Collector. Re-number flask to have unique numbers, segregated by rack.

cfatools/processor/collector.py

+"""
+The collector processor module provides functions to manipulate collector's raw data.
+"""
+import pandas as pd
+
+
+def renumber_flasks(collector_df: pd.DataFrame) -> pd.DataFrame:
+    if "flask" not in collector_df.columns:
+        raise ValueError("The collector_df expected should have a 'flask' column")
+
+    # In the old numbering system, flasks were grouped by 50. Re-number them by 100
+    if collector_df.index[0] < pd.Timestamp(year=2021, month=12, day=16, tz="UTC"):
+        collector_df.loc[(collector_df["flask"] >= 101) & (collector_df["flask"] <= 148)] = collector_df.loc[(collector_df["flask"] >= 101) & (collector_df["flask"] <= 148)] + 300 - 100
+        collector_df.loc[(collector_df["flask"] >= 51) & (collector_df["flask"] <= 98)] = collector_df.loc[(collector_df["flask"] >= 51) & (collector_df["flask"] <= 98)] + 200 - 50
+        collector_df.loc[(collector_df["flask"] >= 1) & (collector_df["flask"] <= 48)] = collector_df.loc[(collector_df["flask"] >= 1) & (collector_df["flask"] <= 48)] + 100
+
+    # Re-number the flasks of the 4th rack and followings.
+    collector_df["cumcount"] = collector_df.groupby("flask").cumcount()
+    collector_df["flask"] = collector_df["flask"] + collector_df["cumcount"] * 300
+
+    return collector_df
+
+
+if __name__ == '__main__':
+    from cfatools.logreader.dataset import DatasetReader
+    dataset = DatasetReader("/homel/ojossoud/_temp/data_cfa", "20210826_ASUMA2016_6-1_sq")
+
+    # collector_df = dataset.get_timeseries("COLLECTR_instant")
+    # renumber_flasks(collector_df)
+
+    collector_df = dataset.get_timeseries("COLLECTR_instant", "flask")
+    collector_df = collector_df.rename(columns={"value": "flask"})
+    renumber_flasks(collector_df)