Clean SDXL expression labels

sdxl_tag_policy.py

@@ -56,6 +56,7 @@ def split_tag_text(text: Any) -> list[str]:
     text = re.sub(r"\bMan [A-Z]'s\b", "man's", text)
     text = re.sub(r"\bWoman [A-Z]\b", "woman", text)
     text = re.sub(r"\bMan [A-Z]\b", "man", text)
+    text = re.sub(r"\b(?:the\s+)?(?:woman|man)\s+has\s+", "", text, flags=re.IGNORECASE)
     text = re.sub(
         r"\b(?:Clothing state|Visual clothing state|visible remaining styling|teaser outfit detail|softcore visual reference|Sexual scene|Role graph):\s*",
         "",

tools/prompt_route_simulation.py

@@ -576,6 +576,12 @@ def _sdxl_composition_tag_issues(name: str, sdxl_prompt: str) -> list[str]:
     return []
 
 
+def _sdxl_expression_label_issues(name: str, sdxl_prompt: str) -> list[str]:
+    if re.search(r"\b(?:woman|man)\s+has\s+", sdxl_prompt, flags=re.IGNORECASE):
+        return [f"{name}.sdxl_prompt: character_has_expression_label"]
+    return []
+
+
 def _trace_dict(formatter_name: str, payload: dict[str, Any]) -> tuple[dict[str, Any], str]:
     trace_text = str(payload.get("route_trace_json") or "")
     if not trace_text:
@@ -713,6 +719,7 @@ def _formatter_issues(
     issues.extend(_krea_grammar_issues(name, krea_prompt))
     issues.extend(_composition_label_issues(name, prompts))
     issues.extend(_sdxl_composition_tag_issues(name, sdxl_prompt))
+    issues.extend(_sdxl_expression_label_issues(name, sdxl_prompt))
 
     for label, value in (
         (f"{name}.krea_negative", krea.get("negative_prompt")),

tools/prompt_smoke.py

@@ -4815,6 +4815,18 @@ def smoke_sdxl_tag_routes() -> None:
     ).as_text()
     _expect("tall-window depth behind them composition" not in composition_label_tags, "SDXL route kept raw composition label tag")
     _expect("tall-window depth behind them" in composition_label_tags, "SDXL route lost composition detail while removing label")
+    expression_label_row = _fixture_hardcore_row(
+        character_expression_text="Woman A has focused gaze; Man A has steady expression",
+        expression="stale expression",
+    )
+    expression_label_tags = sdxl_tag_routes.row_core_tags_result(
+        sdxl_tag_routes.SDXLRowTagRequest(expression_label_row, 1.29),
+        deps,
+    ).as_text()
+    _expect("woman has" not in expression_label_tags.lower(), "SDXL route kept woman-has expression label")
+    _expect("man has" not in expression_label_tags.lower(), "SDXL route kept man-has expression label")
+    _expect("focused gaze" in expression_label_tags, "SDXL route lost cleaned woman expression")
+    _expect("steady expression" in expression_label_tags, "SDXL route lost cleaned man expression")
     stale_prompt_row = _fixture_hardcore_row(
         prompt="stale raw prompt mentions fully nude naked pussy penis oral anal semen penetration",
         item="standing portrait setup",