如何在Python中组织在负值和正值之间交替的数据

Question

I have a list of data points in a file that I am inserting into a linked list class that I built.
The file is organized such that there are a series of negative values followed by a series of positive values alternating back and forth. An example:

-2323 

-2324

-53434

-1027

-34232

 343434

 5657

 6565

 6500

-343434

-3434

-565

5845

4667

5453

98356

This pattern continues for many lines. The number of negative or positive values is never the same for each section.

I would like to separate these values somehow so that the first list object contains the first set of positive to negative numbers, in this case from -2323 to 6500 . The next list object would contain the values from -343434 to 98356 , and so on.

I cannot figure out how to get python to know how to separate these sets of data when reading the file. Any help would be appreciated!

Answer 1

import itertools
groups = itertools.groupby(l, lambda x : x > 0)
result = [list(groups[i][1]) + list(groups[i + 1][1]) for i in range(0, len(groups), 2)]

This will first group the elements by whether they are positive or not, and will then combine adjacent pairs from the groups into a single list, which will then be an element of the result list.

Edit: I keep forgetting that itertools makes objects that don't work like normal iterables.

The following should actually work, if a bit more messily.

import itertools
groups = itertools.groupby(l, lambda x : x > 0)
grouplist = [[i for i in y] for (x, y) in groups]
result = [grouplist[i] + grouplist[i + 1] for i in range(0, len(grouplist), 2)]

Answer 2

def takeSection(sequence):
    it = iter(sequence)
    a = -1
    group = []
    while True:
        try:
            a, last = next(it), a
        except StopIteration:
            if group:
                yield group
            return
        if a < 0 and last >= 0:
            if group:
                yield group
            group = [a]
        else:
            group.append(a)

---

>>> sequence = [-2323, -2324, -53434, -1027, -34232, 343434, 5657, 6565, 6500, -343434, -3434, -565, 5845, 4667, 5453, 98356]
>>> list(takeSection(sequence))
Out[2]: 
[[-2323, -2324, -53434, -1027, -34232, 343434, 5657, 6565, 6500],
 [-343434, -3434, -565, 5845, 4667, 5453, 98356]]

Edit

If you want to filter this on the first value in a pair of values, you can change the if condition to test for this instead. You could for instance change the condition line to if a[0] < 0 and last[0] >=0 , and you'd also need to initialise a as a = (-1, -1)

However I'd be tempted to make a more generalised and useful function instead.

def sections(sequence, key):
    it = iter(sequence)
    a = placeholder = object()
    group = []
    while True:
        try:
            a, last = next(it), a
        except StopIteration:
            if group:
                yield group
            return
        if last is not placeholder and key(a, last):
            if group:
                yield group
            group = [a]
        else:
            group.append(a)

---

>>> sequence = [(-2323, -7465), (-2324, -7687), (-53434, -1027), (-34232, 343434), (5657, 6565), (6500, 978987), (-343434, -987), (-3434, -565), (-98, -8798), (-89898, -898), (5845, 4667), (5453, 98356)]
>>> list(sections(sequence, key=lambda current, last: current[0] < 0 and last[0] >= 0))
Out[1]:
[[(-2323, -7465), (-2324, -7687), (-53434, -1027), (-34232, 343434), (5657, 6565), (6500, 978987)],
 [(-343434, -987), (-3434, -565), (-98, -8798), (-89898, -898), (5845, 4667), (5453, 98356)]]