fix: resolve dict attribute error in MarkdownGenerator deserialization

[MTahirKleem]

Fixes #1880

This PR addresses an AttributeError that occurred when CrawlResult was initialized from a dictionary (e.g., from cache or JSON), where the markdown field remained a raw dict instead of being hydrated into a MarkdownGenerationResult object.

List of files changed and why
crawl4ai/models.py: Updated CrawlResult with a Pydantic alias and validator to ensure proper object hydration while maintaining the backward-compatible string-like interface.

tests/unit/test_crawl_result_markdown_validator.py: New test file to verify that passing a dictionary to the markdown parameter correctly results in a MarkdownGenerationResult object.

How Has This Been Tested?
Manual Verification: Created a reproduction script that initialized CrawlResult with a raw dictionary and verified that property access (e.g., .raw_markdown) worked without error.

Unit Testing: Ran the newly created test case to confirm the validator intercepts and converts dictionary inputs correctly.

Serialization Test: Verified that model_dump() still correctly exports the markdown field for storage.

Checklist:
[x] My code follows the style guidelines of this project

[x] I have performed a self-review of my own code

[x] I have commented my code, particularly in hard-to-understand areas

[x] I have made corresponding changes to the documentation

[x] I have added/updated unit tests that prove my fix is effective or that my feature works

[x] New and existing unit tests pass locally with my changes

[Copilot]

Pull request overview

This PR aims to make CrawlResult resilient to cached/JSON hydration cases where the markdown payload is a raw dict, preventing attribute errors during markdown access/serialization.

Changes:

• Add a markdown_data field (aliased from "markdown") to improve Pydantic hydration robustness.
• Add a field_validator to convert raw dict inputs into MarkdownGenerationResult.
• Add a unit test validating dict-to-model conversion and string-compatible access behavior.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 3 comments.

File	Description
crawl4ai/models.py	Introduces markdown_data alias + validator + post-init syncing to prevent markdown deserialization issues.
tests/unit/test_crawl_result_markdown_validator.py	Adds a unit test ensuring dict markdown input is converted and serialized correctly.

Comments suppressed due to low confidence (1)

crawl4ai/models.py:185

• The NOTE block at lines 181–191 appears dedented to column 0 within CrawlResult. If those lines are truly unindented, that would end the class body and make the subsequent indented def model_post_init(...) an IndentationError (module won’t import). Keep the NOTE block indented at the same level as other class members (or move it fully outside the class along with the following methods).

# NOTE: The StringCompatibleMarkdown class, model_post_init hook, property getters/setters,
# and model_dump override all exist to support a smooth transition from markdown as a string
# to markdown as a MarkdownGenerationResult object, while maintaining backward compatibility.
# 
# This allows code that expects markdown to be a string to continue working, while also

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

validate_markdown only converts dicts. If input markdown is a plain string (which can happen with legacy cache hydration), Pydantic will try to validate that string as MarkdownGenerationResult for markdown_data and raise a validation error. Consider handling str here (e.g., wrap it into a MarkdownGenerationResult with raw_markdown set and other fields empty) to preserve backward compatibility.

Suggested change

	return MarkdownGenerationResult(**v)
	return MarkdownGenerationResult(**v)
	if isinstance(v, str):
	# Legacy backward compatibility: markdown provided as a plain string
	# Wrap it into MarkdownGenerationResult, using raw_markdown
	return MarkdownGenerationResult(raw_markdown=v)

[Copilot]

This test covers dict input, but it doesn’t cover legacy string hydration (e.g., CrawlResult(..., markdown="# Hello")). Adding that case would protect backward compatibility and catch the validation regression mentioned in the model validator.

[Copilot]

The PR metadata links/mentions Issue #123 (“Timeout setting”), but the changes here are focused on CrawlResult markdown deserialization. If this PR is intended to fix #123, the code changes don’t appear to address timeouts; otherwise consider removing/clarifying the issue reference in the PR description.