数据字典逻辑到物理缩写名称转换

Question

我需要将逻辑数据字典转换为物理（缩写）数据字典-我在下面给出了4个用例。你知道吗

需要此psuedo代码/要求的帮助：

# empty dict declaration
refDict = {}

# to catch and report on any 'not-found' dictionary words to replace
noMatchFound = {}

# read from a dictionary of comma delimited dictionary
# with open('dictionary.csv') as inputDict:
#     for line in inputDict:
#         busTerm, busAbbr = line.split(',')
#         refDict[busTerm] = busAbbr.replace("\n","")

# sample data dictionary entries
refDict = {
         'user': 'USR',
         'call': 'CALL', 
         'detail': 'DTL', 
         'record': 'REC', 
         'call detail record': 'CDR',
         'count', 'CNT'}


input_string1="user call detail record"
# output should be "USR_CDR"
# noMatchFound - will be empty - since all are matched and replaced

input_string2="user test call detail record"
# output should be "USR_TEST_CDR"
# noMatchFound - should have an entry "TEST" with a refernce to "user test call detail record"

input_string3="user call count detail record"
# output should be "USR_CALL_CNT_DTL_REC"
# noMatchFound - will be empty - since all are matched and replaced

input_string4="user call  detail record count"
# output should be "USR_CDR_CNT"
# noMatchFound - will be empty - since all are matched and replaced

到目前为止，我可以找出匹配任何可能的单个最大表达式的代码段，如下所示：

import re

# using regular expressions find longest matcing expression
def getLongestSequenceSize(inputStr,  inDict):
    ret_match = ""
    ret_match_len = 0
    ret_abbr = ""
    for inKey in inDict:
        matches = re.findall(r'(?:\b%s\b\s?)+' % inKey.strip().upper(), inputStr.strip().upper())
        if len(matches) > 0:
            longest_match = max(matches)
            if ret_match_len <  len(longest_match):
                ret_match_len = len(longest_match)
                ret_match = longest_match.strip()
                ret_abbr = inDict[inKey]
    return [ret_match.strip(), ret_abbr.strip()]

Answer 1

这个想法是你开始尝试replace()从字典中最大的字符串开始，然后检查给定字典的每一个可能的替换，从长到短。你知道吗

这正是你所期望的：

refDict = {
         'user': 'USR',
         'call': 'CALL',
         'detail': 'DTL',
         'record': 'REC',
         'call detail record': 'CDR',
         'count': 'CNT'}

sorted_ref = sorted( refDict.items(), key=lambda x:len(x[0]), reverse = True )

def do_work(input_string):
    noMatchFound = {}
    rval = input_string[:]
    for key, value in sorted_ref:
        rval = rval.replace(key, value)
    not_founds = [x for x in rval.split() if x.islower()]
    for not_found in not_founds:
        noMatchFound[not_found] = input_string
        rval = rval.replace(not_found, not_found.upper())
    rval = '_'.join( rval.split() )
    return rval, noMatchFound

inputs = ["user call detail record", "user test call detail record",
          "user call count detail record","user call detail record count"]

for inp in inputs:
    print inp
    output, noMatchFound = do_work(inp)
    print output
    print noMatchFound
    print ' -'

输出：

user call detail record
USR_CDR
{}
 -
user test call detail record
USR_TEST_CDR
{'test': 'user test call detail record'}
 -
user call count detail record
USR_CALL_CNT_DTL_REC
{}
 -
user call detail record count
USR_CDR_CNT
{}