jinye_huang/TravelContentCreator
Template
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

修改了结果文本的渲染方法

Browse Source
This commit is contained in:
jinye_huang 2025-05-12 15:43:32 +08:00
parent 6415966034
commit 2c39d981f4
1 changed files with 72 additions and 23 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

95
scripts/extract_and_render.py
View File

@ -10,10 +10,18 @@ import re
import argparse import argparse
from datetime import datetime from datetime import datetime
def convert_json_to_txt_content(json_path): def convert_json_to_txt_content(json_path, prefer_original=False):
""" """
读取 JSON 文件，提取标题、内容和标签，移除 Markdown 格式， 读取 JSON 文件，提取标题、内容和标签，移除 Markdown 格式，
并返回格式化文本。 并返回格式化文本。
根据judge_success字段决定使用原始内容还是审核后内容：
- judge_success为True时使用title/content（除非prefer_original=True）
- judge_success为False时使用original_title/original_content
Args:
json_path: JSON文件路径
prefer_original: 是否优先使用原始内容，无视judge_success结果
""" """
print(f" - 正在读取 JSON: {json_path}") print(f" - 正在读取 JSON: {json_path}")
if not os.path.exists(json_path): if not os.path.exists(json_path):
@ -24,10 +32,35 @@ def convert_json_to_txt_content(json_path):
with open(json_path, 'r', encoding='utf-8') as f_json: with open(json_path, 'r', encoding='utf-8') as f_json:
data = json.load(f_json) data = json.load(f_json)
# 提取字段 # 根据judge_success选择标题和内容
title = data.get('title', '未找到标题') judge_success = data.get('judge_success', None)
content = data.get('content', '未找到内容')
tags = data.get('tags', data.get('tag', '未找到标签')) if prefer_original and 'original_title' in data and 'original_content' in data:
# 优先使用原始内容
title = data.get('original_title', '未找到原始标题')
content = data.get('original_content', '未找到原始内容')
print(f" - 优先使用原始内容 (prefer_original=True)")
elif judge_success is True and not prefer_original:
# 使用审核后的内容
title = data.get('title', '未找到标题')
content = data.get('content', '未找到内容')
print(f" - 使用审核后内容 (judge_success=True)")
elif 'original_title' in data and 'original_content' in data:
# 使用原始内容
title = data.get('original_title', '未找到原始标题')
content = data.get('original_content', '未找到原始内容')
print(f" - 使用原始内容 (judge_success={judge_success})")
else:
# 若无original字段，使用常规字段
title = data.get('title', '未找到标题')
content = data.get('content', '未找到内容')
print(f" - 使用常规内容 (无judge结果)")
# 解决tag/tags字段重复问题，按照修正后的处理逻辑，只使用tags字段
tags = data.get('tags', '')
if not tags and 'tag' in data:
tags = data.get('tag', '未找到标签')
print(f" - 使用tag字段作为标签 (该字段将在后续版本中统一为tags)")
# 移除Markdown格式 # 移除Markdown格式
content_no_format = re.sub(r'\*\*(.*?)\*\*', r'\1', content) content_no_format = re.sub(r'\*\*(.*?)\*\*', r'\1', content)
@ -74,7 +107,7 @@ def load_topic_data(source_dir, run_id):
return topic_data return topic_data
def process_result_directory(source_dir, output_dir, run_id=None): def process_result_directory(source_dir, output_dir, run_id=None, prefer_original=False):
""" """
处理指定的结果目录，提取内容并渲染到输出目录。 处理指定的结果目录，提取内容并渲染到输出目录。
@ -82,6 +115,7 @@ def process_result_directory(source_dir, output_dir, run_id=None):
source_dir: 源目录路径，包含i_j子目录 source_dir: 源目录路径，包含i_j子目录
output_dir: 输出目录路径 output_dir: 输出目录路径
run_id: 可选的运行ID，如果不提供则使用源目录名 run_id: 可选的运行ID，如果不提供则使用源目录名
prefer_original: 是否优先使用原始内容，无视judge_success结果
""" """
if not os.path.isdir(source_dir): if not os.path.isdir(source_dir):
print(f"错误: 源目录不存在: {source_dir}") print(f"错误: 源目录不存在: {source_dir}")
@ -121,7 +155,8 @@ def process_result_directory(source_dir, output_dir, run_id=None):
"AdditionalImagesCount", "AdditionalImagesCount",
"Status", "Status",
"Details", "Details",
"JudgeStatus" "JudgeStatus",
"ContentSource"
] ]
] ]
@ -177,7 +212,8 @@ def process_result_directory(source_dir, output_dir, run_id=None):
"AdditionalImagesCount": 0, "AdditionalImagesCount": 0,
"Status": "Processing", "Status": "Processing",
"Details": "", "Details": "",
"JudgeStatus": "" "JudgeStatus": "",
"ContentSource": "unknown"
} }
# 创建输出条目目录 # 创建输出条目目录
@ -209,7 +245,7 @@ def process_result_directory(source_dir, output_dir, run_id=None):
except Exception as e: except Exception as e:
print(f" - 错误: 读取article.json失败: {e}") print(f" - 错误: 读取article.json失败: {e}")
txt_content, error = convert_json_to_txt_content(json_path) txt_content, error = convert_json_to_txt_content(json_path, prefer_original)
if error: if error:
record["Status"] = "Partial" record["Status"] = "Partial"
record["Details"] += f"文章处理失败: {error}; " record["Details"] += f"文章处理失败: {error}; "
@ -219,6 +255,17 @@ def process_result_directory(source_dir, output_dir, run_id=None):
with open(txt_path, 'w', encoding='utf-8') as f_txt: with open(txt_path, 'w', encoding='utf-8') as f_txt:
f_txt.write(txt_content) f_txt.write(txt_content)
print(f" - 成功写入文本文件: {txt_path}") print(f" - 成功写入文本文件: {txt_path}")
# 记录内容来源
if prefer_original:
record["ContentSource"] = "original_preferred"
elif article_data.get("judge_success") is True:
record["ContentSource"] = "judged"
elif "original_title" in article_data:
record["ContentSource"] = "original"
else:
record["ContentSource"] = "default"
except Exception as e: except Exception as e:
record["Status"] = "Partial" record["Status"] = "Partial"
record["Details"] += f"写入文本文件失败: {e}; " record["Details"] += f"写入文本文件失败: {e}; "
@ -298,30 +345,32 @@ def process_result_directory(source_dir, output_dir, run_id=None):
print(f"结果保存在: {output_dir}") print(f"结果保存在: {output_dir}")
def main(): def main():
# parser = argparse.ArgumentParser(description="从TravelContentCreator结果目录提取内容并渲染到指定目录") parser = argparse.ArgumentParser(description="从TravelContentCreator结果目录提取内容并渲染到指定目录")
# parser.add_argument("--source", type=str, help="源目录路径") parser.add_argument("--source", type=str, help="源目录路径")
# parser.add_argument("--output", type=str, help="输出目录路径") parser.add_argument("--output", type=str, help="输出目录路径")
# parser.add_argument("--run-id", type=str, help="自定义运行ID") parser.add_argument("--run-id", type=str, help="自定义运行ID")
parser.add_argument("--prefer-original", action="store_true", help="优先使用原始内容，忽略审核结果")
# args = parser.parse_args() args = parser.parse_args()
# # 默认值设置 # 默认值设置
# source = args.source if args.source else "/root/autodl-tmp/TravelContentCreator/result/2025-05-11_00-26-30" source = args.source if args.source else "/root/autodl-tmp/TravelContentCreator/result/2025-05-12_09-33-12"
# output = args.output if args.output else "/root/autodl-tmp/TravelContentCreator/output/2025-05-11_00-26-30" output = args.output if args.output else "/root/autodl-tmp/TravelContentCreator/output/2025-05-12_09-33-12"
# run_id = args.run_id if args.run_id else os.path.basename(source) run_id = args.run_id if args.run_id else os.path.basename(source)
prefer_original = args.prefer_original
source = "/root/autodl-tmp/TravelContentCreator/result/2025-05-12_09-33-12"
output = "/root/autodl-tmp/TravelContentCreator/output/2025-05-12_09-33-12"
run_id = os.path.basename(source)
print("-" * 60) print("-" * 60)
print(f"开始提取和渲染流程") print(f"开始提取和渲染流程")
print(f"源目录: {source}") print(f"源目录: {source}")
print(f"输出目录: {output}") print(f"输出目录: {output}")
print(f"运行ID: {run_id}") print(f"运行ID: {run_id}")
if prefer_original:
print("内容模式: 优先使用原始内容")
else:
print("内容模式: 根据审核结果选择内容")
print("-" * 60) print("-" * 60)
process_result_directory(source, output, run_id) process_result_directory(source, output, run_id, prefer_original)
print("\n脚本执行完毕.") print("\n脚本执行完毕.")