片宽度等于文字区域宽度
+            target_width = rect_width - 40  # 左右各留20像素边距
+
+            # 计算等比例缩放后的高度
+            aspect_ratio = title_img.height / title_img.width
+            target_height = int(target_width * aspect_ratio)
+
+            # 调整图片大小
+            resized_img = title_img.resize((int(target_width), target_height), Image.Resampling.LANCZOS)
+
+            # 添加圆角
+            rounded_img = round_corner_image(resized_img, radius=20)  # 可以调整圆角半径
+
+            # 计算居中位置（水平方向）
+            x = rect_x + 20  # 左边距20像素
+            y = rect_y + 20  # 顶部边距20像素
+
+            # 粘贴图片（使用图片自身的alpha通道）
+            background.paste(rounded_img, (x, y), rounded_img)
+
+            return y + target_height + 20  # 返回图片底部位置加上20像素间距
+    except Exception as e:
+        print(f"Error loading title image: {e}")
+        return rect_y + 30
+
+
+class MarkdownParser:
+    """Markdown解析器"""
+
+    def __init__(self):
+        self.reset()
+
+    def reset(self):
+        self.segments = []
+        self.current_section = None  # 当前处理的段落类型
+
+    def parse(self, text: str) -> List[TextSegment]:
+        """解析整个文本"""
+        self.reset()
+        segments = []
+        lines = text.splitlines()
+
+        for i, line in enumerate(lines):
+            line = line.strip()
+            if not line:
+                # 只有当下一行有内容时才添加空行
+                next_has_content = False
+                for next_line in lines[i + 1:]:
+                    if next_line.strip():
+                        next_has_content = True
+                        break
+                if next_has_content:
+                    style = TextStyle(
+                        line_spacing=20 if segments and segments[-1].style.is_title else 15
+                    )
+                    segments.append(TextSegment(text='', style=style))
+                continue
+
+            # 处理常规行
+            line_segments = self.parse_line(line)
+            segments.extend(line_segments)
+
+            # 只在确定有下一行内容时添加空行
+            if i < len(lines) - 1:
+                has_next_content = False
+                for next_line in lines[i + 1:]:
+                    if next_line.strip():
+                        has_next_content = True
+                        break
+                if has_next_content:
+                    style = line_segments[-1].style
+                    segments.append(TextSegment(text='', style=TextStyle(line_spacing=style.line_spacing)))
+
+        # 最后添加签名，不添加任何额外空行
+        if segments:
+            signature = TextSegment(
+                text="                                         —By 99i",
+                style=TextStyle(font_name='regular', indent=0, line_spacing=0)  # 设置 line_spacing=0
+            )
+            segments.append(signature)
+
+        return segments
+
+    def is_category_title(self, text: str) -> bool:
+        """判断是否为分类标题"""
+        return text.strip() in ['国内要闻', '国际动态']
+
+    def process_title_marks(self, text: str) -> str:
+        """处理标题标记"""
+        # 移除 ** 标记
+        text = re.sub(r'\*\*(.+?)\*\*', r'\1', text)
+        # 统一中文冒号
+        text = text.replace(':', '：')
+        return text
+
+    def split_number_and_content(self, text: str) -> Tuple[str, str]:
+        """分离序号和内容"""
+        match = re.match(r'(\d+)\.\s*(.+)', text)
+        if match:
+            return match.group(1), match.group(2)
+        return '', text
+
+    def split_title_and_content(self, text: str) -> Tuple[str, str]:
+        """分离标题和内容"""
+        parts = text.split('：', 1)
+        if len(parts) == 2:
+            return parts[0] + '：', parts[1].strip()
+        return text, ''
+
+    def parse_line(self, text: str) -> List[TextSegment]:
+        """解析单行文本"""
+        if not text.strip():
+            return [TextSegment(text='', style=TextStyle())]
+
+        # 处理一级标题
+        if text.startswith('# '):
+            style = TextStyle(
+                font_name='bold',
+                font_size=40,
+                is_title=True,
+                indent=0
+            )
+            return [TextSegment(text=text[2:].strip(), style=style)]
+
+        # 处理二级标题
+        if text.startswith('## '):
+            style = TextStyle(
+                font_name='bold',
+                font_size=35,
+                is_title=True,
+                line_spacing=25,
+                indent=0
+            )
+            self.current_section = text[3:].strip()
+            return [TextSegment(text=self.current_section, style=style)]
+
+        # 处理分类标题
+        if self.is_category_title(text):
+            style = TextStyle(
+                font_name='bold',
+                font_size=35,
+                is_category=True,
+                line_spacing=25,
+                indent=0
+            )
+            return [TextSegment(text=text.strip(), style=style)]
+
+        # 处理emoji标题格式
+        if text.strip() and emoji.is_emoji(text[0]):
+            # 移除文本中的加粗标记 **
+            content = text.strip()
+            if '**' in content:
+                content = content.replace('**', '')
+
+            style = TextStyle(
+                font_name='bold',
+                font_size=40,  # 使用H1的���体大小
+                is_title=True,
+                line_spacing=25,
+                indent=0
+            )
+            return [TextSegment(text=content, style=style)]
+
+        # 处理带序号的新闻条目
+        number, content = self.split_number_and_content(text)
+        if number:
+            content = self.process_title_marks(content)
+            title, body = self.split_title_and_content(content)
+            segments = []
+
+            title_style = TextStyle(
+                font_name='bold',
+                indent=0,
+                is_title=True,
+                line_spacing=15 if body else 20
+            )
+            segments.append(TextSegment(
+                text=f"{number}. {title}",
+                style=title_style
+            ))
+
+            if body:
+                content_style = TextStyle(
+                    font_name='regular',
+                    indent=40,
+                    line_spacing=20
+                )
+                segments.append(TextSegment(
+                    text=body,
+                    style=content_style
+                ))
+            return segments
+
+        # 处理破折号开头的内容
+        if text.strip().startswith('-'):
+            style = TextStyle(
+                font_name='regular',
+                indent=40,
+                line_spacing=15
+            )
+            return [TextSegment(text=text.strip(), style=style)]
+
+        # 处理普通文本
+        style = TextStyle(
+            font_name='regular',
+            indent=40 if self.current_section else 0,
+            line_spacing=15
+        )
+
+        return [TextSegment(text=text.strip(), style=style)]
+
+
+def parse_line(self, text: str) -> List[TextSegment]:
+    """解析单行文本"""
+    if not text.strip():
+        return [TextSegment(text='', style=TextStyle())]
+
+    # 处理一级标题
+    if text.startswith('# '):
+        style = TextStyle(
+            font_name='bold',
+            font_size=40,
+            is_title=True,
+            indent=0
+        )
+        return [TextSegment(text=text[2:].strip(), style=style)]
+
+    # 处理二级标题
+    if text.startswith('## '):
+        style = TextStyle(
+            font_name='bold',
+            font_size=35,
+            is_title=True,
+            line_spacing=25,
+            indent=0
+        )
+        self.current_section = text[3:].strip()
+        return [TextSegment(text=self.current_section, style=style)]
+
+    # 处理分类标题
+    if self.is_category_title(text):
+        style = TextStyle(
+            font_name='bold',
+            font_size=35,
+            is_category=True,
+            line_spacing=25,
+            indent=0
+        )
+        return [TextSegment(text=text.strip(), style=style)]
+
+    # 处理emoji标题格式
+    # 匹配模式：emoji + 空格 + (可选的**) + 内容 + (可选的**)
+    if text.strip() and emoji.is_emoji(text[0]):
+        # 移除文本中的加粗标记 **
+        content = text.strip()
+        if '**' in content:
+            content = content.replace('**', '')
+
+        style = TextStyle(
+            font_name='bold',
+            font_size=40,  # 使用H1的字体大小
+            is_title=True,
+            line_spacing=25,
+            indent=0
+        )
+        return [TextSegment(text=content, style=style)]
+
+    # 处理带序号的新闻条目
+    number, content = self.split_number_and_content(text)
+    if number:
+        content = self.process_title_marks(content)
+        title, body = self.split_title_and_content(content)
+        segments = []
+
+        title_style = TextStyle(
+            font_name='bold',
+            indent=0,
+            is_title=True,
+            line_spacing=15 if body else 20
+        )
+        segments.append(TextSegment(
+            text=f"{number}. {title}",
+            style=title_style
+        ))
+
+        if body:
+            content_style = TextStyle(
+                font_name='regular',
+                indent=40,
+                line_spacing=20
+            )
+            segments.append(TextSegment(
+                text=body,
+                style=content_style
+            ))
+        return segments
+
+    # 处理破折号开头的内容
+    if text.strip().startswith('-'):
+        style = TextStyle(
+            font_name='regular',
+            indent=40,
+            line_spacing=15
+        )
+        return [TextSegment(text=text.strip(), style=style)]
+
+    # 处理普通文本
+    style = TextStyle(
+        font_name='regular',
+        indent=40 if self.current_section else 0,
+        line_spacing=15
+    )
+
+    return [TextSegment(text=text.strip(), style=style)]
+
+    def parse(self, text: str) -> List[TextSegment]:
+        """解析整个文本"""
+        self.reset()
+        segments = []
+        lines = text.splitlines()
+
+        for i, line in enumerate(lines):
+            line = line.strip()
+            if not line:
+                # 只有当下一行有内容时才添加空行
+                next_has_content = False
+                for next_line in lines[i + 1:]:
+                    if next_line.strip():
+                        next_has_content = True
+                        break
+                if next_has_content:
+                    style = TextStyle(
+                        line_spacing=20 if segments and segments[-1].style.is_title else 15
+                    )
+                    segments.append(TextSegment(text='', style=style))
+                continue
+
+            # 处理常规行
+            line_segments = self.parse_line(line)
+            segments.extend(line_segments)
+
+            # 只在确定有下一行内容时添加空行
+            if i < len(lines) - 1:
+                has_next_content = False
+                for next_line in lines[i + 1:]:
+                    if next_line.strip():
+                        has_next_content = True
+                        break
+                if has_next_content:
+                    style = line_segments[-1].style
+                    segments.append(TextSegment(text='', style=TextStyle(line_spacing=style.line_spacing)))
+
+        # 最后添加签名，不添加任何额外空行
+        if segments:
+            signature = TextSegment(
+                text="                                         —By 嫣然",
+                style=TextStyle(font_name='regular', indent=0, line_spacing=0)  # 设置 line_spacing=0
+            )
+            segments.append(signature)
+
+        return segments
+
+
+class TextRenderer:
+    """文本渲染器"""
+
+    def __init__(self, font_manager: FontManager, max_width: int):
+        self.font_manager = font_manager
+        self.max_width = max_width
+        self.temp_image = Image.new('RGBA', (2000, 100))
+        self.temp_draw = ImageDraw.Draw(self.temp_image)
+
+    def measure_text(self, text: str, font: ImageFont.FreeTypeFont,
+                     emoji_font: Optional[ImageFont.FreeTypeFont] = None) -> Tuple[int, int]:
+        """测量文本尺寸，考虑emoji"""
+        total_width = 0
+        max_height = 0
+
+        for char in text:
+            if emoji.is_emoji(char) and emoji_font:
+                bbox = self.temp_draw.textbbox((0, 0), char, font=emoji_font)
+                width = bbox[2] - bbox[0]
+                height = bbox[3] - bbox[1]
+            else:
+                bbox = self.temp_draw.textbbox((0, 0), char, font=font)
+                width = bbox[2] - bbox[0]
+                height = bbox[3] - bbox[1]
+
+            total_width += width
+            max_height = max(max_height, height)
+
+        return total_width, max_height
+
+    def draw_text_with_emoji(self, draw: ImageDraw.ImageDraw, pos: Tuple[int, int], text: str,
+                             font: ImageFont.FreeTypeFont, emoji_font: ImageFont.FreeTypeFont,
+                             fill: str = "white") -> int:
+        """绘制包含emoji的文本，返回绘制宽度"""
+        x, y = pos
+        total_width = 0
+
+        for char in text:
+            if emoji.is_emoji(char):
+                # 使用emoji字体
+                bbox = draw.textbbox((x, y), char, font=emoji_font)
+                draw.text((x, y), char, font=emoji_font, fill=fill)
+                char_width = bbox[2] - bbox[0]
+            else:
+                # 使用常规字体
+                bbox = draw.textbbox((x, y), char, font=font)
+                draw.text((x, y), char, font=font, fill=fill)
+                char_width = bbox[2] - bbox[0]
+
+            x += char_width
+            total_width += char_width
+
+        return total_width
+
+    def calculate_height(self, processed_lines: List[ProcessedLine]) -> int:
+        """计算总高度，确保不在最后添加额外间距"""
+        total_height = 0
+        prev_line = None
+
+        for i, line in enumerate(processed_lines):
+            if not line.text.strip():
+                # 只有当不是最后一行，且后面还有内容时才添加间距
+                if i < len(processed_lines) - 1 and any(l.text.strip() for l in processed_lines[i + 1:]):
+                    if prev_line:
+                        total_height += prev_line.style.line_spacing
+                continue
+
+            # 计算当前行高度
+            line_height = line.height * line.line_count
+
+            # 添加行间距，但不在最后一行后添加
+            if prev_line and i < len(processed_lines) - 1:
+                if prev_line.style.is_category:
+                    total_height += 30
+                elif prev_line.style.is_title and not line.style.is_title:
+                    total_height += 20
+                else:
+                    total_height += line.style.line_spacing
+
+            total_height += line_height
+            prev_line = line
+
+        return total_height
+
+    def split_text_to_lines(self, segment: TextSegment, available_width: int) -> List[ProcessedLine]:
+        """将文本分割成合适宽度的行，支持emoji"""
+        if not segment.text.strip():
+            return [ProcessedLine(text='', style=segment.style, height=0, line_count=1)]
+
+        font = self.font_manager.get_font(segment.style)
+        emoji_font = self.font_manager.fonts['emoji_30']
+        words = []
+        current_word = ''
+        processed_lines = []
+
+        # 分词处理
+        for char in segment.text:
+            if emoji.is_emoji(char):
+                if current_word:
+                    words.append(current_word)
+                    current_word = ''
+                words.append(char)  # emoji作为单独的词
+            elif char in [' ', '，', '。', '：', '、', '！', '？', '；']:
+                if current_word:
+                    words.append(current_word)
+                words.append(char)
+                current_word = ''
+            else:
+                if ord(char) > 0x4e00:  # 中文字符
+                    if current_word:
+                        words.append(current_word)
+                        current_word = ''
+                    words.append(char)
+                else:
+                    current_word += char
+
+        if current_word:
+            words.append(current_word)
+
+        current_line = ''
+        line_height = 0
+
+        for word in words:
+            test_line = current_line + word
+            width, height = self.measure_text(test_line, font, emoji_font)
+            line_height = max(line_height, height)
+
+            if width <= available_width:
+                current_line = test_line
+            else:
+                if current_line:
+                    processed_lines.append(ProcessedLine(
+                        text=current_line,
+                        style=segment.style,
+                        height=line_height,
+                        line_count=1
+                    ))
+                current_line = word
+
+        if current_line:
+            processed_lines.append(ProcessedLine(
+                text=current_line,
+                style=segment.style,
+                height=line_height,
+                line_count=1
+            ))
+
+        return processed_lines
+
+
+def compress_image(image_path: str, output_path: str, max_size: int = 3145728):  # 3MB in bytes
+    """
+    Compress an image to ensure it's under a certain file size.
+
+    :param image_path: The path to the image to be compressed.
+    :param output_path: The path where the compressed image will be saved.
+    :param max_size: The maximum file size in bytes (default is 3MB).
+    """
+    # Open the image
+    with Image.open(image_path) as img:
+        # Convert to RGB if it's not already
+        if img.mode != 'RGB':
+            img = img.convert('RGB')
+
+        # Define the quality to start with
+        quality = 95  # Start with a high quality
+
+        # Save the image with different qualities until the file size is acceptable
+        while True:
+            # Save the image with the current quality
+            img.save(output_path, "PNG", optimize=True, compress_level=0)
+
+            # Check the file size
+            if os.path.getsize(output_path) <= max_size:
+                break  # The file size is acceptable, break the loop
+
+            # If the file is still too large, decrease the quality
+            quality -= 5
+            if quality < 10:  # To prevent an infinite loop, set a minimum quality
+                break
+
+        # If the quality is too low, you might want to handle it here
+        if quality < 10:
+            print("The image could not be compressed enough to meet the size requirements.")
+
+
+def generate_image(text: str, output_path: str, title_image: Optional[str] = None):
+    """生成图片主函数 - 修复彩色emoji渲染"""
+    try:
+        width = 720
+        current_dir = os.path.dirname(os.path.abspath(__file__))
+        font_paths = {
+            'regular': os.path.join(current_dir, "msyh.ttc"),
+            'bold': os.path.join(current_dir, "msyhbd.ttc"),
+            'emoji': os.path.join(current_dir, "TwitterColorEmoji.ttf")  # 或其他彩色emoji字体
+        }
+
+        # 验证字体文件
+        for font_type, path in font_paths.items():
+            if not os.path.exists(path):
+                raise FileNotFoundError(f"Font file not found: {path}")
+
+        # 初始化组件
+        font_manager = FontManager(font_paths)
+        rect_width = width - 80
+        max_content_width = rect_width - 80
+        parser = MarkdownParser()
+        renderer = TextRenderer(font_manager, max_content_width)
+
+        # 解析文本
+        segments = parser.parse(text)
+        processed_lines = []
+
+        for segment in segments:
+            available_width = max_content_width - segment.style.indent
+            if segment.text.strip():
+                lines = renderer.split_text_to_lines(segment, available_width)
+                processed_lines.extend(lines)
+            else:
+                processed_lines.append(ProcessedLine(
+                    text='',
+                    style=segment.style,
+                    height=0,
+                    line_count=1
+                ))
+
+        # 计算高度
+        title_height = 0
+        if title_image:
+            try:
+                with Image.open(title_image) as img:
+                    aspect_ratio = img.height / img.width
+                    title_height = int((rect_width - 40) * aspect_ratio) + 40
+            except Exception as e:
+                print(f"Title image processing error: {e}")
+
+        content_height = renderer.calculate_height(processed_lines)
+        rect_height = content_height + title_height
+        rect_x = (width - rect_width) // 2
+        rect_y = 40
+        total_height = rect_height + 80
+
+        # 创建RGBA背景
+        background = create_gradient_background(width, total_height)
+        draw = ImageDraw.Draw(background)
+
+        # 获取主题颜色
+        background_color, text_color, is_dark_theme = get_theme_colors()
+        if len(background_color) == 3:
+            background_color = background_color + (128,)  # 添加alpha通道
+
+        # 创建卡片背景
+        create_rounded_rectangle(
+            background, rect_x, rect_y, rect_width, rect_height,
+            radius=30, bg_color=background_color
+        )
+
+        # 绘制内容
+        current_y = rect_y + 30
+        if title_image:
+            current_y = add_title_image(background, title_image, rect_x, rect_y, rect_width)
+
+        # 逐字符绘制文本
+        for i, line in enumerate(processed_lines):
+            if not line.text.strip():
+                if i < len(processed_lines) - 1 and any(l.text.strip() for l in processed_lines[i + 1:]):
+                    current_y += line.style.line_spacing
+                continue
+
+            x = rect_x + 40 + line.style.indent
+            current_x = x
+
+            # 逐字符渲染
+            for char in line.text:
+                if emoji.is_emoji(char):
+                    # emoji字体渲染
+                    emoji_font = font_manager.fonts['emoji_30']
+                    bbox = draw.textbbox((current_x, current_y), char, font=emoji_font)
+                    # 使用RGBA模式绘制emoji
+                    draw.text((current_x, current_y), char, font=emoji_font, embedded_color=True)
+                    current_x += bbox[2] - bbox[0]
+                else:
+                    # 普通文字渲染
+                    font = font_manager.get_font(line.style)
+                    bbox = draw.textbbox((current_x, current_y), char, font=font)
+                    draw.text((current_x, current_y), char, font=font, fill=text_color)
+                    current_x += bbox[2] - bbox[0]
+
+            if i < len(processed_lines) - 1:
+                current_y += line.height + line.style.line_spacing
+            else:
+                current_y += line.height
+
+        # 直接保存为PNG，保持RGBA模式
+        background = background.convert('RGB')
+        background.save(output_path, "PNG", optimize=False, compress_level=0)
+
+    except Exception as e:
+        print(f"Error generating image: {e}")
+        raise

msyh.ttc

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3084f1f88369af6bf9989c909024164d953d1e38d08734f05f28ef24b2f9d577
+size 19701556

msyhbd.ttc

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:519309b7ab0479c4dc3ace5e291de5a8702175be5586e165bc810267bd4619a5
+size 16880832