def test_docdb_required():
with raises(MissingRequiredValue):
Docdb('123', None, None)
with raises(MissingRequiredValue):
Docdb('123', '345', '')
with raises(MissingRequiredValue):
Docdb('', None, None)
def test_docdb_required():
with raises(MissingRequiredValue):
Docdb("123", None, None)
with raises(MissingRequiredValue):
Docdb("123", "345", "")
with raises(MissingRequiredValue):
Docdb("", None, None)
os.chdir(ndf.replace('.dump', ''))
desc, clm, ft = 0, 0, 0
for brevet in lstBrevets:
#tempo =('publication', Docdb(,, ))
ndb = brevet[u'document-id'][u'country']['$'] + brevet[u'document-id'][
u'doc-number']['$']
for content in [u'claims', u'description', u'fulltext']:
if content not in filter(os.path.isdir, os.listdir(os.getcwd())):
os.makedirs(content)
# optional, list of constituents
try:
data = registered_client.published_data(
reference_type='publication',
input=Docdb(brevet[u'document-id'][u'doc-number']['$'],
brevet[u'document-id'][u'country']['$'],
brevet[u'document-id'][u'kind']['$']),
endpoint=content,
constituents=[])
if data.status_code == 403:
#making necessary redirections
print data
if data.ok:
patentCont = data.json()
#withch language ?
#the following could be factorized !!!!!!!!
if content == 'description':
description = []
description = patentCont[u'ops:world-patent-data'][
u'ftxt:fulltext-documents'][u'ftxt:fulltext-document'][
if ndb+'.txt' not in fichier: #hack here as chinese patents seems not be in claims or description endpoint
#, u'fulltext'
temp =('publication', Epodoc(pays+ndb[2:])) #, brevet[u'document-id'][u'kind']['$']))
try:
data = registered_client.published_data(*temp, endpoint = endP) #registered_client.published_data()
if data.ok and content.replace(typeSrc, "").lower() in str(data.json()):
CheckDocDB = False
else:
CheckDocDB = True
except:
CheckDocDB = True
if CheckDocDB:
if isinstance(brevet[u'kind'], list):
tempoData = []
for cc in brevet[u'kind']:
temp =('publication', Docdb(ndb[2:],pays, cc)) # hope all comes from same country
try:
tempoData.append(registered_client.published_data(*temp, endpoint = endP))
except:
data = None
pass
for dat in tempoData:
if dat is not None and dat.ok: #doing the same for all content. This may result in redundancy
contenu = content.replace(typeSrc, "").lower()
patentCont = dat.json()
Langs = MakeIram2(brevet, ndb +'.txt', patentCont, RepDir+ '//'+ typeSrc + contenu+'//', contenu)
if endP == 'biblio':
for contenu in ['claims', 'description']:
Langs = MakeIram2(brevet, ndb +'.txt', patentCont, RepDir+ '//'+ typeSrc + contenu+'//', contenu)
else:
def issue_request(client):
return client.published_data("publication",
Docdb("Quota", "Forbidden", "exceeded"))
desc, clm, ft = 0,0,0
if GatherContent:
for brevet in lstBrevet:
#tempo =('publication', Docdb(,, ))
#if brevet['label'] == 'FR2997041':
ndb =brevet[u'label']#[u'document-id'][u'country']['$']+brevet[u'document-id'][u'doc-number']['$']brevet['publication-ref'][u'document-id'][0][u'kind']['$'])
#check for already gathered patents
lstfic =[] #alreadycollected
for content in [u'claims', u'description']:
lstfic += os.listdir(ResultPathContent+'//'+content+'//')
fichier = [fics[3:] for fics in lstfic]
if fichier.count(ndb+'.txt') < 2: #one or both files claim or desc are missing
tmp = Epodoc(ndb)
tempo2 = ('publication', tmp)
tmp = Docdb(ndb[2:], ndb[0:2],brevet['status'])
tempo = ('publication', tmp)
ndb =brevet[u'label']#[u'document-id'][u'country']['$']+brevet[u'document-id'][u'doc-number']['$']brevet['publication-ref'][u'document-id'][0][u'kind']['$'])
if True: #avoid check of chinese patents since they aren't descibed in english
for content in [u'claims', u'description']: #, u'fulltext'
if content not in os.listdir(RepDir):
os.makedirs(RepDir +'//' +content)
# optional, list of constituents
try :
data = registered_client.published_data(*tempo, endpoint = content)
#registered_client.published_data()
if data.ok:
patentCont=data.json()
def test_docdb_as_api_input():
params = ["US08/921,321", "CC", "B2", "20140122"]
expected = "(CC).(US08/921%2C321).(B2).(20140122)"
assert Docdb(*params).as_api_input() == expected
*temp,
endpoint=endP) #ops_client.published_data()
if data.ok and content.replace(
typeSrc, "").lower() in str(data.json()):
CheckDocDB = False
else:
CheckDocDB = True
except Exception as err:
CheckDocDB = True
if CheckDocDB:
if isinstance(brevet[u'kind'], list):
tempoData = []
for cc in brevet[u'kind']:
temp = (
'publication',
Docdb(ndb[2:], pays, cc)
) # hope all comes from same country
try:
tempoData.append(
ops_client.published_data(
*temp, endpoint=endP))
except:
data = None
pass
for dat in tempoData:
if dat is not None and dat.ok: #doing the same for all content. This may result in redundancy
contenu = content.replace(typeSrc, "")
patentCont = dat.json()
Langs = MakeIram2(
brevet, ndb + '.txt', patentCont,
def OPSChercheAbstractBrevet(pat, DirStockage):
import epo_ops
from epo_ops.models import Docdb
from epo_ops.models import Epodoc
fic = open('../cles-epo.txt', 'r')
key, secret = fic.read().split(',')
key, secret = key.strip(), secret.strip()
fic.close()
ops_client = epo_ops.Client(key, secret)
ops_client.accept_type = 'application/json'
ndb = pat[
'label'] #[u'document-id'][u'country']['$']+brevet[u'document-id'][u'doc-number']['$']brevet['publication-ref'][u'document-id'][0][u'kind']['$'])
Abstracts = dict()
if isinstance(ndb, list):
ndb = ndb[0]
#print("Retrieving ", ndb)
pays = pat['country']
for key in ['label', 'country', 'kind']:
if isinstance(pat[key], list):
pat[key] = list(
set(pat[key])
) # hum some problem (again) in cleaning data within the family gatherer... 22/12/15
if isinstance(pays, list):
pays = pays[0]
content = 'Abstract'
endP = 'biblio'
#
temp = ('publication', Epodoc(pays + ndb[2:])
) #, brevet[u'document-id'][u'kind']['$']))
try:
data = ops_client.published_data(
*temp, endpoint=endP) #ops_client.published_data()
if data.ok and 'abstract' in str(data.json()):
CheckDocDB = False
else:
CheckDocDB = True
except Exception as err:
CheckDocDB = True
if CheckDocDB:
if isinstance(pat['kind'], list):
tempoData = []
for cc in pat['kind']:
temp = ('publication', Docdb(ndb[2:], pays, cc)
) # hope all comes from same country
try:
tempoData.append(
ops_client.published_data(*temp, endpoint=endP))
except:
data = None
pass
for dat in tempoData:
if dat is not None and dat.ok:
contenu = content
patentCont = dat.json()
Abstracts = MakeIram4(pat, patentCont, contenu)
# Make2Iram2 devrait formater le brevet dans un fichier txt au format Iramuteq dans le bon repertoire
# Lang est un truc :-) (je crois que cela renvoit la langue de l'abstract récupéré))
else:
temp = ('publication',
Docdb(pat['label'][2:], pat['country'], pat['kind']))
if data is not None and data.ok:
contenu = content
patentCont = data.json()
Abstracts = MakeIram4(pat, patentCont, contenu)
# if ops:world-patent-data exchange-documents exchange-documents abstract
return Abstracts
import re
import requests
from epo_ops.models import Docdb
from epo_ops.models import Epodoc
from epo_ops.models import Original
data = ('publication', Docdb('1000000', 'EP', 'A1'))
rdata = ('publication', Epodoc('EP1000000'))
def find_range(document, pattern):
return re.search("range.*{0}".format(pattern), document)
def assert_request_success(response):
assert response.status_code == requests.codes.ok
assert response.headers['X-API'] == 'ops-v3.1'
def assert_family_success(client):
response = client.family(*data)
assert_request_success(response)
assert 'patent-family' in response.text
return response
def issue_published_data_request(client):
return client.published_data(*data)
return None, DejaLa, BiblioPatents
if GatherBibli and GatherBiblio:
registered_client = epo_ops.RegisteredClient(key, secret)
# data = registered_client.family('publication', , 'biblio')
registered_client.accept_type = 'application/json'
for brevet in lstBrevets:
YetGathered = [u['label'] for u in BiblioPatents]
# may be current patent has already be gathered in a previous attempt
# should add a condition here to check in os.listdir()
tempo =('publication', Docdb(brevet[u'document-id'][u'doc-number']['$'],brevet[u'document-id'][u'country']['$'], brevet[u'document-id'][u'kind']['$']))
tempo2 =('publication', Epodoc(brevet[u'document-id'][u'country']['$']+brevet[u'document-id'][u'doc-number']['$']))#, brevet[u'document-id'][u'kind']['$']))
ndb =brevet[u'document-id'][u'country']['$']+brevet[u'document-id'][u'doc-number']['$'] #nameOfPatent
if ndb not in YetGathered:
try: #trying Epodoc first, unused due to response format (multi document instead of one only)
data = registered_client.published_data(*tempo2, endpoint = 'biblio')
patentBib = data.json()
data2 = registered_client.published_data(*tempo, endpoint = 'biblio')
if data.ok and data2.ok:
patentBibtemp = data.json()
patentBibtemp2= data2.json()
if len(str(patentBibtemp)) > len(str(patentBibtemp2)):
patentBib = patentBibtemp
else:
patentBib = patentBibtemp2
def issue_request(client):
return client.published_data(
'publication',
Docdb('Quota', 'Forbidden', 'exceeded')
)
def test_docdb_as_api_input():
params = ['US08/921,321', 'CC', 'B2', '20140122']
expected = '(CC).(US08/921%2C321).(B2).(20140122)'
assert Docdb(*params).as_api_input() == expected
brevet = CleanPatent(brevet)
brevet = CleanPatent(brevet)
ndb =brevet[u'label']#[u'document-id'][u'country']['$']+brevet[u'document-id'][u'doc-number']['$']brevet['publication-ref'][u'document-id'][0][u'kind']['$'])
#check for already gathered patents
lstfic = os.listdir(ResultPathContent+'//Abstracts/')
fichier = [fics[3:] for fics in lstfic]
if ndb+'.txt' not in fichier:
for content in [u'abstract']:#claims', u'description']: #, u'fulltext'
try :
tmp = Epodoc(ndb)
tempo = ('publication', tmp)
data = registered_client.published_data(*tempo, endpoint = content) #registered_client.published_data()
if 'abstract' not in str(data.json()):
tmp = Docdb(ndb[2:], ndb[0:2],brevet['status'])
tempo = ('publication', tmp)
data = registered_client.published_data(*tempo, endpoint = content)
#brevet['publication-ref'][u'document-id'][0][u'doc-number']['$'],#brevet[u'document-id'][u'doc-number']['$'],
#brevet['publication-ref'][u'document-id'][0][u'country']['$'],#brevet[u'document-id'][u'country']['$'],
#brevet['publication-ref'][u'document-id'][0][u'kind']['$']), endpoint = content, constituents = [])
except:
try:
tmp = Docdb(ndb[2:], ndb[0:2],brevet['status'])
tempo = ('publication', tmp)
data = registered_client.published_data(*tempo, endpoint = content) #registered_client.published_data()
except:#from there totally fun... may be we do not get there...
try:
# print 'yes we get'
tmp = Epodoc(brevet['publication-ref'][u'document-id'][1][u'doc-number']['$'])
def assert_bulk_service_retrival_success(client):
input_list = [Docdb("1000000", "EP", "A1"), Epodoc("US2018265402")]
response = client.published_data("publication", input=input_list)
assert response.status_code == requests.codes.ok
import re
import requests
from epo_ops.models import Docdb, Epodoc, Original
data = ("publication", Docdb("1000000", "EP", "A1"))
rdata = ("publication", Epodoc("EP1000000"))
idata = ("published-data/images/EP/1000000/A1/fullimage", 1)
# idata path is the result @path from images published-data json request
def find_range(document, pattern):
return re.search("range.*{0}".format(pattern), document)
def assert_request_success(response):
assert response.status_code == requests.codes.ok
assert response.headers["X-API"] == "ops-v3.2"
def assert_family_success(client):
response = client.family(*data)
assert_request_success(response)
assert "patent-family" in response.text
return response
def assert_family_biblio_success(client):
response = client.family(*data, constituents=["biblio"])
assert_request_success(response)