start using markdown representation for read writable context (CF-687)

[mohammedahmed18]

this still WIP, most tests are expected to fail

PR Type

Enhancement, Tests

---

Description

• Switch read/write context extraction to markdown

• Introduce CodeStringsMarkdown with splitter and cache

• Update CodeOptimizationContext model type

• Adjust tests for markdown splitter and __str__

---

Changes diagram

flowchart LR
  A["get_code_optimization_context()"] --> B["extract_code_markdown_context_from_files()"]
  B --> C["CodeStringsMarkdown (with splitter)"]
  C --> D["encoded_tokens_len(__str__)"]
  C --> E["find_preexisting_objects(__str__)"]

Loading

---

Changes walkthrough 📝

	Relevant files
Enhancement	code_context_extractor.py Switch to markdown context extraction                                        codeflash/context/code_context_extractor.py Replaced extract_code_string_context_from_files call Now uses extract_code_markdown_context_from_files Token length uses .__str__ of markdown context Preexisting objects extracted via .__str__ +6/-5      models.py Add markdown code strings model                                                    codeflash/models/models.py Added get_code_block_splitter helper Enhanced CodeStringsMarkdown with cache and __str__ Changed read_writable_code type to CodeStringsMarkdown Removed unused Field import +16/-2
Tests	test_code_context_extractor.py Update tests for markdown splitter                                              tests/test_code_context_extractor.py Imported get_code_block_splitter Updated expected context with splitter f-string Assert on read_writable_code.__str__ +4/-3

---

Need help?

Type /help how to ... in the comments thread for any questions about PR-Agent usage.

Check out the documentation for more information.

[github-actions]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

⏱️ Estimated effort to review: 3 🔵🔵🔵⚪⚪

🧪 PR contains tests

🔒 No security concerns identified

⚡ Recommended focus areas for review Magic method misuse Defining __str__ as a property may override or conflict with Python's built-in __str__ behavior; consider implementing the actual __str__ method instead of a property @property def __str__(self) -> str: if self.cached_code is not None: return self.cached_code self.cached_code = "\n\n".join( get_code_block_splitter(block.file_path) + "\n" + block.code for block in self.code_strings ) return self.cached_code Property invocation The code passes final_read_writable_code.__str__ to functions without invoking it; ensure the string is obtained (e.g., via str(...) or accessing the property) rather than passing the property object final_read_writable_tokens = encoded_tokens_len(final_read_writable_code.__str__) if final_read_writable_tokens > optim_token_limit: raise ValueError("Read-writable code has exceeded token limit, cannot proceed") # Setup preexisting objects for code replacer preexisting_objects = set( chain( find_preexisting_objects(final_read_writable_code.__str__),

[github-actions]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
General	Normalize splitter path Use file_path.as_posix() in the f-string to ensure the splitter uses a consistent POSIX path, avoiding backslashes on Windows and matching test expectations. codeflash/models/models.py [142-143] def get_code_block_splitter(file_path: Path) -> str: - return f"# codeflash-splitter__{file_path}" + return f"# codeflash-splitter__{file_path.as_posix()}" Suggestion importance[1-10]: 7 __ Why: Using file_path.as_posix() in the splitter guarantees forward-slash paths on all OSes, avoiding backslashes on Windows and preventing cross-platform test failures.	Medium
	Properly implement __str__ Replace the @property on str with a standard special method override so calling str(obj) invokes it correctly and doesn’t shadow built-in behavior. codeflash/models/models.py [150-157] -@property def __str__(self) -> str: if self.cached_code is not None: return self.cached_code self.cached_code = "\n\n".join( get_code_block_splitter(block.file_path) + "\n" + block.code for block in self.code_strings ) return self.cached_code Suggestion importance[1-10]: 6 __ Why: Defining __str__ with @property prevents str(obj) from invoking this logic; converting it to a normal special method override ensures the custom formatter is used when calling str(), improving usability.	Low