Update -> ro-crate

src/ro_crate.py

-
+import json
+import glob
+import os
 
 class RO_Crate:
     def __init__(self, workflow):
-        self.nextflow_file = workflow
-        self.directroy = '/'.join(workflow.get_file_address().split('/')[:-1])
+        self.workflow = workflow
+        self.directory = '/'.join(workflow.get_file_address().split('/')[:-1])
+        self.files = []
         self.dico = {}
 
+    def get_files(self):
+        self.files = glob.glob(f'{self.directory}/**/*.*', recursive=True)
+        tab_files = []
+        for file in self.files:
+            tab_files.append({"@id":file[len(self.directory)+1:]})
+        return tab_files
+
     def initialise_dico(self):
-        None
+        self.dico["@context"] = "https://w3id.org/ro/crate/1.1/context"
+        self.dico["@graph"] = []
+        #GENERAL
+        general = {}
+        general["@id"] = f"ro-crate-metadata-{self.workflow.get_name()}.json"
+        general["@type"] = "CreativeWork"
+        general["about"] = {"@id":"./"}
+        general["conformsTo"] = [{"@id":"https://w3id.org/ro/crate/1.1"}
+                                 #, {"@id":"https://w3id.org/workflowhub/workflow-ro-crate/1.0"}#This description does not conform 
+                                 ]
+        self.dico["@graph"].append(general)
+        #ROOT
+        root = {}
+        root["@id"] = "./"
+        root["@type"] = "Dataset"
+        root["name"] = self.workflow.get_name()
+        root["datePublished"] = self.workflow.get_datePublished()
+        root["description"] = self.workflow.get_description()
+        root["mainEntity"] = {"@id": self.workflow.get_main_file(),
+                              "@type":["File", "SoftwareSourceCode"]} #We do not consider a File as a "ComputationalWorkflow" since multiple (sub)workflows can be defined in a same file
+        root["license"] = {"@id":self.workflow.get_license()}
+        authors = self.workflow.get_authors()
+        tab_authors = []
+        for author in authors:
+            tab_authors.append({"@id":author["@id"]})
+        root["author"] = tab_authors
+        root["maintainer"] = tab_authors #Right now i'm assuming that all the authors are maintainers
+        files = self.get_files()
+        tab_files = []
+        for file in files:
+            tab_files.append({"@id":file["@id"]})
+        root["hasPart"] = tab_files
+        root["publisher"] = {"@id":self.workflow.get_publisher()}
+        #subjectOf TODO
+        root["subjectOf"] = None
+        root["creativeWorkStatus"] = self.workflow.get_creativeWorkStatus()
+        root["@version"] = self.workflow.get_version()
+        root["keywords"] = self.workflow.get_keywords()
+        root["producer"] = self.workflow.get_producer()
+        self.dico["@graph"].append(root)
+
+    #TODO 
+    def get_programming_language(self, file):
+        if(file[-3:]==".nf"):
+            return "https://w3id.org/workflowhub/workflow-ro-crate#nextflow"
+        return None
+    
+    def get_contentSize(self, file):
+        file_stats = os.stat(file)
+        return file_stats.st_size/1e3
+    
+    #TODO
+    def get_dateCreated(self, file):
+        return "TODO"
+    
+    #TODO
+    def get_dateModified(self, file):
+        return "TODO"
+    
+    #TODO
+    def get_url(self, file):
+        return "TODO"
+    
+    #TODO
+    def get_creators(self, file):
+        return [{"@id": "George"}]
+    
+    #TODO
+    def get_types(self, file):
+        types = ["File"]
+        if(file[-3:]==".nf"):
+            types.append("SoftwareSourceCode")
+        return types
+        
+
+    def initialise_file(self, file):
+        key = file[len(self.directory)+1:]
+        dico = {}
+        dico["@id"] = key
+        dico["name"] = key
+        dico["@type"] = self.get_types(file)
+        dico["programmingLanguage"] = {"@id":self.get_programming_language(file)}
+        dico["contentSize"] = self.get_contentSize(file)
+        dico["dateCreated"] = self.get_dateCreated(file)
+        dico["dateModified"] = self.get_dateModified(file)
+        dico["url"] = self.get_url(file)
+        creators = self.get_creators(file)
+        dico["creator"] = []
+        for creator in creators:
+            dico["creator"].append({"@id": creator["@id"]})
+        dico["isPartOf"] = []
+        dico["hasPart"] = []
+        self.dico["@graph"].append(dico)    
 
     def initialise(self):
         self.initialise_dico()
-        print("i'm initialised")
\ No newline at end of file
+        for file in self.files:
+            self.initialise_file(file)
+
+        with open(f"{self.workflow.get_output_dir()}/ro-crate-metadata-{self.workflow.get_name()}.json", 'w') as output_file :
+            json.dump(self.dico, output_file, indent=2)
\ No newline at end of file

src/workflow.py

@@ -8,6 +8,57 @@ class Workflow:
         self.nextflow_file = Nextflow_File(file, duplicate = duplicate, display_info = display_info)
         self.rocrate = None
 
+    #TODO
+    def get_name(self):
+        return "TODO"
+    
+    #TODO
+    def get_datePublished(self):
+        return "TODO"
+    
+    #TODO
+    def get_description(self):
+        return "TODO"
+    
+    #TODO
+    def get_main_file(self):
+        return "TODO"
+    
+    #TODO
+    def get_license(self):
+        return "TODO"
+    
+    #TODO
+    def get_main_license(self):
+        return "TODO"
+    
+    #TODO
+    def get_creativeWorkStatus(self):
+        return "TODO"
+    
+    #TODO
+    def get_version(self):
+        return "TODO"
+    
+    #TODO
+    def get_authors(self):
+        return [{"@id": "George"}]
+    
+    #TODO
+    #Need to follow this format : "rna-seq, nextflow, bioinformatics, reproducibility, workflow, reproducible-research, bioinformatics-pipeline"
+    def get_keywords(self):
+        return "TODO"
+    
+    #TODO
+    def get_producer(self):
+        return "TODO"
+    
+    #TODO
+    def get_publisher(self):
+        return "TODO"
+    
+    def get_output_dir(self):
+        return self.nextflow_file.get_output_dir()
 
     def get_file_address(self):
         return self.nextflow_file.get_file_address()