def adhoc_interface(options):
"""
Patent2Net ad-hoc mode interface
"""
# Read OPS credentials from designated file
credentials = OPSCredentials(credentials_file='cles-epo.txt')
key, secret = credentials.read()
# Create Patent2Net instance
patent2net = Patent2Net(key, secret)
# Gather data
results = patent2net.gather(options['expression'],
with_family=options['with-family'],
with_register=options['with-register'])
# Display search results for given query expression in raw OPS format (JSON), e.g. run::
# p2n adhoc search --expression='TA=lentille'
if options['search']:
payload = patent2net.response_data
print(json.dumps(payload, cls=JsonObjectEncoder))
# Display full results for given query expression in OpsExchangeDocument or Patent2NetBrevet format (JSON), e.g. run::
# p2n adhoc dump --expression='TA=lentille'
if options['dump']:
if options['format'] == 'ops':
payload = [result.as_dict() for result in results.documents]
elif options['format'] == 'brevet':
payload = results.brevets
else:
logger.error('Unknown format "{}" for dumping.'.format(
options['format']))
sys.exit(1)
print(json.dumps(payload, cls=JsonObjectEncoder))
# Display list of values from single field for given query expression, e.g. run::
# p2n adhoc list --expression='TA=lentille'
if options['list']:
documents = results.documents
output = []
for document in documents:
try:
values = [operator.attrgetter(options['field'])(document)]
output += values
except AttributeError:
pass
print(json.dumps(output, indent=4))
# Generate world map over given field, e.g. run::
# p2n adhoc worldmap --expression='TA=lentille' --country-field='applicants'
if options['worldmap']:
mapdata = results.worldmap(options['country-field'])
print(json.dumps(mapdata))
# Generate data for PivotTable.js
if options['pivot']:
mapdata = results.pivot(options['format'])
print(json.dumps(mapdata))
#from P2N_Lib import EcritContenu, coupeEnMots from Patent2Net.P2N_Config import LoadConfig from p2n.config import OPSCredentials import datetime import epo_ops import os import sys #from epo_ops.models import Docdb #from epo_ops.models import Epodoc os.environ['REQUESTS_CA_BUNDLE'] = 'cacert.pem' global key global secret # put your credential from epo client in this file... # chargement clés de client c = OPSCredentials(credentials_file='../cles-epo.txt') key, secret = c.read() DureeBrevet = 20 SchemeVersion = '20140101' # for the url to the classification scheme import os ListeBrevet = [] # LA iste de brevets # ouverture fichier de travail ficOk = False cptNotFound = 0 nbTrouves = 0 lstBrevets = [] # The patent List BiblioPatents = [] # The bibliographic data
def classic_interface(options):
"""
Patent2Net classic interface
"""
# running flask server
#run_script('app.py', configfile=None, directory='.')
# Convenience: Write OPS API credentials to file "cles-epo.txt"
if options['ops'] and options['init']:
if options['key'] and options['secret']:
credentials = OPSCredentials()
credentials.write(options['key'], options['secret'])
sys.exit()
# All tasks from here require a configuration file.
configfile = options['config']
if not configfile:
configfile = os.environ.get('P2N_CONFIG')
if not configfile:
logger.error(
'No configuration file given. Either use --config commandline argument or P2N_CONFIG environment variable.'
)
sys.exit(1)
# Patent2Net classic steps, aggregated
if options['acquire'] or options['run']:
#print(configfile)
run_script('OPSGatherPatentsv2.py', configfile)
run_script('PatentListFiltering.py', configfile)
if options['with-family']:
run_script('OPSGatherAugment-Families.py', configfile)
# the following will fail if Families aren't gathered yet....
run_script('preProcessNormalisationNames.py', configfile)
# normalisation features New 08/2020
if options['with-family'] or options['run']:
run_script('PatentListFiltering.py', configfile)
run_script('OPSGatherAugment-Families.py', configfile)
if options['normalisation'] or options['run']:
run_script('PatentListFiltering.py', configfile)
run_script('preProcessNormalisationNames.py', configfile)
if options['maps'] or options['run']:
run_script('FormateExportCountryCartography.py', configfile)
run_script('FormateExportAttractivityCartography.py', configfile)
if options['bibfile'] or options['run']:
run_script('FormateExportBiblio.py', configfile)
if options['tables'] or options['run']:
run_script('FormateExportDataTableFamilies.py', configfile)
AnnonceProgres(Appli='p2n_tables', valMax=100, valActu=50) # approx
run_script('FormateExportDataTable.py', configfile)
AnnonceProgres(Appli='p2n_tables', valMax=100, valActu=75) # approx
run_script('FormateExportPivotTable.py', configfile)
AnnonceProgres(Appli='p2n_tables', valMax=100, valActu=100) # approx
if options['networks'] or options['run']:
run_script('P2N-Nets-new.py', configfile)
# run_script('P2N-NetworksBis.py {network}'.format(network=network), configfile)
networks = [
"CountryCrossTech",
"CrossTech",
"InventorsCrossTech",
"Applicants_CrossTech",
"Inventors",
"ApplicantInventor",
"Applicants",
"References",
"Citations",
"Equivalents",
]
compt = 0
for network in networks:
compt += 1
# run_script('P2N-PreNetworks.py {network}'.format(network=network), configfile)
run_script('P2N-NetworksBis.py {network}'.format(network=network),
configfile)
AnnonceProgres(Appli='p2n_network',
valMax=100,
valActu=compt * 100 /
len(networks)) # valMax is unused in front end
# run_script('P2N-NetworksJS.py {network}'.format(network=network), configfile)
#adding the complete net a buggy very very slow process
# run_script('P2N-NetworksFull.py', configfile)
if options['freeplane'] or options['run']:
run_script('P2N-FreePlane.py', configfile)
# AnnonceProgres (Appli = 'p2n_network', valMax = len(networks), valActu = compt*100/len(networks)) # valMax is unused in front end
#Gather more textual data if it exsts (Claims, Descriptions) an format them
#for Iramuteq
if options['iramuteq'] or options['run']:
if options['with-family']:
run_script('OPSGatherAugment-Families.py', configfile)
run_script('OPSGatherContentsV2-Iramuteq.py', configfile)
run_script('FusionIramuteq2.py', configfile)
#format xml files for Carrot2
if options['carrot'] or options['run']:
if options['with-family']:
run_script('OPSGatherAugment-Families.py', configfile)
run_script('FusionCarrot2.py', configfile)
run_script('P2N-Indexer.py', configfile)
#Gather images
if options['images'] or options['run']:
run_script('OPSGatherContentsV2-Images.py', configfile)
run_script('FusionImages.py', configfile)
#Cluster processing
if options['cluster'] or options['run']:
if options['with-family']:
run_script('OPSGatherAugment-Families.py', configfile)
run_script('IPC-WS-metrics.py', configfile)
run_script('ClusterPreProcess.py', configfile)
run_script('P2N-Cluster.py', configfile)
#Present data through HTML
if options['interface'] or options['run']:
run_script('Interface2.py', configfile)
def autom_request_spliter_time(RequestOrig, directory):
today = datetime.datetime.today()
jourOk, moisOk, ipcOk = False, False, False
Request = RequestOrig + ' AND PD=date'
DataDir = directory + '/segments/' + directory
targetDirectory = '../RequestsAuto/' + directory
fic = open("REQUESTS/requestModel.cql", 'r') #requestModel.cql
DataReq = targetDirectory
data = fic.read()
fic.close()
if 'RequestsAuto' not in os.listdir('..') or directory not in os.listdir(
'../RequestsAuto'):
os.makedirs(targetDirectory)
lstFicOk = []
else:
lstFicOk = os.listdir(targetDirectory)
Months = {
'jan': 31,
'fev': 29,
'mar': 31,
'apr': 30,
'may': 31,
'jun': 30,
'jui': 31,
'aug': 31,
'sep': 30,
'octo': 31,
'nov': 30,
'dec': 31
}
IPC = {"A", "B", "C", "D", "E", "F", "G", "H"}
os.environ['REQUESTS_CA_BUNDLE'] = 'cacert.pem'
global key
global secret
# put your credential from epo client in this file...
# chargement clés de client
c = OPSCredentials(credentials_file='../cles-epo.txt')
key, secret = c.read()
def checkRequest(req):
ops_client = epo_ops.Client(key, secret)
# data = ops_client.family('publication', , 'biblio')
ops_client.accept_type = 'application/json'
try:
lstBrevets2, nbTrouves = PatentSearch(ops_client, req)
return nbTrouves
except:
return 0
Total = 0
nbFiles = 0
toBeFound = checkRequest(RequestOrig)
if toBeFound > 2000:
print(
'wow ', toBeFound,
' (or more) patents to retreive... A good reason to use this script'
)
Need = True
else:
print("no need to split, gather directly your request '", RequestOrig,
"' with p2n")
Need = False
dateDeb = int(
input("please enter the stardate year for gathering you request"))
if Need:
for AN in range(dateDeb, today.year + 1, 1):
Trouves = checkRequest(Request.replace('=date', '=' + str(AN)))
if 2000 > Trouves > 0:
Total += Trouves
# a request for that year is ok
monthOk = False
ipcOk = False
Request2 = Request.replace('=date', '=' + str(AN))
data2 = data.replace("***requete***", Request2)
data2 = data2.replace("***dataDir***", DataDir + str(AN))
NameFic = str(AN) + 'Request.cql'
with open(DataReq + "/" + NameFic, "w") as ficRes: #+"-"+ipc
print(ficRes.name.split('/')[1])
if ficRes.name.split('/')[1] not in lstFicOk:
ficRes.write(data2)
nbFiles += 1
print(ficRes.name, 'file written, ', Trouves,
' patents expected and ', Total, ' cumulative.')
if Trouves == 0:
monthOk = False
ipcOk = False
jourOk = False
#nothing to do
if Trouves >= 2000:
# we have to split by monthes
monthOk = True
jourOk = False
cpt = 0 #used as monthes
for month in Months.keys():
cpt += 1
if len(
str(cpt)
) < 2: # monthes are numbered thanks to cpt (ugly isn't it ?)
mois = '0' + str(cpt)
else:
mois = str(cpt)
Request2 = Request.replace('=date', '=' + str(AN) + mois)
Trouves = checkRequest(Request2)
if 2000 > Trouves > 0:
Total += Trouves
# OK less than 2000 and more than 0 go ahead for that request
ipcOk = False
jourOk = False
data2 = data.replace("***requete***", Request2)
data2 = data2.replace("***dataDir***",
DataDir + str(AN) + mois)
NameFic = str(AN) + mois + 'Request.cql'
if NameFic not in lstFicOk:
with open(DataReq + "/" + NameFic,
"w") as ficRes: #+"-"+ipc
ficRes.write(data2)
nbFiles += 1
print(ficRes.name, 'file written, ', Trouves,
' patents expected and ', Total,
' cumulative.')
if Trouves == 0:
ipcOk = False
jourOk = False
#nothing to do
if Trouves >= 2000:
monthOk = True
jourOk = True
ipcOk = False
# spliting days for that month
for day in range(1, Months[month] + 1):
if len(str(day)) < 2:
jour = '0' + str(day)
else:
jour = str(day)
Request2 = Request.replace(
'=date', '=' + str(AN) + mois + jour)
Trouves = checkRequest(Request2)
if 2000 > Trouves > 0:
Total += Trouves
# go ahead for that day
ipcOk = False
data2 = data.replace("***requete***", Request2)
data2 = data2.replace(
"***dataDir***",
DataDir + str(AN) + mois + jour)
NameFic = str(AN) + mois + jour + 'Request.cql'
if NameFic not in lstFicOk:
with open(DataReq + "/" + NameFic,
"w") as ficRes: #+"-"+ipc
ficRes.write(data2)
nbFiles += 1
print(ficRes.name, 'file written, ',
Trouves, ' patents expected and ',
Total, ' cumulative.')
if Trouves == 0:
ipcOk = False
jourOk = False
#nothing to do
if Trouves >= 2000:
monthOk = True
jourOk = True
# bad days for EPO... we need to split again
# last solution IPC splitting
# for that day only
for ipc in IPC:
Request3 = Request2 + " AND IC=" + ipc
Trouves = checkRequest(Request3)
if Trouves > 2000:
print(
"thats bad... the request : " +
Request3 +
" should be splitted and the limits of this script are reached"
)
break
Total += Trouves
data2 = data.replace(
"***requete***", Request3)
data2 = data2.replace(
"***dataDir***",
DataDir + str(AN) + mois + jour + ipc)
if NameFic not in lstFicOk:
with open(
DataReq + "/" + str(AN) +
mois + '-' + jour + '-' + ipc +
'Request.cql',
"w") as ficRes: #+"-"+ipc
ficRes.write(data2)
nbFiles += 1
print(ficRes.name, 'file written, ',
Trouves,
' patents expected and ', Total,
' cumulative.')
print("request splitted in ", nbFiles, " files")
print("Gathering with P2N all this request should lead to ", Total,
" patents")