Incorporated review feedback

Author: UmerHA

Acknowledgements.md

@@ -0,0 +1,5 @@
+# We thank the following people for inspiration
+
+| Person | Content | File(s) | Source |
+|----|---|---|---|
+| Paul Gauthier | The prompt for the `improve code` step is strongly based on Paul's prompt in Aider | /preprompts/improve.txt | https://github.com/paul-gauthier/aider/blob/main/aider/coders/editblock_coder.py

docs/intro/chat_parsing.md

@@ -8,7 +8,9 @@ Here are the functions defined in the `chat_to_files.py` module:
 
 `parse_chat(chat)`: This function takes the chat as an argument and returns a list of tuples. Each tuple contains a filename and the corresponding file content. The function uses regular expressions to extract the filenames and the file contents from the chat. The filenames are cleaned up to remove any non-allowed characters.
 
-`to_files(chat, dbs)`: This function takes the chat and the DBs as arguments. DBs contains the workspace and memory path. The function first saves the entire chat as a text file in the memory path. Then it calls the parse_chat function to parse the chat and get the files. Each file is then saved to the workspace.
+`to_files_and_memory(chat, dbs)`: This function takes the chat and the DBs as arguments. DBs contains the workspace and memory path. The function first saves the entire chat as a text file in the memory path. Then it calls the to_files function to  write each file to the workspace.
+
+`to_files(chat, db)`: This function takes the chat and workspace DB as arguments. It calls the parse_chat function to parse the chat and get the files. Each file is then saved to the workspace.
 
 <br>
 

docs/intro/quick_overview.md

@@ -15,7 +15,9 @@ This module contains two main functions:
 
 `parse_chat(chat)`: This function takes a chat conversation and extracts all the code blocks and preceding filenames. It returns a list of tuples, where each tuple contains a filename and the corresponding code block.
 
-`to_files(chat, dbs)`: This function takes the chat and the DBs as arguments. DBs contains the workspace and memory path. The function first saves the entire chat as a text file in the memory path. Then it calls the parse_chat function to parse the chat and get the files. Each file is then saved to the workspace.
+`to_files_and_memory(chat, dbs)`: This function takes the chat and the DBs as arguments. DBs contains the workspace and memory path. The function first saves the entire chat as a text file in the memory path. Then it calls the to_files function to  write each file to the workspace.
+
+`to_files(chat, db)`: This function takes the chat and workspace DB as arguments. It calls the parse_chat function to parse the chat and get the files. Each file is then saved to the workspace.
 
 <br>
 

gpt_engineer/chat_to_files.py

@@ -1,6 +1,6 @@
 import os
-from pathlib import Path
 import re
+import logging
 
 from dataclasses import dataclass
 from typing import List, Tuple
@@ -9,6 +9,9 @@
 from gpt_engineer.file_selector import FILE_LIST_NAME
 
 
+logger = logging.getLogger(__name__)
+
+
 def parse_chat(chat) -> List[Tuple[str, str]]:
     """
     Extracts all code blocks from a chat and returns them
@@ -56,7 +59,22 @@ def parse_chat(chat) -> List[Tuple[str, str]]:
     return files
 
 
-def to_files(chat: str, dbs: DBs):
+def to_files_and_memory(chat: str, dbs: DBs):
+    """
+    Save chat to memory, and parse chat to extracted file and save them to the workspace.
+
+    Parameters
+    ----------
+    chat : str
+        The chat to parse.
+    dbs : DBs
+        The databases that include the memory and workspace database
+    """
+    dbs.memory["all_output.txt"] = chat
+    to_files(chat, dbs.workspace)
+
+
+def to_files(chat: str, workspace: DB):
     """
     Parse the chat and add all extracted files to the workspace.
 
@@ -67,11 +85,9 @@ def to_files(chat: str, dbs: DBs):
     workspace : DB
         The database containing the workspace.
     """
-    dbs.memory["all_output.txt"] = chat
-
     files = parse_chat(chat)
     for file_name, file_content in files:
-        dbs.workspace[file_name] = file_content
+        workspace[file_name] = file_content
 
 
 def overwrite_files(chat: str, dbs: DBs) -> None:
@@ -218,10 +234,18 @@ def parse_all_edits(txt):
 
 def apply_edits(edits: List[Edit], workspace: DB):
     for edit in edits:
-        filename = edit.filename.replace("workspace/", "")
+        filename = edit.filename
         if edit.before == "":
+            if workspace.get(filename) is not None:
+                logger.warn(
+                    f"The edit to be applied wants to create a new file `{filename}`, but that already exists. The file will be overwritten. See `.gpteng/memory` for previous version."
+                )
             workspace[filename] = edit.after  # new file
         else:
+            if workspace[filename].count(edit.before) > 1:
+                logger.warn(
+                    f"While applying an edit to `{filename}`, the code block to be replaced was found multiple times. All instances will be replaced."
+                )
             workspace[filename] = workspace[filename].replace(
                 edit.before, edit.after
             )  # existing file

gpt_engineer/main.py

@@ -80,6 +80,8 @@ def main(
 
     if lite_mode:
         assert not improve_mode, "Lite mode cannot improve code"
+        if steps_config == StepsConfig.DEFAULT:
+            steps_config = StepsConfig.LITE
 
     if improve_mode:
         assert (
@@ -97,7 +99,7 @@ def main(
 
     project_path = os.path.abspath(
         project_path
-    )  # resolve the string to a valid path (eg a/b/../c to a/c)
+    )  # resolve the string to a valid path (eg "a/b/../c" to "a/c")
     path = Path(project_path).absolute()
     print("Running gpt-engineer in", path, "\n")
 

gpt_engineer/preprompts/ht.txt

@@ -25,7 +25,7 @@ Here is an example reponse:
 ---
 PLANNING:
 We need to change ... because ..., therefore I will add the line `a=a+1` to the function `add_one`.
-Also, in the class `DB`, we need to update the ... 
+Also, in the class `DB`, we need to update the ...
 
 OUTPUT:
 ```python

gpt_engineer/preprompts/improve

@@ -13,7 +13,7 @@
     format_file_to_input,
     get_code_strings,
     overwrite_files_with_edits,
-    to_files,
+    to_files_and_memory,
 )
 from gpt_engineer.db import DBs
 from gpt_engineer.file_selector import FILE_LIST_NAME, ask_for_files
@@ -62,14 +62,14 @@ def lite_gen(ai: AI, dbs: DBs) -> List[Message]:
     messages = ai.start(
         dbs.input["prompt"], dbs.preprompts["file_format"], step_name=curr_fn()
     )
-    to_files(messages[-1].content.strip(), dbs)
+    to_files_and_memory(messages[-1].content.strip(), dbs)
     return messages
 
 
 def simple_gen(ai: AI, dbs: DBs) -> List[Message]:
     """Run the AI on the default prompts and save the results"""
     messages = ai.start(setup_sys_prompt(dbs), dbs.input["prompt"], step_name=curr_fn())
-    to_files(messages[-1].content.strip(), dbs)
+    to_files_and_memory(messages[-1].content.strip(), dbs)
     return messages
 
 
@@ -130,7 +130,7 @@ def gen_clarified_code(ai: AI, dbs: DBs) -> List[dict]:
         step_name=curr_fn(),
     )
 
-    to_files(messages[-1].content.strip(), dbs)
+    to_files_and_memory(messages[-1].content.strip(), dbs)
     return messages
 
 
@@ -211,7 +211,7 @@ def use_feedback(ai: AI, dbs: DBs):
     ]
     if dbs.input["feedback"]:
         messages = ai.next(messages, dbs.input["feedback"], step_name=curr_fn())
-        to_files(messages[-1].content.strip(), dbs)
+        to_files_and_memory(messages[-1].content.strip(), dbs)
         return messages
     else:
         print(

gpt_engineer/steps.py

@@ -6,7 +6,7 @@
 import typer
 
 from gpt_engineer.ai import AI
-from gpt_engineer.chat_to_files import to_files
+from gpt_engineer.chat_to_files import to_files_and_memory
 
 app = typer.Typer()
 
@@ -30,7 +30,7 @@ def main(
 
     if out_path:
         # ToDo: to_files expects a DBs as 2nd argument. But do we need this entire file?
-        to_files(messages[-1]["content"], out_path)
+        to_files_and_memory(messages[-1]["content"], out_path)
         with open(pathlib.Path(out_path) / "all_output.txt", "w") as f:
             json.dump(messages[-1]["content"], f)
 

scripts/rerun_edited_message_logs.py

@@ -1,7 +1,7 @@
 from dataclasses import dataclass
 import textwrap
 
-from gpt_engineer.chat_to_files import to_files
+from gpt_engineer.chat_to_files import to_files_and_memory
 
 
 @dataclass
@@ -15,7 +15,7 @@ class DummyDBs:
     project_metadata = {}
 
 
-def test_to_files():
+def test_to_files_and_memory():
     chat = textwrap.dedent(
         """
     This is a sample program.
@@ -34,7 +34,7 @@ def add(a, b):
     )
 
     dbs = DummyDBs()
-    to_files(chat, dbs)
+    to_files_and_memory(chat, dbs)
 
     assert dbs.memory["all_output.txt"] == chat
 
@@ -66,7 +66,7 @@ def add(a, b):
     """
     )
     dbs = DummyDBs()
-    to_files(chat, dbs)
+    to_files_and_memory(chat, dbs)
 
     assert dbs.memory["all_output.txt"] == chat
 
@@ -93,7 +93,7 @@ def test_files_with_brackets_in_name():
     )
 
     dbs = DummyDBs()
-    to_files(chat, dbs)
+    to_files_and_memory(chat, dbs)
 
     assert dbs.memory["all_output.txt"] == chat
 
@@ -119,7 +119,7 @@ def test_files_with_file_colon():
     )
 
     dbs = DummyDBs()
-    to_files(chat, dbs)
+    to_files_and_memory(chat, dbs)
 
     assert dbs.memory["all_output.txt"] == chat
 
@@ -145,7 +145,7 @@ def test_files_with_back_tick():
     )
 
     dbs = DummyDBs()
-    to_files(chat, dbs)
+    to_files_and_memory(chat, dbs)
 
     assert dbs.memory["all_output.txt"] == chat
 
@@ -172,7 +172,7 @@ def test_files_with_newline_between():
     )
 
     dbs = DummyDBs()
-    to_files(chat, dbs)
+    to_files_and_memory(chat, dbs)
 
     assert dbs.memory["all_output.txt"] == chat
 
@@ -199,7 +199,7 @@ def test_files_with_newline_between_header():
     )
 
     dbs = DummyDBs()
-    to_files(chat, dbs)
+    to_files_and_memory(chat, dbs)
 
     assert dbs.memory["all_output.txt"] == chat