def specification_2_linkset_subset(specs, activated=False):
if activated is True:
print Ut.headings("EXECUTING LINKSET SUBSET SPECS...")
else:
print Ut.headings(
"THE FUNCTION [specification_2_linkset_subset] IS NOT ACTIVATED")
return {St.message: Ec.ERROR_CODE_0, St.error_code: 0, St.result: None}
# ACCESS THE TASK SPECIFIC PREDICATE COUNT
specs[St.sameAsCount] = Qry.get_same_as_count(specs[St.mechanism])
# UPDATE THE QUERY THAT IS GOING TO BE EXECUTED
if specs[St.sameAsCount]:
source = specs[St.source]
target = specs[St.target]
# UPDATE THE SPECS OF SOURCE AND TARGETS
update_specification(source)
update_specification(target)
# GENERATE THE NAME OF THE LINKSET
Ls.set_subset_name(specs)
# SETTING SOME GENERIC METADATA INFO
specs[St.link_name] = "same"
specs[St.linkset_name] = specs[St.linkset_name]
specs[St.link] = "http://risis.eu/linkset/predicate/{}".format(
specs[St.link_name])
specs[
St.
link_subpropertyof] = "http://risis.eu/linkset/predicate/{}".format(
specs[St.link_name])
specs[St.linkset] = "{}{}".format(Ns.linkset, specs[St.linkset_name])
specs[St.assertion_method] = "{}{}".format(Ns.method,
specs[St.linkset_name])
specs[St.justification] = "{}{}".format(Ns.justification,
specs[St.linkset_name])
# COMMENT ON THE LINK PREDICATE
specs[St.link_comment] = "The predicate <{}> is used in replacement of the linktype <{}> used in the " \
"original <{}> dataset.".format(
specs[St.link], specs[St.source][St.link_old], specs[St.source][St.graph])
# COMMENT ON THE JUSTIFICATION FOR THIS LINKSET
specs[St.justification_comment] = "In OrgRef's a set of entities are linked to GRID. The linking method " \
"used by OrgRef is unknown. Here we assume that it is a curated work " \
"and extracted it as a linkset.",
# COMMENT ON THE LINKSET ITSELF
specs[St.linkset_comment] = "The current linkset is a subset of the <{0}> dataset that links <{0}> to " \
"<{1}>. The methodology used by <{0}> to generate this builtin linkset in " \
"unknown.".format(specs[St.source][St.graph], specs[St.target][St.graph])
source[St.entity_ns] = str(source[St.entity_datatype]).replace(
source[St.entity_name], '')
target[St.entity_ns] = str(target[St.entity_datatype]).replace(
target[St.entity_name], '')
# GENERATE THE LINKSET
inserted_linkset = spa_linkset_subset(specs, activated)
# print "LINKSET SUBSET RESULT:", inserted_linkset
if inserted_linkset[St.message].__contains__("ALREADY EXISTS"):
return inserted_linkset
if specs[St.triples] > "0":
# REGISTER THE ALIGNMENT
if inserted_linkset[St.message].__contains__("ALREADY EXISTS"):
Urq.register_alignment_mapping(specs, created=False)
else:
Urq.register_alignment_mapping(specs, created=True)
return inserted_linkset
else:
print Ec.ERROR_CODE_1
return {St.message: Ec.ERROR_CODE_1, St.error_code: 5, St.result: None}
def refining(specs, insert_query, activated=False):
refined = {St.message: Ec.ERROR_CODE_1, St.error_code: 5, St.result: None}
diff = {St.message: Ec.ERROR_CODE_4, St.error_code: 1, St.result: None}
# UPDATE THE SPECS VARIABLE
# print "UPDATE THE SPECS VARIABLE"
update_specification(specs)
update_specification(specs[St.source])
update_specification(specs[St.target])
# ACCESS THE TASK SPECIFIC PREDICATE COUNT
specs[St.sameAsCount] = Qry.get_same_as_count(specs[St.mechanism])
# print "sameAsCount:", specs[St.sameAsCount]
if specs[St.sameAsCount] is None:
return {'refined': refined, 'difference': diff}
# GENERATE THE NAME OF THE LINKSET
Ls.set_refined_name(specs)
# print "\nREFINED NAME:", specs[St.refined]
# print "LINKSET TO REFINE BEFORE CHECK:", specs[St.linkset]
# CHECK WHETHER OR NOT THE LINKSET WAS ALREADY CREATED
check = Ls.run_checks(specs, check_type="refine")
# print "\nREFINED NAME:", specs[St.refined]
# print "LINKSET TO REFINE:", specs[St.linkset]
if check[St.message] == "NOT GOOD TO GO":
# refined = check[St.refined]
# difference = check["difference"]
return check
# print "\nREFINED:", specs[St.refined]
# print "LINKSET TO REFINE:", specs[St.linkset]
# print "CHECK:", check
# THE LINKSET DOES NOT EXIT, LETS CREATE IT NOW
print Ls.refined_info(specs, specs[St.sameAsCount])
# POINT TO THE LINKSET THE CURRENT LINKSET WAS DERIVED FROM
print "1. wasDerivedFrom {}".format(specs[St.linkset])
specs[St.derivedfrom] = "\t\tprov:wasDerivedFrom\t\t\t<{}> ;".format(
specs[St.linkset])
# print "REFINED NAME:", specs[St.refined_name]
# print "REFINED:", specs[St.refined]
# print "LINKSET TO BE REFINED:", specs[St.linkset]
print "\n2. RETRIEVING THE METADATA ABOUT THE GRAPH TO REFINE"
# metadata_q = Qry.q_linkset_metadata(specs[St.linkset])
metadata_q = """
prefix ll: <{}>
SELECT DISTINCT ?type ?singletonGraph
{{
# LINKSET METADATA
<{}>
a ?type ;
ll:singletonGraph ?singletonGraph .
}}
""".format(Ns.alivocab, specs[St.linkset])
print "QUERY:", metadata_q
matrix = Qry.sparql_xml_to_matrix(metadata_q)
# print "\nMETA DATA: ", matrix
if matrix:
if matrix[St.message] == "NO RESPONSE":
print Ec.ERROR_CODE_1
print matrix[St.message]
return {'refined': refined, 'difference': diff}
elif matrix[St.result] is None:
print matrix[St.message]
returned = {
St.message: matrix[St.message],
St.error_code: 666,
St.result: None
}
return {'refined': returned, 'difference': diff}
else:
print Ec.ERROR_CODE_1
return {'refined': refined, 'difference': diff}
# GET THE SINGLETON GRAPH OF THE LINKSET TO BE REFINED
print "\n3. GETTING THE SINGLETON GRAPH OF THE GRAPH TO REFINE"
specs[St.singletonGraph] = matrix[St.result][1][1]
# print matrix[St.result][1][0]
specs[St.insert_query] = insert_query(specs)
print specs[St.insert_query]
if type(specs[St.insert_query]) == str:
is_run = Qry.boolean_endpoint_response(specs[St.insert_query])
else:
print "\n4. RUNNING THE EXTRACTION QUERY"
print specs[St.insert_query][0]
# is_run = Qry.boolean_endpoint_response(specs[St.insert_query][0])
Qry.boolean_endpoint_response(specs[St.insert_query][0])
print "\n5. RUNNING THE FINDING QUERY"
print specs[St.insert_query][1]
is_run = Qry.boolean_endpoint_response(specs[St.insert_query][1])
print "\n>>> RUN SUCCESSFULLY:", is_run.upper()
# NO INSERTION HAPPENED
if is_run == "true" or is_run == Ec.ERROR_STARDOG_1:
# GENERATE THE
# (1) LINKSET METADATA
# (2) LINKSET OF CORRESPONDENCES
# (3) SINGLETON METADATA
# AND WRITE THEM ALL TO FILE
print "GENERATING THE METADATA"
pro_message = refine_metadata(specs)
# SET THE RESULT ASSUMING IT WENT WRONG
refined = {
St.message: Ec.ERROR_CODE_4,
St.error_code: 4,
St.result: None
}
diff = {St.message: Ec.ERROR_CODE_4, St.error_code: 4, St.result: None}
server_message = "Linksets created as: [{}]".format(specs[St.refined])
message = "The linkset was created as [{}]. <br/>{}".format(
specs[St.refined], pro_message)
# MESSAGE ABOUT THE INSERTION STATISTICS
print "\t", server_message
if int(specs[St.triples]) > 0:
# UPDATE THE REFINED VARIABLE AS THE INSERTION WAS SUCCESSFUL
refined = {
St.message: message,
St.error_code: 0,
St.result: specs[St.linkset]
}
print "REGISTERING THE ALIGNMENT"
if refined[St.message].__contains__("ALREADY EXISTS"):
register_alignment_mapping(specs, created=False)
else:
register_alignment_mapping(specs, created=True)
try:
print "\nCOMPUTE THE DIFFERENCE AND DOCUMENT IT"
diff_lens_specs = {
St.researchQ_URI: specs[St.researchQ_URI],
St.subjectsTarget: specs[St.linkset],
St.objectsTarget: specs[St.refined]
}
diff = Df.difference(diff_lens_specs, activated=activated)
message_2 = "\t>>> {} CORRESPONDENCES INSERTED AS THE DIFFERENCE".format(
diff_lens_specs[St.triples])
print message_2
except Exception as err:
print "THE DIFFERENCE FAILED: ", str(err.message)
print "\tLinkset created as: ", specs[St.refined]
print "\t*** JOB DONE! ***"
return {'refined': refined, 'difference': diff}
else:
print ">>> NO TRIPLE WAS INSERTED BECAUSE NO MATCH COULD BE FOUND"
return {'refined': refined, 'difference': diff}
else:
print "NO MATCH COULD BE FOUND."
def lens_transitive(specs, activated=False):
# CHECK BOTH DATASETS FOR SAME MECHANISM
print "GENERATE THE LENS NAME"
Lu.composition_lens_name(specs)
print "GET THE SAME AS COUNT"
specs[St.sameAsCount] = Qry.get_same_as_count(specs[St.lens_operation])
# print same_as_count
# GENERATE THE INSERT QUERY FOR TRANSITIVITY
# transitive_analyses = lens_transitive_query(specs)
# if transitive_analyses is None:
# return
# specs[St.insert_query] = transitive_analyses[1]
# print insert_query
# exit(0)
# specs['is_transitive_by'] = transitive_analyses[0]
ln = get_uri_local_name(specs[St.lens])
sg = specs[St.subjectsTarget]
tg = specs[St.objectsTarget]
ssg = "{}{}".format(Ns.singletons, get_uri_local_name(sg))
tsg = "{}{}".format(Ns.singletons, get_uri_local_name(tg))
print "SOURCE: {}".format(sg)
print "TARGET: {}".format(tg)
print "1. GENERATING THE INSERT QUERY"
specs[St.insert_query] = transitive_insert_query(ln, sg, tg, ssg, tsg)
if activated is True:
# RUN THE QUERY AT THE END POINT
print "2. RUNNING THE INSERT QUERY"
Qry.boolean_endpoint_response(specs[St.insert_query])
# GET THE SIZE OF THE LENS JUST CREATED ABOVE
print "3. ETTING THE SIZE OF THE LENS JUST INSERTED"
size = Qry.get_namedgraph_size(specs[St.lens], isdistinct=False)
# IF ACTIVATED, INSERT THE METADATA
if size > 0:
# GENERATE THE METADATA ABOUT THE LENS JUST CREATED
print "4. SOME {} TRANSITIVE TRIPLES WERE FOUND".format(size)
metadata = transitive_metadata(specs, size)
# print metadata
print "5. INSERTING THE METADATA"
Qry.boolean_endpoint_response(metadata)
print "6. REGISTER THE LENS"
Urq.register_lens(specs, is_created=True)
# RUN A CORRESPONDENCE CONSTRUCT QUERY FOR BACKING UP THE DATA TO DISC
print "7. GENERATE THE CONSTRUCT FOR FILE DUMP"
construct_correspondence = Qry.endpointconstruct(
Qry.construct_namedgraph(specs[St.lens]))
if construct_correspondence is not None:
construct_correspondence = construct_correspondence.replace(
'{', "<{}>\n{{".format(specs[St.lens]), 1)
# RUN A SINGLETON METADATA CONSTRUCT QUERY FOR BACKING UP THE DATA TO DISC
construct_singletons = Qry.endpointconstruct(
Qry.construct_namedgraph("{}{}".format(Ns.singletons,
specs[St.lens_name])))
if construct_singletons is not None:
construct_singletons = construct_singletons. \
replace('{', "<{}{}>\n{{".format(Ns.singletons, specs[St.lens_name]), 1)
# WRITE TO FILE
print "WRITING TO FILE"
write_to_file(graph_name=ln,
metadata=metadata,
directory=DIRECTORY,
correspondences=construct_correspondence,
singletons=construct_singletons)
# return specs[St.lens]
message = "THE LENS WAS CREATED!<br/>URI = {}".format(
specs[St.lens])
print message
print "\t*** JOB DONE! ***"
return {
St.message: message,
St.error_code: 0,
St.result: specs[St.lens]
}
if activated is False:
logger.warning(
"THE FUNCTION IS NOT ACTIVATED BUT THE METADATA THAT IS "
"SUPPOSED TO BE ENTERED IS WRITEN TO THE CONSOLE.")
def refine_lens(specs, activated=False, check_file=False):
try:
message = Ec.ERROR_CODE_0.replace('\n', "<br/>")
if activated is False:
print Ut.headings("THE FUNCTION [refine_lens] IS NOT ACTIVATED")
return {St.message: message, St.error_code: 4, St.result: None}
# 1. UPDATING THE SPECS BY CHANGING LINKSET TO TENS
specs[St.refined] = specs['linkset']
specs.pop('linkset')
Ut.update_specification(specs)
# CHECKING WHETHER THE LENS IS REFINENABLE
# Refine.is_refinable(specs[St.refined])
# PRINTING THE SPECIFICATIONS
# lensUt.print_specs(specs)
# ASSIGN THE SAME AS COUNT
specs[St.sameAsCount] = Qry.get_same_as_count(specs[St.mechanism])
message = Ec.ERROR_CODE_4.replace('\n', "<br/>")
if specs[St.sameAsCount]:
source = specs[St.source]
target = specs[St.target]
# 2. SET THE LENS NAME
# *******************************
print "\n2. SET THE LENS NAME"
# *******************************
lensUt.lens_refine_name(specs, 'refine')
#*******************************
# GOOD TO GO CHECK
# *******************************
query = """
SELECT *
{{
<{}> ?predicate ?object .
}}
""".format(specs[St.lens])
check = Lens_Union.run_checks(specs, query, operator="refine")
# NOT GOOD TO GO, IT ALREADY EXISTS
if check[St.message].__contains__("ALREADY EXISTS"):
return {
St.message: check[St.message],
St.error_code: 71,
St.result: specs[St.lens]
}
# *******************************
# GOOD TO GO
# *******************************
else:
lens_start = time.time()
# UPDATE THE SPECIFICATION
Ut.update_specification(specs[St.source])
Ut.update_specification(specs[St.target])
# PRINTING THE SPECIFICATIONS
lensUt.print_specs(specs)
########################################################################
print """\n4. EXECUTING THE GEO-MATCH """
########################################################################
geo_match(specs)
########################################################################
print """\n5. EXTRACT THE NUMBER OF TRIPLES """
########################################################################
specs[St.triples] = Qry.get_namedgraph_size("{0}{1}".format(
Ns.lens, specs[St.lens_name]))
########################################################################
print """\n6. ASSIGN THE SPARQL INSERT QUERY """
########################################################################
specs[St.insert_query] = "{} ;\n{};\n{}".format(
geo_load_query(specs, True), geo_load_query(specs, False),
geo_match_query(specs))
lens_end = time.time()
diff = lens_end - lens_start
print "\n\t>>> Executed so far in : {:<14}".format(
str(datetime.timedelta(seconds=diff)))
if int(specs[St.triples]) > 0:
########################################################################
print """\n4. INSERTING THE GENERIC METADATA """
########################################################################
metadata = Gn.lens_refine_geo_metadata(specs)
Qry.boolean_endpoint_response(metadata)
########################################################################
print """\n5. WRITING TO FILE """
########################################################################
src = [source[St.graph_name], "", source[St.entity_ns]]
trg = [target[St.graph_name], "", target[St.entity_ns]]
# linkset_path = "D:\datasets\Linksets\ExactName"
linkset_path = DIRECTORY
writelinkset(src,
trg,
specs[St.lens_name],
linkset_path,
metadata,
check_file=check_file)
server_message = "Linksets created as: {}".format(
specs[St.lens])
message = "The linkset was created as [{}] with {} triples found!".format(
specs[St.lens], specs[St.triples])
print "\n\t", server_message
Urq.register_lens(specs, is_created=True)
ls_end_2 = time.time()
diff = ls_end_2 - lens_end
print ">>> Executed in : {:<14}".format(
str(datetime.timedelta(seconds=diff)))
print "\t*** JOB DONE! ***"
return {
St.message: message,
St.error_code: 0,
St.result: specs[St.lens]
}
else:
print "\tThe linkset was not generated as no match could be found"
print "\t*** JOB DONE! ***"
return {
St.message: message,
St.error_code: 4,
St.result: None
}
except Exception as err:
traceback.print_exc()
return {St.message: Ec.ERROR_CODE_1, St.error_code: 5, St.result: None}
# print geo_load_query(specs, is_source=True)
# print geo_load_query(specs, is_source=False)
# geo_match_query(specs)
# traceback.print_exception()
# import Alignments.Manage.AdminGraphs as adm
# adm.drop_a_lens("http://risis.eu/lens/refine_union_Grid_20170712_Eter_2014_N291690309", display=True, activated=True)
# refine_lens(specs_example, activated=True, check_file=False)
#
# adm.drop_a_lens("http://risis.eu/lens/refine_union_Orgreg_20170718_Eter_2014_P1061032980", display=True, activated=True)
# refine_lens(specs_example_2, activated=True, check_file=False)
#
# adm.drop_a_lens("http://risis.eu/lens/refine_union_Orgreg_20170718_Grid_20170712_N1966224323", display=True, activated=True)
# refine_lens(specs_example_3, activated=True, check_file=False)