Nomad changes

docs/snippets/python/plugins/pdf_metadata_extractor.md

@@ -0,0 +1,42 @@
+```python title="Python"
+from kreuzberg import register_post_processor, ExtractionResult
+import logging
+
+logger = logging.getLogger(__name__)
+
+class PdfMetadataExtractor:
+    def __init__(self):
+        self.processed_count: int = 0
+
+    def name(self) -> str:
+        return "pdf_metadata_extractor"
+
+    def version(self) -> str:
+        return "1.0.0"
+
+    def description(self) -> str:
+        return "Extracts and enriches PDF metadata"
+
+    def processing_stage(self) -> str:
+        return "early"
+
+    def should_process(self, result: ExtractionResult) -> bool:
+        return result.mime_type == "application/pdf"
+
+    def process(self, result: ExtractionResult) -> ExtractionResult:
+        self.processed_count += 1
+        result.metadata["pdf_processed"] = True
+        result.metadata["pdf_order"] = self.processed_count
+        result.metadata["content_length"] = len(result.content)
+        result.metadata["pdf_processor_version"] = "1.0.0"
+        return result
+
+    def initialize(self) -> None:
+        logger.info("PDF metadata extractor initialized")
+
+    def shutdown(self) -> None:
+        logger.info(f"Processed {self.processed_count} PDFs")
+
+processor: PdfMetadataExtractor = PdfMetadataExtractor()
+register_post_processor(processor)
+```