For loop headache

[TMS]

I'm working on the definitive igpay atinlay assignment. I've defined a module that has one function: def igpay(word): Its sole purpose is to process a word into pig latin. It seems to work well.

Now, the 2nd part of the assignment: Define a module that takes an argument on the command line (thanks to previous questions, this is complete) and processes the entire file into pig latin.

First I went through some basic tests to process the file: Find a space (for the delimiter), find punctuation and test for a capital. When I put this into a loop my logic doesn't seem to get it past the first word.

import sys
import igpay

filename = sys.argv[1]
data = open( filename ) .read()
print data

def atinlay(data):
    for i in range(len(data)):             #begin the loop
        space = data.find(" ")            #find the first space
        period = data.find(".")            #determine punctuation to handle later
        comma = data.find(",")          
        temp = data[0:space]           #get the first slice
        a = temp[:1]                        #copy the first letter to see if it is capital
        capital = a.isupper()
        if capital:
            temp = temp.lower()         #make it lower case if it is capital
        newData = igpay.igpay(temp)    #new temp variable 
        if capital:                             #capital flag set? Handle it
            a = newData[:1]
            a = a.upper()
        newData = a + newData[1:]  #put new cap back on word
        space += 1                         #increment to new space?
        i = space                            #increment i?
        temp = newData[space:]      #thought I needed another variable here.... :(
    return newData    

c = atinlay(data)
print c

I think part of my problem is the temp assignment at the end. But I could use a gentle nudge to get this loop going because all it will do is process the first word right now.

Thank you

[bartonc]

I'll look at your code in a bit. In the mean time, something to consider is that file objects, themselves, are iterable. Text files appear as lists of end-of-line terminated strings. So you can write:

dateFile = open("filename")
for line in dataFile:
    for word in line.split():    # split() defaults to any whitespace
        print word
dataFile.close()

[TMS]

I'll look at your code in a bit. In the mean time, something to consider is that file objects, themselves, are iterable. Text files appear as lists of end-of-line terminated strings. So you can write:

dateFile = open("filename")
for line in dataFile:
    for word in line.split():    # split() defaults to any whitespace
        print word
dataFile.close()

But when its converted to a list python see's each word as 1 unit, not individual chars like a string. I tried something similar but wasn't able to process the word, so I gave up and went this direction. I appreciate your help.

[bartonc]

import sys
import igpay

filename = sys.argv[1]
data = open( filename ) .read()
print data

def atinlay(data):
    pos = 0    # need to keep track of where you are in order to "move" through data
### you can then add pos to args of find()
    for i in range(len(data)):             #begin the loop
        space = data.find(" ")            #find the first space
        period = data.find(".")            #determine punctuation to handle later
        comma = data.find(",")          

### data[0] should be data[pos] or you'll always start at the beginning
        #### temp assigned at the end of the loop gets reassigned here ####
        temp = data[0:space]           #get the first slice
        a = temp[:1]                        #copy the first letter to see if it is capital
        capital = a.isupper()

### you can use str.capitalize() to upper the first letter
        if capital:
            temp = temp.lower()         #make it lower case if it is capital
        newData = igpay.igpay(temp)    #new temp variable 
        if capital:                             #capital flag set? Handle it
            a = newData[:1]
            a = a.upper()
        newData = a + newData[1:]  #put new cap back on word

### This would be you next Pos
        pos = space + 1
#        space += 1                         #increment to new space?

### Even if this works, it's bad practice to change for's variable
### I'm not actully sure what happens
#        i = space                            #increment i?
### and besides, I don't see it used anywhere


        temp = newData[space:]      # this will be replaced at the top of the next loop
    return newData    

c = atinlay(data)
print c

[TMS]

I am still only getting one word processed (at least printed).

The text file I'm working with is nonsensical, intended for testing. The result I get is this:

C:\Python25>ati nlay.py someFile.txt
NewFile and, more new file.

Ewfilenay

C:\Python25>

It is capitalizing appropriately the first letter of the only word it processes. This is the same problem I was having before. Any ideas? :)

[bartonc]

This'll get you started. Create an empty list, then append() to it and return the list.

import sys
import igpay

filename = sys.argv[1]
data = open( filename ) .read()
print data

def atinlay(data):
    pos = 0    # need to keep track of where you are in order to "move" through data
    resultList = []
### you can then add pos to args of find()
    for i in range(len(data)):             #begin the loop
        space = data.find(" ")            #find the first space
        period = data.find(".")            #determine punctuation to handle later
        comma = data.find(",")          

### data[0] should be data[pos] or you'll always start at the beginning
        #### temp assigned at the end of the loop gets reassigned here ####
        temp = data[0:space]           #get the first slice
        a = temp[:1]                        #copy the first letter to see if it is capital
        capital = a.isupper()

### you can use str.capitalize() to upper the first letter
        if capital:
            temp = temp.lower()         #make it lower case if it is capital
        newData = igpay.igpay(temp)    #new temp variable 
        if capital:                             #capital flag set? Handle it
            a = newData[:1]
            a = a.upper()
        newData = a + newData[1:]  #put new cap back on word

### This would be you next Pos
        pos = space + 1
#        space += 1                         #increment to new space?

### Even if this works, it's bad practice to change for's variable
### I'm not actully sure what happens
#        i = space                            #increment i?
### and besides, I don't see it used anywhere

        resultList.append(newData + " ")
#        temp = newData[space:]      # this will be replaced at the top of the next loop
    return resultList

c = atinlay(data)
print c

[/QUOTE]

[dshimer]

I haven't studied every snip of code, but based on what I understand, can I interject something as seen from another direction. As I understand it igpay() is supposed to take any word you send it and convert to the new string, and atinlay() should read through a whole file of text converting each word and capitalizing if the word falls after a period (or is already capitalized).

Could it possibly be easier to

1) write igpay so that if you send it a properly capitalized word it translates it to a properly capitalized word in the new language, or if you send it a word that has punctuation at the end it returns the translated word with the same punctuation.

Then..

2) Just take the whole data stream, split at white spaces (which will keep the puctuation with the word proceeding it), process it in a linear fashion from beginning to end. It could even test a word so that if a period is found in this string then make sure the next is capitalized before sending to igpay().

It seems if it were approached from this directon igpay() would need a couple more lines, but atinlay would just be a simple..

read data
split it
for each word in that list
convert it and append to the output capitalizing if the previous word contained a period.

[TMS]

OK, so, now I've changed igpay to do capitalization. The same problem still remains. I can't seem to process the list. It only does one word, the first word. If I use split() it will be made into a list (the text file) and I won't be able to process it because it will no longer be a string. Is there a way to change it back into a string after making it a list? Lists are tuples, right?

Here is my code:

import sys
import igpay

filename = sys.argv[1]
data = open( filename ) .read()
print data

def atinlay(data):
    pos = 0                             # begin position
    for i in range(len(data)):       # begin loop
        space = data.find(" ")      # find space for delimiter
        #period = data.find(".")      # set a flag for punctuation period
        #comma = data.find(",")    # set a flag for punctuation comma
        temp = data[0:space]      # slice the first word
        newData = igpay.igpay(temp)   #place to put processed words
        pos = space + 1
        temp = newData[space:]
    return newData    

c = atinlay(data)
print c

[bvdet]

OK, so, now I've changed igpay to do capitalization. The same problem still remains. I can't seem to process the list. It only does one word, the first word. If I use split() it will be made into a list (the text file) and I won't be able to process it because it will no longer be a string. Is there a way to change it back into a string after making it a list? Lists are tuples, right?

Here is my code:

import sys
import igpay

filename = sys.argv[1]
data = open( filename ) .read()
print data

def atinlay(data):
    pos = 0                             # begin position
    for i in range(len(data)):       # begin loop
        space = data.find(" ")      # find space for delimiter
        #period = data.find(".")      # set a flag for punctuation period
        #comma = data.find(",")    # set a flag for punctuation comma
        temp = data[0:space]      # slice the first word
        newData = igpay.igpay(temp)   #place to put processed words
        pos = space + 1
        temp = newData[space:]
    return newData    

c = atinlay(data)
print c

Lists and tuples are similar but different. Lists are mutable and tuples are not. To make a string from a list:

>>> lst = ['I', 'am', 'a', 'detailer']
>>> " ".join(lst)
'I am a detailer'
>>>

It looks like you are only processing the first word in each loop. I do not see where you are accumulating an output string. You could do something like this, but your igpay function would need to handle the capitalization and punctuation:

def process_file(fn):
    f = open(fn)
    outStr = ""
    for line in f:
        lineList = line.split(" ") # split on space character
        lineListOut = []
        for word in lineList:
            lineListOut.append(igpay.igpay(word))
        outStr += " ".join(lineListOut)
    print outStr

HTH

[dshimer]

You could do something like this, but your igpay function would need to handle the capitalization and punctuation:

Very clean, now the for loop is simply doing what it does best, working through the sequence of words, and since split should send in the punctuation along with the single word that preceeds it, "handling" it could be as simple as...
Test if it's there.
If so remove it.
Process the string.
Replace punctuation and return.

[TMS]

Wow... very nice. It processes the list, but appends a bunch of stuff. The list after processing looks like this:

NewaywFiwaylewa yway wayawayd,way wyamovwayrewayw ay waynewaywway wayfiwayleway.

LOL, its an entirely new language. I should name it...

I need to go through and see what is happening, but it is processing the list. I think I can handle it from here. Thank you!

[dshimer]

One more thing, it blows me away the little things you can miss as you go along. In case you didn't see it, look at the last few posts in the
how to convert gpr file to csv format: using python thread. The fileinput tip is worth the price of admission all by itself and I had never looked at it before ghostdog74 mentioned it.

I need to go through and see what is happening, but it is processing the list. I think I can handle it from here. Thank you!

[bvdet]

One more thing, it blows me away the little things you can miss as you go along. In case you didn't see it, look at the last few posts in the
how to convert gpr file to csv format: using python thread. The fileinput tip is worth the price of admission all by itself and I had never looked at it before ghostdog74 mentioned it.

The fileinput was new to me also. Good tip. One more thing - it is good practice to close each file you open when you are through with it:

f.close()

[TMS]

Its all done. Thank you so much for your help. It works very well, thanks to all your help!

[bartonc]

Its all done. Thank you so much for your help. It works very well, thanks to all your help!

Thanks for the update. I'm glad the experts here were of help to you, Keep posting.