moved the text_sep, prefix and suffix

from models.py to config
also move language to config
also moves the masking char to scrub

openadapt/config.py

@@ -63,8 +63,18 @@ def getenv_fallback(var_name):
             logger.info(f"{key}={val}")
 
 
+# ACTION EVENT CONFIGURATIONS
+TEXT_SEP = "-"
+TEXT_NAME_PREFIX = "<"
+TEXT_NAME_SUFFIX = ">"
+
+
+
+
 # SCRUBBING CONFIGURATIONS
 
+SCRUB_CHAR = "*"
+SCRUB_LANGUAGE = "en"
 SCRUB_CONFIG_TRF = {
     "nlp_engine_name": "spacy",
     "models": [{"lang_code": "en", "model_name": "en_core_web_trf"}],

openadapt/models.py

@@ -6,7 +6,7 @@
 import numpy as np
 import sqlalchemy as sa
 
-from openadapt import db, utils, window
+from openadapt import config, db, utils, window
 
 
 # https://groups.google.com/g/sqlalchemy/c/wlr7sShU6-k
@@ -61,6 +61,9 @@ def processed_action_events(self):
 
 class ActionEvent(db.Base):
     __tablename__ = "action_event"
+    _text_sep = config.TEXT_SEP
+    _text_name_prefix = config.TEXT_NAME_PREFIX
+    _text_name_suffix = config.TEXT_NAME_SUFFIX
 
     id = sa.Column(sa.Integer, primary_key=True)
     name = sa.Column(sa.String)
@@ -201,10 +204,6 @@ def __str__(self):
         rval = " ".join(attrs)
         return rval
 
-    _text_sep = "-"
-    _text_name_prefix = "<"
-    _text_name_suffix = ">"
-
     @classmethod
     def from_children(cls, children_dicts):
         children = [

openadapt/scrub.py

@@ -31,20 +31,20 @@ def scrub_text(text: str, is_hyphenated: bool = False) -> str:
         return None
 
     if is_hyphenated and not (
-        text.startswith("<") or text.endswith(">")
+        text.startswith(config.TEXT_NAME_PREFIX) or text.endswith(config.TEXT_NAME_SUFFIX)
     ):
-        text = "".join(text.split("-"))
+        text = "".join(text.split(config.TEXT_SEP))
 
     analyzer_results = config.ANALYZER_TRF.analyze(
-        text=text, entities=config.SCRUBBING_ENTITIES, language="en"
+        text=text, entities=config.SCRUBBING_ENTITIES, language=config.SCRUB_LANGUAGE
     )
 
     operators = {}
     for entity in analyzer_results:
         operators[entity.entity_type] = OperatorConfig(
             "mask",
             {
-                "masking_char": "*",
+                "masking_char": config.SCRUB_CHAR,
                 "chars_to_mask": entity.end - entity.start,
                 "from_end": True,
             },
@@ -57,9 +57,9 @@ def scrub_text(text: str, is_hyphenated: bool = False) -> str:
     )
 
     if is_hyphenated and not (
-        text.startswith("<") or text.endswith(">")
+        text.startswith(config.TEXT_NAME_PREFIX) or text.endswith(config.TEXT_NAME_SUFFIX)
     ):
-        anonymized_results.text = "-".join(anonymized_results.text)
+        anonymized_results.text = config.TEXT_SEP.join(anonymized_results.text)
 
     return anonymized_results.text