Refactor reading of pcomp model

scripts/get_indices_of_model_parameters_in_ode.py

@@ -8,79 +8,7 @@ from loggingUtility import set_log_level, add_log_level_to_argument_parser
 
 from src.read_dataset import read_datasets_from_directory
 from src.Dataset import ModelType, Dataset
-
-
-from dataclasses import dataclass
-
-@dataclass
-class PCOMPModel:
-    variables: list[str]
-    functions: list[str]
-    
-
-def remove_comments(modelFileContent: str) -> str:
-    modelFileLines = modelFileContent.splitlines()
-    modelFileLinesWithoutComments = [line for line in modelFileLines if not line.startswith(("c", "C"))]
-    return '\n'.join(modelFileLinesWithoutComments)
-
-
-def get_new_pcomp_special_line_pattern() -> re.Pattern:
-    return re.compile(r"^\*", flags=re.IGNORECASE|re.MULTILINE)
-
-
-def read_pcomp_block(modelFileContent: str,  blockStartIndex: int) -> str:
-    findNewSpecialLinePattern = get_new_pcomp_special_line_pattern()
-
-    blockEndMatch = findNewSpecialLinePattern.search(modelFileContent[blockStartIndex + 1:])
-    if blockEndMatch:
-        blockEndIndex = blockStartIndex + blockEndMatch.end()
-    else:
-        blockEndIndex = -1
-
-    return modelFileContent[blockStartIndex:blockEndIndex]
-
-
-def read_pcomp_model_infos(modelFile: Path) -> PCOMPModel:
-    modelFileContent = modelFile.read_text()
-    modelFileContent = remove_comments(modelFileContent)
-
-    variables: list[str] = []
-    functions: list[str] = []
-
-    indexSets: list[tuple[str, str]] = []
-
-    findNewSpecialLinePattern = get_new_pcomp_special_line_pattern()
-    findNewSetOfIndicesPattern = re.compile(r"^(\*\s*SET OF INDICES)", flags=re.IGNORECASE|re.MULTILINE)
-    findNewVariablePattern = re.compile(r"^(\*\s*VARIABLE)", flags=re.IGNORECASE|re.MULTILINE)
-    findNewFunctionPattern = re.compile(r"^(\*\s*FUNCTION)", flags=re.IGNORECASE|re.MULTILINE)
-        
-    for specialLineStartMatch in findNewSpecialLinePattern.finditer(modelFileContent):
-        if findNewSetOfIndicesPattern.match(modelFileContent[specialLineStartMatch.start():]):
-            indexSetBlock = read_pcomp_block(modelFileContent, specialLineStartMatch.start())
-            indexSetBlock = findNewSetOfIndicesPattern.sub("", indexSetBlock)
-            indexSetBlock = indexSetBlock.strip()
-            indexSetBlock = indexSetBlock.strip('\n')
-            indexSetBlock = indexSetBlock.strip()
-            for line in indexSetBlock.splitlines():
-                splitIndexSet = line.split('=', maxsplit=1)
-                assert len(splitIndexSet) == 2
-                indexSets.append((splitIndexSet[0].strip(), splitIndexSet[1].strip()))
-            continue
-
-        if findNewFunctionPattern.match(modelFileContent[specialLineStartMatch.start():]):
-            functions.append(read_pcomp_block(modelFileContent, specialLineStartMatch.start()))
-            continue
-
-        if findNewVariablePattern.match(modelFileContent[specialLineStartMatch.start():]):
-            variablesBlock = read_pcomp_block(modelFileContent, specialLineStartMatch.start())
-            variablesBlock = findNewVariablePattern.sub("", variablesBlock)
-            for line in variablesBlock.splitlines():
-                if any([indexSet[0] in line for indexSet in indexSets]):
-                    variables.append(line.strip())
-                else:
-                    variables.extend([var for var in line.replace(' ', '').split(',') if var])
-            
-    return PCOMPModel(variables, functions)
+from src.read_pcomp_model import read_pcomp_model
 
 
 def find_whole_word(targetString: str, sourceString: str, ignoreCase: bool = True) -> bool:
@@ -100,7 +28,7 @@ def any_function_contains_parameter(functions: list[str], parameterName: str) ->
 
 
 def get_indices_of_model_parameters_in_ode(modelFileName: Path, dataset: Dataset) -> list[tuple[int, str]]:
-    pcompModelInfos = read_pcomp_model_infos(modelFileName)
+    pcompModelInfos = read_pcomp_model(modelFileName)
 
     numberOfTimeVariables = 1 
     numberOfConcentrationVariables = 1 if dataset.numberOfConcentrations > 0 else 0
@@ -110,13 +38,12 @@ def get_indices_of_model_parameters_in_ode(modelFileName: Path, dataset: Dataset
 
     if len(pcompModelInfos.functions) != expectedNumberOfFunctions:
         logging.warning(f"{dataset.name}: wrong number of functions detected, expected: {expectedNumberOfFunctions}, observed: {len(pcompModelInfos.functions)}. Return no detected parameters.")
-        modelParametersInOde: list[tuple[int, str]] = []
+        return []
+    elif len(pcompModelInfos.variables) != expectedNumberOfVariables:
+        logging.warning(f"{dataset.name}: wrong number of variables detected, expected: {expectedNumberOfVariables}, observed: {len(pcompModelInfos.variables)}. Return no detected parameters")
+        return []
     else:
-        if len(pcompModelInfos.variables) == expectedNumberOfVariables:
-            modelParameterNames = pcompModelInfos.variables[0:dataset.numberOfModelParameters]
-        else:
-            logging.warning(f"{dataset.name}: wrong number of variables detected, expected: {expectedNumberOfVariables}, observed: {len(pcompModelInfos.variables)}. Use model parameter names from dataset.")
-            modelParameterNames = [info.name for info in dataset.modelParameterInfos]
+        modelParameterNames = pcompModelInfos.variables[0:dataset.numberOfModelParameters]
 
         odeFunctions = pcompModelInfos.functions[0:dataset.numberOfOdes]
         modelParametersInOde = [(index, parameter) for index, parameter in enumerate(modelParameterNames) 

scripts/src/read_pcomp_model.py

+from dataclasses import dataclass
+from pathlib import Path
+import re
+
+
+@dataclass
+class PCOMPModel:
+    variables: list[str]
+    functions: list[str]
+    
+
+def remove_comments(modelFileContent: str) -> str:
+    modelFileLines = modelFileContent.splitlines()
+    modelFileLinesWithoutComments = [line for line in modelFileLines if not line.startswith(("c", "C"))]
+    return '\n'.join(modelFileLinesWithoutComments)
+
+
+def get_new_pcomp_special_line_pattern() -> re.Pattern:
+    return re.compile(r"^\*", flags=re.IGNORECASE|re.MULTILINE)
+
+
+def read_pcomp_block(modelFileContent: str,  blockStartIndex: int) -> str:
+    findNewSpecialLinePattern = get_new_pcomp_special_line_pattern()
+
+    blockEndMatch = findNewSpecialLinePattern.search(modelFileContent[blockStartIndex + 1:])
+    if blockEndMatch:
+        blockEndIndex = blockStartIndex + blockEndMatch.end()
+    else:
+        blockEndIndex = -1
+
+    return modelFileContent[blockStartIndex:blockEndIndex]
+
+
+def read_pcomp_model(modelFile: Path) -> PCOMPModel:
+    modelFileContent = modelFile.read_text()
+    modelFileContent = remove_comments(modelFileContent)
+
+    variables: list[str] = []
+    functions: list[str] = []
+
+    indexSets: list[tuple[str, str]] = []
+
+    findNewSpecialLinePattern = get_new_pcomp_special_line_pattern()
+    findNewSetOfIndicesPattern = re.compile(r"^(\*\s*SET OF INDICES)", flags=re.IGNORECASE|re.MULTILINE)
+    findNewVariablePattern = re.compile(r"^(\*\s*VARIABLE)", flags=re.IGNORECASE|re.MULTILINE)
+    findNewFunctionPattern = re.compile(r"^(\*\s*FUNCTION)", flags=re.IGNORECASE|re.MULTILINE)
+        
+    for specialLineStartMatch in findNewSpecialLinePattern.finditer(modelFileContent):
+        if findNewSetOfIndicesPattern.match(modelFileContent[specialLineStartMatch.start():]):
+            indexSetBlock = read_pcomp_block(modelFileContent, specialLineStartMatch.start())
+            indexSetBlock = findNewSetOfIndicesPattern.sub("", indexSetBlock)
+            indexSetBlock = indexSetBlock.strip()
+            indexSetBlock = indexSetBlock.strip('\n')
+            indexSetBlock = indexSetBlock.strip()
+            for line in indexSetBlock.splitlines():
+                splitIndexSet = line.split('=', maxsplit=1)
+                assert len(splitIndexSet) == 2
+                indexSets.append((splitIndexSet[0].strip(), splitIndexSet[1].strip()))
+            continue
+
+        if findNewFunctionPattern.match(modelFileContent[specialLineStartMatch.start():]):
+            functions.append(read_pcomp_block(modelFileContent, specialLineStartMatch.start()))
+            continue
+
+        if findNewVariablePattern.match(modelFileContent[specialLineStartMatch.start():]):
+            variablesBlock = read_pcomp_block(modelFileContent, specialLineStartMatch.start())
+            variablesBlock = findNewVariablePattern.sub("", variablesBlock)
+            for line in variablesBlock.splitlines():
+                if any([indexSet[0] in line for indexSet in indexSets]):
+                    variables.append(line.strip())
+                else:
+                    variables.extend([var for var in line.replace(' ', '').split(',') if var])
+            
+    return PCOMPModel(variables, functions)
\ No newline at end of file