Support for GitHub issue/prs to markdown

pyproject.toml

@@ -39,6 +39,7 @@ dependencies = [
   "youtube-transcript-api",
   "SpeechRecognition",
   "pathvalidate",
+  "pygithub"
 ]
 
 [project.urls]

src/markitdown/_markitdown.py

@@ -44,6 +44,14 @@
 except ModuleNotFoundError:
     pass
 
+# Optional GitHub issue support
+try:
+    from github import Github
+
+    IS_GITHUB_ISSUE_CAPABLE = True
+except ModuleNotFoundError:
+    IS_GITHUB_ISSUE_CAPABLE = False
+
 
 class _CustomMarkdownify(markdownify.MarkdownConverter):
     """
@@ -837,6 +845,75 @@
         return response.choices[0].message.content
 
 
+class GitHubIssueConverter(DocumentConverter):
+    """Converts GitHub issues to Markdown."""
+
+    def convert(self, issue_url, github_token) -> Union[None, DocumentConverterResult]:
+
+        # Bail if not a valid GitHub issue URL
+        if issue_url:
+            parsed_url = urlparse(issue_url)
+            path_parts = parsed_url.path.strip("/").split("/")
+            if len(path_parts) < 4 or path_parts[2] != "issues":
+                return None
+
+            if not github_token:
+                raise ValueError("GitHub token is not set. Cannot convert GitHub issue.")
+
+            return self._convert_github_issue(issue_url, github_token)
+
+        return None
+
+    def _convert_github_issue(
+        self, issue_url: str, github_token: str
+    ) -> DocumentConverterResult:
+        """
+        Convert a GitHub issue to a markdown document.
+        Args:
+            issue_url (str): The URL of the GitHub issue to convert.
+            github_token (str): A GitHub token with access to the repository.
+        Returns:
+            DocumentConverterResult: The result containing the issue title and markdown content.
+        Raises:
+            ImportError: If the PyGithub library is not installed.
+            ValueError: If the provided URL is not a valid GitHub issue URL.
+        """
+        if not IS_GITHUB_ISSUE_CAPABLE:
+            raise ImportError(
+                "PyGithub is not installed. Please install it to use this feature."
+            )
+
+        # Parse the issue URL
+        parsed_url = urlparse(issue_url)
+        path_parts = parsed_url.path.strip("/").split("/")
+        if len(path_parts) < 4 or path_parts[2] != "issues":
+            raise ValueError("Invalid GitHub issue URL")
+
+        owner, repo, _, issue_number = path_parts[:4]
+
+        # Authenticate with GitHub
+        g = Github(github_token)
+        repo = g.get_repo(f"{owner}/{repo}")
+        issue = repo.get_issue(int(issue_number))
+
+        # Convert issue details to markdown
+        markdown_content = f"# {issue.title}\n\n{issue.body}\n\n"
+        markdown_content += f"**State:** {issue.state}\n"
+        markdown_content += f"**Created at:** {issue.created_at}\n"
+        markdown_content += f"**Updated at:** {issue.updated_at}\n"
+        markdown_content += f"**Comments:**\n"
+
+        for comment in issue.get_comments():
+            markdown_content += (
+                f"- {comment.user.login} ({comment.created_at}): {comment.body}\n"
+            )
+
+        return DocumentConverterResult(
+            title=issue.title,
+            text_content=markdown_content,
+        )
+
+
 class FileConversionException(BaseException):
     pass
 
@@ -889,6 +966,12 @@
             - source: can be a string representing a path or url, or a requests.response object
             - extension: specifies the file extension to use when interpreting the file. If None, infer from source (path, uri, content-type, etc.)
         """
+        # Handle GitHub issue URLs directly
+        if isinstance(source, str) and "github.com" in source and "/issues/" in source:
+            github_token = kwargs.get("github_token", os.getenv("GITHUB_TOKEN"))
+            if not github_token:
+                raise ValueError("GitHub token is required for GitHub issue conversion.")
+            return GitHubIssueConverter().convert(issue_url=source, github_token=github_token)
 
         # Local path or url
         if isinstance(source, str):

tests/test_markitdown.py

@@ -87,6 +87,9 @@
     "data:image/svg+xml,%3Csvg%20width%3D",
 ]
 
+GITHUB_ISSUE_URL = "https://github.com/microsoft/autogen/issues/1421"
+GITHUB_TOKEN = os.environ.get("GITHUB_TOKEN", "")
+
 
 @pytest.mark.skipif(
     skip_remote,
@@ -179,8 +182,22 @@ def test_markitdown_exiftool() -> None:
         assert target in result.text_content
 
 
+@pytest.mark.skipif(
+    not GITHUB_TOKEN,
+    reason="GitHub token not provided",
+)
+def test_markitdown_github_issue() -> None:
+    markitdown = MarkItDown()
+    result = markitdown.convert(GITHUB_ISSUE_URL, github_token=GITHUB_TOKEN)
+    print(result.text_content)
+    assert "User-Defined Functions" in result.text_content
+    assert "closed" in result.text_content
+    assert "Comments:" in result.text_content
+
+
 if __name__ == "__main__":
     """Runs this file's tests from the command line."""
     test_markitdown_remote()
     test_markitdown_local()
     test_markitdown_exiftool()
+    test_markitdown_github_issue()