Add ability to remove directives

sphinx_llms_txt/__init__.py

@@ -58,6 +58,7 @@ def build_finished(app: Sphinx, exception):
             "llms_txt_full_max_size": app.config.llms_txt_full_max_size,
             "llms_txt_directives": app.config.llms_txt_directives,
             "llms_txt_exclude": app.config.llms_txt_exclude,
+            "llms_txt_rm_directives": app.config.llms_txt_rm_directives,
             "html_baseurl": getattr(app.config, "html_baseurl", ""),
         }
         _manager.set_config(config)
@@ -86,6 +87,7 @@ def setup(app: Sphinx) -> Dict[str, Any]:
     app.add_config_value("llms_txt_title", None, "env")
     app.add_config_value("llms_txt_summary", None, "env")
     app.add_config_value("llms_txt_exclude", [], "env")
+    app.add_config_value("llms_txt_rm_directives", False, "env")
 
     # Connect to Sphinx events
     app.connect("doctree-resolved", doctree_resolved)

sphinx_llms_txt/processor.py

@@ -50,8 +50,33 @@ class DocumentProcessor:
         # Then process path directives (image, figure, etc.)
         content = self._process_path_directives(content, source_path)
 
+        # Remove directives if configured to do so
+        if self.config.get("llms_txt_rm_directives", False):
+            content = self._remove_directives(content)
+
         return content
 
+    def _remove_directives(self, content: str) -> str:
+        """Remove directives from content.
+
+        Args:
+            content: The source content from which to remove directives
+
+        Returns:
+            Content with all directives removed
+        """
+        # Match any directive pattern (starting with .. followed by ::)
+        directive_pattern = re.compile(r'^\s*\.\.\s+[\w\-]+::.*?$(?:\n\s+.*?$)*',
+                                      re.MULTILINE | re.DOTALL)
+
+        # Replace all directives with an empty string
+        processed_content = directive_pattern.sub('', content)
+
+        # Clean up any consecutive blank lines that might result from directive removal
+        processed_content = re.sub(r'\n{3,}', '\n\n', processed_content)
+
+        return processed_content
+
     def _extract_relative_document_path(
         self, source_path: Path
     ) -> Tuple[Optional[str], Optional[str], Optional[List[str]]]: