def load_citation_data(self):
"""Adds citation links to the paper objects """
print "Loading paper citations"
dirs = get_dir_names(self.releasePath)
#remove directories that are not years
for dir in dirs:
dirName = dir.split("/")[len(dir.split("/"))-1]
if dirName.isdigit():
continue
else:
dirs.remove(dir)
for dir in dirs:
citationFileReader = open(dir+"/acl.txt")
for line in citationFileReader:
decodedLine = line.decode("iso-8859-1")
encodedLine = decodedLine.encode("utf-8", 'ignore')
cleanedLine = line.replace('\n', '').rstrip()
if cleanedLine == '':
# if it is a blank line skip it
continue
lineParts = cleanedLine.split("==>")
sourcePaperId = lineParts[0].strip()
citedPaperId = lineParts[1].strip()
if self.paperInfo.has_key(citedPaperId):
paper = self.paperInfo[citedPaperId]
if paper.has_key(self.paperFields.CITATION_LIST):
paper[self.paperFields.CITATION_LIST].append(sourcePaperId)
else:
paper[self.paperFields.CITATION_LIST] = []
paper[self.paperFields.CITATION_LIST].append(sourcePaperId)
citationFileReader.close()
def load_paper_metadata(self):
"""creates initial paper objects with metadata from acl-metadata.txt """
print "Loading paper metadata"
dirs = get_dir_names(self.releasePath)
#remove directories that are not years
for dir in dirs:
dirName = dir.split("/")[len(dir.split("/"))-1]
if dirName.isdigit():
continue
else:
dirs.remove(dir)
currentPaper = {}
isLineContinuation = False
for dir in dirs:
metaDataReader = open(dir+"/acl-metadata.txt", 'r')
prevLine = ""
prevValue = ""
prevField = ""
for line in metaDataReader:
decodedLine = line.decode("iso-8859-1")
encodedLine = decodedLine.encode("utf-8", 'ignore')
cleanedLine = encodedLine.replace("\n", '').rstrip()
if not isLineContinuation:
#checking for continuation lines
if cleanedLine == '':
#if line is blank create new paper object and skip line
if prevLine == '' and cleanedLine == '':
prevLine = cleanedLine
continue
paperId = currentPaper[self.paperFields.AAN_ID]
self.paperInfo[paperId] = currentPaper
currentPaper = {}
prevLine = cleanedLine
continue
lineParts = cleanedLine.split("=")
if len(lineParts) == 2:
field = lineParts[0].strip()
value = lineParts[1].strip()
if '}' in value:
self.load_field_value(currentPaper, field, value)
isLineContinuation = False
else:
prevField = field
prevValue = value
isLineContinuation = True
else:
#continued line
field = prevField
value = prevValue +" "+cleanedLine
if '}' in value:
self.load_field_value(currentPaper, field, value)
isLineContinuation = False
else:
prevField = field
prevValue = value
isLineContinuation = True
prevLine = cleanedLine
metaDataReader.close()