Python Regex Find Word with Random White Space Mixed in

Question

How do you write a regular expression to match a specific word in a string, when the string has white space added in random places?

I've got a string that has been extracted from a pdf document that has a table structure. As a consequence of that structure the extracted string contains randomly inserted new lines and white spaces. The specific words and phrases that I'm looking for are there with characters all in the correct order, but chopped randomly with white spaces. For example: "sta ck over flow".

The content of the pdf document was extracted with PyPDF2 as this is the only option available on my company's python library.

I know that I can write a specific string match for this with a possible white space after every character, but there must be a better way of searching for it.

Here's an example of what I've been trying to do.

my_string = "find the ans weron sta ck over flow" 
# r's\s*t\s*a\s*c\s*k\s*'  # etc
my_cleaned_string = re.sub(r's\s*t\s*a\s*c\s*k\s*', '', my_string)

Any suggestions?

What do you mean by "better"? If you need to match any amount of whitespaces between each letter, it is the only way. — Wiktor Stribiżew, May 28 '20 at 08:12
Thanks @Wiktor Stribiżew. I was hoping there was a wildcard approach that would accept any number of \s occurrences... — PhilTregidgo, May 28 '20 at 08:24
Also, see https://stackoverflow.com/questions/4590298/how-to-ignore-whitespace-in-a-regular-expression-subject-string, exactly Tim's answer [here](https://stackoverflow.com/a/53777903/3832970). — Wiktor Stribiżew, May 28 '20 at 08:31

score 2 · Answer 1 · answered May 28 '20 at 08:14

The best you can probably do here is to just strip all whitespace and then search for the target string inside the stripped text:

my_string = "find the ans weron sta ck over flow"
my_string = re.sub(r'\s+', '', my_string)
if 'stack' in my_string:
    print("MATCH")

The reason I use "best" above is that in general you won't know if a space is an actual word boundary, or just random whitespace which has been inserted. So, you can really only do as good as finding your target as a substring in the stripped text. Note that the input text 'rust acknowledge' would now match positive for stack.

score 2 · Accepted Answer · answered May 28 '20 at 08:15

Actually what you're doing is the best way. The only addition I can suggest is to dynamically construct such regexp from a word:

word = "stack"
regexp = r'\s*'.join(word)
my_string = "find the ans weron sta ck over flow" 
my_cleaned_string = re.sub(regexp, '', my_string)

Python Regex Find Word with Random White Space Mixed in

2 Answers2