fix shape URIs when generating shex from void

Author: vemonet

deploy.sh

@@ -18,7 +18,11 @@
 
 
 # NOTE: if OOM error, check `dmsg` on server and search for `oom`
-# sudo -u podman bash -c 'journalctl -k'
+# sudo -u podman bash -c 'journalctl -k '
+
+# sudo -u podman bash -c 'cd /var/containers/podman/sparql-llm ; tail -f data/logs/expasygpt_podman.log'
+
+
 
 ssh_cmd() {
     ssh -t expasychat "sudo -u podman bash -c 'cd /var/containers/podman/sparql-llm ; XDG_RUNTIME_DIR=/run/user/1001 $1'"

src/sparql-llm/src/sparql_llm/sparql_void_shapes_loader.py

@@ -47,7 +47,8 @@ def get_shex_dict_from_void(
             subj = prefix_converter.compress(subject_cls)
         except Exception as _e:
             subj = f"<{subject_cls}>"
-        shex_dict[subject_cls] = {"shex": f"{subj} {{\n  a [ {subj} ] ;\n"}
+        shape_iri = f"shape:{subj.replace(':', '_')}"
+        shex_dict[subject_cls] = {"shex": f"{shape_iri} {{\n  a [ {subj} ] ;\n"}
 
         for predicate, object_list in predicates.items():
             try:

tutorial/graph.py

@@ -162,13 +162,13 @@ async def validate_output(
 ) -> dict[str, bool | list[tuple[str, str]] | int]:
     """Node to validate the output of a LLM call, e.g. SPARQL queries generated."""
     recall_messages = []
-    print(state["messages"])
+    # print(state["messages"])
     last_msg = next(
         msg.content
         for msg in reversed(state["messages"])
         if isinstance(msg, AIMessage) and msg.content
     )
-    print(last_msg)
+    # print(last_msg)
     # last_msg = state["messages"][-1].content
     validation_outputs = validate_sparql_in_msg(
         last_msg, prefixes_map, endpoints_void_dict