How do I prune all the '' from these strings generated from a CSV?

Question

I've made this formula which turns a CSV into a dictionary:

def CSVtoDict(BDF, mode):
    saved={}
    with open('%s%s.csv' % (dataDir,BDF), mode='r') as infile:
        reader = csv.reader(infile)
        for row in reader:
            if mode is 'prune' and row == '':
                break
            else:
                pass
            key = row[0]
            saved[key]=row[1:]
    return saved

I'm trying to make it so that when the mode is 'prune', any string from the CSV such as ['data1', 'data2', '', '', ''] is returned without any ''. But the break statement doesn't seem to work for some reason and things are returned as if there was no if else.

Also do you know what would be the pythonic way of doing this? I feel like this could be more efficient somehow...

Example line from the CSV:

Awareness,ASD,ASD2,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

Can you show us a line or two of your input .csv file. I'm having trouble understanding your issue with single quotes. Are you saying the single quotes are in the file? — BoarGules, Jun 09 '17 at 17:35
wait a minute, for this list: ['data1', 'data2', '', '', ''], is that the list result that you want? -> ['data1', 'data2']? — developer_hatch, Jun 09 '17 at 17:52
Are there multiple blank entries per row, with multiple rows that can have blanks? — Tom Sitter, Jun 09 '17 at 17:59
Could you edit your post with a small snippet straight from your CSV file? Something that would demonstrate what your file looks like? I seriously doubt that your actual csv file contains ['data1','data2','','','']. — ashbygeek, Jun 09 '17 at 18:01
@BoarGules A line from the csv would be: Awareness,ASD,ASD2,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,, I want to prune all those '' — Miguel Alberola Cano, Jun 09 '17 at 18:10
@TomSitter Every row has a lot of blank entries. But they're all at the end, none in the middle or start. — Miguel Alberola Cano, Jun 09 '17 at 18:10
@MiguelAlberolaCano Okey perfect, I updated the answer, don't forget voting and accepting if that's what you were looking at :) — developer_hatch, Jun 09 '17 at 18:20

OldGeeksGuide · Answer 1 · 2017-06-09T17:47:18.367

If I understand your question correctly, you're trying to filter out empty rows when mode == 'prune', is that correct? Or are you trying to stop as soon as you hit an empty row (which is what the break would achieve if your conditional was correct)?

The problem is that csv rows are returned as lists, not strings, i.e. a line in a csv file that looks like this:

a,b,1,2

will be returned as ['a','b','1','2'] when iterating through reader. An empty row comes up as [], not ''. If you want to check for an empty row, you can check for len(row) == 0 instead. Also, your else: pass doesn't really do anything so you can leave it out. Result should be something like this:

for row in reader:
    if mode is 'prune' and len(row) == 0:
        break # or continue, if you want to proceed with the rest of the file
    key = row[0]
    saved[key]=row[1:]

If as another answer suggests, you're trying to prune out any row that has an empty field, then change len(row) == 0 to '' in row.

Thanks! I just want to prune the [] rows, not the entire row. Lemme try it out! — Miguel Alberola Cano, Jun 09 '17 at 18:25

score 1 · Answer 2 · answered Jun 09 '17 at 17:33

1

I'm thinking you may want to use continue rather than break. Break will take you out of the if/else statement, but continue will take you to the next iteration of the for loop.

answered Jun 09 '17 at 17:33

Tyler R

474
2
6
15

1

Break interrupts loops, not if/else blocks. I think you have the right idea, just revise the terminology. – Davy M Jun 09 '17 at 17:49

score 1 · Answer 3 · answered Jun 09 '17 at 17:39

1

The condition row == '' means the whole row is an empty string. YOu could substitute it with '' in row so:

if mode is 'prune' and '' in row:

answered Jun 09 '17 at 17:39

user2314737

27,088
20
102
114

I think the op want anything else :( – developer_hatch Jun 09 '17 at 18:22
1

@MiguelAlberolaCano this will skip all the rows that contain an empty field somewhere, I also doubt that you might need that – user2314737 Jun 09 '17 at 18:33
I just want to prune the '', not the whole row. You're right – Miguel Alberola Cano Jun 09 '17 at 18:33

developer_hatch · Answer 4 · 2017-06-09T18:20:06.720

1

wait a minute, is that what you want? if you want to remove all the '' if mode == "prune" just do:

def CSVtoDict(BDF, mode):
    with open('%s%s.csv' % (dataDir,BDF), mode='r') as infile:
        reader = csv.reader(infile)
        return [ row for row in reader if row != '' ] if (mode == "prune") else reader

edited Jun 09 '17 at 18:20

answered Jun 09 '17 at 17:46

developer_hatch

15,898
3
42
75

I think that's it! But where would I put it within my function? – Miguel Alberola Cano Jun 09 '17 at 18:29
Whoopsie. I need a dictionary not a list. I'll try to see if I can adapt it. Thanks :) – Miguel Alberola Cano Jun 09 '17 at 18:31

Tom Sitter · Accepted Answer · 2017-06-09T18:38:10.103

1

I think the problem is that you still have to loop through each entry, not just each row. Here is my solution:

# CSV contents
#  'data1', 'data2', 'data3', ''
#  'data4', 'data5', '' ''
#  'data6', '', '', ''

def CSVtoDict(BDF, mode):
    saved={}
    with open('%s%s.csv' % (dataDir,BDF), mode='r') as infile:
        reader = csv.reader(infile)
        for row in reader:
            key = row[0]
            saved[key] = [r for r in row[1:] if not (r is '' and mode is 'prune')]
    return saved


# saved will equal
# {
#   'data1': ['data2', 'data3']
#   'data4': ['data5'], 
#   'data6': [], 
# }

edited Jun 09 '17 at 18:38

answered Jun 09 '17 at 18:29

Tom Sitter

1,082
1
10
23

Trying this out! – Miguel Alberola Cano Jun 09 '17 at 18:37
I've turned it into a function, it works! Thanks! :) def CSVtoDict(BDF, mode): with open('%s%s.csv' % (dataDir,BDF), mode='r') as infile: reader = csv.reader(infile) saved = {} for row in reader: key = row[0] saved[key] = [r for r in row[1:] if not (r is '' and mode is 'prune')] return saved – Miguel Alberola Cano Jun 09 '17 at 18:40
Sorry if it's a mess, I don't know how to put it in a clearer format in the comments :( – Miguel Alberola Cano Jun 09 '17 at 18:41
You did it and it looks identical to mine! yay – Miguel Alberola Cano Jun 09 '17 at 18:43
No problem! Glad to help – Tom Sitter Jun 09 '17 at 19:28

How do I prune all the '' from these strings generated from a CSV?

5 Answers5