removing duplicates, or, converting Set() to string

[maphew@gmail.com]

Hello,

I have some lists for which I need to remove duplicates. I found the
sets.Sets() module which does exactly this, but how do I get the set
back out again?

# existing input: A,B,B,C,D
# desired result: A,B,C,D

import sets
dupes = ['A','B','B','C' ,'D']
clean = sets.Set(dupes)

out = open('clean-list.txt','w')
out.write(clean )
out.close

---
out.write(clean ) fails with "TypeError: argument 1 must be string or
read-only character buffer, not Set" and out.write( str(clean) )
creates "Set(['A', 'C', 'B', 'D'])" instead of just A,B,C,D.

thanks in advance for your time,

-matt

[bearophileHUGS@lycos.com]

Re: removing duplicates, or, converting Set() to string

The write accepts strings only, so you may do:

out.write( repr(list(clean )) )

Notes:
- If you need the strings in a nice order, you may sort them before
saving them:
out.write( repr(sorted(cle an)) )
- If you need them in the original order you need a stable method, you
can extract the relevant code from this large blob:

http://aspn.activestate.com/ASPN/Cookbook/Python/Recipe/438599

Bye,
bearophile

[Simon Forman]

Re: removing duplicates, or, converting Set() to string


maphew@gmail.co m wrote:

Hello,
>
I have some lists for which I need to remove duplicates. I found the
sets.Sets() module which does exactly this, but how do I get the set
back out again?
>
# existing input: A,B,B,C,D
# desired result: A,B,C,D
>
import sets
dupes = ['A','B','B','C' ,'D']
clean = sets.Set(dupes)
>
out = open('clean-list.txt','w')
out.write(clean )
out.close
>
---
out.write(clean ) fails with "TypeError: argument 1 must be string or
read-only character buffer, not Set" and out.write( str(clean) )
creates "Set(['A', 'C', 'B', 'D'])" instead of just A,B,C,D.
>
thanks in advance for your time,
>
-matt

Do ','.join(clean) to make a single string with commas between the
items in the set. (If the items aren't all strings, you'll need to
convert them to strings first.)

Peace,
~Simon

[John Machin]

Re: removing duplicates, or, converting Set() to string

maphew@gmail.co m wrote:

Hello,
>
I have some lists for which I need to remove duplicates. I found the
sets.Sets() module which does exactly this

I think you mean that you found the sets.Set() constructor in the set
module.
If you are using Python 2.4, use the built-in set() function instead.
If you are using Python 2.3, consider upgrading if you can.

but how do I get the set
back out again?

>
# existing input: A,B,B,C,D
# desired result: A,B,C,D
>
import sets
dupes = ['A','B','B','C' ,'D']
clean = sets.Set(dupes)
>
out = open('clean-list.txt','w')
out.write(clean )
out.close
>
---
out.write(clean ) fails with "TypeError: argument 1 must be string or
read-only character buffer, not Set"

as expected

and out.write( str(clean) )
creates "Set(['A', 'C', 'B', 'D'])" instead of just A,B,C,D.

again as expected.

BTW, in practice you'd probably want to append '\n' to the string that
you're writing.

You should be able to get a (possibly unsorted) list of the contents of
*any* container like this (but note that dictionaries divulge only
their keys):

>>dupes = ['A','B','B','C' ,'D']
>>clean = set(dupes) # the Python 2.4+ way
>>clean

set(['A', 'C', 'B', 'D'])

>>[x for x in clean]

['A', 'C', 'B', 'D']

>>>

list(clean) would work as well.

If you want the output sorted, then use the list.sort() method. Details
in the manual.

HTH,
John

[John Machin]

Re: removing duplicates, or, converting Set() to string

Simon Forman wrote:

>
Do ','.join(clean) to make a single string with commas between the
items in the set. (If the items aren't all strings, you'll need to
convert them to strings first.)
>

And if the items themselves could contain commas, or quote characters,
you might like to look at the csv module.

[maphew]

Re: removing duplicates, or, converting Set() to string

thank you everybody for your help! That worked perfectly. :) I really
appreciate the time you spent answering what is probably a pretty basic
question for you. It's nice not to be ignored.

be well,

-matt