kopipasta 0.13.0__py3-none-any.whl → 0.15.0__py3-none-any.whl

kopipasta/main.py

@@ -4,14 +4,15 @@ import io
 import json
 import os
 import argparse
-import ast
 import re
-from textwrap import dedent
+from typing import Dict, List, Optional, Tuple
 import pyperclip
 import fnmatch
 
 import requests
 
+FileTuple = Tuple[str, bool, Optional[List[str]], str]
+
 def read_gitignore():
     default_ignore_patterns = [
         '.git', 'node_modules', 'venv', '.venv', 'dist', '.idea', '__pycache__',
@@ -425,7 +426,7 @@ def print_char_count(count):
     token_estimate = count // 4
     print(f"\rCurrent prompt size: {count} characters (~ {token_estimate} tokens)", flush=True)
 
-def select_files_in_directory(directory, ignore_patterns, current_char_count=0):
+def select_files_in_directory(directory: str, ignore_patterns: List[str], current_char_count: int = 0) -> Tuple[List[FileTuple], int]:
     files = [f for f in os.listdir(directory)
              if os.path.isfile(os.path.join(directory, f)) and not is_ignored(os.path.join(directory, f), ignore_patterns) and not is_binary(os.path.join(directory, f))]
 
@@ -445,8 +446,8 @@ def select_files_in_directory(directory, ignore_patterns, current_char_count=0):
     while True:
         print_char_count(current_char_count)
         choice = input("(y)es add all / (n)o ignore all / (s)elect individually / (q)uit? ").lower()
+        selected_files: List[FileTuple] = []
         if choice == 'y':
-            selected_files = []
             for file in files:
                 file_path = os.path.join(directory, file)
                 if is_large_file(file_path):
@@ -456,13 +457,13 @@ def select_files_in_directory(directory, ignore_patterns, current_char_count=0):
                             break
                         print("Invalid choice. Please enter 'f' or 's'.")
                     if snippet_choice == 's':
-                        selected_files.append((file, True))
+                        selected_files.append((file, True, None, get_language_for_file(file)))
                         current_char_count += len(get_file_snippet(file_path))
                     else:
-                        selected_files.append((file, False))
+                        selected_files.append((file, False, None, get_language_for_file(file)))
                         current_char_count += os.path.getsize(file_path)
                 else:
-                    selected_files.append((file, False))
+                    selected_files.append((file, False, None, get_language_for_file(file)))
                     current_char_count += os.path.getsize(file_path)
             print(f"Added all files from {directory}")
             return selected_files, current_char_count
@@ -470,7 +471,6 @@ def select_files_in_directory(directory, ignore_patterns, current_char_count=0):
             print(f"Ignored all files from {directory}")
             return [], current_char_count
         elif choice == 's':
-            selected_files = []
             for file in files:
                 file_path = os.path.join(directory, file)
                 file_size = os.path.getsize(file_path)
@@ -489,13 +489,13 @@ def select_files_in_directory(directory, ignore_patterns, current_char_count=0):
                                     break
                                 print("Invalid choice. Please enter 'f' or 's'.")
                             if snippet_choice == 's':
-                                selected_files.append((file, True))
+                                selected_files.append((file, True, None, get_language_for_file(file_path)))
                                 current_char_count += len(get_file_snippet(file_path))
                             else:
-                                selected_files.append((file, False))
+                                selected_files.append((file, False, None, get_language_for_file(file_path)))
                                 current_char_count += file_char_estimate
                         else:
-                            selected_files.append((file, False))
+                            selected_files.append((file, False, None, get_language_for_file(file_path)))
                             current_char_count += file_char_estimate
                         break
                     elif file_choice == 'n':
@@ -503,7 +503,7 @@ def select_files_in_directory(directory, ignore_patterns, current_char_count=0):
                     elif file_choice == 'p':
                         chunks, char_count = select_file_patches(file_path)
                         if chunks:
-                            selected_files.append((file_path, False, chunks))
+                            selected_files.append((file_path, False, chunks, get_language_for_file(file_path)))
                             current_char_count += char_count
                         break
                     elif file_choice == 'q':
@@ -520,7 +520,7 @@ def select_files_in_directory(directory, ignore_patterns, current_char_count=0):
             print("Invalid choice. Please try again.")
 
 def process_directory(directory, ignore_patterns, current_char_count=0):
-    files_to_include = []
+    files_to_include: List[FileTuple] = []
     processed_dirs = set()
 
     for root, dirs, files in os.walk(directory):
@@ -530,15 +530,26 @@ def process_directory(directory, ignore_patterns, current_char_count=0):
         if root in processed_dirs:
             continue
 
-        selected_files, current_char_count = select_files_in_directory(root, ignore_patterns, current_char_count)
-        for file_tuple in selected_files:
-            if len(file_tuple) == 3:
-                f, use_snippet, chunks = file_tuple
-                files_to_include.append((os.path.join(root, f), use_snippet, chunks))
-            else:
-                f, use_snippet = file_tuple
-                files_to_include.append((os.path.join(root, f), use_snippet))
-        processed_dirs.add(root)
+        print(f"\nExploring directory: {root}")
+        choice = input("(y)es explore / (n)o skip / (q)uit? ").lower()
+        if choice == 'y':
+            selected_files, current_char_count = select_files_in_directory(root, ignore_patterns, current_char_count)
+            for file_tuple in selected_files:
+                if len(file_tuple) == 3:
+                    f, use_snippet, chunks = file_tuple
+                    files_to_include.append((os.path.join(root, f), use_snippet, chunks))
+                else:
+                    f, use_snippet = file_tuple
+                    files_to_include.append((os.path.join(root, f), use_snippet))
+            processed_dirs.add(root)
+        elif choice == 'n':
+            dirs[:] = []  # Skip all subdirectories
+            continue
+        elif choice == 'q':
+            break
+        else:
+            print("Invalid choice. Skipping this directory.")
+            continue
 
     return files_to_include, processed_dirs, current_char_count
 
@@ -659,30 +670,29 @@ def handle_env_variables(content, env_vars):
 
     return content
 
-def generate_prompt(files_to_include, ignore_patterns, web_contents, env_vars):
+def generate_prompt(files_to_include: List[FileTuple], ignore_patterns: List[str], web_contents: Dict[str, Tuple[str, str]], env_vars: Dict[str, str]) -> str:
     prompt = "# Project Overview\n\n"
     prompt += "## Project Structure\n\n"
     prompt += "```\n"
     prompt += get_project_structure(ignore_patterns)
     prompt += "\n```\n\n"
     prompt += "## File Contents\n\n"
-    for file_tuple in files_to_include:
-        if len(file_tuple) == 4:
-            file, content, is_snippet, content_type = file_tuple
-            chunks = None
-        else:
-            file, content, is_snippet, content_type, chunks = file_tuple
+    for file, use_snippet, chunks, content_type in files_to_include:
         relative_path = get_relative_path(file)
-        language = get_language_for_file(file) if content_type == 'text' else content_type
+        language = content_type if content_type else get_language_for_file(file)
 
         if chunks is not None:
             prompt += f"### {relative_path} (selected patches)\n\n```{language}\n"
             for chunk in chunks:
                 prompt += f"{chunk}\n"
             prompt += "```\n\n"
+        elif use_snippet:
+            file_content = get_file_snippet(file)
+            prompt += f"### {relative_path} (snippet)\n\n```{language}\n{file_content}\n```\n\n"
         else:
-            content = handle_env_variables(content, env_vars)
-            prompt += f"### {relative_path}{' (snippet)' if is_snippet else ''}\n\n```{language}\n{content}\n```\n\n"
+            file_content = read_file_contents(file)
+            file_content = handle_env_variables(file_content, env_vars)
+            prompt += f"### {relative_path}\n\n```{language}\n{file_content}\n```\n\n"
     
     if web_contents:
         prompt += "## Web Content\n\n"
@@ -711,79 +721,38 @@ def main():
     ignore_patterns = read_gitignore()
     env_vars = read_env_file()
 
-    files_to_include = []
+    files_to_include:List[FileTuple] = []
+    processed_dirs = set()
     web_contents = {}
-
-    def process_directory(directory):
-        files = [f for f in os.listdir(directory)
-                 if os.path.isfile(os.path.join(directory, f)) and not is_ignored(os.path.join(directory, f), ignore_patterns) and not is_binary(os.path.join(directory, f))]
-
-        if not files:
-            return []
-
-        print(f"\nDirectory: {directory}")
-        print("Files:")
-        for file in files:
-            file_path = os.path.join(directory, file)
-            file_size = os.path.getsize(file_path)
-            file_size_readable = get_human_readable_size(file_size)
-            print(f"- {file} ({file_size_readable})")
-
-        while True:
-            choice = input("(y)es add all / (n)o ignore all / (s)elect individually / (q)uit? ").lower()
-            if choice == 'y':
-                return [(os.path.join(directory, f), False) for f in files]
-            elif choice == 'n':
-                return []
-            elif choice == 's':
-                selected_files = []
-                for file in files:
-                    file_path = os.path.join(directory, file)
-                    while True:
-                        file_choice = input(f"{file} (y/n/p/q)? ").lower()
-                        if file_choice == 'y':
-                            selected_files.append((file_path, False))
-                            break
-                        elif file_choice == 'n':
-                            break
-                        elif file_choice == 'p':
-                            chunks, _ = select_file_patches(file_path)
-                            if chunks:
-                                selected_files.append((file_path, True, chunks))
-                            break
-                        elif file_choice == 'q':
-                            return selected_files
-                        else:
-                            print("Invalid choice. Please enter 'y', 'n', 'p', or 'q'.")
-                return selected_files
-            elif choice == 'q':
-                return []
-            else:
-                print("Invalid choice. Please try again.")
+    current_char_count = 0
 
     for input_path in args.inputs:
         if input_path.startswith(('http://', 'https://')):
-            content, content_type = fetch_web_content(input_path)
-            if content:
-                content, is_snippet = handle_content(content, content_type, input_path)
-                web_contents[input_path] = (content, is_snippet, content_type)
+            full_content, snippet = fetch_web_content(input_path)
+            if full_content:
+                web_contents[input_path] = (full_content, snippet)
+                current_char_count += len(snippet if len(full_content) > 10000 else full_content)
                 print(f"Added web content from: {input_path}")
         elif os.path.isfile(input_path):
             if not is_ignored(input_path, ignore_patterns) and not is_binary(input_path):
                 while True:
                     file_choice = input(f"{input_path} (y)es include / (n)o skip / (p)atches / (q)uit? ").lower()
                     if file_choice == 'y':
-                        content, content_type = read_file_content(input_path)
-                        content, is_snippet = handle_content(content, content_type, input_path)
-                        files_to_include.append((input_path, content, is_snippet, content_type))
-                        print(f"Added file: {input_path}{' (snippet)' if is_snippet else ''}")
+                        use_snippet = is_large_file(input_path)
+                        files_to_include.append((input_path, use_snippet, None, get_language_for_file(input_path)))
+                        if use_snippet:
+                            current_char_count += len(get_file_snippet(input_path))
+                        else:
+                            current_char_count += os.path.getsize(input_path)
+                        print(f"Added file: {input_path}{' (snippet)' if use_snippet else ''}")
                         break
                     elif file_choice == 'n':
                         break
                     elif file_choice == 'p':
-                        chunks, _ = select_file_patches(input_path)
+                        chunks, char_count = select_file_patches(input_path)
                         if chunks:
-                            files_to_include.append((input_path, None, False, 'text', chunks))
+                            files_to_include.append((input_path, False, chunks, get_language_for_file(input_path)))
+                            current_char_count += char_count
                         break
                     elif file_choice == 'q':
                         print("Quitting.")
@@ -793,16 +762,9 @@ def main():
             else:
                 print(f"Ignored file: {input_path}")
         elif os.path.isdir(input_path):
-            selected_files = process_directory(input_path)
-            for file_info in selected_files:
-                if len(file_info) == 2:
-                    file_path, use_snippet = file_info
-                    content, content_type = read_file_content(file_path)
-                    content, is_snippet = handle_content(content, content_type, file_path)
-                    files_to_include.append((file_path, content, is_snippet, content_type))
-                else:
-                    file_path, _, chunks = file_info
-                    files_to_include.append((file_path, None, False, 'text', chunks))
+            dir_files, dir_processed, current_char_count = process_directory(input_path, ignore_patterns, current_char_count)
+            files_to_include.extend(dir_files)
+            processed_dirs.update(dir_processed)
         else:
             print(f"Warning: {input_path} is not a valid file, directory, or URL. Skipping.")
 
@@ -811,7 +773,8 @@ def main():
         return
 
     print("\nFile and web content selection complete.")
-    print(f"Summary: Added {len(files_to_include)} files and {len(web_contents)} web sources.")
+    print_char_count(current_char_count)
+    print(f"Summary: Added {len(files_to_include)} files from {len(processed_dirs)} directories and {len(web_contents)} web sources.")
 
     prompt = generate_prompt(files_to_include, ignore_patterns, web_contents, env_vars)
     print("\n\nGenerated prompt:")

{kopipasta-0.13.0.dist-info → kopipasta-0.15.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: kopipasta
-Version: 0.13.0
+Version: 0.15.0
 Summary: A CLI tool to generate prompts with project structure and file contents
 Home-page: https://github.com/mkorpela/kopipasta
 Author: Mikko Korpela

kopipasta-0.15.0.dist-info/RECORD

@@ -0,0 +1,8 @@
+kopipasta/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+kopipasta/main.py,sha256=51k6jZXChwZY_CYVnqCqeY_hoHCAj3flN8qnr0vEVQI,31652
+kopipasta-0.15.0.dist-info/LICENSE,sha256=xw4C9TAU7LFu4r_MwSbky90uzkzNtRwAo3c51IWR8lk,1091
+kopipasta-0.15.0.dist-info/METADATA,sha256=ehTVHvX8IWh7hISa0O3bxJqj4a7gBqX9H0JtU1xU2ys,5646
+kopipasta-0.15.0.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
+kopipasta-0.15.0.dist-info/entry_points.txt,sha256=but54qDNz1-F8fVvGstq_QID5tHjczP7bO7rWLFkc6Y,50
+kopipasta-0.15.0.dist-info/top_level.txt,sha256=iXohixMuCdw8UjGDUp0ouICLYBDrx207sgZIJ9lxn0o,10
+kopipasta-0.15.0.dist-info/RECORD,,

kopipasta-0.13.0.dist-info/RECORD

@@ -1,8 +0,0 @@
-kopipasta/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-kopipasta/main.py,sha256=Z5_iI0dDp8WeRLp_FMDBPzNWrUiHNIteOcPl_fh7e4k,32860
-kopipasta-0.13.0.dist-info/LICENSE,sha256=xw4C9TAU7LFu4r_MwSbky90uzkzNtRwAo3c51IWR8lk,1091
-kopipasta-0.13.0.dist-info/METADATA,sha256=80uGAXCQckifJnEkXCsgu5uNkFRFO7CWuwSi9wAVbDE,5646
-kopipasta-0.13.0.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
-kopipasta-0.13.0.dist-info/entry_points.txt,sha256=but54qDNz1-F8fVvGstq_QID5tHjczP7bO7rWLFkc6Y,50
-kopipasta-0.13.0.dist-info/top_level.txt,sha256=iXohixMuCdw8UjGDUp0ouICLYBDrx207sgZIJ9lxn0o,10
-kopipasta-0.13.0.dist-info/RECORD,,