implement trigger matching and double counting removal in DL selector

hbw/columnflow_patches.py

@@ -83,8 +83,12 @@ def patch_csp_versioning():
     """
     Patches the TaskArrayFunction to add the version to the string representation of the task.
     """
+
     def TaskArrayFunction_str(self):
-        version_str = f"V{self.version}" if hasattr(self, "version") else ""
+        version = self.version() if callable(getattr(self, "version", None)) else getattr(self, "version", None)
+        if version and not isinstance(version, int):
+            raise Exception(f"version must be an integer, but is {version}")
+        version_str = f"V{version}" if version is not None else ""
         return f"{self.cls_name}{version_str}"
 
     TaskArrayFunction.__str__ = TaskArrayFunction_str

hbw/config/datasets.py

@@ -22,10 +22,12 @@
 
 
 def hbw_dataset_names(config: od.Config, as_list: bool = False) -> DotDict[str: list[str]] | list[str]:
+    # define data streams based on the run
+    # NOTE: the order of the streams is important for data trigger filtering (removing double counting)
     if config.campaign.x.run == 2:
-        data_streams = ["mu", "e"]
+        config.x.data_streams = ["mu", "e"]
     elif config.campaign.x.run == 3:
-        data_streams = ["mu", "egamma", "muoneg"]
+        config.x.data_streams = ["mu", "egamma", "muoneg"]
 
     data_eras = {
         "2017": "cdef",
@@ -38,7 +40,7 @@ def hbw_dataset_names(config: od.Config, as_list: bool = False) -> DotDict[str:
             f"data_{stream}_{era}"
             for era in data_eras
         ]
-        for stream in data_streams
+        for stream in config.x.data_streams
     }
 
     dataset_names = DotDict.wrap({