Release prep (#152)

* Adds knowledge to ac exec env

* Removes linebreaks for knowledge terms

Author: JWittmeyer

controller/attribute/util.py

@@ -19,6 +19,7 @@
 from submodules.model.models import Attribute
 from submodules.s3 import controller as s3
 from util import daemon, notification
+from controller.knowledge_base import util as knowledge_base
 
 client = docker.from_env()
 image = os.getenv("AC_EXEC_ENV_IMAGE")
@@ -83,6 +84,7 @@ def run_attribute_calculation_exec_env(
 
     prefixed_function_name = f"{attribute_id}_fn"
     prefixed_payload = f"{attribute_id}_payload.json"
+    prefixed_knowledge_base = f"{attribute_id}_knowledge"
     project_item = project.get(project_id)
     org_id = str(project_item.organization_id)
 
@@ -91,9 +93,15 @@ def run_attribute_calculation_exec_env(
         project_id + "/" + prefixed_function_name,
         attribute_item.source_code,
     )
+    s3.put_object(
+        org_id,
+        project_id + "/" + prefixed_knowledge_base,
+        knowledge_base.build_knowledge_base_from_project(project_id),
+    )
     command = [
         s3.create_access_link(org_id, project_id + "/" + doc_bin),
         s3.create_access_link(org_id, project_id + "/" + prefixed_function_name),
+        s3.create_access_link(org_id, project_id + "/" + prefixed_knowledge_base),
         project_item.tokenizer_blank,
         s3.create_file_upload_link(org_id, project_id + "/" + prefixed_payload),
         attribute_item.data_type,
@@ -170,7 +178,6 @@ def read_container_logs_thread(
     attribute_id: str,
     docker_container: Any,
 ) -> None:
-
     ctx_token = general.get_ctx_token()
     # needs to be refetched since it is not thread safe
     attribute_item = attribute.get(project_id, attribute_id)

controller/knowledge_base/util.py

@@ -24,17 +24,19 @@ def build_knowledge_base_from_project(project_id: str) -> str:
     for knowledge_base_item in knowledge_base.get_all_by_project_id(project_id):
         knowledge_bases_dict[resolve_name_as_variable(knowledge_base_item.name)] = []
 
-    for term, knowledge_base_item in knowledge_term.get_terms_with_base_names(
+    for term, knowledge_base_name in knowledge_term.get_terms_with_base_names(
         project_id
     ):
-        knowledge_bases_dict[resolve_name_as_variable(knowledge_base_item)].append(
-            term
-        )  # use here knowledge base name in standard format (underscore and )
+        # use here knowledge base name in standard format (underscore and )
+        knowledge_bases_dict[resolve_name_as_variable(knowledge_base_name)].append(term)
 
     for knowledge_base_item, values in knowledge_bases_dict.items():
         knowledge_base_source += f"\n{knowledge_base_item} = [\n"
         for value in values:
-            value: str = value.replace("'", "\\'")  # e.g. "You're too good to me"
+            # e.g. "You're too good to me"
+            value: str = value.replace("'", "\\'")
+            # safety net: knowledge entries are single line, linebreaks can only come through extraction task auto generation which now should remove linebreaks
+            value = value.replace("\n", " ").replace("\r", "")
             knowledge_base_source += f"\t'{value}',\n"
         knowledge_base_source += "]"
     return knowledge_base_source

controller/knowledge_term/manager.py

@@ -72,6 +72,8 @@ def create_term_in_named_knowledge_base(project_id: str, name: str, value: str)
     if not base:
         base = knowledge_base.create(project_id, name)
     try:
+        # ensure linebreaks aren't part of the entry
+        value = value.replace("\n", " ").replace("\r", "")
         knowledge_term.create(project_id, base.id, value, None, with_commit=True)
     except EntityAlreadyExistsException:
         pass  # TODO EXCEPTION HANDLING