Add changes from PR #26

.pre-commit-config.yaml

@@ -14,7 +14,7 @@ repos:
     -   id: check-added-large-files
     -   id: check-merge-conflict
 -   repo: https://github.com/psf/black
-    rev: 22.0.0
+    rev: 24.1.0
     hooks:
     -   id: black
 # We should add some linter to here at some point

pyproject.toml

@@ -13,9 +13,9 @@ packages = [
 ]
 
 [tool.poetry.dependencies]
-python = ">=3.9,<3.11"
+python = ">=3.9,<3.13"
 urllib3 = "^1.2.26"
-pandas = "^1.0.0"
+pandas = ">=1.0.0,<3.0.0"
 xxhash = "^2.0.0"
 rdchiral = "^1.1.0"
 PyYAML = "^6.0.1"
@@ -30,6 +30,7 @@ cgrtools = "^4.1.35"
 scipy = "^1.11.4"
 pydantic = "^2.8.2"
 apted = "^1.0.3"
+dask = ">=2024.4.1"
 onnxruntime = {version = "<1.17.0", optional=true}
 
 [tool.poetry.group.dev.dependencies]

rxnutils/pipeline/actions/dataframe_mod.py

@@ -31,7 +31,7 @@ class DropColumns:
     columns: list[str]
 
     def __call__(self, data: pd.DataFrame) -> pd.DataFrame:
-        return data.drop(self.columns, 1)
+        return data.drop(self.columns, axis=1)
 
     def __str__(self) -> str:
         return f"{self.pretty_name} (drop columns in specified list, keeps the rest)"
@@ -100,7 +100,7 @@ def __call__(self, data: pd.DataFrame) -> pd.DataFrame:
         for name in self.columns:
             drop_columns.remove(name)
 
-        return data.drop(drop_columns, 1)
+        return data.drop(drop_columns, axis=1)
 
     def __str__(self) -> str:
         return f"{self.pretty_name} (keeps columns in specified list, drops the rest)"
@@ -187,7 +187,7 @@ def __call__(self, data: pd.DataFrame) -> pd.DataFrame:
             value_vars=self.in_columns,
             var_name=var_column_name,
             value_name=self.out_column,
-        ).drop(var_column_name, 1)
+        ).drop(var_column_name, axis=1)
 
     def __str__(self) -> str:
         return f"{self.pretty_name} (stacks two columns into a single column)"

rxnutils/pipeline/actions/reaction_mod.py

@@ -262,8 +262,10 @@ class IsotopeInfo:
 
     def __call__(self, data: pd.DataFrame) -> pd.DataFrame:
         isotop_info = data[self.in_column].str.extract(self.match_regex)
-        isotop_col = isotop_info.mass + isotop_info.symbol
-        rsmi_col = data[self.in_column].str.replace(self.match_regex, self.sub_regex)
+        isotop_col = isotop_info["mass"] + isotop_info["symbol"]
+        rsmi_col = data[self.in_column].str.replace(
+            self.match_regex, self.sub_regex, regex=True
+        )
         return data.assign(**{self.isotope_column: isotop_col, self.out_column: rsmi_col})
 
     def __str__(self) -> str:

rxnutils/routes/base.py

@@ -306,7 +306,7 @@ def _assign_mapping(self, overwrite: bool = False, only_rxnmapper: bool = False)
             df["mapped_smiles"] = df["RxnmapperRxnSmiles"]
         else:
             sel = df["NMC"] == "0.0"
-            df["mapped_smiles"].mask(sel, df["RxnmapperRxnSmiles"], inplace=True)
+            df.loc[sel, "mapped_smiles"] = df.loc[sel, "RxnmapperRxnSmiles"]
         datamap = df.set_index("smiles").to_dict("index")
         _copy_mapping_from_datamap(self.reaction_tree, datamap)
 

rxnutils/routes/readers.py

@@ -106,7 +106,8 @@ def reaction_dataframe2routes(subdata):
         return route
 
     metadata_columns = metadata_columns or []
-    return data.groupby(group_by).apply(reaction_dataframe2routes)
+    grouped = data.groupby(group_by)[data.columns.to_list()]
+    return grouped.apply(reaction_dataframe2routes)
 
 
 def reactions2route(reactions: Sequence[str], metadata: Sequence[Dict[str, Any]] = None) -> SynthesisRoute: