string.count issue (i'm stupid?)

[Matteo Rattotti]

Hi all,

i've noticed a strange beaviour of string.count:

in my mind this code must work in this way:

str = "a_a_a_a_"
howmuch = str.count("_a_" )
print howmuch -> 3

but the count return only 2

Ok this can be fine, but why? The doc string tell that count will
return the number of substring in the master string, if we spoke about
substring i count 3 substring...

Can someone explain me this? And in which way i can count all the
occurrence of a substring in a master string? (yes all occurrence
reusing already counter character if needed)

Thanks a lot

Matteo Rattotti www.rknet.it
Powered by:
- MacOsX
- Gnu / Linux Debian Sarge
- Amiga Os 3.9
- Milk

[Dirk Hagemann]

Re: string.count issue (i'm stupid?)

I think I can tell you WHY this happens, but I don't know a work-around
at the moment.
It seems as if only the following "_a_" (A) are counted: a_A_a_A_

regards
Dirk

[Diez B. Roggisch]

Re: string.count issue (i'm stupid?)

Matteo Rattotti wrote:
[color=blue]
> Hi all,
>
> i've noticed a strange beaviour of string.count:
>
> in my mind this code must work in this way:
>
> str = "a_a_a_a_"
> howmuch = str.count("_a_" )
> print howmuch -> 3
>
> but the count return only 2
>
> Ok this can be fine, but why? The doc string tell that count will
> return the number of substring in the master string, if we spoke about
> substring i count 3 substring...[/color]

It appears to be a documentation bug. It should say something about
non-overlapping occurences.

Better would be of course to introduce a parameter that defines the
behavior, either overlapping or not.

Diez

[Alexandre Fayolle]

Re: string.count issue (i'm stupid?)

Le 22-05-2006, Matteo <matteo.rattott i@gmail.com> nous disait:[color=blue]
> Hi all,
>
> i've noticed a strange beaviour of string.count:
>
> in my mind this code must work in this way:
>
> str = "a_a_a_a_"
> howmuch = str.count("_a_" )
> print howmuch -> 3
>
> but the count return only 2
>
> Ok this can be fine, but why? The doc string tell that count will
> return the number of substring in the master string, if we spoke about
> substring i count 3 substring...
>
> Can someone explain me this? And in which way i can count all the
> occurrence of a substring in a master string? (yes all occurrence
> reusing already counter character if needed)[/color]

Use the optional start argument of find or index in a loop, such as:
[color=blue][color=green][color=darkred]
>>> def count_all(strin g, substring):[/color][/color][/color]
.... index = 0
.... count = 0
.... while True:
.... index = string.find(sub string, index)
.... if index < 0:
.... return count
.... else:
.... count += 1
.... index += 1
....[color=blue][color=green][color=darkred]
>>> count_all("a_a_ a_a_", '_a_')[/color][/color][/color]
3


--
Alexandre Fayolle LOGILAB, Paris (France)
Formations Python, Zope, Plone, Debian: http://www.logilab.fr/formations
Développement logiciel sur mesure: http://www.logilab.fr/services
Python et calcul scientifique: http://www.logilab.fr/science

[Alexander Schmolck]

Re: string.count issue (i'm stupid?)

"Dirk Hagemann" <DirkHagemann@g mail.com> writes:
[color=blue]
> I think I can tell you WHY this happens, but I don't know a work-around
> at the moment.[/color]

len(re.findall( '_(?=a_)', '_a_a_a_a_'))

# untested
def countWithOverla ps(s, pat):
return len(re.findall( "%s(?=%s)" % (re.escape(pat[0]), re.escape(pat[1:])),s))

'as

[bruno at modulix]

Re: string.count issue (i'm stupid?)

Matteo Rattotti wrote:[color=blue]
> Hi all,
>
> i've noticed a strange beaviour of string.count:
>
> in my mind this code must work in this way:
>
> str = "a_a_a_a_"[/color]

dont use 'str' as an identifier, it shadows the builtin str type.
[color=blue]
> howmuch = str.count("_a_" )
> print howmuch -> 3
>
> but the count return only 2
>
> Ok this can be fine, but why? The doc string tell that count will
> return the number of substring in the master string, if we spoke about
> substring i count 3 substring...[/color]

depends on how you define "number of substring", I mean, overlapping or
not. FWIW, I agree that this may be somewhat unintuitive, and would at
least require a little bit more precision in the docstring.
[color=blue]
> Can someone explain me this?[/color]

It seems obvious that str.count counts non-overlapping substrings.
[color=blue]
> And in which way i can count all the
> occurrence of a substring in a master string? (yes all occurrence
> reusing already counter character if needed)[/color]

Look at the re module.

--
bruno desthuilliers
python -c "print '@'.join(['.'.join([w[::-1] for w in p.split('.')]) for
p in 'onurb@xiludom. gro'.split('@')])"

[Tim Chase]

Re: string.count issue (i'm stupid?)

I agree the docstring is a bit confusing and could be clarified
as to what's happening
[color=blue]
> Can someone explain me this? And in which way i can count all
> the occurrence of a substring in a master string? (yes all
> occurrence reusing already counter character if needed)[/color]


You should be able to use something like

s = "a_a_a_a_"
count = len([i for i in range(len(s)) if s.startswith("_ a_", i)])

which will count the way you wanted, rather than the currently
existing count() behavior.

-tkc

[BartlebyScrivener]

Re: string.count issue (i'm stupid?)

We were doing something like this last week

thestring = "a_a_a_a_"[color=blue][color=green][color=darkred]
>>> for x in range(len(thest ring)):[/color][/color][/color]
.... try:
.... thestring.count ("_a_", x, x + 3)
.... except ValueError:
.... pass