Splitting escapes

escape_splitter.py

# escape_splitter.py
# Callum Howard, 2019

escapes = {r'\a', r'\b', r'\f', r'\n', r'\r', r'\t', r'\v', r'\\', r"\'", r'\"', r'\?'}

def split_on_escape(input, prefix=''):
    digraphs = [''.join(pair) for pair in zip(input, input[1:])]
    for i, digraph in enumerate(digraphs):
        if digraph in escapes:
            return [prefix + input[:i]] + split_on_escape(input[i+2:], prefix=digraph)
    return [prefix + input]

def split_on_escape1(input):
    result = []
    prefix = ''
    skip = 0

    digraphs = [''.join(pair) for pair in zip(input, input[1:])]
    for i, digraph in enumerate(digraphs):
        if skip > 0:
            skip -= 1
            continue

        if digraph not in escapes:
            prefix += input[i]
        else:
            result.append(prefix)
            prefix = digraph
            skip = 1

    if skip == 0:
        prefix += input[-1]
    result.append(prefix)

    return result

# tests
f = split_on_escape1

print(f(r'Hello\nworld'))
assert(f(r'Hello\nworld') == ['Hello', r'\nworld'])

print(f(r'\nHello world'))
assert(f(r'\nHello world') == ['', r'\nHello world'])

print(f(r'Hello world\n'))
assert(f(r'Hello world\n') == [r'Hello world', r'\n'])

print(f(r'Hell\o\nworld'))
assert(f(r'Hell\o\nworld') == [r'Hell\o', r'\nworld'])

print(f(r'Hello\\nworld'))
assert(f(r'Hello\\nworld') == ['Hello', r'\\nworld'])

print(f(r'Hello\nworld\nfoo\nbar\nbaz'))
assert(f(r'Hello\nworld\nfoo\nbar\nbaz') ==
        ['Hello', r'\nworld', r'\nfoo', r'\nbar', r'\nbaz'])

>>> a = 'hello'

>>> list( zip(a, a[1:]) )
[('h', 'e'), ('e', 'l'), ('l', 'l'), ('l', 'o')]

>>> (first + second for (first, second) in zip(a, a[1:]))
<generator object <genexpr> at 0x1030555e8>

>>> list(first + second for (first, second) in zip(a, a[1:]))
['he', 'el', 'll', 'lo']

>>> enumerate(first + second for (first, second) in zip(a, a[1:]))
<enumerate object at 0x103bffea0>

>>> list(enumerate(first + second for (first, second) in zip(a, a[1:])))
[(0, 'he'), (1, 'el'), (2, 'll'), (3, 'lo')]