jarvis-ai-assistant 0.1.132__py3-none-any.whl → 0.1.138__py3-none-any.whl

jarvis/jarvis_tools/read_code.py

@@ -1,8 +1,10 @@
 from typing import Dict, Any
 import os
 
+from pkg_resources import add_activation_listener
 from yaspin import yaspin
 
+from jarvis.jarvis_utils.globals import add_read_file_record
 from jarvis.jarvis_utils.output import OutputType, PrettyOutput
 
 class ReadCodeTool:
@@ -30,17 +32,18 @@ class ReadCodeTool:
 
     def _handle_single_file(self, filepath: str, start_line: int = 1, end_line: int = -1) -> Dict[str, Any]:
         """处理单个文件的读取操作
-        
+
         Args:
             filepath (str): 文件路径
             start_line (int): 起始行号，默认为1
             end_line (int): 结束行号，默认为-1表示文件末尾
-            
+
         Returns:
             Dict[str, Any]: 包含成功状态、输出内容和错误信息的字典
         """
         try:
             abs_path = os.path.abspath(filepath)
+            add_read_file_record(abs_path)
             with yaspin(text=f"正在读取文件: {abs_path}...", color="cyan") as spinner:
                 # 文件存在性检查
                 if not os.path.exists(abs_path):
@@ -49,7 +52,7 @@ class ReadCodeTool:
                         "stdout": "",
                         "stderr": f"文件不存在: {abs_path}"
                     }
-                
+
                 # 文件大小限制检查（10MB）
                 if os.path.getsize(abs_path) > 10 * 1024 * 1024:
                     return {
@@ -57,21 +60,21 @@ class ReadCodeTool:
                         "stdout": "",
                         "stderr": "文件过大 (>10MB)"
                     }
-                
+
                 # 读取文件内容
                 with open(abs_path, 'r', encoding='utf-8', errors="ignore") as f:
                     lines = f.readlines()
-                
+
                 total_lines = len(lines)
-                
+
                 # 处理特殊值-1表示文件末尾
                 if end_line == -1:
                     end_line = total_lines
                 else:
                     end_line = max(1, min(end_line, total_lines)) if end_line >= 0 else total_lines + end_line + 1
-                
+
                 start_line = max(1, min(start_line, total_lines)) if start_line >= 0 else total_lines + start_line + 1
-                
+
                 if start_line > end_line:
                     spinner.fail("❌")
                     return {
@@ -79,20 +82,19 @@ class ReadCodeTool:
                         "stdout": "",
                         "stderr": f"无效的行范围 [{start_line}-{end_line}] (总行数: {total_lines})"
                     }
-                
+
                 # 添加行号并构建输出内容
                 selected_lines = lines[start_line-1:end_line]
                 numbered_content = "".join(
-                    [f"{i:4d}:{line}" 
+                    [f"{i:4d}:{line}"
                      for i, line in enumerate(selected_lines, start=start_line)]
                 )
-                
+
                 # 构建输出格式
                 output = (
                     f"\n🔍 文件: {abs_path}\n"
-                    f"📄 原始行号: {start_line}-{end_line} (共{end_line - start_line + 1}行) | 显示行号: 1-{len(selected_lines)}\n\n"
-                    f"{numbered_content}\n"
-                    f"{'='*80}\n"
+                    f"📄 原始行号: {start_line}-{end_line} (共{total_lines}行) \n\n"
+                    f"{numbered_content}\n\n"
                 )
                 spinner.text = f"文件读取完成: {abs_path}"
                 spinner.ok("✅")
@@ -104,7 +106,7 @@ class ReadCodeTool:
                     "stdout": output,
                     "stderr": ""
                 }
-                
+
         except Exception as e:
             PrettyOutput.print(str(e), OutputType.ERROR)
             return {
@@ -115,10 +117,10 @@ class ReadCodeTool:
 
     def execute(self, args: Dict) -> Dict[str, Any]:
         """执行代码读取操作
-        
+
         Args:
             args (Dict): 包含文件列表的参数字典
-            
+
         Returns:
             Dict[str, Any]: 包含成功状态、输出内容和错误信息的字典
         """
@@ -129,32 +131,32 @@ class ReadCodeTool:
                     "stdout": "",
                     "stderr": "参数中必须包含文件列表"
                 }
-            
+
             all_outputs = []
             overall_success = True
-            
+
             for file_info in args["files"]:
                 if not isinstance(file_info, dict) or "path" not in file_info:
                     continue
-                
+
                 result = self._handle_single_file(
                     file_info["path"].strip(),
                     file_info.get("start_line", 1),
                     file_info.get("end_line", -1)
                 )
-                
+
                 if result["success"]:
                     all_outputs.append(result["stdout"])
                 else:
                     all_outputs.append(f"❌ {file_info['path']}: {result['stderr']}")
                     overall_success = False
-                
+
             return {
                 "success": overall_success,
                 "stdout": "\n".join(all_outputs),
                 "stderr": ""
             }
-            
+
         except Exception as e:
             PrettyOutput.print(str(e), OutputType.ERROR)
             return {

jarvis/jarvis_tools/read_webpage.py

@@ -24,38 +24,38 @@ class WebpageTool:
         """Read webpage content using Playwright to handle JavaScript-rendered pages"""
         try:
             url = args["url"].strip()
-            
+
             with sync_playwright() as p:
                 # Launch browser
                 browser = p.chromium.launch(
                     headless=True,
                     args=['--disable-gpu', '--no-sandbox', '--disable-dev-shm-usage']
                 )
-                
+
                 # Create a new page with appropriate settings
                 page = browser.new_page(
                     user_agent='Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/121.0.0.0 Safari/537.36',
                     viewport={'width': 1920, 'height': 1080}
                 )
-                
+
                 # Set timeout to avoid long waits
                 page.set_default_timeout(30000)  # 30 seconds
-                
+
                 try:
                     # Navigate to URL and wait for page to load
                     response = page.goto(url, wait_until="domcontentloaded")
-                    
+
                     # Additional wait for network to be idle (with a timeout)
                     try:
                         page.wait_for_load_state("networkidle", timeout=10000)
                     except PlaywrightTimeoutError:
                         # Continue even if network doesn't become completely idle
                         pass
-                        
+
                     # Make sure we got a valid response
                     if not response or response.status >= 400:
                         raise Exception(f"Failed to load page: HTTP {response.status if response else 'No response'}")
-                    
+
                     # Get page title safely
                     title = "No title"
                     try:
@@ -68,26 +68,26 @@ class WebpageTool:
                                 title = title_element.text_content() or "No title"
                         except Exception:
                             pass
-                    
+
                     # Get the HTML content after JavaScript execution
                     html_content = page.content()
-                    
+
                 except Exception as e:
                     raise Exception(f"Error navigating to page: {str(e)}")
                 finally:
                     # Always close browser
                     browser.close()
-                
+
                 # Parse with BeautifulSoup and convert to markdown
                 markdown_content = self._html_to_markdown(html_content, url)
-                
+
                 # Build output in markdown format
                 output = [
                     f"# {title}",
                     f"Url: {url}",
                     markdown_content
                 ]
-                
+
                 return {
                     "success": True,
                     "stdout": "\n".join(output),
@@ -101,7 +101,7 @@ class WebpageTool:
                 "stdout": "",
                 "stderr": f"Failed to parse webpage: {str(e)}"
             }
-    
+
     def _create_soup_element(self, content):
         """Safely create a BeautifulSoup element, ensuring it's treated as markup"""
         if isinstance(content, str):
@@ -112,15 +112,15 @@ class WebpageTool:
             # Return an empty list if the div is None
             return []
         return content
-    
+
     def _html_to_markdown(self, html_content: str, base_url: str) -> str:
         """Convert HTML to Markdown format preserving the content structure"""
         soup = BeautifulSoup(html_content, 'html.parser')
-        
+
         # Remove unwanted elements
         for element in soup(['script', 'style', 'meta', 'noscript', 'head']):
             element.decompose()
-        
+
         # Process headings
         for level in range(1, 7):
             for heading in soup.find_all(f'h{level}'):
@@ -128,14 +128,14 @@ class WebpageTool:
                 heading_md = "\n\n" + "#" * level + " " + text + "\n\n"
                 new_element = self._create_soup_element(heading_md)
                 heading.replace_with(*new_element)
-        
+
         # Process paragraphs
         for p in soup.find_all('p'):
             text = p.get_text().strip()
             if text:
                 new_element = self._create_soup_element("\n\n" + text + "\n\n")
                 p.replace_with(*new_element)
-        
+
         # Process unordered lists
         for ul in soup.find_all('ul'):
             items = []
@@ -143,7 +143,7 @@ class WebpageTool:
                 items.append("* " + li.get_text().strip())
             new_element = self._create_soup_element("\n\n" + "\n".join(items) + "\n\n")
             ul.replace_with(*new_element)
-        
+
         # Process ordered lists
         for ol in soup.find_all('ol'):
             items = []
@@ -151,7 +151,7 @@ class WebpageTool:
                 items.append(str(i) + ". " + li.get_text().strip())
             new_element = self._create_soup_element("\n\n" + "\n".join(items) + "\n\n")
             ol.replace_with(*new_element)
-        
+
         # Process links (first pass)
         for a in soup.find_all('a', href=True):
             try:
@@ -166,7 +166,7 @@ class WebpageTool:
                     a.replace_with(*new_element)
             except (KeyError, AttributeError):
                 continue
-        
+
         # Process images
         for img in soup.find_all('img', src=True):
             try:
@@ -180,37 +180,37 @@ class WebpageTool:
                 img.replace_with(*new_element)
             except (KeyError, AttributeError, UnboundLocalError):
                 continue
-        
+
         # Process code blocks
         for pre in soup.find_all('pre'):
             code = pre.get_text().strip()
             pre_md = "\n\n```\n" + code + "\n```\n\n"
             new_element = self._create_soup_element(pre_md)
             pre.replace_with(*new_element)
-        
+
         # Process inline code
         for code in soup.find_all('code'):
             text = code.get_text().strip()
             code_md = "`" + text + "`"
             new_element = self._create_soup_element(code_md)
             code.replace_with(*new_element)
-        
+
         # Process line breaks
         for br in soup.find_all('br'):
             new_element = self._create_soup_element('\n')
             br.replace_with(*new_element)
-        
+
         # Get the full text
         markdown_text = soup.get_text()
-        
+
         # Clean up extra whitespace and line breaks
         markdown_text = re.sub(r'\n{3,}', '\n\n', markdown_text)
         markdown_text = re.sub(r'\s{2,}', ' ', markdown_text)
-        
+
         # Process links again (for any that might have been missed)
         link_pattern = r'\[([^\]]+)\]\(([^)]+)\)'
         all_links = re.findall(link_pattern, markdown_text)
-        
+
         # Add a section with all links at the end
         if all_links:
             link_section = ["", "## Links", ""]
@@ -220,7 +220,7 @@ class WebpageTool:
                 if link_entry not in seen_links:
                     link_section.append(link_entry)
                     seen_links.add(link_entry)
-            
+
             markdown_text += "\n\n" + "\n".join(link_section)
-        
-        return markdown_text.strip() 
+
+        return markdown_text.strip()