diff --git a/src/DBfile.py b/src/DBfile.py
index a416c3bebce8b668008df18d4632f5437ad5eb5d..66e5510fcf68740095737f236617f1c2a4724d73 100644
--- a/src/DBfile.py
+++ b/src/DBfile.py
@@ -2,10 +2,19 @@ import json
import glob
import os
import re
-from .ro_crate import RO_Crate
+import ctypes
+from collections import OrderedDict
+
+
+from .ro_crate import RO_Crate
from . import constant
+
+def get_object(address):
+ address = int(re.findall(r"\dx\w+", address)[0], base=16)
+ return ctypes.cast(address, ctypes.py_object).value
+
class DBfile(RO_Crate):
def __init__(self, workflow,
personnal_acces_token = None,
@@ -21,8 +30,6 @@ class DBfile(RO_Crate):
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
@prefix efo: <https://www.ebi.ac.uk/gwas/efotraits/> ."""
- print(self.info_dico_workflow)
- #print(self.log)
def initialise(self):
@@ -44,20 +51,45 @@ class DBfile(RO_Crate):
steps_string.append(f"step{p.get_alias()}")
steps_string = ", ".join(steps_string)
- #TODO update the input and output of the workflow
+ link_dico_processes = self.workflow.get_link_dico_processes()
+ first_processes, last_processes = list(link_dico_processes.keys()), []
+ for p in link_dico_processes:
+ if(link_dico_processes[p]==[]):
+ last_processes.append(get_object(p))
+ else:
+ for p2 in link_dico_processes[p]:
+ try:
+ first_processes.remove(p2)
+ except:
+ None
+ temp = []
+ for p in first_processes:
+ temp.append(get_object(p))
+ first_processes = temp
+
+ inputs = []
+ for p in first_processes:
+ inputs+= p.get_inputs_DBfile()
+ outputs = []
+ for p in last_processes:
+ outputs+= p.get_outputs_DBfile()
+
+
header = f"""### Main workflow
sf:{workflow_name}Workflow rdf:type sf:Workflow ;
- schema:name "{workflow_long_name}" ;
- schema:description "{description}" ;
- schema:creator <{creator}> ;
- schema:maintainer <{maintainer}> ;
- schema:dateCreated "f{date_created}"^^xsd:date ;
- schema:programmingLanguage <https://w3id.org/workflowhub/workflow-ro-crate#nextflow> ;
- schema:step {steps_string} ;
- sf:inputVariable sf:samplesheet, sf:ch_versions, sf:ch_fasta,
- sf:ch_fasta_index, sf:ch_bismark_index, sf:ch_bwameth_index ;
- sf:outputVariable sf:output_bam, sf:output_bai, sf:output_qualimap, sf:output_preseq, sf:output_versions ;
- schema:keywords "{keywords}" ."""
+\tschema:name "{workflow_long_name}" ;
+\tschema:description "{description}" ;
+\tschema:creator <{creator}> ;
+\tschema:maintainer <{maintainer}> ;
+\tschema:dateCreated "f{date_created}"^^xsd:date ;
+\tschema:programmingLanguage <https://w3id.org/workflowhub/workflow-ro-crate#nextflow> ;
+\tschema:step {steps_string} ;"""
+ if(inputs!=[]):
+ header+=f"\n\tsf:inputVariable {', '.join(inputs)} ;"
+ if(outputs!=[]):
+ header+=f"\n\tsf:outputVariable {', '.join(outputs)} ;"
+ header+=f'\n\tschema:keywords "{keywords}" .'
+
self.file_contents+=f"\n\n{header}"
@@ -66,8 +98,21 @@ sf:{workflow_name}Workflow rdf:type sf:Workflow ;
self.file_contents+="""\n\n<https://w3id.org/workflowhub/workflow-ro-crate#nextflow> rdf:type schema:ComputerLanguage ;
rdfs:label "Nextflow" ."""
- print(self.file_contents)
- #TODO ADD Inputs values
- #TODO ADD Outputs value
+ for p in self.workflow.get_processes_called():
+ self.file_contents+="\n\n"+p.get_DBfile_description()
+
+ all_calls = self.workflow.get_workflow_main().get_all_calls_in_subworkflow()
+ for c in all_calls:
+ if(c.get_first_element_called().get_type()=="Subworkflow"):
+ sub = c.get_first_element_called()
+ self.file_contents+="\n\n"+sub.get_DBfile_description(workflow_name)
+
+ #This is to remove multiple lines
+ self.file_contents = "\n\n".join(list(OrderedDict.fromkeys(self.file_contents.split("\n\n"))))
+
+
+ with open(f"{self.workflow.get_output_dir()}/DBfile.ttl", 'w') as output_file :
+ output_file.write(self.file_contents)
+
diff --git a/src/graph.py b/src/graph.py
index a06890ff0de98a98570ca61194711f459af9943c..c03f4f342f66219f9530dc78dbc67c96b3d13c42 100644
--- a/src/graph.py
+++ b/src/graph.py
@@ -24,6 +24,7 @@ class Graph():
# json.dump(self.full_dico, output_file, indent=4)
#This dico give for the nodes its sister nodes
self.link_dico = None
+ self.link_dico_processes = None
#Dico to graph without operations
self.dico_process_dependency_graph = {}
self.user_view = {}
@@ -147,6 +148,11 @@ class Graph():
if(self.link_dico==None):
self.link_dico = initia_link_dico_rec(self.full_dico)
+ #Method that initalisise the link process link dico
+ def intia_link_dico_processes(self):
+ if(self.link_dico_processes==None):
+ self.link_dico_processes = initia_link_dico_rec(self.dico_process_dependency_graph)
+
def get_specification_graph(self, dirc = 'graphs', filename = "specification_graph", render_graphs = True):
generate_graph(self.get_output_dir()/ dirc /filename, self.full_dico, render_graphs = render_graphs)
#generate_graph(self.get_output_dir()/ dirc /(filename+"_without_artificial_nodes"), remove_artificial_nodes(self.full_dico), render_graphs = render_graphs)
@@ -340,6 +346,7 @@ class Graph():
add_edges(dico, condition="", checking_conditions=False)
self.dico_process_dependency_graph = dico
+ self.intia_link_dico_processes()
with open(f"{self.get_output_dir()}/graphs/process_dependency_graph.json", 'w') as output_file :
json.dump(self.dico_process_dependency_graph, output_file, indent=4)
diff --git a/src/nextflow_building_blocks.py b/src/nextflow_building_blocks.py
index c27c2d09ceb854ebb17ea4b02d1b3c6c90da57eb..28591c5a3e35be94e03a109b15f598171d2c161b 100644
--- a/src/nextflow_building_blocks.py
+++ b/src/nextflow_building_blocks.py
@@ -57,6 +57,13 @@ class Nextflow_Building_Blocks:
except:
return self.nextflow_file
+ def get_link_dico_processes(self):
+ try:
+ return self.origin.get_link_dico_processes()
+ except:
+ return self.nextflow_file.get_link_dico_processes()
+
+
def get_display_info(self):
return self.origin.get_display_info()
diff --git a/src/nextflow_file.py b/src/nextflow_file.py
index f4aef0e3aa0268a230e6b60e4215ce3fb5774699..2f06666fc238b09213fb37904265877e7ab9c448 100644
--- a/src/nextflow_file.py
+++ b/src/nextflow_file.py
@@ -38,6 +38,9 @@ class Nextflow_File(Nextflow_Building_Blocks):
#GENERAL
#----------------------
+ def get_link_dico_processes(self):
+ return self.workflow.get_link_dico_processes()
+
def get_cycle_status(self):
return self.workflow.get_cycle_status()
diff --git a/src/process.py b/src/process.py
index 3ffb410d3c6f8d5291160ad7e74794a5aef11f2f..6a97f93e16efecef436a3ec7bc3e18e37fa99da0 100644
--- a/src/process.py
+++ b/src/process.py
@@ -1,6 +1,7 @@
import re
import glob
import copy
+import ctypes
from .code_ import Code
from .condition import Condition
@@ -10,6 +11,10 @@ from .bioflowinsighterror import BioFlowInsightError
from . import constant
+def get_object(address):
+ address = int(re.findall(r"\dx\w+", address)[0], base=16)
+ return ctypes.cast(address, ctypes.py_object).value
+
class Process(Nextflow_Building_Blocks):
def __init__(self, code, nextflow_file):
self.nextflow_file = nextflow_file
@@ -648,6 +653,69 @@ class Process(Nextflow_Building_Blocks):
if(not check_if_element_in_tab_rocrate(dico_process["isPartOf"], parent_key)):
dico_process["isPartOf"].append({"@id":parent_key})
self.get_nextflow_file().add_to_has_part(dico, process_key)
+
+
+ def get_inputs_DBfile(self):
+ input_params = self.get_input_parameters()
+ for i in range(len(input_params)):
+ input_params[i] = f"sf:{input_params[i]}"
+ return list(set(input_params))
+
+ def get_outputs_DBfile(self):
+ output_params = self.get_outputs()
+ for i in range(len(output_params)):
+ output_params[i] = output_params[i].replace(",", "\\,")
+ output_params[i] = output_params[i].replace("'", "\\'")
+ output_params[i] = output_params[i].replace('"', '\\"')
+ output_params[i] = f'sf:"{output_params[i]}"'
+ return list(set(output_params))
+
+ def get_DBfile_description(self):
+ text = ""
+ text+= f"# {self.get_alias()} Step"
+ text+= f"\nsf:step{self.get_alias()} rdf:type sf:Step"
+
+ input_params = self.get_inputs_DBfile()
+ if(len(input_params)!=0):
+ text+= f" ;\n\tsf:inputVariable {', '.join(input_params)}"
+
+ output_params = self.get_outputs_DBfile()
+ if(len(output_params)!=0):
+ text+= f" ;\n\tsf:outputVariable {', '.join(output_params)}"
+
+ tools = self.get_tools()
+ for i in range(len(tools)):
+ tools[i] = f"sf:TOOL{tools[i]}"
+ if(len(tools)!=0):
+ text+= f" ;\n\tsf:use {', '.join(tools)}"
+
+ isFollowedBy = self.get_link_dico_processes()[str(self)]
+ for i in range(len(isFollowedBy)):
+ isFollowedBy[i] = f'sf:step{get_object(isFollowedBy[i]).get_alias()}'
+ if(len(isFollowedBy)!=0):
+ text+= f" ;\n\tsf:isFollowedBy {', '.join(isFollowedBy)}"
+
+ text+=" ."
+
+ for t in tools:
+ name = t[7:]#To remove the sf:TOOL"
+ text+= f"""\n\n{t} rdf:type sf:Command ;
+\tschema:name "{name}" ;
+\tschema:url <https://bio.tools/t?page=1&q={name}&sort=score> ."""
+
+
+
+ for input in input_params:
+ text+=f"\n\n{input} rdf:type sf:Variable ."
+
+ for output in output_params:
+ text+=f"\n\n{output} rdf:type sf:Variable ."
+
+ return text
+
+
+
+
diff --git a/src/subworkflow.py b/src/subworkflow.py
index fc16dd2381a903d70c9632f3985ebf163417f96b..23fd62ca8c0c9fc2f8b46c7aa90045a1faf11c3f 100644
--- a/src/subworkflow.py
+++ b/src/subworkflow.py
@@ -524,4 +524,52 @@ class Subworkflow(Main):
dico_sub["isPartOf"].append({"@id":parent_key})
self.get_nextflow_file().add_to_has_part(dico, sub_key)
+
+ def get_DBfile_description(self, workflow_name):
+ text = ""
+ text+= f"# {self.get_alias()} Subworkflow"
+ text+=f"""\nsf:subworkflow{self.get_alias()} rdf:type sf:Subworkflow ;
+\tp-plan:isSubPlanOf sf:{workflow_name}"""
+
+ processes_called, subworkflows_called = [], []
+ all_calls = self.get_all_calls_in_subworkflow()
+ for c in all_calls:
+ if(c.get_first_element_called().get_type()=="Process"):
+ processes_called.append(c.get_first_element_called())
+ if(c.get_first_element_called().get_type()=="Subworkflow"):
+ subworkflows_called.append(c.get_first_element_called())
+
+ for i in range(len(processes_called)):
+ processes_called[i] = f'sf:step{processes_called[i].get_alias()}'
+ if(len(processes_called)!=0):
+ text+= f" ;\n\tschema:step {', '.join(processes_called)}"
+
+ input_params = []
+ for input in self.get_takes():
+ input_params.append(f"sf:{input.get_code(get_OG = True)}")
+ if(len(input_params)!=0):
+ text+= f" ;\n\tsf:inputVariable {', '.join(input_params)}"
+
+ output_params = []
+ for output in self.get_emit():
+ val = output.get_code(get_OG = True)
+ val = val.replace(",", "\\,")
+ val = val.replace("'", "\\'")
+ val = val.replace('"', '\\"')
+ output_params.append(f"sf:{val}")
+ if(len(output_params)!=0):
+ text+= f" ;\n\tsf:outputVariable {', '.join(output_params)}"
+ text+=" ."
+
+ for input in input_params:
+ text+=f"\n\n{input} rdf:type sf:Variable ."
+
+ for output in output_params:
+ text+=f"\n\n{output} rdf:type sf:Variable ."
+
+ for sub in subworkflows_called:
+ text+=f'\n\n{sub.get_DBfile_description(f"subworkflow{self.get_alias()}")}'
+
+ return text
+
\ No newline at end of file
diff --git a/src/workflow.py b/src/workflow.py
index bb4bdaab995fc1c0ce666b3f9af6f3885db3a5e9..e8e0d711f77e942a71df99b7fa56bad567d45e82 100644
--- a/src/workflow.py
+++ b/src/workflow.py
@@ -180,7 +180,10 @@ George Marchment, Bryan Brancotte, Marie Schmit, Frédéric Lemoine, Sarah Cohen
tools = p.get_tools()
self.alias_2_tools[p.get_alias()] = tools
self.scripts_2_tools[p.get_script_code()] = tools
-
+
+ def get_link_dico_processes(self):
+ self.graph.initialise()
+ return self.graph.link_dico_processes
def iniatilise_tab_processes_2_remove(self):
if(self.processes_2_remove==None):