def getWords():
inputPath = '/home/aishwary/Desktop/ltrc/inference'
fileList = ssfAPI.folderWalk(inputPath)
newFileList = []
for fileName in fileList:
xFileName = fileName.split('/')[-1]
if xFileName == 'err.txt' or xFileName.split('.')[-1] in [
'comments', 'bak'
] or xFileName[:4] == 'task':
continue
else:
newFileList.append(fileName)
words = []
for fileName in newFileList:
d = ssfAPI.Document(fileName)
for tree in d.nodeList:
for chunkNode in tree.nodeList:
for node in chunkNode.nodeList:
words.append(node.type.encode('utf-8'))
return words
def readFilesAndComputeHeadOfChunks(inputPath, chunkTagPOSMapping, outputPath):
if not os.path.isdir(inputPath):
d = ssfAPI.Document(inputPath)
headComputedSentences = computeHeadOfChunks(d, chunkTagPOSMapping)
writeListToFile(headComputedSentences, outputPath)
else:
fileList = ssfAPI.folderWalk(inputPath)
newFileList = list()
for fileName in fileList:
xFileName = fileName.split('/')[-1]
if xFileName == 'err.txt' or xFileName.split('.')[-1] in ['comments', 'bak'] or xFileName[:4] == 'task':
continue
else:
newFileList.append(fileName)
for fl in newFileList:
d = ssfAPI.Document(fl)
headComputedSentences = computeHeadOfChunks(d, chunkTagPOSMapping)
print( fl[fl.rfind('/') + 1:] + '.head')
writeListToFile(headComputedSentences, os.path.join(outputPath, fl[fl.rfind('/') + 1:] + '.head'))
return fileList
if __name__ == '__main__' :
inputPath = sys.argv[1]
fileList = folderWalk(inputPath)
newFileList = []
for fileName in fileList :
xFileName = fileName.split('/')[-1]
if xFileName == 'err.txt' or xFileName.split('.')[-1] in ['comments','bak'] or xFileName[:4] == 'task' :
continue
else :
newFileList.append(fileName)
for fileName in newFileList :
d = ssfAPI.Document(fileName)
writeTo = './finalTest/' + fileName.split('/')[-1]
f = open(writeTo,'w')
#print f
sentIndex = 0
for tree in d.nodeList :
try :
flag_pbrel = 0
tree.populateNodes(naming='strict')
tree.populateEdges()
for chunkNode in tree.nodeList :
if chunkNode.parentPB != '0' :
flag_pbrel=1
break
if flag_pbrel == 1:
string = ""