chore: normalize frozenset literals and myscale typing (#34327)

api/core/helper/csv_sanitizer.py

@@ -17,7 +17,7 @@ class CSVSanitizer:
     """
 
     # Characters that can start a formula in Excel/LibreOffice/Google Sheets
-    FORMULA_CHARS = frozenset({"=", "+", "-", "@", "\t", "\r"})
+    FORMULA_CHARS = frozenset(("=", "+", "-", "@", "\t", "\r"))
 
     @classmethod
     def sanitize_value(cls, value: Any) -> str:

api/core/rag/datasource/keyword/jieba/jieba_keyword_table_handler.py

@@ -122,6 +122,6 @@ class JiebaKeywordTableHandler:
             results.add(token)
             sub_tokens = re.findall(r"\w+", token)
             if len(sub_tokens) > 1:
-                results.update({w for w in sub_tokens if w not in list(STOPWORDS)})
+                results.update({w for w in sub_tokens if w not in STOPWORDS})
 
         return results

api/core/rag/datasource/vdb/myscale/myscale_vector.py

@@ -4,7 +4,7 @@ import uuid
 from enum import StrEnum
 from typing import Any
 
-from clickhouse_connect import get_client
+from clickhouse_connect import get_client  # type: ignore[import-untyped]
 from pydantic import BaseModel
 
 from configs import dify_config

api/core/rag/extractor/pdf_extractor.py

@@ -35,7 +35,7 @@ class PdfExtractor(BaseExtractor):
     """
 
     # Magic bytes for image format detection: (magic_bytes, extension, mime_type)
-    IMAGE_FORMATS = [
+    IMAGE_FORMATS: tuple[tuple[bytes, str, str], ...] = (
         (b"\xff\xd8\xff", "jpg", "image/jpeg"),
         (b"\x89PNG\r\n\x1a\n", "png", "image/png"),
         (b"\x00\x00\x00\x0c\x6a\x50\x20\x20\x0d\x0a\x87\x0a", "jp2", "image/jp2"),
@@ -45,7 +45,7 @@ class PdfExtractor(BaseExtractor):
         (b"MM\x00*", "tiff", "image/tiff"),
         (b"II+\x00", "tiff", "image/tiff"),
         (b"MM\x00+", "tiff", "image/tiff"),
-    ]
+    )
     MAX_MAGIC_LEN = max(len(m) for m, _, _ in IMAGE_FORMATS)
 
     def __init__(self, file_path: str, tenant_id: str, user_id: str, file_cache_key: str | None = None):

api/core/trigger/constants.py

@@ -5,11 +5,11 @@ TRIGGER_SCHEDULE_NODE_TYPE: Final[str] = "trigger-schedule"
 TRIGGER_PLUGIN_NODE_TYPE: Final[str] = "trigger-plugin"
 
 TRIGGER_NODE_TYPES: Final[frozenset[str]] = frozenset(
-    {
+    (
         TRIGGER_WEBHOOK_NODE_TYPE,
         TRIGGER_SCHEDULE_NODE_TYPE,
         TRIGGER_PLUGIN_NODE_TYPE,
-    }
+    )
 )
 
 

api/core/workflow/nodes/trigger_webhook/entities.py

@@ -8,24 +8,20 @@ from pydantic import BaseModel, Field, field_validator
 
 from core.trigger.constants import TRIGGER_WEBHOOK_NODE_TYPE
 
-_WEBHOOK_HEADER_ALLOWED_TYPES = frozenset(
-    {
-        SegmentType.STRING,
-    }
-)
+_WEBHOOK_HEADER_ALLOWED_TYPES: frozenset[SegmentType] = frozenset((SegmentType.STRING,))
 
-_WEBHOOK_QUERY_PARAMETER_ALLOWED_TYPES = frozenset(
-    {
+_WEBHOOK_QUERY_PARAMETER_ALLOWED_TYPES: frozenset[SegmentType] = frozenset(
+    (
         SegmentType.STRING,
         SegmentType.NUMBER,
         SegmentType.BOOLEAN,
-    }
+    )
 )
 
 _WEBHOOK_PARAMETER_ALLOWED_TYPES = _WEBHOOK_HEADER_ALLOWED_TYPES | _WEBHOOK_QUERY_PARAMETER_ALLOWED_TYPES
 
-_WEBHOOK_BODY_ALLOWED_TYPES = frozenset(
-    {
+_WEBHOOK_BODY_ALLOWED_TYPES: frozenset[SegmentType] = frozenset(
+    (
         SegmentType.STRING,
         SegmentType.NUMBER,
         SegmentType.BOOLEAN,
@@ -35,7 +31,7 @@ _WEBHOOK_BODY_ALLOWED_TYPES = frozenset(
         SegmentType.ARRAY_BOOLEAN,
         SegmentType.ARRAY_OBJECT,
         SegmentType.FILE,
-    }
+    )
 )