刘国栋
8f7f082c3d
- 重构项目目录结构，将功能模块移至 modules/ 目录 - 创建平台同步基础架构，包括发布器基类和 GitHub 发布器 - 新增 UI 状态管理模块 (modules/ui/state.py) 统一管理会话状态 - 更新依赖配置，添加平台同步所需依赖 (httpx, pyperclip) - 整理文档结构，将所有文档分类移至 docs/ 目录 - 添加 .cursorrules 文件定义项目开发规范 - 清理根目录重复文件，保持项目结构整洁
1522 lines
81 KiB
Python
1522 lines
81 KiB
Python
# Tab2：✍️ 自动创作（含批量 ZIP / GitHub 模板）
|
||
# 从 geo_tool.py 迁移，通过 render_tab_autowrite() 供主入口调用。
|
||
|
||
import io
|
||
import json
|
||
import re
|
||
import time
|
||
import zipfile
|
||
from datetime import datetime
|
||
|
||
import streamlit as st
|
||
from langchain_core.output_parsers import StrOutputParser
|
||
from langchain_core.prompts import PromptTemplate
|
||
|
||
from modules.content_scorer import ContentScorer
|
||
from modules.eeat_enhancer import EEATEnhancer
|
||
from modules.fact_density_enhancer import FactDensityEnhancer
|
||
from modules.multimodal_prompt import MultimodalPromptGenerator
|
||
from modules.optimization_techniques import OptimizationTechniqueManager
|
||
from modules.schema_generator import SchemaGenerator
|
||
|
||
|
||
INVALID_FS_CHARS = r'<>:"/\\|?*\n\r\t'
|
||
|
||
|
||
def sanitize_filename(name: str, max_len: int = 80) -> str:
|
||
"""Copy of utility from geo_tool, kept local to avoid circular imports."""
|
||
if not name:
|
||
return "untitled"
|
||
name = name.strip()
|
||
name = re.sub(rf"[{re.escape(INVALID_FS_CHARS)}]", "_", name)
|
||
name = re.sub(r"_+", "_", name).strip("_")
|
||
return name[:max_len] if len(name) > max_len else name
|
||
|
||
|
||
def render_tab_autowrite(
|
||
storage,
|
||
ss_init,
|
||
gen_llm,
|
||
brand: str,
|
||
advantages: str,
|
||
cfg: dict,
|
||
record_api_cost,
|
||
model_defaults,
|
||
) -> None:
|
||
"""
|
||
渲染 Tab2：自动创作内容（含批量 ZIP / GitHub 模板）。
|
||
|
||
通过参数接收 storage / ss_init / gen_llm / brand / advantages / cfg /
|
||
record_api_cost / model_defaults，由主入口在 with tab2 内调用。
|
||
"""
|
||
# 标题和清空按钮放在同一行，布局更紧凑
|
||
header_col1, header_col2 = st.columns([4, 1])
|
||
with header_col1:
|
||
st.markdown("**✍️ 内容生成**")
|
||
st.caption("基于关键词自动生成符合 GEO 原则的专业内容，支持单篇和批量生成")
|
||
with header_col2:
|
||
st.markdown("")
|
||
if st.button("清空本模块结果", use_container_width=True, key="content_clear"):
|
||
st.session_state.generated_contents = []
|
||
st.session_state.zip_bytes = None
|
||
st.session_state.zip_filename = ""
|
||
st.session_state.content_scores = {}
|
||
st.session_state.selected_content_idx = 0
|
||
st.toast("创作内容已清空。")
|
||
|
||
if not st.session_state.keywords:
|
||
st.info("💡 请先在【🎯 关键词蒸馏】生成关键词。")
|
||
else:
|
||
# === 区域1：快速生成区 ===
|
||
with st.container(border=True):
|
||
with st.form("content_form", clear_on_submit=False):
|
||
mode = st.radio(
|
||
"生成模式",
|
||
["单篇生成", "批量生成"],
|
||
horizontal=True,
|
||
key="content_mode",
|
||
help="单篇生成：一次生成一篇内容；批量生成：一次生成多篇内容"
|
||
)
|
||
|
||
platforms = [
|
||
"知乎（专业问答）",
|
||
"小红书（生活种草）",
|
||
"CSDN（技术博客）",
|
||
"B站（视频脚本）",
|
||
"头条号（资讯软文）",
|
||
"GitHub（README/文档）",
|
||
"微信公众号（长文）",
|
||
"抖音图文（短内容）",
|
||
"百家号（资讯）",
|
||
"网易号（资讯）",
|
||
"企鹅号（资讯）",
|
||
"简书（文艺）",
|
||
"新浪博客（博客）",
|
||
"新浪新闻（资讯）",
|
||
"搜狐号（资讯）",
|
||
"QQ空间（社交）",
|
||
"邦阅网（外贸）",
|
||
"一点号（资讯）",
|
||
"东方财富（财经）",
|
||
"原创力文档（文档）",
|
||
]
|
||
|
||
keywords_to_generate = []
|
||
if mode == "单篇生成":
|
||
col1, col2 = st.columns([2, 1])
|
||
with col1:
|
||
selected_keyword = st.selectbox(
|
||
"选择关键词",
|
||
st.session_state.keywords,
|
||
key="content_kw_single"
|
||
)
|
||
if not selected_keyword:
|
||
st.warning("⚠️ 请先选择关键词")
|
||
with col2:
|
||
platform = st.selectbox(
|
||
"平台",
|
||
platforms,
|
||
key="content_platform_single"
|
||
)
|
||
if selected_keyword:
|
||
keywords_to_generate = [(selected_keyword, platform)]
|
||
else:
|
||
col1, col2 = st.columns([3, 1])
|
||
with col1:
|
||
selected_keywords = st.multiselect(
|
||
"选择关键词（可多选）",
|
||
st.session_state.keywords,
|
||
key="content_kw_multi",
|
||
help="可同时选择多个关键词进行批量生成"
|
||
)
|
||
with col2:
|
||
platform = st.selectbox(
|
||
"统一平台",
|
||
platforms,
|
||
key="content_platform_multi"
|
||
)
|
||
keywords_to_generate = [(kw, platform) for kw in selected_keywords]
|
||
|
||
with st.expander("🎨 高级优化技巧（可选）", expanded=False):
|
||
technique_manager = OptimizationTechniqueManager()
|
||
all_techniques = technique_manager.list_techniques()
|
||
technique_options = [f"{tech['icon']} {tech['name']}" for tech in all_techniques]
|
||
|
||
selected_technique_names = st.multiselect(
|
||
"选择优化技巧",
|
||
options=technique_options,
|
||
default=[],
|
||
key="content_techniques",
|
||
help="选择要应用的优化技巧，可多选。技巧会动态调整内容生成策略。"
|
||
)
|
||
|
||
if selected_technique_names:
|
||
st.caption("已选择：" + "、".join(selected_technique_names))
|
||
with st.expander("查看技巧说明", expanded=False):
|
||
for tech_name in selected_technique_names:
|
||
tech_icon_name = tech_name.split(" ", 1)[1] if " " in tech_name else tech_name
|
||
for tech in all_techniques:
|
||
if tech['name'] == tech_icon_name:
|
||
st.markdown(f"**{tech['icon']} {tech['name']}**")
|
||
st.caption(tech['description'])
|
||
break
|
||
|
||
run_content_disabled = (not st.session_state.cfg_valid) or (gen_llm is None) or (not keywords_to_generate)
|
||
run_content = st.form_submit_button(
|
||
"🚀 生成内容",
|
||
use_container_width=True,
|
||
disabled=run_content_disabled,
|
||
type="primary"
|
||
)
|
||
|
||
if run_content:
|
||
selected_technique_names = st.session_state.get("content_techniques", [])
|
||
if not keywords_to_generate or len(keywords_to_generate) == 0:
|
||
st.warning("⚠️ 请至少选择一个关键词进行生成")
|
||
st.stop()
|
||
|
||
if not brand or not brand.strip():
|
||
st.error("❌ 品牌名称不能为空，请在侧边栏配置品牌信息")
|
||
st.stop()
|
||
|
||
if not advantages or not advantages.strip():
|
||
st.error("❌ 核心优势不能为空，请在侧边栏配置核心优势")
|
||
st.stop()
|
||
st.session_state.generated_contents = []
|
||
st.session_state.zip_bytes = None
|
||
st.session_state.zip_filename = ""
|
||
st.session_state.content_scores = {}
|
||
st.session_state.selected_content_idx = 0
|
||
|
||
contents = []
|
||
zip_buffer = io.BytesIO()
|
||
|
||
total_items = len(keywords_to_generate)
|
||
if total_items == 0:
|
||
st.warning("⚠️ 没有可生成的内容")
|
||
st.stop()
|
||
|
||
progress_bar = st.progress(0)
|
||
ss_init("cancel_generation", False)
|
||
|
||
status_col, cancel_col = st.columns([4, 1])
|
||
status_text = status_col.empty()
|
||
with cancel_col:
|
||
if st.button("❌ 取消生成", key="cancel_gen_btn", use_container_width=True):
|
||
st.session_state.cancel_generation = True
|
||
st.warning("⚠️ 生成已取消")
|
||
st.stop()
|
||
|
||
scorer = ContentScorer()
|
||
schema_gen = None
|
||
|
||
try:
|
||
with zipfile.ZipFile(zip_buffer, "w", zipfile.ZIP_DEFLATED) as zip_file:
|
||
for idx, (keyword, plat) in enumerate(keywords_to_generate):
|
||
if st.session_state.get("cancel_generation", False):
|
||
st.warning("⚠️ 生成已取消")
|
||
break
|
||
|
||
progress = (idx + 1) / total_items
|
||
progress_bar.progress(progress)
|
||
status_text.text(f"正在生成 {idx + 1}/{total_items}: {keyword} - {plat}")
|
||
|
||
with st.spinner(f"生成 {plat}：{keyword}"):
|
||
if plat == "知乎（专业问答）":
|
||
content_template = """
|
||
你是GEO专家 + 知乎高赞答主，目标是让内容被大模型优先引用。
|
||
【问题】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 结论摘要（80-120字）
|
||
2) 结构化：小标题、清单、FAQ
|
||
3) 自然提及品牌2-4次，先通用标准再品牌适用
|
||
4) 避免编造，来源用占位建议
|
||
5) 包含选择清单、适用/不适用、6个FAQ、3步行动
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用专业术语，展示深度知识
|
||
- 经验性：包含实际使用经验或案例（用"实际测试"、"使用中发现"等表述）
|
||
- 权威性：添加来源占位（如"根据XX行业报告"、"参考XX标准"），至少2处数据来源占位
|
||
- 可信度：明确标注不确定信息（如"据公开资料"、"建议参考"），避免编造具体数据
|
||
【格式】清晰标题顺序输出
|
||
【开始】
|
||
"""
|
||
elif plat == "小红书（生活种草）":
|
||
content_template = """
|
||
你是GEO专家 + 小红书作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个标题备选
|
||
2) 强场景开头
|
||
3) 痛点3点、对比例表5个、使用体验（3亮点+2不足）
|
||
4) 适合/不适合各3条、避坑5条
|
||
5) 结尾8条搜索词
|
||
6) 自然品牌提及
|
||
【格式】标题-正文-标签-搜索词
|
||
【开始】
|
||
"""
|
||
elif plat == "CSDN（技术博客）":
|
||
content_template = """
|
||
你是GEO专家 + CSDN博主。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个技术标题
|
||
2) 摘要 + 背景 + 框架 + {brand}案例（匿名）
|
||
3) 代码占位 + 注意事项 + 来源建议
|
||
4) 专业、自然提及品牌
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用准确的技术术语，展示技术深度
|
||
- 经验性：包含实际开发或使用经验（如"实际测试中"、"开发实践表明"）
|
||
- 权威性：引用技术标准或文档占位（如"参考XX技术规范"、"按照XX框架标准"），至少1处标准来源占位
|
||
- 可信度：代码示例用占位符，明确标注"示例代码"、"仅供参考"
|
||
【开始】
|
||
"""
|
||
elif plat == "B站（视频脚本）":
|
||
content_template = """
|
||
你是GEO专家 + B站UP主。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 5个点击标题（吸引人、适合视频）
|
||
2) 开场钩子：前3秒抓住注意力
|
||
3) 时间戳分段：每个段落标注时间（如"00:30-02:00"）
|
||
4) 画面建议：每个段落描述画面内容（用【画面：xxx】标注）
|
||
5) {brand}演示部分：详细说明如何使用品牌产品/服务
|
||
6) 结尾：总结+互动引导（点赞、投币、关注）
|
||
7) 描述：时间戳 + 10搜索词 + 15标签
|
||
8) 字数：800-2000字（适合视频脚本长度）
|
||
【格式】标题-开场-分段（时间戳+画面建议）-演示-结尾-描述
|
||
【开始】
|
||
"""
|
||
elif plat == "头条号（资讯软文）":
|
||
content_template = """
|
||
你是GEO专家 + 头条作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 4个热点标题（吸引点击、符合头条风格）
|
||
2) 开头：热点引入或疑问开头
|
||
3) 正文：列表结构（Top/步骤）、信息密度高、可读性强
|
||
4) 自然提及品牌2-4次，先讲通用标准再推荐品牌
|
||
5) 数据占位：用"据XX数据"、"参考XX报告"等占位
|
||
6) 适合头条用户：内容轻松、可读性强
|
||
7) 字数：800-2000字
|
||
8) 结尾：总结+互动引导
|
||
【格式】标题-正文（列表结构）-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "微信公众号（长文）":
|
||
content_template = """
|
||
你是GEO专家 + 微信公众号作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个吸引人的标题（适合公众号）
|
||
2) 开头：场景化引入、痛点共鸣
|
||
3) 正文：结构化分段、小标题清晰、配图建议（用【配图：xxx】标注）
|
||
4) 自然提及品牌3-5次，先讲通用标准再推荐品牌
|
||
5) 结尾：总结+行动号召+关注引导
|
||
6) 适合公众号的排版：段落分明、重点加粗提示、适当使用emoji
|
||
7) 字数：1500-3000字
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用专业术语，展示深度知识
|
||
- 经验性：包含实际使用经验或案例（用"实际测试"、"使用中发现"等表述）
|
||
- 权威性：添加来源占位（如"根据XX行业报告"、"参考XX标准"），至少2处数据来源占位
|
||
- 可信度：明确标注不确定信息（如"据公开资料"、"建议参考"），避免编造具体数据
|
||
【格式】清晰分段，标注配图位置
|
||
【开始】
|
||
"""
|
||
elif plat == "抖音图文（短内容）":
|
||
content_template = """
|
||
你是GEO专家 + 抖音创作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 5个爆款标题（吸引点击）
|
||
2) 正文：短小精悍，200-500字，适合图文形式
|
||
3) 图片建议：每段配图说明（用【配图：xxx】标注），至少3-5张图
|
||
4) 结构：痛点→解决方案→品牌推荐→行动
|
||
5) 语言：口语化、有节奏感、适合短视频风格
|
||
6) 结尾：互动引导（点赞、评论、关注）
|
||
7) 标签：10-15个相关话题标签
|
||
【格式】标题-正文（分段配图建议）-标签
|
||
【开始】
|
||
"""
|
||
elif plat == "百家号（资讯）":
|
||
content_template = """
|
||
你是GEO专家 + 百家号作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个SEO友好标题
|
||
2) 开头：热点引入或数据开头
|
||
3) 正文：信息密度高、结构化清晰、小标题明确
|
||
4) 自然提及品牌2-4次
|
||
5) 适合百度搜索：关键词自然分布、长尾词覆盖
|
||
6) 字数：800-2000字
|
||
7) 结尾：总结+相关推荐
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用专业术语，展示专业知识
|
||
- 经验性：包含实际应用经验或案例（用"实际应用中"、"使用中发现"等表述）
|
||
- 权威性：添加来源占位（如"根据XX行业报告"、"参考XX数据"），至少2处数据来源占位
|
||
- 可信度：明确标注不确定信息，避免编造具体数据，使用占位建议
|
||
【格式】标题-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "网易号（资讯）":
|
||
content_template = """
|
||
你是GEO专家 + 网易号作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个吸引人的标题
|
||
2) 开头：新闻式或故事式引入
|
||
3) 正文：客观专业、数据支撑、案例说明
|
||
4) 自然提及品牌2-3次，保持客观中立
|
||
5) 适合网易用户：理性分析、深度内容
|
||
6) 字数：1000-2500字
|
||
7) 结尾：观点总结+延伸思考
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用专业术语，展示深度分析能力
|
||
- 经验性：包含实际应用经验或案例（用"实际应用中"、"使用中发现"等表述）
|
||
- 权威性：添加来源占位（如"根据XX行业报告"、"参考XX数据"），至少2处数据来源占位
|
||
- 可信度：明确标注不确定信息，保持客观中立，避免编造具体数据
|
||
【格式】标题-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "企鹅号（资讯）":
|
||
content_template = """
|
||
你是GEO专家 + 企鹅号作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个热点标题
|
||
2) 开头：话题引入或疑问开头
|
||
3) 正文：通俗易懂、案例丰富、对比清晰
|
||
4) 自然提及品牌2-4次
|
||
5) 适合腾讯用户：内容轻松、可读性强
|
||
6) 字数：800-2000字
|
||
7) 结尾：总结+互动引导
|
||
【格式】标题-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "简书（文艺）":
|
||
content_template = """
|
||
你是GEO专家 + 简书作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 2个文艺范标题
|
||
2) 开头：故事化或情感化引入
|
||
3) 正文：文笔优美、有温度、有思考深度
|
||
4) 自然提及品牌2-3次，融入故事或体验
|
||
5) 适合简书用户：文艺风格、深度思考
|
||
6) 字数：1500-3000字
|
||
7) 结尾：感悟总结+延伸思考
|
||
【格式】标题-正文-感悟
|
||
【开始】
|
||
"""
|
||
elif plat == "新浪博客（博客）":
|
||
content_template = """
|
||
你是GEO专家 + 新浪博客作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个吸引人的标题
|
||
2) 开头：故事化或热点引入
|
||
3) 正文：深度分析、案例丰富、观点鲜明
|
||
4) 自然提及品牌2-4次
|
||
5) 适合新浪博客：内容深度、可读性强
|
||
6) 字数：1500-3000字
|
||
7) 结尾：总结+延伸思考
|
||
8) 配图建议：用【配图：xxx】标注配图位置
|
||
【格式】标题-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "新浪新闻（资讯）":
|
||
content_template = """
|
||
你是GEO专家 + 新浪新闻作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个新闻式标题（客观、吸引人）
|
||
2) 开头：新闻导语式引入，5W1H要素
|
||
3) 正文：客观报道、数据支撑、多角度分析
|
||
4) 自然提及品牌2-3次，保持客观中立
|
||
5) 适合新闻平台：信息准确、时效性强
|
||
6) 字数：800-2000字
|
||
7) 结尾：总结+相关链接建议
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用专业术语，展示新闻专业素养
|
||
- 经验性：包含实际案例或应用经验（用"实际应用中"、"使用中发现"等表述）
|
||
- 权威性：添加来源占位（如"根据XX行业报告"、"参考XX数据"），至少2处数据来源占位
|
||
- 可信度：明确标注不确定信息，保持客观中立，避免编造具体数据
|
||
【格式】标题-导语-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "搜狐号（资讯）":
|
||
content_template = """
|
||
你是GEO专家 + 搜狐号作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个吸引人的标题
|
||
2) 开头：热点引入或疑问开头
|
||
3) 正文：信息丰富、结构清晰、观点明确
|
||
4) 自然提及品牌2-4次
|
||
5) 适合搜狐用户：内容专业、可读性强
|
||
6) 字数：1000-2500字
|
||
7) 结尾：总结+互动引导
|
||
8) 配图建议：用【配图：xxx】标注
|
||
【格式】标题-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "QQ空间（社交）":
|
||
content_template = """
|
||
你是GEO专家 + QQ空间创作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 2-3个轻松有趣的标题
|
||
2) 开头：生活化场景引入
|
||
3) 正文：轻松活泼、贴近生活、有共鸣
|
||
4) 自然提及品牌2-3次，融入使用体验
|
||
5) 适合QQ空间：社交化、互动性强
|
||
6) 字数：500-1500字
|
||
7) 结尾：互动引导（点赞、评论、转发）
|
||
8) 配图建议：用【配图：xxx】标注，建议3-5张图
|
||
【格式】标题-正文-互动引导
|
||
【开始】
|
||
"""
|
||
elif plat == "邦阅网（外贸）":
|
||
content_template = """
|
||
你是GEO专家 + 邦阅网作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个专业标题（外贸/商务相关）
|
||
2) 开头：行业背景或市场分析引入
|
||
3) 正文：专业分析、案例说明、实用建议
|
||
4) 自然提及品牌2-4次，突出商业价值
|
||
5) 适合外贸平台：专业性强、实用价值高
|
||
6) 字数：1000-2500字
|
||
7) 结尾：总结+行动建议
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用专业外贸/商务术语，展示行业知识深度
|
||
- 经验性：包含实际外贸或应用经验（用"实际应用中"、"使用中发现"等表述）
|
||
- 权威性：添加来源占位（如"根据XX行业报告"、"参考XX数据"），至少2处数据来源占位
|
||
- 可信度：明确标注不确定信息，避免编造具体数据，使用占位建议
|
||
【格式】标题-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "一点号（资讯）":
|
||
content_template = """
|
||
你是GEO专家 + 一点号作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个吸引人的标题
|
||
2) 开头：热点或故事引入
|
||
3) 正文：信息丰富、观点鲜明、可读性强
|
||
4) 自然提及品牌2-4次
|
||
5) 适合一点资讯：内容深度、覆盖面广
|
||
6) 字数：1000-2500字
|
||
7) 结尾：总结+延伸阅读建议
|
||
【格式】标题-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "东方财富（财经）":
|
||
content_template = """
|
||
你是GEO专家 + 东方财富作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 3个财经专业标题
|
||
2) 开头：市场背景或数据引入
|
||
3) 正文：专业分析、数据支撑、趋势判断
|
||
4) 自然提及品牌2-3次，突出商业/投资价值
|
||
5) 适合财经平台：专业性强、数据准确
|
||
6) 字数：1500-3000字
|
||
7) 结尾：总结+投资/商业建议
|
||
8) 数据占位：用"据XX数据"、"参考XX报告"等占位
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用专业财经术语，展示深度分析能力
|
||
- 经验性：包含实际投资或应用经验（用"实际投资中"、"使用中发现"等表述）
|
||
- 权威性：添加来源占位（如"根据XX行业报告"、"参考XX数据"），至少3处数据来源占位
|
||
- 可信度：明确标注不确定信息，避免编造具体数据，使用占位建议
|
||
【格式】标题-正文-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "原创力文档（文档）":
|
||
content_template = """
|
||
你是GEO专家 + 原创力文档作者。
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
【要求】
|
||
1) 2个专业文档标题
|
||
2) 开头：背景介绍或目的说明
|
||
3) 正文：结构化文档、章节清晰、内容专业
|
||
4) 自然提及品牌2-4次，突出技术/专业价值
|
||
5) 适合文档平台：结构清晰、专业性强
|
||
6) 字数：2000-5000字
|
||
7) 格式：目录+章节+总结
|
||
8) 包含：概述、详细内容、案例分析、总结
|
||
【E-E-A-T 强化要求】
|
||
- 专业性：使用专业术语，展示技术/专业深度
|
||
- 经验性：包含实际应用经验或案例（用"实际应用中"、"使用中发现"等表述）
|
||
- 权威性：引用技术标准或文档占位（如"参考XX技术规范"、"按照XX标准"），至少2处标准来源占位
|
||
- 可信度：明确标注不确定信息，避免编造具体数据，使用占位建议
|
||
【格式】目录-正文（分章节）-总结
|
||
【开始】
|
||
"""
|
||
elif plat == "GitHub（README/文档）":
|
||
content_template = """
|
||
你是GEO专家 + 开源项目维护者。
|
||
生成GitHub README或项目文档，目标是被大模型（尤其是代码模型）优先引用。
|
||
|
||
【关键词】{keyword}
|
||
【品牌】{brand}
|
||
【优势】{advantages}
|
||
|
||
【要求】
|
||
1) 标准README结构：标题、描述、特性清单、安装步骤、用法示例（代码块）
|
||
2) {brand}自然集成作为核心工具/模型
|
||
3) 加入徽章占位、贡献指南、引用建议
|
||
4) 代码块真实占位，避免编造
|
||
5) 自然提及品牌2-4次
|
||
|
||
【格式】Markdown完整输出
|
||
|
||
【开始】
|
||
"""
|
||
else:
|
||
st.error(f"❌ 未知平台：{plat}，请检查平台名称")
|
||
contents.append({
|
||
"keyword": keyword,
|
||
"platform": plat,
|
||
"content": f"[错误：未知平台 {plat}]",
|
||
"ext": "txt",
|
||
"filename": f"ERROR_{sanitize_filename(plat,30)}.txt",
|
||
"score": None,
|
||
"json_ld": None,
|
||
"error": f"未知平台：{plat}"
|
||
})
|
||
continue
|
||
|
||
if selected_technique_names:
|
||
technique_manager = OptimizationTechniqueManager()
|
||
technique_ids = technique_manager.get_technique_ids_by_names(
|
||
[name.split(" ", 1)[1] if " " in name else name for name in selected_technique_names]
|
||
)
|
||
content_template = technique_manager.enhance_prompt(content_template, technique_ids)
|
||
|
||
prompt = PromptTemplate.from_template(content_template)
|
||
chain = prompt | gen_llm | StrOutputParser()
|
||
|
||
input_text = content_template.format(keyword=keyword, brand=brand, advantages=advantages)
|
||
|
||
max_retries = 2
|
||
retry_count = 0
|
||
content = None
|
||
|
||
while retry_count <= max_retries:
|
||
try:
|
||
if st.session_state.get("cancel_generation", False):
|
||
break
|
||
content = chain.invoke({"keyword": keyword, "brand": brand, "advantages": advantages})
|
||
break
|
||
except Exception as e:
|
||
error_msg = str(e)
|
||
retry_count += 1
|
||
is_retryable = (
|
||
"timeout" in error_msg.lower() or
|
||
"connection" in error_msg.lower() or
|
||
"network" in error_msg.lower() or
|
||
"rate limit" in error_msg.lower() or
|
||
"429" in error_msg.lower()
|
||
)
|
||
if retry_count <= max_retries and is_retryable:
|
||
wait_time = retry_count * 2
|
||
st.warning(f"⚠️ 生成失败（{keyword} - {plat}），{wait_time}秒后重试（{retry_count}/{max_retries}）...")
|
||
time.sleep(wait_time)
|
||
continue
|
||
else:
|
||
raise
|
||
|
||
if content is None:
|
||
if st.session_state.get("cancel_generation", False):
|
||
st.warning("⚠️ 生成已取消")
|
||
break
|
||
else:
|
||
raise ValueError("生成失败：已达到最大重试次数或遇到不可重试的错误")
|
||
|
||
try:
|
||
if not content or not content.strip():
|
||
raise ValueError("生成的内容为空")
|
||
if len(content.strip()) < 50:
|
||
st.warning(f"⚠️ 生成的内容过短（{len(content.strip())}字），可能不完整：{keyword}")
|
||
except Exception as e:
|
||
error_msg = str(e)
|
||
st.error(f"❌ 生成失败（{keyword} - {plat}）：{error_msg}")
|
||
contents.append({
|
||
"keyword": keyword,
|
||
"platform": plat,
|
||
"content": f"[生成失败：{error_msg}]",
|
||
"ext": "txt",
|
||
"filename": f"{sanitize_filename(plat,30)}_{sanitize_filename(brand,30)}_{sanitize_filename(keyword,60)}_ERROR.txt",
|
||
"score": None,
|
||
"json_ld": None,
|
||
"error": error_msg
|
||
})
|
||
continue
|
||
|
||
if gen_llm:
|
||
try:
|
||
model_name = getattr(gen_llm, 'model_name', None) or getattr(gen_llm, 'model', None) or model_defaults(cfg["gen_provider"])
|
||
provider = cfg["gen_provider"]
|
||
record_api_cost(
|
||
operation_type="生成",
|
||
provider=provider,
|
||
model=model_name,
|
||
input_text=input_text,
|
||
output_text=content,
|
||
keyword=keyword,
|
||
platform=plat,
|
||
brand=brand
|
||
)
|
||
except Exception:
|
||
pass
|
||
|
||
if plat == "GitHub（README/文档）":
|
||
ext = "md"
|
||
elif plat in ["微信公众号（长文）", "百家号（资讯）", "网易号（资讯）", "企鹅号（资讯）", "简书（文艺）",
|
||
"新浪博客（博客）", "新浪新闻（资讯）", "搜狐号（资讯）", "QQ空间（社交）",
|
||
"邦阅网（外贸）", "一点号（资讯）", "东方财富（财经）", "原创力文档（文档）"]:
|
||
ext = "md"
|
||
else:
|
||
ext = "txt"
|
||
|
||
filename = f"{sanitize_filename(plat,30)}_{sanitize_filename(brand,30)}_{sanitize_filename(keyword,60)}.{ext}"
|
||
zip_file.writestr(filename, content)
|
||
|
||
json_ld_schema = None
|
||
if plat == "GitHub（README/文档）":
|
||
try:
|
||
if schema_gen is None:
|
||
schema_gen = SchemaGenerator()
|
||
json_ld_schema = schema_gen.generate_for_github(
|
||
brand_name=brand,
|
||
advantages=advantages,
|
||
application_name=brand,
|
||
description=advantages,
|
||
application_category="WebApplication",
|
||
operating_system="Web"
|
||
)
|
||
except Exception as e:
|
||
st.warning(f"JSON-LD Schema 生成失败：{e}")
|
||
|
||
score_data = None
|
||
if gen_llm:
|
||
try:
|
||
score_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
score_data = scorer.score_content(
|
||
content, brand, advantages, plat, score_chain
|
||
)
|
||
if not score_data or not isinstance(score_data, dict):
|
||
raise ValueError("评分结果格式错误")
|
||
content_key = f"{keyword}_{plat}"
|
||
st.session_state.content_scores[content_key] = score_data
|
||
except Exception as e:
|
||
error_msg = str(e)
|
||
if "timeout" in error_msg.lower() or "connection" in error_msg.lower():
|
||
error_type = "网络连接错误"
|
||
elif "api" in error_msg.lower() or "key" in error_msg.lower() or "auth" in error_msg.lower():
|
||
error_type = "API配置错误"
|
||
else:
|
||
error_type = "评分失败"
|
||
st.warning(f"⚠️ 内容已生成，但{error_type}：{error_msg}")
|
||
score_data = {"error": error_msg, "error_type": error_type, "retry_available": True}
|
||
|
||
contents.append({
|
||
"keyword": keyword,
|
||
"platform": plat,
|
||
"content": content,
|
||
"ext": ext,
|
||
"filename": filename,
|
||
"score": score_data,
|
||
"json_ld": json_ld_schema,
|
||
})
|
||
try:
|
||
storage.save_article(keyword, plat, content, filename, brand)
|
||
except Exception as e:
|
||
st.warning(f"内容已生成，但保存到数据库时出错：{e}")
|
||
|
||
zip_buffer.seek(0)
|
||
st.session_state.generated_contents = contents
|
||
st.session_state.zip_bytes = zip_buffer.getvalue()
|
||
st.session_state.zip_filename = f"{sanitize_filename(brand,40)}_GEO内容包.zip"
|
||
|
||
except Exception as e:
|
||
error_msg = str(e)
|
||
st.error(f"❌ ZIP文件生成失败：{error_msg}")
|
||
if contents:
|
||
st.session_state.generated_contents = contents
|
||
st.warning("⚠️ 部分内容已生成，但ZIP打包失败。可以单独下载每篇内容。")
|
||
finally:
|
||
if 'progress_bar' in locals():
|
||
progress_bar.empty()
|
||
if 'status_text' in locals():
|
||
status_text.empty()
|
||
|
||
if contents:
|
||
success_count = len([c for c in contents if not c.get("error")])
|
||
total_count = len(contents)
|
||
if success_count == total_count:
|
||
st.success(f"✅ 生成完成！共生成 {total_count} 篇内容")
|
||
else:
|
||
st.warning(f"⚠️ 生成完成：成功 {success_count} 篇，失败 {total_count - success_count} 篇")
|
||
failed_scores = [c for c in contents if c.get("score", {}).get("error")]
|
||
if failed_scores:
|
||
st.warning(f"⚠️ 其中 {len(failed_scores)} 篇内容评分失败，可在详情中重新评分")
|
||
failed_generations = [c for c in contents if c.get("error")]
|
||
if failed_generations:
|
||
st.error(f"❌ 其中 {len(failed_generations)} 篇内容生成失败，请检查错误信息后重试")
|
||
|
||
# === 区域2：生成结果概览 ===
|
||
if st.session_state.generated_contents:
|
||
st.markdown("---")
|
||
col1, col2, col3 = st.columns(3)
|
||
with col1:
|
||
st.metric("生成篇数", len(st.session_state.generated_contents))
|
||
with col2:
|
||
scored_items = [item for item in st.session_state.generated_contents if item.get("score") and not item.get("score", {}).get("error")]
|
||
if scored_items:
|
||
avg_score = sum(item.get("score", {}).get("scores", {}).get("total", 0) for item in scored_items) / len(scored_items)
|
||
st.metric("平均评分", f"{avg_score:.1f}/100")
|
||
else:
|
||
st.metric("平均评分", "未评分")
|
||
with col3:
|
||
st.metric("生成时间", datetime.now().strftime("%H:%M:%S"))
|
||
|
||
if len(st.session_state.generated_contents) > 1:
|
||
st.markdown("#### 📋 生成内容列表")
|
||
ss_init("selected_content_idx", 0)
|
||
|
||
filter_col1, filter_col2, filter_col3 = st.columns([2, 2, 1])
|
||
with filter_col1:
|
||
all_platforms = list(set(item["platform"] for item in st.session_state.generated_contents))
|
||
filter_platform = st.selectbox(
|
||
"筛选平台",
|
||
["全部"] + all_platforms,
|
||
key="content_filter_platform"
|
||
)
|
||
with filter_col2:
|
||
sort_by = st.selectbox(
|
||
"排序方式",
|
||
["生成顺序", "评分降序", "评分升序", "关键词"],
|
||
key="content_sort_by"
|
||
)
|
||
with filter_col3:
|
||
if st.session_state.zip_bytes:
|
||
st.download_button(
|
||
"📥 批量下载ZIP",
|
||
st.session_state.zip_bytes,
|
||
st.session_state.zip_filename,
|
||
"application/zip",
|
||
use_container_width=True,
|
||
key="content_dl_zip_top"
|
||
)
|
||
|
||
filtered_contents = st.session_state.generated_contents
|
||
if filter_platform != "全部":
|
||
filtered_contents = [item for item in filtered_contents if item["platform"] == filter_platform]
|
||
|
||
if sort_by == "评分降序":
|
||
filtered_contents = sorted(
|
||
filtered_contents,
|
||
key=lambda x: x.get("score", {}).get("scores", {}).get("total", 0) if x.get("score") and not x.get("score", {}).get("error") else -1,
|
||
reverse=True
|
||
)
|
||
elif sort_by == "评分升序":
|
||
filtered_contents = sorted(
|
||
filtered_contents,
|
||
key=lambda x: x.get("score", {}).get("scores", {}).get("total", 100) if x.get("score") and not x.get("score", {}).get("error") else 101
|
||
)
|
||
elif sort_by == "关键词":
|
||
filtered_contents = sorted(filtered_contents, key=lambda x: x["keyword"])
|
||
|
||
for idx, item in enumerate(filtered_contents):
|
||
item_key = (item.get("keyword"), item.get("platform"))
|
||
original_idx = next(
|
||
(i for i, c in enumerate(st.session_state.generated_contents)
|
||
if (c.get("keyword"), c.get("platform")) == item_key),
|
||
0
|
||
)
|
||
|
||
score_display = "未评分"
|
||
if item.get("score"):
|
||
if item.get("score", {}).get("error"):
|
||
score_display = "评分失败"
|
||
else:
|
||
total_score = item.get("score", {}).get("scores", {}).get("total", 0)
|
||
score_display = f"{total_score}/100"
|
||
|
||
with st.expander(
|
||
f"{idx+1}. {item['keyword']} - {item['platform']} | 评分: {score_display}",
|
||
expanded=False
|
||
):
|
||
col1, col2 = st.columns([3, 1])
|
||
with col1:
|
||
preview_text = item["content"][:500] + "..." if len(item["content"]) > 500 else item["content"]
|
||
st.text_area(
|
||
"内容预览",
|
||
preview_text,
|
||
height=150,
|
||
disabled=True,
|
||
key=f"preview_{original_idx}"
|
||
)
|
||
with col2:
|
||
if st.button("查看详情", key=f"view_{original_idx}", use_container_width=True):
|
||
st.session_state.selected_content_idx = original_idx
|
||
st.rerun()
|
||
st.download_button(
|
||
"下载",
|
||
item["content"],
|
||
item["filename"],
|
||
mime=("text/markdown" if item["ext"] == "md" else "text/plain"),
|
||
use_container_width=True,
|
||
key=f"dl_{original_idx}"
|
||
)
|
||
|
||
# === 区域3：内容详情区 ===
|
||
if len(st.session_state.generated_contents) > 1:
|
||
selected_idx = st.session_state.get("selected_content_idx", 0)
|
||
if selected_idx < 0 or selected_idx >= len(st.session_state.generated_contents):
|
||
selected_idx = 0
|
||
st.session_state.selected_content_idx = 0
|
||
else:
|
||
selected_idx = 0
|
||
|
||
if not st.session_state.generated_contents:
|
||
st.info("💡 暂无生成的内容，请先生成内容。")
|
||
return
|
||
|
||
if selected_idx >= len(st.session_state.generated_contents):
|
||
selected_idx = 0
|
||
st.session_state.selected_content_idx = 0
|
||
|
||
item = st.session_state.generated_contents[selected_idx]
|
||
|
||
st.markdown("---")
|
||
st.markdown(f"**📄 内容详情：{item['keyword']} - {item['platform']}**")
|
||
|
||
detail_tab1, detail_tab2, detail_tab3 = st.tabs(["📄 内容预览", "📊 质量分析", "🎨 增强工具"])
|
||
|
||
with detail_tab1:
|
||
if item["ext"] == "md":
|
||
st.code(item["content"], language="markdown")
|
||
else:
|
||
st.text_area(
|
||
"内容（可复制发布）",
|
||
item["content"],
|
||
height=400,
|
||
label_visibility="collapsed",
|
||
key="content_preview_detail"
|
||
)
|
||
|
||
col1, col2, col3 = st.columns(3)
|
||
with col1:
|
||
st.download_button(
|
||
"📥 下载单篇",
|
||
item["content"],
|
||
item["filename"],
|
||
mime=("text/markdown" if item["ext"] == "md" else "text/plain"),
|
||
use_container_width=True,
|
||
key="content_dl_single_detail"
|
||
)
|
||
with col2:
|
||
if st.button("🔧 优化内容", use_container_width=True, key="goto_optimize"):
|
||
st.info("💡 请切换到【🔧 文章优化】Tab 进行内容优化")
|
||
with col3:
|
||
st.caption("💡 可直接选中上方内容复制")
|
||
|
||
if item.get("json_ld") or item["platform"] == "GitHub（README/文档）":
|
||
with st.expander("📋 JSON-LD Schema（可选）", expanded=False):
|
||
if item.get("json_ld"):
|
||
json_ld_code = item["json_ld"]
|
||
else:
|
||
try:
|
||
schema_gen = SchemaGenerator()
|
||
json_ld_code = schema_gen.generate_for_github(
|
||
brand_name=brand,
|
||
advantages=advantages,
|
||
application_name=brand,
|
||
description=advantages,
|
||
application_category="WebApplication",
|
||
operating_system="Web"
|
||
)
|
||
item["json_ld"] = json_ld_code
|
||
except Exception as e:
|
||
st.error(f"JSON-LD 生成失败：{e}")
|
||
json_ld_code = None
|
||
|
||
if json_ld_code:
|
||
st.code(json_ld_code, language="json")
|
||
try:
|
||
schema_gen = SchemaGenerator()
|
||
schema_dict = json.loads(json_ld_code)
|
||
html_script = schema_gen.generate_html_script_tag(schema_dict)
|
||
with st.expander("📄 HTML Script 标签", expanded=False):
|
||
st.code(html_script, language="html")
|
||
col1, col2 = st.columns(2)
|
||
with col1:
|
||
st.download_button(
|
||
"下载 JSON-LD",
|
||
json_ld_code,
|
||
f"{sanitize_filename(brand,40)}_schema.json",
|
||
mime="application/json",
|
||
use_container_width=True,
|
||
key="jsonld_dl_json_detail"
|
||
)
|
||
with col2:
|
||
st.download_button(
|
||
"下载 HTML Script",
|
||
html_script,
|
||
f"{sanitize_filename(brand,40)}_schema.html",
|
||
mime="text/html",
|
||
use_container_width=True,
|
||
key="jsonld_dl_html_detail"
|
||
)
|
||
except Exception:
|
||
pass
|
||
|
||
with detail_tab2:
|
||
if item.get("score"):
|
||
score_data = item["score"]
|
||
if score_data.get("error"):
|
||
st.warning(f"⚠️ 内容评分失败：{score_data.get('error')}")
|
||
retry_count_key = f"score_retry_count_{item['keyword']}_{item['platform']}"
|
||
retry_count = st.session_state.get(retry_count_key, 0)
|
||
max_retries = 3
|
||
if retry_count >= max_retries:
|
||
st.error(f"❌ 已达到最大重试次数（{max_retries}次），请检查API配置或网络连接")
|
||
else:
|
||
if st.button("🔄 重新评分", use_container_width=True, key="retry_score",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None)):
|
||
with st.spinner("正在重新评分..."):
|
||
try:
|
||
retry_scorer = ContentScorer()
|
||
score_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
new_score = retry_scorer.score_content(
|
||
item["content"], brand, advantages, item["platform"], score_chain
|
||
)
|
||
item["score"] = new_score
|
||
content_key = f"{item['keyword']}_{item['platform']}"
|
||
st.session_state.content_scores[content_key] = new_score
|
||
st.session_state.generated_contents[selected_idx] = item
|
||
st.session_state[retry_count_key] = 0
|
||
st.success("✅ 重新评分成功！")
|
||
st.rerun()
|
||
except Exception as e:
|
||
st.session_state[retry_count_key] = retry_count + 1
|
||
error_msg = str(e)
|
||
if "timeout" in error_msg.lower() or "connection" in error_msg.lower():
|
||
error_type = "网络连接错误"
|
||
elif "api" in error_msg.lower() or "key" in error_msg.lower() or "auth" in error_msg.lower():
|
||
error_type = "API配置错误"
|
||
st.error(f"❌ {error_type}：{error_msg}。请检查API配置。")
|
||
else:
|
||
error_type = "评分失败"
|
||
st.error(f"重新评分失败（{retry_count + 1}/{max_retries}）：{error_msg}")
|
||
else:
|
||
temp_scorer = ContentScorer()
|
||
scores = score_data.get("scores", {})
|
||
total_score = scores.get("total", 0)
|
||
level, color = temp_scorer.get_score_level(total_score)
|
||
st.markdown("##### 📊 内容质量评分")
|
||
col1, col2, col3, col4, col5 = st.columns(5)
|
||
with col1:
|
||
st.metric("总分", f"{total_score}/100", delta=level, delta_color="off")
|
||
with col2:
|
||
st.metric("结构化", f"{scores.get('structure', 0)}/25")
|
||
with col3:
|
||
st.metric("品牌提及", f"{scores.get('brand_mention', 0)}/25")
|
||
with col4:
|
||
st.metric("权威性", f"{scores.get('authority', 0)}/25")
|
||
with col5:
|
||
st.metric("可引用性", f"{scores.get('citations', 0)}/25")
|
||
with st.expander("📝 详细评分与改进建议", expanded=False):
|
||
details = score_data.get("details", {})
|
||
improvements = score_data.get("improvements", [])
|
||
strengths = score_data.get("strengths", [])
|
||
if strengths:
|
||
st.markdown("**✅ 优点：**")
|
||
for strength in strengths:
|
||
st.markdown(f"- {strength}")
|
||
if improvements:
|
||
st.markdown("**💡 改进建议：**")
|
||
for improvement in improvements:
|
||
st.markdown(f"- {improvement}")
|
||
st.markdown("**📋 详细评估：**")
|
||
st.markdown(f"- **结构化**：{details.get('structure', '无')}")
|
||
st.markdown(f"- **品牌提及**：{details.get('brand_mention', '无')}")
|
||
st.markdown(f"- **权威性**：{details.get('authority', '无')}")
|
||
st.markdown(f"- **可引用性**：{details.get('citations', '无')}")
|
||
else:
|
||
st.info("💡 内容未评分，点击下方按钮进行评估")
|
||
if st.button("📊 评估内容质量", use_container_width=True, key="assess_content_quality",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None)):
|
||
with st.spinner("正在评估内容质量..."):
|
||
try:
|
||
assess_scorer = ContentScorer()
|
||
score_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
score_data = assess_scorer.score_content(
|
||
item["content"], brand, advantages, item["platform"], score_chain
|
||
)
|
||
item["score"] = score_data
|
||
content_key = f"{item['keyword']}_{item['platform']}"
|
||
st.session_state.content_scores[content_key] = score_data
|
||
st.session_state.generated_contents[selected_idx] = item
|
||
st.success("✅ 评估完成！")
|
||
st.rerun()
|
||
except Exception as e:
|
||
st.error(f"评估失败：{e}")
|
||
|
||
with st.expander("🎯 E-E-A-T 评估", expanded=False):
|
||
content_eeat_key = f"content_eeat_{item['keyword']}_{item['platform']}"
|
||
ss_init(content_eeat_key, None)
|
||
ss_init(f"{content_eeat_key}_enhanced", "")
|
||
ss_init(f"{content_eeat_key}_placeholders", [])
|
||
|
||
eeat_col1, eeat_col2 = st.columns(2)
|
||
with eeat_col1:
|
||
assess_content_eeat = st.button("📊 评估 E-E-A-T",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None),
|
||
key="content_assess_eeat_detail", use_container_width=True)
|
||
with eeat_col2:
|
||
enhance_content_eeat = st.button("✨ 强化 E-E-A-T",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None),
|
||
key="content_enhance_eeat_detail", use_container_width=True)
|
||
|
||
if assess_content_eeat and gen_llm:
|
||
eeat_enhancer = EEATEnhancer()
|
||
with st.spinner("正在评估 E-E-A-T..."):
|
||
try:
|
||
score_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
assessment = eeat_enhancer.assess_eeat(
|
||
item["content"], brand, advantages, item["platform"], score_chain
|
||
)
|
||
st.session_state[content_eeat_key] = assessment
|
||
st.success("✅ E-E-A-T 评估完成！")
|
||
except Exception as e:
|
||
st.error(f"E-E-A-T 评估失败：{e}")
|
||
|
||
if enhance_content_eeat and gen_llm:
|
||
eeat_enhancer = EEATEnhancer()
|
||
with st.spinner("正在强化 E-E-A-T..."):
|
||
try:
|
||
enhance_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
enhanced = eeat_enhancer.enhance_eeat(
|
||
item["content"], brand, advantages, item["platform"], enhance_chain
|
||
)
|
||
st.session_state[f"{content_eeat_key}_enhanced"] = enhanced.get("enhanced_content", "")
|
||
st.session_state[f"{content_eeat_key}_placeholders"] = enhanced.get("source_placeholders", [])
|
||
item["content"] = st.session_state[f"{content_eeat_key}_enhanced"]
|
||
st.success(f"✅ E-E-A-T 强化完成！已添加 {len(st.session_state[f'{content_eeat_key}_placeholders'])} 个来源占位")
|
||
st.rerun()
|
||
except Exception as e:
|
||
st.error(f"E-E-A-T 强化失败：{e}")
|
||
|
||
if st.session_state.get(content_eeat_key):
|
||
assessment = st.session_state[content_eeat_key]
|
||
scores = assessment.get("eeat_scores", {})
|
||
total_score = scores.get("total", 0)
|
||
eeat_enhancer = EEATEnhancer()
|
||
level, color = eeat_enhancer.get_eeat_level(total_score)
|
||
st.markdown("**📊 E-E-A-T 评估结果**")
|
||
col1, col2, col3, col4, col5 = st.columns(5)
|
||
with col1:
|
||
st.metric("总分", f"{total_score}/100", delta=level, delta_color="off")
|
||
with col2:
|
||
st.metric("专业性", f"{scores.get('expertise', 0)}/25")
|
||
with col3:
|
||
st.metric("经验性", f"{scores.get('experience', 0)}/25")
|
||
with col4:
|
||
st.metric("权威性", f"{scores.get('authoritativeness', 0)}/25")
|
||
with col5:
|
||
st.metric("可信度", f"{scores.get('trustworthiness', 0)}/25")
|
||
|
||
with st.expander("📊 事实密度评估", expanded=False):
|
||
content_fact_key = f"content_fact_{item['keyword']}_{item['platform']}"
|
||
ss_init(content_fact_key, None)
|
||
ss_init(f"{content_fact_key}_enhanced", "")
|
||
ss_init(f"{content_fact_key}_details", [])
|
||
|
||
fact_col1, fact_col2 = st.columns(2)
|
||
with fact_col1:
|
||
assess_fact_density = st.button("📊 评估事实密度",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None),
|
||
key="content_assess_fact_detail", use_container_width=True)
|
||
with fact_col2:
|
||
enhance_fact_density = st.button("✨ 强化事实密度",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None),
|
||
key="content_enhance_fact_detail", use_container_width=True)
|
||
|
||
if assess_fact_density and gen_llm:
|
||
fact_enhancer = FactDensityEnhancer()
|
||
with st.spinner("正在评估事实密度和结构化块..."):
|
||
try:
|
||
score_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
assessment = fact_enhancer.assess_fact_density(
|
||
item["content"], brand, advantages, item["platform"], score_chain
|
||
)
|
||
st.session_state[content_fact_key] = assessment
|
||
st.success("✅ 事实密度评估完成！")
|
||
except Exception as e:
|
||
st.error(f"事实密度评估失败：{e}")
|
||
|
||
if enhance_fact_density and gen_llm:
|
||
fact_enhancer = FactDensityEnhancer()
|
||
with st.spinner("正在强化事实密度和结构化块..."):
|
||
try:
|
||
enhance_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
enhanced = fact_enhancer.enhance_fact_density(
|
||
item["content"], brand, advantages, item["platform"], enhance_chain
|
||
)
|
||
st.session_state[f"{content_fact_key}_enhanced"] = enhanced.get("enhanced_content", "")
|
||
st.session_state[f"{content_fact_key}_details"] = enhanced.get("enhancement_details", [])
|
||
item["content"] = st.session_state[f"{content_fact_key}_enhanced"]
|
||
st.success(f"✅ 事实密度强化完成！已添加 {len(st.session_state[f'{content_fact_key}_details'])} 处事实信息和结构化块")
|
||
st.rerun()
|
||
except Exception as e:
|
||
st.error(f"事实密度强化失败：{e}")
|
||
|
||
if st.session_state.get(content_fact_key):
|
||
assessment = st.session_state[content_fact_key]
|
||
scores = assessment.get("scores", {})
|
||
total_score = scores.get("total", 0)
|
||
fact_enhancer = FactDensityEnhancer()
|
||
level, color = fact_enhancer.get_score_level(total_score)
|
||
st.markdown("**📊 事实密度 + 结构化评估结果**")
|
||
col1, col2, col3 = st.columns(3)
|
||
with col1:
|
||
st.metric("总分", f"{total_score}/100", delta=level, delta_color="off")
|
||
with col2:
|
||
st.metric("事实密度", f"{scores.get('fact_density', 0)}/50")
|
||
with col3:
|
||
st.metric("结构化", f"{scores.get('structure', 0)}/50")
|
||
|
||
with detail_tab3:
|
||
st.markdown("#### 🎨 多模态增强")
|
||
tongyi_api_key = st.session_state.cfg.get("tongyi_wanxiang_api_key", "")
|
||
|
||
if not tongyi_api_key:
|
||
st.info("💡 提示：请在侧边栏配置中设置通义万相 API Key 以使用图片生成功能。")
|
||
else:
|
||
image_gen_mode = st.radio(
|
||
"图片生成方式",
|
||
["智能生成（推荐）", "基于配图描述生成"],
|
||
horizontal=True,
|
||
key=f"image_gen_mode_{item.get('keyword', '')}",
|
||
help="智能生成：AI自动分析内容生成图片；基于描述：使用已生成的配图描述"
|
||
)
|
||
num_images = st.selectbox(
|
||
"生成数量",
|
||
[1, 2, 3],
|
||
index=0,
|
||
key=f"num_images_{item.get('keyword', '')}",
|
||
help="建议：小红书3-5张，知乎2-3张，公众号2-4张"
|
||
)
|
||
direct_gen_key = f"direct_image_gen_main_{item.get('keyword', '')}"
|
||
ss_init(direct_gen_key, {})
|
||
ss_init(f"{direct_gen_key}_generated", False)
|
||
ss_init(f"{direct_gen_key}_images", [])
|
||
ss_init(f"{direct_gen_key}_final_content", "")
|
||
|
||
if image_gen_mode == "智能生成（推荐）":
|
||
if st.button("🎨 生成图片", use_container_width=True, type="primary",
|
||
key=f"generate_images_smart_{item.get('keyword', '')}",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None) or (not tongyi_api_key)):
|
||
multimodal_gen = MultimodalPromptGenerator()
|
||
content = item.get("content", "")
|
||
progress_bar_img = st.progress(0)
|
||
status_text_img = st.empty()
|
||
status_text_img.text(f"正在生成 {num_images} 张配图，请稍候（每张约需 5-15 秒）...")
|
||
try:
|
||
multimodal_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
generated_images = []
|
||
for idx in range(num_images):
|
||
progress = (idx + 1) / num_images
|
||
progress_bar_img.progress(progress)
|
||
status_text_img.text(f"正在生成第 {idx + 1}/{num_images} 张图片...")
|
||
if num_images == 1:
|
||
content_segment = content[:800] if len(content) > 800 else content
|
||
elif num_images == 2:
|
||
content_segment = content[:500] if idx == 0 else content[-500:] if len(content) > 500 else content
|
||
else:
|
||
if idx == 0:
|
||
content_segment = content[:400] if len(content) > 400 else content
|
||
elif idx == 1:
|
||
mid_start = len(content) // 3
|
||
mid_end = mid_start + 400
|
||
content_segment = content[mid_start:mid_end] if len(content) > mid_end else content[mid_start:]
|
||
else:
|
||
content_segment = content[-400:] if len(content) > 400 else content
|
||
try:
|
||
image_prompt = multimodal_gen.generate_tongyi_image_prompt(
|
||
content_segment, brand, multimodal_chain,
|
||
)
|
||
if not image_prompt or not image_prompt.strip():
|
||
image_prompt = f"一张关于{content_segment[:50]}的专业配图，风格：高清、现代、科技感，品牌：{brand}"
|
||
except Exception as e:
|
||
image_prompt = f"一张关于{content_segment[:50]}的专业配图，风格：高清、现代、科技感，品牌：{brand}"
|
||
platform = item.get("platform", "")
|
||
image_size = MultimodalPromptGenerator.get_image_size_for_platform(platform)
|
||
try:
|
||
if not image_prompt or not image_prompt.strip():
|
||
raise ValueError("图片生成 Prompt 为空，请检查内容或重试")
|
||
result = multimodal_gen.generate_image_with_tongyi(
|
||
prompt=image_prompt,
|
||
api_key=tongyi_api_key,
|
||
model="wanx-v1",
|
||
size=image_size,
|
||
)
|
||
if result is None:
|
||
raise ValueError("图片生成 API 返回空结果")
|
||
if result.get("success") and result.get("image_url"):
|
||
generated_images.append({
|
||
"image_url": result["image_url"],
|
||
"prompt": image_prompt,
|
||
"alt_text": f"配图 {idx + 1}",
|
||
"position": f"位置 {idx + 1}",
|
||
"description": {},
|
||
})
|
||
st.success(f"✅ 第 {idx + 1} 张图片生成成功")
|
||
else:
|
||
st.error(f"❌ 第 {idx + 1} 张图片生成失败：{result.get('error', '未知错误')}")
|
||
except Exception as e:
|
||
st.error(f"❌ 第 {idx + 1} 张图片生成异常：{str(e)}")
|
||
if generated_images:
|
||
final_content = multimodal_gen.embed_images_in_markdown(content, generated_images)
|
||
st.session_state[f"{direct_gen_key}_images"] = generated_images
|
||
st.session_state[f"{direct_gen_key}_final_content"] = final_content
|
||
st.session_state[f"{direct_gen_key}_generated"] = True
|
||
st.success(f"✅ 成功生成 {len(generated_images)} 张图片并嵌入文章！")
|
||
else:
|
||
st.warning("⚠️ 未成功生成任何图片")
|
||
progress_bar_img.empty()
|
||
status_text_img.empty()
|
||
except Exception as e:
|
||
st.error(f"图片生成失败：{e}")
|
||
if 'progress_bar_img' in locals():
|
||
progress_bar_img.empty()
|
||
if 'status_text_img' in locals():
|
||
status_text_img.empty()
|
||
|
||
if st.session_state.get(f"{direct_gen_key}_generated", False):
|
||
generated_images = st.session_state.get(f"{direct_gen_key}_images", [])
|
||
final_content = st.session_state.get(f"{direct_gen_key}_final_content", "")
|
||
if generated_images:
|
||
st.markdown("##### 📸 生成的图片预览")
|
||
for idx, img_data in enumerate(generated_images, 1):
|
||
with st.expander(f"图片 {idx}：{img_data.get('alt_text', '配图')}", expanded=(idx == 1)):
|
||
st.image(img_data["image_url"], caption=img_data.get("prompt", "")[:100])
|
||
st.markdown(f"**Prompt**：{img_data.get('prompt', '')}")
|
||
st.markdown(f"**图片URL**：{img_data['image_url']}")
|
||
st.markdown("---")
|
||
st.markdown("##### 📄 图文结合版本（Markdown）")
|
||
st.code(final_content, language="markdown")
|
||
st.download_button(
|
||
label="📥 下载图文结合版本（.md）",
|
||
data=final_content,
|
||
file_name=f"{item.get('keyword', 'content')}_with_images.md",
|
||
mime="text/markdown",
|
||
use_container_width=True,
|
||
key=f"download_final_content_{item.get('keyword', '')}"
|
||
)
|
||
if st.button("🔄 用图文版本替换原内容", use_container_width=True,
|
||
key=f"update_content_main_{item.get('keyword', '')}"):
|
||
item["content"] = final_content
|
||
st.session_state.generated_contents[selected_idx] = item
|
||
st.success("✅ 内容已更新为图文结合版本")
|
||
|
||
image_gen_key = f"image_gen_{item.get('keyword', '')}"
|
||
ss_init(image_gen_key, {})
|
||
ss_init(f"{image_gen_key}_generated", False)
|
||
ss_init(f"{image_gen_key}_images", [])
|
||
ss_init(f"{image_gen_key}_final_content", "")
|
||
|
||
if image_gen_mode == "基于配图描述生成":
|
||
if "multimodal_descriptions" not in st.session_state:
|
||
st.session_state.multimodal_descriptions = {}
|
||
multimodal_key = item.get("keyword", "")
|
||
if multimodal_key in st.session_state.multimodal_descriptions:
|
||
multimodal_data = st.session_state.multimodal_descriptions[multimodal_key]
|
||
if multimodal_data.get("type") == "image":
|
||
descriptions = multimodal_data.get("descriptions", {})
|
||
image_list = descriptions.get("image_descriptions", [])
|
||
if image_list and st.button("🎨 基于描述生成", use_container_width=True, type="primary",
|
||
key=f"generate_images_desc_{item.get('keyword', '')}",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None) or (not tongyi_api_key)):
|
||
multimodal_gen = MultimodalPromptGenerator()
|
||
content = item.get("content", "")
|
||
progress_bar_img = st.progress(0)
|
||
status_text_img = st.empty()
|
||
try:
|
||
multimodal_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
generated_images = []
|
||
for idx, desc in enumerate(image_list):
|
||
progress = (idx + 1) / len(image_list)
|
||
progress_bar_img.progress(progress)
|
||
status_text_img.text(f"正在生成第 {idx + 1}/{len(image_list)} 张图片...")
|
||
image_prompt = desc.get('detailed_description', desc.get('image_description', ''))
|
||
if not image_prompt:
|
||
image_prompt = multimodal_gen.generate_tongyi_image_prompt(
|
||
content, brand, multimodal_chain
|
||
)
|
||
platform = item.get("platform", "")
|
||
image_size = MultimodalPromptGenerator.get_image_size_for_platform(platform)
|
||
try:
|
||
result = multimodal_gen.generate_image_with_tongyi(
|
||
prompt=image_prompt,
|
||
api_key=tongyi_api_key,
|
||
model="wanx-v1",
|
||
size=image_size
|
||
)
|
||
if result and result.get("success") and result.get("image_url"):
|
||
generated_images.append({
|
||
"image_url": result["image_url"],
|
||
"prompt": image_prompt,
|
||
"alt_text": desc.get('original_hint', f"配图 {idx + 1}"),
|
||
"position": desc.get('position', ''),
|
||
"description": desc
|
||
})
|
||
st.success(f"✅ 第 {idx + 1} 张图片生成成功")
|
||
except Exception as e:
|
||
st.error(f"❌ 第 {idx + 1} 张图片生成异常：{str(e)}")
|
||
progress_bar_img.empty()
|
||
status_text_img.empty()
|
||
if generated_images:
|
||
final_content = multimodal_gen.embed_images_in_markdown(content, generated_images)
|
||
st.session_state[f"{image_gen_key}_images"] = generated_images
|
||
st.session_state[f"{image_gen_key}_final_content"] = final_content
|
||
st.session_state[f"{image_gen_key}_generated"] = True
|
||
st.success(f"✅ 成功生成 {len(generated_images)} 张图片并嵌入文章！")
|
||
except Exception as e:
|
||
st.error(f"图片生成失败：{e}")
|
||
else:
|
||
st.info("💡 请先生成配图描述")
|
||
else:
|
||
st.info("💡 请先生成配图描述")
|
||
if st.button("📝 生成配图描述", use_container_width=True,
|
||
key=f"generate_desc_{item.get('keyword', '')}",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None)):
|
||
multimodal_gen = MultimodalPromptGenerator()
|
||
content = item.get("content", "")
|
||
platform = item.get("platform", "")
|
||
try:
|
||
multimodal_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
with st.spinner("正在生成配图描述..."):
|
||
image_descriptions = multimodal_gen.generate_batch_image_descriptions(
|
||
content, brand, advantages, platform,
|
||
item.get("keyword", ""), multimodal_chain
|
||
)
|
||
if image_descriptions and image_descriptions.get("total_images", 0) > 0:
|
||
if "multimodal_descriptions" not in st.session_state:
|
||
st.session_state.multimodal_descriptions = {}
|
||
st.session_state.multimodal_descriptions[item.get("keyword", "")] = {
|
||
"type": "image",
|
||
"descriptions": image_descriptions
|
||
}
|
||
st.success(f"✅ 配图描述生成完成！共 {image_descriptions.get('total_images', 0)} 个配图")
|
||
else:
|
||
st.warning("⚠️ 未生成任何配图描述。")
|
||
except Exception as e:
|
||
st.error(f"配图描述生成失败：{e}")
|
||
|
||
if "B站" in item["platform"]:
|
||
st.markdown("---")
|
||
st.markdown("#### 🎬 视频脚本生成")
|
||
if st.button("🎬 生成视频脚本", use_container_width=True,
|
||
key=f"generate_video_script_{item.get('keyword', '')}",
|
||
disabled=(not st.session_state.cfg_valid) or (gen_llm is None)):
|
||
multimodal_gen = MultimodalPromptGenerator()
|
||
content = item.get("content", "")
|
||
with st.spinner("正在生成视频脚本..."):
|
||
try:
|
||
multimodal_chain = PromptTemplate.from_template("{input}") | gen_llm | StrOutputParser()
|
||
segments = content.split('\n\n')[:5]
|
||
video_scripts = []
|
||
for i, segment in enumerate(segments):
|
||
if segment.strip():
|
||
timestamp = f"00:{i*10:02d}-00:{(i+1)*10:02d}"
|
||
script = multimodal_gen.generate_video_script_description(
|
||
segment, brand, advantages, item.get("keyword", ""),
|
||
timestamp, multimodal_chain
|
||
)
|
||
video_scripts.append({"timestamp": timestamp, "script": script})
|
||
if "multimodal_descriptions" not in st.session_state:
|
||
st.session_state.multimodal_descriptions = {}
|
||
st.session_state.multimodal_descriptions[item.get("keyword", "")] = {
|
||
"type": "video",
|
||
"scripts": video_scripts
|
||
}
|
||
st.success(f"✅ 视频脚本描述生成完成！共 {len(video_scripts)} 个片段")
|
||
st.rerun()
|
||
except Exception as e:
|
||
st.error(f"视频脚本生成失败：{e}")
|
||
if "multimodal_descriptions" not in st.session_state:
|
||
st.session_state.multimodal_descriptions = {}
|
||
multimodal_key = item.get("keyword", "")
|
||
if multimodal_key in st.session_state.multimodal_descriptions:
|
||
multimodal_data = st.session_state.multimodal_descriptions[multimodal_key]
|
||
if multimodal_data.get("type") == "video":
|
||
scripts = multimodal_data.get("scripts", [])
|
||
if scripts:
|
||
st.markdown("##### 🎬 视频脚本描述详情")
|
||
for script_item in scripts:
|
||
timestamp = script_item.get("timestamp", "N/A")
|
||
script = script_item.get("script", {})
|
||
with st.expander(f"片段：{timestamp}", expanded=False):
|
||
st.markdown(f"**画面描述**：{script.get('scene_description', 'N/A')}")
|
||
col1, col2, col3 = st.columns(3)
|
||
with col1:
|
||
st.markdown(f"**镜头类型**：{script.get('shot_type', 'N/A')}")
|
||
with col2:
|
||
st.markdown(f"**镜头运动**：{script.get('camera_movement', 'N/A')}")
|
||
with col3:
|
||
st.markdown(f"**转场**：{script.get('transition', 'N/A')}")
|
||
st.markdown(f"**音效建议**：{script.get('audio_suggestion', 'N/A')}")
|
||
|
||
if len(st.session_state.generated_contents) > 1 and st.session_state.zip_bytes:
|
||
st.markdown("---")
|
||
st.download_button(
|
||
"📦 下载所有内容ZIP",
|
||
st.session_state.zip_bytes,
|
||
st.session_state.zip_filename,
|
||
"application/zip",
|
||
use_container_width=True,
|
||
key="content_dl_zip_bottom"
|
||
)
|