else: #Retrocompatibility
print("please use Comptatibilizer")
print("Nice, ", len(DataBrevet["brevets"]),
" patents found. Pre-formating ", sys.argv[1], " net.")
for brev in DataBrevet["brevets"]:
#tempo = pickle.load(fic) # we only memorize needed nfo
pat = OrderedDict()
if "date" not in list(brev.keys()):
brev['date'] = '1-1-1'
if isinstance(brev['label'], list):
brev['label'] = brev['label'][0]
for key in NeededInfo:
if key.count('nice') > 0 and isinstance(brev[key], list):
brev[key] = flatten(brev[key])
machin = []
for truc in brev[key]:
Tt = Cleaning(truc)
if Tt is not None and len(Tt) > 0:
machin.append(Tt)
if len(machin) > 0:
pat[key] = machin #can you do more ugly ?
else:
pat[key] = ''
elif isinstance(brev[key], list):
pat[key] = flatten(brev[key])
if key.count('Date') == 0:
pat[key] = [
cont for cont in brev[key]
Done = []
if 0 <= len(Done) <= len(ListeBrevet):
tempoList = []
try:
#ndfLstBrev = open(ResultPath+'//Families'+ ndf, 'r')
BrevetFam = LoadBiblioFile(ResultPath, "Families" + ndf)
ListeBrevetAug = BrevetFam['brevets']
#adding already gathered
for bre in ListeBrevetAug:
DoneLab.append(bre['label'])
# if isinstance(data, collections.Mapping):
# ListeBrevetAug = data['brevets']
# else:
# ListeBrevetAug = data
flatten(DoneLab)
print(len(ListeBrevetAug),
" patents loaded, already in families list")
if len(ListeBrevetAug) == 0:
Done = []
else:
for k in ListLab: #filtering
if k not in DoneLab:
for brev in ListeBrevet:
if brev['label'] == k:
tempoList.append(brev)
ListeBrevet = tempoList
print(len(DoneLab), ' patents treated yet... doing others : ',
len(ListeBrevet))
if len(ListeBrevet) == 0:
print("Good, nothing to do!")
pickle.dump(DataBrevets, ficRes)
print ('deleted ', cpt, ' abstracts')
fic = 'Families' + ndf
if 'Description' + fic in os.listdir(ResultBiblioPath):
with open(ListBiblioPath + '//' + fic, 'r', encoding ="utf8") as data:
dico = LoadBiblioFile(ListBiblioPath, fic)
os.rename(ResultBiblioPath + '//' + fic, ResultBiblioPath + '//Old' + fic)
#the following should be done by GatherFamilies process.
# just in case...
dat = dico['brevets']
labs = [bre ['label'] for bre in dat]
labs = flatten(labs) # some patents have multiples labels
cpt1, cpt2 = 0,0
if len(labs) != len(set(labs)):
DejaVus = []
with open(ResultBiblioPath+'//Families'+ ndf, 'wb') as ndfLstBrev:
for bre in dat:
for cle in bre.keys():
if isinstance(bre[cle], list): #cleaning
bre[cle] = list(set(bre[cle]))
if isinstance(bre['label'], str):
if bre['label'] not in DejaVus:
pickle.dump(bre , ndfLstBrev)
DejaVus.append(bre['label'])
cpt1 +=1
else:
pass
Done = []
if 0 <= len(Done) <= len(ListeBrevet):
tempoList = []
try:
#ndfLstBrev = open(ResultPath+'//Families'+ ndf, 'r')
BrevetFam = LoadBiblioFile(ResultPath, "Families" + ndf)
ListeBrevetAug = BrevetFam['brevets']
#adding already gathered
for bre in ListeBrevetAug:
DoneLab.append(bre['label'])
# if isinstance(data, collections.Mapping):
# ListeBrevetAug = data['brevets']
# else:
# ListeBrevetAug = data
DoneLab = flatten(DoneLab)
DoneLab = flatten(DoneLab)
print(len(ListeBrevetAug),
" patents loaded, already in families list")
if len(ListeBrevetAug) == 0:
Done = []
else:
for k in ListLab: #filtering
if k not in DoneLab:
for brev in ListeBrevet:
if brev['label'] == k:
tempoList.append(brev)
ListeBrevet = tempoList
print(len(DoneLab), ' patents treated yet... doing others : ',
len(ListeBrevet))
if len(ListeBrevet) == 0: