collapsing whitespace in a string

Question

I have a string that kind of looks like this:

\"stuff   .  // : /// more-stuff .. .. ...$%$% stuff -> DD\"

and I want to strip off all p

Answer 1

works in python3 this will retain the same whitespace character you collapsed. So if you have a tab and a space next to each other they wont collapse into a single character.

def collapse_whitespace_characters(raw_text):
    ret = ''
    if len(raw_text) > 1:
        prev_char = raw_text[0]
        ret += prev_char
        for cur_char in raw_text[1:]:
            if not cur_char.isspace() or cur_char != prev_char:
                ret += cur_char
            prev_char = cur_char
    else:
        ret = raw_text
    return ret

this one will collapse whitespace sets into the first whitespace character it sees

def collapse_whitespace(raw_text):
    ret = ''
    if len(raw_text) > 1:
        prev_char = raw_text[0]
        ret += prev_char
        for cur_char in raw_text[1:]:
            if not cur_char.isspace() or \
                    (cur_char.isspace() and not prev_char.isspace()):
                ret += cur_char
            prev_char = cur_char
    else:
        ret = raw_text
    return ret

>>> collapse_whitespace_characters('we like    spaces  and\t\t TABS   AND WHATEVER\xa0\xa0IS')
'we like spaces and\t TABS\tAND WHATEVER\xa0IS'

>>> collapse_whitespace('we like    spaces  and\t\t TABS   AND WHATEVER\xa0\xa0IS')
'we like spaces and\tTABS\tAND WHATEVER\xa0IS'

for punctuation

def collapse_punctuation(raw_text):
    ret = ''
    if len(raw_text) > 1:
        prev_char = raw_text[0]
        ret += prev_char
        for cur_char in raw_text[1:]:
            if cur_char.isalnum() or cur_char != prev_char:
                ret += cur_char
            prev_char = cur_char
    else:
        ret = raw_text
    return ret

to actually answer the question

orig = 'stuff   .  // : /// more-stuff .. .. ...$%$% stuff -> DD'
collapse_whitespace(''.join([(c.upper() if c.isalnum() else ' ') for c in orig]))

as said, the regexp would be something like

re.sub('\W+', ' ', orig).upper()