extract_info.py

#coding=utf-8
 
import re
 
def extract_info(pattern, input_file_name, output_file_name, prefix_list):
  output_file = open(output_file_name, 'w')
	for line in open(input_file_name,'r'):
		m = re.match(pattern, line)
		if m:
			origin_field = m.group(1)
			for prefix in prefix_list:
				if origin_field.startswith(prefix):
					origin_field = origin_field[len(prefix):]
					break
			table_field = m.group(1)
			table_type = m.group(2)
			comment = m.group(3)
			sample = m.group(4)
			result_line = '\t'.join([origin_field, table_field, table_type, comment, sample])+'\n'
			output_file.write(result_line)

	output_file.close()
 
if __name__ == '__main__':
	pattern = r"\s*<Attribute\s*name\s*=\s*\"(.*?)\".*type\s*=\s*\"(.*?)\"\s*comment\s*=\s*\"(.*?)\"\s*sample\s*=\s*\"(.*?)\".*"
	input_file_name = 'udwetl_nsclick.xml'
	output_file_name = 'out1'
	prefix_list = ['event_', 'nsclick_']
	extract_info(pattern, input_file_name, output_file_name, prefix_list)
	print 'ok'