added functionnality to extract libraires and packages in R and python in...

added functionnality to extract libraires and packages in R and python in process scripts -> both internal and external

src/bioflowinsighterror.py

@@ -53,6 +53,7 @@ class BioFlowInsightError(Exception):
 #* [7] -> Tuple with emit (ch1, ch2) = emit.out 
 #* [9] -> Tuple with call (ch1, ch2) = wf()
 #* [11] -> Failed to extract the operation or call at the line x. Try rewriting it in a simplified version.
+#* [13] -> Multiple scripts with the same name were defined in the source code -> don't know which one to extract then when calling 'get_external_scripts_code'
             
         
 

src/outils.py

@@ -846,4 +846,49 @@ def check_if_element_in_tab_rocrate(tab, id):
     for ele in tab:
         if(ele["@id"]==id):
             return True
-    return False
\ No newline at end of file
+    return False
+
+
+#Function that parses python script and extracts the packages which are imported 
+def get_python_packages(script):
+    packages = []
+    #Examples that i need to consider: 
+    # from fibo import *
+    # from sound.effects.echo import echofilter
+    # import fibo
+    # import fibo, sys
+    # import sound.effects.echo
+    # import numpy as np
+
+    #STEP1
+    patterns_from = [r"fr(om)\s+(\w+)\s+import.+",
+                     r"from\s+((\w+)(\.\w+)+)\s+import.+",]
+    #First step is to extract the packages which are imported from the from and then removing them from the string
+    froms = []
+    for pattern in patterns_from:
+        for match in re.finditer(pattern, script):
+            packages.append(match.group(2))
+            froms.append(match.group(0))
+    for f in froms:
+        script = script.replace(f, "")
+
+    #STEP2
+    #Remove the rest of the 'simple' imports
+    def remove_commas(text):
+        tab = text.split(',')
+        words = []
+        for t in tab:
+            words.append(t.strip())
+        return words
+    for match in re.finditer(r"import\s+(\w+(\s*\,\s*\w+)+|(\w+))", script):
+        packages+= remove_commas(match.group(1))
+
+    return packages
+
+
+#Function that parses R script and extracts the libraries which are loaded 
+def get_R_libraries(script):
+    libraries = []
+    for match in re.finditer(r"library\s*\(\s*(\w+)\s*\)", script):
+        libraries.append(match.group(1))
+    return libraries
\ No newline at end of file

src/process.py

@@ -3,7 +3,7 @@ import glob
 
 from .code_ import Code
 from .nextflow_building_blocks import Nextflow_Building_Blocks
-from .outils import remove_jumps_inbetween_parentheses, remove_jumps_inbetween_curlies, sort_and_filter, get_dico_from_tab_from_id, check_if_element_in_tab_rocrate
+from .outils import remove_jumps_inbetween_parentheses, remove_jumps_inbetween_curlies, sort_and_filter, get_dico_from_tab_from_id, check_if_element_in_tab_rocrate, get_python_packages, get_R_libraries
 from .bioflowinsighterror import BioFlowInsightError
 
 from . import constant
@@ -67,20 +67,65 @@ class Process(Nextflow_Building_Blocks):
         #print(workflow_directory)
         #import os
         #print(os.getcwd(), self.origin.get_address(), self.get_workflow_address())
+        scripts = []
 
         for call in calls:
             #Check first if the file is in the bin
             file = glob.glob(f'{self.get_workflow_address()}/bin/**/{call}', recursive=True)
             if(len(file)>1):
-                print(file)
-                print("More than one file found!")
+                raise BioFlowInsightError(f"More than one script named '{call}' in the workflow source code bin, don't know which one to use when using the process '{self.get_name()}'", num = 13, origin=self)
             #If not we search again
-            if(len(file)==0):
+            elif(len(file)==0):
                 file = glob.glob(f'{self.get_workflow_address()}/**/{call}', recursive=True)
                 if(len(file)>1):
-                    print(file)
-                    print("More than one file found!")
+                    raise BioFlowInsightError(f"More than one script named '{call}' in the workflow source code, don't know which one to use when using the process '{self.get_name()}'", num = 13, origin=self)
                 
+            for f in file:
+                with open(f, 'r') as s:
+                    scripts.append(s.read())
+        
+        return scripts
+    
+
+    def get_python_packages_imported_internal_script(self):
+        packages = []
+        packages+= get_python_packages(self.get_script_code())
+        return packages
+    
+    def get_python_packages_imported_external_scripts(self):
+        packages = []
+        for s in self.get_external_scripts_code():
+            packages+= get_python_packages(s)
+        return packages
+
+    #This methods checks the script and the external script calls for python packages imports
+    def get_python_packages_imported(self):
+        packages = []
+        packages+= self.get_python_packages_imported_internal_script()
+        packages+= self.get_python_packages_imported_external_scripts()
+        return packages
+
+
+    def get_R_libraries_loaded_internal_script(self):
+        libraries = []
+        libraries+= get_R_libraries(self.get_script_code())
+        return libraries
+    
+    def get_R_libraries_loaded_external_scripts(self):
+        libraries = []
+        for s in self.get_external_scripts_code():
+            libraries+= get_R_libraries(s)
+        return libraries
+
+    #This methods checks the script and the external script calls for python packages imports
+    def get_R_libraries_loaded(self):
+        libraries = []
+        libraries+= self.get_R_libraries_loaded_internal_script()
+        libraries+= self.get_R_libraries_loaded_external_scripts()
+        return libraries
+            
+    #TODO -> do the same with Ruby/PErl scripts
+