import os
import re

vault_path = "vault"
report_file = os.path.join(vault_path, "linking_audit_report.md")

def apply_recommendations():
    if not os.path.exists(report_file):
        print(f"Report file {report_file} not found. Run audit_links.py first.")
        return
        
    with open(report_file, "r", encoding="utf-8") as f:
        report_text = f.read()
        
    sec3_match = re.search(r"## 3\. Khuyến nghị Bổ sung Liên kết Nhân quả \(Linking Recommendations\)\n(.*)", report_text, re.DOTALL)
    if not sec3_match:
        print("Could not find Section 3 in the report.")
        return
        
    rec_lines = sec3_match.group(1).strip().split("\n")
    applied_count = 0
    
    for line in rec_lines:
        if not line.strip().startswith("-"):
            continue
            
        match = re.search(r"-\s+\*\*`?\[\[([^\]]+)\]\]`?\*\*\s+(?:trong phần `?([^`]+)`?\s+)?.*?[kK]huyến nghị.*?\s+`?\[\[([^\]]+)\]\]`?", line)
        if match:
            source_file = match.group(1).strip()
            section_group = match.group(2).strip() if match.group(2) else None
            target_link = match.group(3).strip()
            
            file_path = os.path.join(vault_path, f"{source_file}.md")
            if not os.path.exists(file_path):
                print(f"Source file {file_path} does not exist. Skipping.")
                continue
                
            with open(file_path, "r", encoding="utf-8") as sf:
                content = sf.read()
                
            target_name = target_link.split("|")[0].strip()
            
            # Check if link already exists in file
            if re.search(r"\[\[" + re.escape(target_name) + r"(?:\|[^\]]+)?\]\]", content, re.IGNORECASE):
                print(f"Link to [[{target_name}]] already exists in {source_file}. Skipping.")
                continue
                
            if source_file == "open_research_questions" and section_group:
                # Handle special placement for open_research_questions Nhóm sections
                # We want to find the header for Nhóm X and append the link inside that section
                # Sections are separated by "---"
                # Let us locate the header: "## ... section_group ..."
                header_regex = r"(## [^\n]*" + re.escape(section_group) + r"[^\n]*)"
                header_match = re.search(header_regex, content)
                if header_match:
                    header_text = header_match.group(1)
                    pos = content.find(header_text)
                    # find the next divider "---" or "## " to locate end of section
                    next_div = content.find("\n---", pos)
                    if next_div != -1:
                        # insert right before the "---" divider
                        new_content = content[:next_div].rstrip() + f"\n*   **Tài liệu liên đới**: [[{target_link}]]\n" + content[next_div:]
                    else:
                        new_content = content.rstrip() + f"\n*   **Tài liệu liên đới**: [[{target_link}]]\n"
                    
                    with open(file_path, "w", encoding="utf-8") as sf:
                        sf.write(new_content)
                    print(f"Successfully added [[{target_link}]] to open_research_questions.md under {section_group}")
                    applied_count += 1
                else:
                    print(f"Could not find section header for {section_group} in open_research_questions.md. Skipping.")
            else:
                # General file links section append
                links_section_header = "## 🔗 Liên kết Tài liệu liên quan (Obsidian Links)"
                alternative_header = "## 🔗 Liên kết Tài liệu liên quan"
                
                if links_section_header in content:
                    pos = content.find(links_section_header)
                    next_section = content.find("\n## ", pos + len(links_section_header))
                    if next_section != -1:
                        new_content = content[:next_section].rstrip() + f"\n*   [[{target_link}]]\n\n" + content[next_section:]
                    else:
                        new_content = content.rstrip() + f"\n*   [[{target_link}]]\n"
                elif alternative_header in content:
                    pos = content.find(alternative_header)
                    next_section = content.find("\n## ", pos + len(alternative_header))
                    if next_section != -1:
                        new_content = content[:next_section].rstrip() + f"\n*   [[{target_link}]]\n\n" + content[next_section:]
                    else:
                        new_content = content.rstrip() + f"\n*   [[{target_link}]]\n"
                else:
                    new_content = content.rstrip() + f"\n\n---\n\n{links_section_header}\n\n*   [[{target_link}]]\n"
                    
                with open(file_path, "w", encoding="utf-8") as sf:
                    sf.write(new_content)
                    
                print(f"Successfully added [[{target_link}]] to {source_file}.md")
                applied_count += 1
            
    print(f"Applied {applied_count} linking recommendations.")

if __name__ == "__main__":
    apply_recommendations()