def procDir(args,paramLst):
vector_f = args[2]
dirlst = args[3:]
for dir in dirlst:
docs = listDir(dir)
if dir[-1] != '/':
idx_base = dir.rfind('/')
label = dir[idx_base+1:]
else:
idx_base = dir[:-1].rfind('/')
label = dir[idx_base+1:-1]
dir = dir[:-1]
num = 0
for doc in docs:
num += 1
fname = dir+'/'+doc
procFile(paramLst,fname,label,vector_f)
return [fname.rstrip() for fname in p.stdout.readlines()]
##read in the dir names, process files and output to train_vector_file and test_vector_file
def procDir(args,paramLst):
vector_f = args[2]
dirlst = args[3:]
for dir in dirlst:
docs = listDir(dir)
if dir[-1] != '/':
idx_base = dir.rfind('/')
label = dir[idx_base+1:]
else:
idx_base = dir[:-1].rfind('/')
label = dir[idx_base+1:-1]
dir = dir[:-1]
num = 0
for doc in docs:
num += 1
fname = dir+'/'+doc
procFile(paramLst,fname,label,vector_f)
# print 'docs',num
if __name__ == "__main__":
option,args = readCommand(argv)
paramLst = getParams(args)
if option.procFile: ##if '-f', process a single file
procFile(paramLst,*args[2:]) #exclude paramfname from args
else: ##without '-f', process all files under given dirs
procDir(args,paramLst)