从Git仓库移除.md文件但保留本地副本

source/extract_group_info.py

@@ -0,0 +1,101 @@
+import argparse
+import re
+
+def parse_args() -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description="提取UP分组信息")
+    parser.add_argument(
+        "--input",
+        default="source/19_53_no_titles.md",
+        help="输入报告路径",
+    )
+    parser.add_argument(
+        "--output",
+        help="输出报告路径（默认覆盖输入）",
+    )
+    return parser.parse_args()
+
+def main():
+    args = parse_args()
+    input_file = args.input
+    output_file = args.output or input_file
+
+    with open(input_file, 'r', encoding='utf-8') as f:
+        content = f.read()
+
+    lines = content.split('\n')
+    section_starts = []
+    for i, line in enumerate(lines):
+        if line.startswith('## '):
+            section_starts.append(i)
+
+    if len(section_starts) < 2:
+        print('No sections found')
+        return 1
+
+    header = '\n'.join(lines[:section_starts[0]])
+    sections = []
+
+    for idx in range(len(section_starts)):
+        start = section_starts[idx]
+        end = section_starts[idx + 1] if idx + 1 < len(section_starts) else len(lines)
+        section = '\n'.join(lines[start:end])
+        sections.append(section)
+
+    sections = sections[1:]
+
+    parsed = []
+    for sec in sections:
+        match = re.match(r'^## (\d+)\. (.+) \(mid: (\d+)\)', sec)
+        if match:
+            num = int(match.group(1))
+            name = match.group(2)
+            mid = match.group(3)
+            
+            group_m = re.search(r'- 预设分组: (.+)', sec)
+            action_m = re.search(r'- 建议动作: (.+)', sec)
+            reason_m = re.search(r'- 判断依据: (.+)', sec)
+            error_m = re.search(r'AI返回未知group: (.+)', sec)
+            
+            group = group_m.group(1).strip() if group_m else ""
+            action = action_m.group(1).strip() if action_m else ""
+            reason = reason_m.group(1).strip() if reason_m else ""
+            error = error_m.group(1).strip() if error_m else ""
+            
+            parsed.append({
+                'num': num,
+                'name': name,
+                'mid': mid,
+                'group': group,
+                'action': action,
+                'reason': reason,
+                'error': error
+            })
+
+    parsed.sort(key=lambda x: (x['name'].casefold(), int(x['mid'])))
+
+    lines_out = [header, ""]
+
+    for p in parsed:
+        lines_out.append(f"## {p['num']}. {p['name']} (mid: {p['mid']})")
+        lines_out.append("")
+        if p['group']:
+            lines_out.append(f"- 预设分组: {p['group']}")
+        if p['action']:
+            lines_out.append(f"- 建议动作: {p['action']}")
+        if p['reason']:
+            lines_out.append(f"- 判断依据: {p['reason']}")
+        if p['error']:
+            lines_out.append(f"- 异常: {p['error']}")
+        lines_out.append("")
+
+    result = '\n'.join(lines_out)
+    result = re.sub(r'\n{3,}', '\n\n', result)
+
+    with open(output_file, 'w', encoding='utf-8') as f:
+        f.write(result)
+
+    print(f'Extracted {len(parsed)} sections')
+    return 0
+
+if __name__ == "__main__":
+    raise SystemExit(main())

source/remove_10content.py

@@ -1,51 +1,67 @@
+import argparse
 import re
-with open(r'D:\Code\Python\followlist\source\output\reports\up_analysis_full_auto.md', 'r', encoding='utf-8') as f:
-    content = f.read()
-# Pattern: Remove any bullet lines between '## N. xxx' section header and the next '### ' section
-# Match: '## N.' followed by any content (not starting with '- '), then multiple ' - ' lines, then stop at '### '
-# Use a pattern that captures from '## ' to the next '### '
-# More specifically: match the block from '## N.' until the next '### ' that is NOT '### 最近10条标题'
-# and remove any lines starting with ' - ' in between
-lines = content.split('\n')
-new_lines = []
-i = 0
-while i < len(lines):
-    line = lines[i]
-    new_lines.append(line)
 
-    # If we just added a section header '## N.'
+def parse_args() -> argparse.Namespace:
-    if line.startswith('## '):
+    parser = argparse.ArgumentParser(description="删除最近10条标题内容")
-        i += 1
+    parser.add_argument(
-        # Skip lines until we hit either '### 最近10条标题' or another section marker
+        "--input",
-        while i < len(lines):
+        default="source/output/reports/up_analysis_full_auto.md",
-            curr = lines[i]
+        help="输入报告路径",
-            if curr.startswith('## '):
+    )
-                break
+    parser.add_argument(
-            if curr.startswith('### '):
+        "--output",
-                if '最近10条标题' in curr:
+        help="输出报告路径（默认覆盖输入）",
-                    # Skip this title section and its bullets
+    )
-                    i += 1  # skip '### 最近10条标题'
+    return parser.parse_args()
-                    while i < len(lines) and lines[i].startswith(' - '):
+
-                        i += 1
+def main():
-                    continue
+    args = parse_args()
-                else:
+    input_file = args.input
-                    # This is another section like ### AI分析 - keep it
+    output_file = args.output or input_file
-                    break
+
-            # Skip lines that are just basic info (主页, 标签, mid)
+    with open(input_file, 'r', encoding='utf-8') as f:
-            if curr.startswith('- ') and not curr.startswith(' - '):
+        content = f.read()
-                i += 1
+
-                continue
+    lines = content.split('\n')
-            # Skip actual bullet lines (the orphaned ones)
+    new_lines = []
-            if curr.startswith(' - '):
+    i = 0
-                i += 1
+    while i < len(lines):
-                continue
+        line = lines[i]
-            new_lines.append(curr)
+        new_lines.append(line)
+        
+        if line.startswith('## '):
             i += 1
-    else:
+            while i < len(lines):
-        i += 1
+                curr = lines[i]
-result = '\n'.join(new_lines)
+                if curr.startswith('## '):
-# Clean multiple blank lines
+                    break
-result = re.sub(r'\n{3,}', '\n\n', result)
+                if curr.startswith('### '):
-with open(r'D:\Code\Python\followlist\source\output\reports\up_analysis_full_auto.md', 'w', encoding='utf-8') as f:
+                    if '最近10条标题' in curr:
-    f.write(result)
+                        i += 1
-print('Done')
+                        while i < len(lines) and lines[i].startswith(' - '):
+                            i += 1
+                        continue
+                    else:
+                        break
+                if curr.startswith('- ') and not curr.startswith(' - '):
+                    i += 1
+                    continue
+                if curr.startswith(' - '):
+                    i += 1
+                    continue
+                new_lines.append(curr)
+                i += 1
+        else:
+            i += 1
+
+    result = '\n'.join(new_lines)
+    result = re.sub(r'\n{3,}', '\n\n', result)
+
+    with open(output_file, 'w', encoding='utf-8') as f:
+        f.write(result)
+
+    print(f'Done: {output_file}')
+    return 0
+
+if __name__ == "__main__":
+    raise SystemExit(main())