Dateien und Funktionen ausgelagert

verarbeitung/Processing.py → verarbeitung/construct_new_graph/Processing.py

@@ -20,8 +20,10 @@ from os import error
 sys.path.append("../")
 
 from input.publication import Publication
-from get_pub_from_input import get_pub
-from json_demo import output_to_json
+from verarbeitung.get_pub_from_input import get_pub
+from .export_to_json import output_to_json
+from .add_citations_rec import add_citations, create_global_lists_cit
+from .add_references_rec import add_references, create_global_lists_ref
 
 
 def initialize_nodes_list(doi_input_list, search_depth_max, search_height_max, test_var):
@@ -65,12 +67,12 @@ def initialize_nodes_list(doi_input_list, search_depth_max, search_height_max, t
 
         # inserts references as publication objects into list and 
         # inserts first depth references into nodes/edges if maximum search depth > 0
-        for reference in create_graph_structure_references(pub, 0, search_depth_max, test_var):
+        for reference in create_global_lists_ref(nodes, edges, pub, 0, search_depth_max, test_var):
             references_pub_obj_list.append(reference)
 
         # inserts citations as publication objects into list and 
         # inserts first height citations into nodes if maximum search height > 0
-        for citation in create_graph_structure_citations(pub, 0, search_height_max, test_var):
+        for citation in create_global_lists_cit(nodes, edges, pub, 0, search_height_max, test_var):
             citations_pub_obj_list.append(citation)
 
     return(references_pub_obj_list, citations_pub_obj_list)
@@ -98,147 +100,6 @@ def complete_inner_edges(test_var):
                         edges.append([node.doi_url,reference])
 
 
-   
-def create_graph_structure_references(pub, search_depth, search_depth_max, test_var):
-    '''
-        :param pub:                 publication which references will be added
-        :type pub:                  Publication
-
-        :param search_depth:        current depth to search for references
-        :type search_depth:         int
-
-        :param search_depth_max:    maximum depth to search for references
-        :type search_depth_max:     int
-
-        :param test_var:            variable to differenciate between test and url call
-        :type test_var:             boolean
-
-        adds a node for every referenced publication unknown
-        adds edges to added references
-    '''
-
-    references_pub_obj_list = []
-    for reference in pub.references: #iterates over the references of the considered paper
-        not_in_nodes = True #boolean Value to ensure that there will be no dublicates in the set of nodes
-        for node in nodes: #iterates over all nodes in set of nodes #
-            if (reference == node.doi_url): #determines that the node already exists
-                not_in_nodes = False #boolean false --> node will not be created
-                break
-        if (not_in_nodes): #checks that there is no node with this doi 
-            if (search_depth < search_depth_max): #checks that the recursion step is smaller than the limit
-                reference_pub_obj = get_pub(reference, test_var)
-                if (type(reference_pub_obj) != Publication):
-                    print(pub)
-                    continue 
-
-                reference_pub_obj.group = "depth"
-                nodes.append(reference_pub_obj) # appends the object to the set of nodes
-                edges.append([pub.doi_url,reference_pub_obj.doi_url]) #appends the edge to the set of edges
-                references_pub_obj_list.append(reference_pub_obj) #appends the node to the set of references
-
-        # adds edge only if citation already exists           
-        elif [pub.doi_url,reference] not in edges:
-            edges.append([pub.doi_url,reference])  
-    return references_pub_obj_list 
-
-
-def process_references_rec(references_pub_obj_list, search_depth, search_depth_max, test_var):  
-    '''
-        :param references_pub_obj_list: list of publications which references will be added
-        :type references_pub_obj_list:  List[Publication]
-
-        :param search_depth:            current depth to search for references
-        :type search_depth:             int
-
-        :param search_depth_max:        maximum depth to search for references
-        :type search_depth_max:         int
-
-        :param test_var:                variable to differenciate between test and url call
-        :type test_var:                 boolean
-
-        recursive function to implement height-first-search on references
-    '''
-
-    # adds next level to nodes/edges
-    for pub in references_pub_obj_list:
-        new_reference_pub_obj_list = create_graph_structure_references(pub, search_depth, search_depth_max, test_var)    
-
-        # If the maximum height has not yet been reached, calls function recursivly with increased height     
-        if (search_depth < search_depth_max):
-            process_references_rec(new_reference_pub_obj_list, search_depth+1, search_depth_max, test_var)
-
-    
-  
-def create_graph_structure_citations(pub, search_height, search_height_max, test_var):
-    '''
-        :param pub:                 publication which citations will be added
-        :type pub:                  Publication
-
-        :param search_height:       current height to search for citations
-        :type search_height_max:    int
-
-        :param search_height_max:   maximum height to search for citations
-        :type search_height_max:    int
-
-        :param test_var:            variable to differenciate between test and url call
-        :type test_var:             boolean
-
-        adds a node for every citing publication unknown
-        adds edges to added citations
-    '''
-
-    citations_pub_obj_list = []
-    for citation in pub.citations:
-        not_in_nodes = True
-        for node in nodes: # checks every citation for duplication 
-            if (citation == node.doi_url):
-                not_in_nodes = False
-                break
-        if (not_in_nodes):
-            if (search_height < search_height_max): #checks if its a test and chooses input function accordingly
-                citation_pub_obj = get_pub(citation, test_var)
-                if (type(citation_pub_obj) != Publication):
-                    print(pub)
-                    continue 
-
-                citation_pub_obj.group = "height"
-                nodes.append(citation_pub_obj)
-                edges.append([citation_pub_obj.doi_url,pub.doi_url])
-                citations_pub_obj_list.append(citation_pub_obj)
-
-        # adds only edge if citation already exists         
-        elif [citation,pub.doi_url] not in edges:
-            edges.append([citation,pub.doi_url])   
-    return citations_pub_obj_list
-
-
-def process_citations_rec(citations_pub_obj_list, search_height, search_height_max, test_var):  
-    '''
-        :param citations_pub_obj_list:  list of publications which citations will be added
-        :type citations_pub_obj_list:   List[Publication]
-
-        :param search_height:       current height to search for citations
-        :type search_height_max:    int
-
-        :param search_height_max:   maximum height to search for citations
-        :type search_height_max:    int
-
-        :param test_var:            variable to differenciate between test and url call
-        :type test_var:             boolean
-
-        recursive function to implement depth-first-search on citations
-    '''
-
-    # adds next level to nodes/edges
-    for pub in citations_pub_obj_list:
-        new_citation_pub_obj_list = create_graph_structure_citations(pub, search_height, search_height_max, test_var)   
-
-        # If the maximum height has not yet been reached, calls function recursivly with increased height 
-        if (search_height < search_height_max):
-            process_citations_rec(new_citation_pub_obj_list, search_height+1, search_height_max, test_var)
-
-
-
 def process_main(doi_input_list, search_height, search_depth, test_var = False):
     '''
         :param doi_input_list:  input list of doi from UI
@@ -278,8 +139,8 @@ def process_main(doi_input_list, search_height, search_depth, test_var = False):
     references_obj_list, citations_obj_list = initialize_nodes_list(doi_input_list,search_depth, search_height, test_var)
 
     # function calls to begin recursive processing up to max depth/height
-    process_citations_rec(citations_obj_list, 1, search_height, test_var)
-    process_references_rec(references_obj_list, 1, search_depth, test_var)
+    add_citations(nodes, edges, citations_obj_list, 1, search_height, test_var)
+    add_references(nodes, edges, references_obj_list, 1, search_depth, test_var)
 
     # adds edges between reference group and citation group of known publications
     complete_inner_edges(test_var)

verarbeitung/construct_new_graph/add_citations_rec.py

+# -*- coding: utf-8 -*-
+"""
+Functions to add citations recursivly for multiple ACS/Nature journals
+
+"""
+
+__authors__ = "Donna Löding, Alina Molkentin, Xinyi Tang, Judith Große, Malte Schokolowski"
+__email__ = "cis-project2021@zbh.uni-hamburg.de"
+__status__ = "Production"
+#__copyright__ = ""
+#__credits__ = ["", "", "", ""]
+#__license__ = ""
+#__version__ = ""
+#__maintainer__ = ""
+
+
+import sys  
+from pathlib import Path
+from os import error
+sys.path.append("../../")
+
+from input.publication import Publication
+from verarbeitung.get_pub_from_input import get_pub
+
+def create_global_lists_cit(input_nodes, input_edges, pub, search_height, search_height_max, test_var):
+    '''
+        :param input_nodes:         list of nodes from Processing
+        :type input_nodes:          List[Publication]
+
+        :param input_edges:         list of edges from Processing
+        :type input_edges:          List[String, String]
+
+        :param pub:                 Publication which citations will be added
+        :type pub:                  Publication
+
+        :param search_height:       current height to search for citations
+        :type search_height_max:    int
+
+        :param search_height_max:   maximum height to search for citations
+        :type search_height_max:    int
+
+        :param test_var:            variable to differenciate between test and url call
+        :type test_var:             boolean
+
+        function to create nodes and edges and call create_graph_structure_citations
+    '''
+
+    global nodes, edges
+    nodes = input_nodes
+    edges = input_edges
+
+    return create_graph_structure_citations(pub, search_height, search_height_max, test_var)
+
+
+def create_graph_structure_citations(pub, search_height, search_height_max, test_var):
+    '''
+        :param pub:                 publication which citations will be added
+        :type pub:                  Publication
+
+        :param search_height:       current height to search for citations
+        :type search_height_max:    int
+
+        :param search_height_max:   maximum height to search for citations
+        :type search_height_max:    int
+
+        :param test_var:            variable to differenciate between test and url call
+        :type test_var:             boolean
+
+        adds a node for every citing publication unknown
+        adds edges to added citations
+    '''
+
+    citations_pub_obj_list = []
+    for citation in pub.citations:
+        not_in_nodes = True
+        for node in nodes: # checks every citation for duplication 
+            if (citation == node.doi_url):
+                not_in_nodes = False
+                break
+        if (not_in_nodes):
+            if (search_height < search_height_max): #checks if its a test and chooses input function accordingly
+                citation_pub_obj = get_pub(citation, test_var)
+                if (type(citation_pub_obj) != Publication):
+                    print(pub)
+                    continue 
+
+                citation_pub_obj.group = "height"
+                nodes.append(citation_pub_obj)
+                edges.append([citation_pub_obj.doi_url,pub.doi_url])
+                citations_pub_obj_list.append(citation_pub_obj)
+
+        # adds only edge if citation already exists         
+        elif [citation,pub.doi_url] not in edges:
+            edges.append([citation,pub.doi_url])   
+    return citations_pub_obj_list
+
+
+def process_citations_rec(citations_pub_obj_list, search_height, search_height_max, test_var):  
+    '''
+        :param citations_pub_obj_list:  list of publications which citations will be added
+        :type citations_pub_obj_list:   List[Publication]
+
+        :param search_height:       current height to search for citations
+        :type search_height_max:    int
+
+        :param search_height_max:   maximum height to search for citations
+        :type search_height_max:    int
+
+        :param test_var:            variable to differenciate between test and url call
+        :type test_var:             boolean
+
+        recursive function to implement depth-first-search on citations
+    '''
+
+    # adds next level to nodes/edges
+    for pub in citations_pub_obj_list:
+        new_citation_pub_obj_list = create_graph_structure_citations(pub, search_height, search_height_max, test_var)   
+
+        # If the maximum height has not yet been reached, calls function recursivly with increased height 
+        if (search_height < search_height_max):
+            process_citations_rec(new_citation_pub_obj_list, search_height+1, search_height_max, test_var)
+
+
+def add_citations(input_nodes, input_edges, citations_pub_obj_list, search_height, search_height_max, test_var):
+    '''
+        :param input_nodes:             list of nodes from Processing
+        :type input_nodes:              List[Publication]
+
+        :param input_edges:             list of edges from Processing
+        :type input_edges:              List[String, String]
+
+        :param citations_pub_obj_list:  list of publications which citations will be added
+        :type citations_pub_obj_list:   List[Publication]
+
+        :param search_height:       current height to search for citations
+        :type search_height_max:    int
+
+        :param search_height_max:   maximum height to search for citations
+        :type search_height_max:    int
+
+        :param test_var:            variable to differenciate between test and url call
+        :type test_var:             boolean
+
+        function to call recursive depth-first-search of citations
+    '''
+    global nodes, edges
+    nodes = input_nodes
+    edges = input_edges
+
+    process_citations_rec(citations_pub_obj_list, search_height, search_height_max, test_var)
+    return(nodes, edges)
\ No newline at end of file

verarbeitung/construct_new_graph/add_references_rec.py

+# -*- coding: utf-8 -*-
+"""
+Functions to add references recursivly for multiple ACS/Nature journals
+
+"""
+
+__authors__ = "Donna Löding, Alina Molkentin, Xinyi Tang, Judith Große, Malte Schokolowski"
+__email__ = "cis-project2021@zbh.uni-hamburg.de"
+__status__ = "Production"
+#__copyright__ = ""
+#__credits__ = ["", "", "", ""]
+#__license__ = ""
+#__version__ = ""
+#__maintainer__ = ""
+
+
+import sys  
+from pathlib import Path
+from os import error
+sys.path.append("../../")
+
+from input.publication import Publication
+from verarbeitung.get_pub_from_input import get_pub
+
+def create_global_lists_ref(input_nodes, input_edges, pub, search_depth, search_depth_max, test_var):
+    '''
+        :param input_nodes:         list of nodes from Processing
+        :type input_nodes:          List[Publication]
+
+        :param input_edges:         list of edges from Processing
+        :type input_edges:          List[String, String]
+
+        :param pub:                 Publication which references will be added
+        :type pub:                  Publication
+
+        :param search_height:       current height to search for references
+        :type search_height_max:    int
+
+        :param search_height_max:   maximum height to search for references
+        :type search_height_max:    int
+
+        :param test_var:            variable to differenciate between test and url call
+        :type test_var:             boolean
+
+        function to create nodes and edges and call create_graph_structure_references 
+    '''
+
+    global nodes, edges
+    nodes = input_nodes
+    edges = input_edges
+
+    return create_graph_structure_references(pub, search_depth, search_depth_max, test_var)
+
+
+def create_graph_structure_references(pub, search_depth, search_depth_max, test_var):
+    '''
+        :param pub:                 publication which references will be added
+        :type pub:                  Publication
+
+        :param search_depth:        current depth to search for references
+        :type search_depth:         int
+
+        :param search_depth_max:    maximum depth to search for references
+        :type search_depth_max:     int
+
+        :param test_var:            variable to differenciate between test and url call
+        :type test_var:             boolean
+
+        adds a node for every referenced publication unknown
+        adds edges to added references
+    '''
+
+    references_pub_obj_list = []
+    for reference in pub.references: #iterates over the references of the considered paper
+        not_in_nodes = True #boolean Value to ensure that there will be no dublicates in the set of nodes
+        for node in nodes: #iterates over all nodes in set of nodes #
+            if (reference == node.doi_url): #determines that the node already exists
+                not_in_nodes = False #boolean false --> node will not be created
+                break
+        if (not_in_nodes): #checks that there is no node with this doi 
+            if (search_depth < search_depth_max): #checks that the recursion step is smaller than the limit
+                reference_pub_obj = get_pub(reference, test_var)
+                if (type(reference_pub_obj) != Publication):
+                    print(pub)
+                    continue 
+
+                reference_pub_obj.group = "depth"
+                nodes.append(reference_pub_obj) # appends the object to the set of nodes
+                edges.append([pub.doi_url,reference_pub_obj.doi_url]) #appends the edge to the set of edges
+                references_pub_obj_list.append(reference_pub_obj) #appends the node to the set of references
+
+        # adds edge only if citation already exists           
+        elif [pub.doi_url,reference] not in edges:
+            edges.append([pub.doi_url,reference])  
+    return references_pub_obj_list 
+
+
+def process_references_rec(references_pub_obj_list, search_depth, search_depth_max, test_var):  
+    '''
+        :param references_pub_obj_list: list of publications which references will be added
+        :type references_pub_obj_list:  List[Publication]
+
+        :param search_depth:            current depth to search for references
+        :type search_depth:             int
+
+        :param search_depth_max:        maximum depth to search for references
+        :type search_depth_max:         int
+
+        :param test_var:                variable to differenciate between test and url call
+        :type test_var:                 boolean
+
+        recursive function to implement height-first-search on references
+    '''
+
+    # adds next level to nodes/edges
+    for pub in references_pub_obj_list:
+        new_reference_pub_obj_list = create_graph_structure_references(pub, search_depth, search_depth_max, test_var)    
+
+        # If the maximum height has not yet been reached, calls function recursivly with increased height     
+        if (search_depth < search_depth_max):
+            process_references_rec(new_reference_pub_obj_list, search_depth+1, search_depth_max, test_var)
+
+
+def add_references(input_nodes, input_edges, references_pub_obj_list, search_height, search_height_max, test_var):
+    '''
+        :param input_nodes:             list of nodes from Processing
+        :type input_nodes:              List[Publication]
+
+        :param input_edges:             list of edges from Processing
+        :type input_edges:              List[String, String]
+
+        :param references_pub_obj_list:  list of publications which references will be added
+        :type references_pub_obj_list:   List[Publication]
+
+        :param search_height:       current height to search for references
+        :type search_height_max:    int
+
+        :param search_height_max:   maximum height to search for references
+        :type search_height_max:    int
+
+        :param test_var:            variable to differenciate between test and url call
+        :type test_var:             boolean
+
+        function to call recursive depth-first-search of references
+    '''
+    global nodes, edges
+    nodes = input_nodes
+    edges = input_edges
+
+    process_references_rec(references_pub_obj_list, search_height, search_height_max, test_var)
+    return(nodes, edges)
\ No newline at end of file