Dealing with a dirty table

Question

I have to deal with a really ugly looking response I get back from a particular device. The string represents a table. Each section of the string represents the form:

["id", "date_time", "path", "type","duration","count", "rate"]

Which as an example looks like:

"1,2014-4-3 10:1:10,somerandompath,some/screwy!stringwithstufflike[0,0,0,0]inIt,0,0;"

all table entries as you can see are terminated by ; and have their elements separated by ,

But some entries like in some/screwy!stringwithstufflike[0,0,0,0]inIt have commas within their data,.. and times/dates 10:1:10 aren't padded with 0's to be consistent

This is what I have so far, but it doesn't feel very optimized...

def GetEventTable(self):
        # After I split by ',' commas can be inside of [..], 
        # so I combine entries until their number of starting/end braces match
        def split_entry(entry):
            splitentry = entry.split(',')
            for ii in range(len(splitentry))[::-1]:
                start_bracket = splitentry[ii].count('[')
                end_bracket = splitentry[ii].count(']')
                if not start_bracket == end_bracket:
                    splitentry[ii-1:ii+1] = [splitentry[ii-1] + ',' + splitentry[ii]]
            return splitentry
        event_lable = ["id", "date_time", "path", "type","duration","count", "rate"]
        ## This section just talks to the remote device and gets the unformatted string ###
        print "Sending: GET EVENT TABLE"
        stat, raw_data = SendRecieve(self, "GET EVENT TABLE")
        if IsError(stat):
            raise Exception("Error type '{}': {}".format(stat, raw_data))
        print raw_data
        ##################################################################################
        events = raw_data.split(';')[:-1]
        event_table = [{key:value for key,value in zip(event_lable,split_entry(entry) )} for entry in events ]
        return event_table

How can I improve the efficiency and also the readability?

It returns a list of dictionaries with keys ["id", "date_time", "path", "type","duration","count", "rate"]

Answer 1

Disclaimer

Because a lot of context is missing, here's the code I will review :

def split_entry(entry):
    splitentry = entry.split(',')
    for ii in range(len(splitentry))[::-1]:
        start_bracket = splitentry[ii].count('[')
        end_bracket = splitentry[ii].count(']')
        if not start_bracket == end_bracket:
            splitentry[ii-1:ii+1] = [splitentry[ii-1] + ',' + splitentry[ii]]
    return splitentry

event_lable = ["id", "date_time", "path", "type","duration","count", "rate"]
raw_data = "1,2014-4-3 10:1:10,somerandompath,some/screwy!stringwithstufflike[0,0,0,0]inIt,0,0;"
events = raw_data.split(';')[:-1]
event_table = [{key:value for key,value in zip(event_lable,split_entry(entry) )} for entry in events ]
print event_table

---

ii is a pretty bad name. i is twice shorter and so many times easier to understand.

---

Instead of creating a new list by reversing the return of range, you could just use all the potential of range (or xrange):

---

Instead of using not A == B, you could write a more natural A != B

---

Looping over an array using range(len(a)) is an anti-pattern in python. To loop on an array while keeping track of the index, the usual solution is to use enumerate(). In your case, because you are updating the array anyway, it is quite tricky to loop on it.

---

Final point : I had trouble understanding your list slicing stuff but as I was messing a bit with it, I obtained a result different from yours ('rate' appears in the final result so it might actually be a good thing) :

def split_entry(entry):
    splitentry = entry.split(',')
    for i in xrange(len(splitentry)-1,0,-1):
        chunk = splitentry[i]
        if chunk.count('[') != chunk.count(']'):
                splitentry[i-1] += ',' + chunk
                splitentry[i] = '0'
    return splitentry

event_lable = ["id", "date_time", "path", "type","duration","count", "rate"]
raw_data = "1,2014-4-3 10:1:10,somerandompath,some/screwy!stringwithstufflike[0,0,0,0]inIt,0,0;"
events = raw_data.split(';')[:-1]
event_table = [{key:value for key,value in zip(event_lable,split_entry(entry) )} for entry in events]
print event_table