Dynamic Lists, or...?

[Lorn]

I'm trying to figure out a way to create dynamic lists or possibly
antother solution for the following problem. I have multiple lines in a
text file (every line is the same format) that are iterated over and
which need to be compared to previous lines in the file in order to
perform some simple math. Each line contains 3 fileds: a descriptor and
two integers. Here is an example:

rose, 1, 500
lilac, 1, 300
lilly, 1, 400
rose, 0, 100

The idea is that the 0/1 values are there to let the program know
wether to add or subtract the second integer value for a specific
descriptor (flower in this case). So the program comes upon rose, adds
the 500 to an empty list, waits for the next appearance of the rose
descriptor and then (in this case) subtracts 100 from 500 and prints
the value. If the next rose was a 1 then it would have added 100.

I'm uncertain on how to approach doing this though. My idea was to
somehow be able to create lists dynamically upon each new occurence of
a descriptor that currently has no list and then perform the
calculations from there. Unfortunately, the list of descriptors is
potentially infinte, so I'm unable to previously create lists with the
descriptor names. Could anyonw give any suggestions on how to best
approach this problem, hopefully I've been clear enough? Any help would
be very gratly appreciated.

Best regards,
Lorn

[tiissa]

Lorn wrote:

I'm trying to figure out a way to create dynamic lists or possibly
antother solution for the following problem. I have multiple lines in a
text file (every line is the same format) that are iterated over and
which need to be compared to previous lines in the file in order to
perform some simple math. Each line contains 3 fileds: a descriptor and
two integers. Here is an example:

rose, 1, 500
lilac, 1, 300
lilly, 1, 400
rose, 0, 100

Do you have to maintain a list or just the current value?
Also I did not find it clear why you have to compare with previous line?

Anyway, I think dictionnary may be useful there:

input = """rose, 1, 500 ... lilac, 1, 300
... lilly, 1, 400
... rose, 0, 100"""
entries = [l.split(',') for l in input.split('\n ')]

d = {}
for k, op, n in entries: ... if int(op) == 1:
... d[k] = d.get(k, 0) + int(n)
... else:
... d[k] = d.get(k, 0) - int(n)
... print d {'rose': 400, 'lilly': 400, 'lilac': 300}

That may not be what you want but there may be some things for you to use.
And you can of course keep the full list of operation in d.

[Raymond Hettinger]

# Professional driver on a closed course.
# Don't try this at home.

data = """
rose, 1, 500
lilac, 1, 300
lilly, 1, 400
rose, 0, 100
"""

data = data.replace(', 1, ', ' += ')
data = data.replace(', 0, ', ' -= ')

class DefaultDict(dic t):
def __getitem__(sel f, key):
return self.get(key, 0)

d = DefaultDict()
exec data in {}, d
print d.items()

[Steven D'Aprano]

On Sat, 11 Jun 2005 10:30:53 -0700, Lorn wrote:

I'm trying to figure out a way to create dynamic lists or possibly
antother solution for the following problem. I have multiple lines in a
text file (every line is the same format) that are iterated over and
which need to be compared to previous lines in the file in order to
perform some simple math. Each line contains 3 fileds: a descriptor and
two integers. Here is an example:

rose, 1, 500
lilac, 1, 300
lilly, 1, 400
rose, 0, 100

The idea is that the 0/1 values are there to let the program know
wether to add or subtract the second integer value for a specific
descriptor (flower in this case). So the program comes upon rose, adds
the 500 to an empty list, waits for the next appearance of the rose
descriptor and then (in this case) subtracts 100 from 500 and prints
the value. If the next rose was a 1 then it would have added 100.
Why not just use a leading minus sign for the number if it is to be
subtracted?
I'm uncertain on how to approach doing this though. My idea was to
somehow be able to create lists dynamically upon each new occurence of a
descriptor that currently has no list and then perform the calculations
from there. Unfortunately, the list of descriptors is potentially
infinte, so I'm unable to previously create lists with the descriptor
names. Could anyonw give any suggestions on how to best approach this
problem, hopefully I've been clear enough? Any help would be very gratly
appreciated.

Use a dictionary:

def update_dict(D, key, sign, value):
"""Update dictionary D item with key by adding
or subtracting value."""
if not sign:
value = -value
x = D.get(key, 0)
D[key] = x + value

def split_line(s):
"""Split a string s into a tuple (key, sign, value),
ignoring whitespace."""
L = s.split(",")
return L[0].strip(), L[1].strip(), L[2].strip()
# WARNING: insufficient error checking for real world use.

Then call them like this:

D = {}
for line in sourcetext:
key, sign, value = split_line(line )
update_dict(D, key, sign, value)
Oh, by the way... just in case this is homework, which I'm sure it isn't
<wink>, I've deliberately left a teeny tiny bug in the code. You'll find
the bug pretty much the first time you run it, and the fix is very simple.
--
Steven.

[Christos TZOTZIOY Georgiou]

On 13 Jun 2005 01:04:53 -0700, rumours say that "Raymond Hettinger"
<py****@rcn.com > might have written:

# Professional driver on a closed course.
# Don't try this at home.

data = """
rose, 1, 500
lilac, 1, 300
lilly, 1, 400
rose, 0, 100
"""

data = data.replace(', 1, ', ' += ')
data = data.replace(', 0, ', ' -= ')

class DefaultDict(dic t):
def __getitem__(sel f, key):
return self.get(key, 0)

d = DefaultDict()
exec data in {}, d
print d.items()

For kids trying this at home, note that it only works if the descriptor
is a valid Python identifier.
--
TZOTZIOY, I speak England very best.
"Be strict when sending and tolerant when receiving." (from RFC1958)
I really should keep that in mind when talking with people, actually...