Use alphanumeric well IDs in .fluidics DataFrame index

Part of #38

Author: michaelosthege

bletl/__init__.py

@@ -19,4 +19,4 @@
     NoMeasurementData,
 )
 
-__version__ = "1.2.3"
+__version__ = "1.3.0"

bletl/parsing/blpro.py

@@ -25,6 +25,28 @@
 logger = logging.getLogger("blpro")
 
 
+_MF_WELL_NUMC_TO_ID = {
+    (colnumber + rownumber * 8): f"{row}{column:02d}"
+    for colnumber, column in enumerate(range(1, 9))
+    for rownumber, row in enumerate("CDEF")
+}
+"""Maps 0-based well numbers in MF mode by **C-style** counting order (→ then ↓) to alphanumeric well IDs."""
+
+_MF_WELL_NUMF_TO_ID = {
+    (rownumber + colnumber * 4): f"{row}{column:02d}"
+    for rownumber, row in enumerate("CDEF")
+    for colnumber, column in enumerate(range(1, 9))
+}
+"""Maps 0-based well numbers in MF mode by **Fortran-style** counting order (↓ then →) to alphanumeric well IDs."""
+
+_MF_WELL_NUMM_TO_ID = {
+    1 + (colnumber + rownumber * 8): f"{row}{column:02d}"
+    for rownumber, row in enumerate("CDEF")
+    for colnumber, column in enumerate(reversed(range(1, 9)) if rownumber % 2 else range(1, 9))
+}
+"""Maps 1-based well numbers in MF mode by **measurement counting order** (→→→ ↓ ←←← ↓ →→→ ...) to alphanumeric well IDs."""
+
+
 class BioLectorProParser(BLDParser):
     def parse(
         self,
@@ -329,6 +351,7 @@ def extract_fluidics(dfraw):
         ("Temp_Ch4", "volume", float),
     ]
     df = utils.__to_typed_cols__(dfraw[dfraw["Type"] == "F"], ocol_ncol_type)
+    df["well"] = [_MF_WELL_NUMM_TO_ID[w] for w in df["well"]]
     df = df.sort_values(["well", "cycle"]).set_index(["well"])
     return standardize(df)
 
@@ -348,6 +371,7 @@ def extract_valves_module(dfraw):
     df_valves.columns = ["cycle", "valve", "well", "acid", "base"]
     df_valves["valve"] = df_valves["valve"].str.replace("Valve ", "").astype(int)
     df_valves["well"] = df_valves["well"].str.replace("Well", "").astype(int)
+    # TODO: Which numbering style is this?
     df_valves["acid"] = df_valves["acid"].str.replace("Sollvolumen (Acid) ", "", regex=False).astype(float)
     df_valves["base"] = df_valves["base"].str.replace("Sollvolumen (Base) ", "", regex=False).astype(float)
     df_valves = standardize(df_valves).set_index(["well", "valve", "cycle"])
@@ -357,6 +381,7 @@ def extract_valves_module(dfraw):
     df_module.columns = ["cycle", "module", "valve", "well", "volume"]
     df_module["valve"] = df_module["valve"].str.replace("Valve ", "").astype(int)
     df_module["well"] = df_module["well"].str.replace("Well ", "").astype(int)
+    # TODO: Which numbering style is this?
     df_module["volume"] = df_module["volume"].str.replace("Volume ", "").astype(float)
     df_module = standardize(df_module).set_index(["well", "valve", "cycle"])
 

tests/test_core.py

@@ -9,7 +9,7 @@
 
 import bletl
 from bletl import core, parsing, utils
-from bletl.parsing import bl1
+from bletl.parsing import bl1, blpro
 
 dir_testfiles = pathlib.Path(pathlib.Path(__file__).absolute().parent, "data")
 
@@ -595,3 +595,43 @@ def test_comments(self):
 
         assert list(d_1.comments.columns) == list(d_p.comments.columns)
         return
+
+
+class TestBLProMF:
+    def test_well_num_mappings(self):
+        # C-style counts right then down
+        assert blpro._MF_WELL_NUMC_TO_ID[0] == "C01"
+        assert blpro._MF_WELL_NUMC_TO_ID[1] == "C02"
+        assert blpro._MF_WELL_NUMC_TO_ID[7] == "C08"
+        assert blpro._MF_WELL_NUMC_TO_ID[8] == "D01"
+        assert blpro._MF_WELL_NUMC_TO_ID[31] == "F08"
+        # F-style counts down then right
+        assert blpro._MF_WELL_NUMF_TO_ID[0] == "C01"
+        assert blpro._MF_WELL_NUMF_TO_ID[1] == "D01"
+        assert blpro._MF_WELL_NUMF_TO_ID[2] == "E01"
+        assert blpro._MF_WELL_NUMF_TO_ID[3] == "F01"
+        assert blpro._MF_WELL_NUMF_TO_ID[4] == "C02"
+        assert blpro._MF_WELL_NUMF_TO_ID[31] == "F08"
+        # Measurement order style is 1-based and goes left/right vice versa
+        assert blpro._MF_WELL_NUMM_TO_ID[1] == "C01"
+        assert blpro._MF_WELL_NUMM_TO_ID[2] == "C02"
+        assert blpro._MF_WELL_NUMM_TO_ID[8] == "C08"
+        assert blpro._MF_WELL_NUMM_TO_ID[9] == "D08"
+        assert blpro._MF_WELL_NUMM_TO_ID[16] == "D01"
+        assert blpro._MF_WELL_NUMM_TO_ID[32] == "F01"
+        pass
+
+    def test_issue_38(self):
+        fp = dir_testfiles / "BLPro" / "18-FZJ-Test2--2018-02-07-10-01-11.csv"
+        bldata = bletl.parse(fp)
+        assert bldata.fluidics.index.name == "well"
+        assert bldata.module.index.names == ["well", "valve", "cycle"]
+        assert bldata.valves.index.names == ["well", "valve", "cycle"]
+        # Check some initial and final well volumes against values shown in the BioLection
+        assert bldata.fluidics.loc["C01", "volume"][0] == 800
+        assert bldata.fluidics.loc["C01", "volume"][-1] == 1201.776
+        assert bldata.fluidics.loc["D01", "volume"][-1] == 1204.892
+        assert bldata.fluidics.loc["D02", "volume"][-1] == 954.68
+        assert bldata.fluidics.loc["E06", "volume"][-1] == 913.16
+        assert bldata.fluidics.loc["F01", "volume"][-1] == 1202.719
+        pass