batch docs update

Makefile

@@ -33,12 +33,15 @@ EXT    ?=
 PYTHON ?=
 export PYTHON
 
-# doocus (document extraction) knobs. DOCS_OUT defaults to docs-output; the input
-# tree is mirrored under it. DOC_EXTRA forwards one-off flags to process_doc.py
-# (e.g. --render for pdf page thumbnails, --ocr for images/scanned pdfs).
+# doocus (tree indexing) knobs. DOCS_OUT defaults to docs-output; the whole input
+# tree is replicated into index.json under it. DOC_EXTRA forwards one-off flags to
+# process_tree.py (e.g. --render for pdf page thumbnails, --ocr for scanned pdfs).
 DOCS_OUT  ?= docs-output
 DOC_EXTRA ?=
 
+# Python interpreter for doocus (override PYTHON, else python3 — use `uv run make`).
+PY := $(if $(PYTHON),$(PYTHON),python3)
+
 .PHONY: batch dry docs docs-dry help
 
 batch:
@@ -48,16 +51,16 @@ batch:
 dry:
 	@ctrl/batch.sh -i "$(IN)" -o "$(OUT)" $(if $(EXT),-e "$(EXT)") -n
 
-# Document extraction (doocus), mirrors the input tree into DOCS_OUT.
-#   make docs IN="/mnt/win/drive"                    # extract all docs
-#   make docs IN="..." DOC_EXTRA="--render --ocr"    # thumbnails + OCR
-#   make docs-dry IN="..."                           # list what would run
+# Document indexing (doocus): replicate the whole IN tree into DOCS_OUT/index.json,
+# extracting text sidecars only for docx/pdf/pptx/xlsx.
+#   uv run make docs IN="/mnt/win/drive"             # index the tree
+#   uv run make docs IN="..." DOC_EXTRA="--ocr"      # + OCR scanned pdfs
+#   make docs-dry IN="..."                           # counts only, no writes
 docs:
-	@ctrl/batch_docs.sh -i "$(IN)" -o "$(DOCS_OUT)" $(if $(EXT),-e "$(EXT)") -- \
-		$(DOC_EXTRA)
+	@$(PY) process_tree.py "$(IN)" --output "$(DOCS_OUT)" $(DOC_EXTRA)
 
 docs-dry:
-	@ctrl/batch_docs.sh -i "$(IN)" -o "$(DOCS_OUT)" $(if $(EXT),-e "$(EXT)") -n
+	@$(PY) process_tree.py "$(IN)" --output "$(DOCS_OUT)" --dry-run
 
 help:
 	@sed -n '3,14p' Makefile