Select a string in regex with ruby

Question

I have to clean a string passed in parameter, and remove all lowercase letters, and all special character except :

+
|
^
space
=>
<=>

so i have this string passed in parameter:

aA azee + B => C=

and i need to clean this string to have this result:

A + B => C

I do

string.gsub(/[^[:upper:][+|^ ]]/, "")

output: "A + B C"

I don't know how to select the => (and for <=>) string's with regex in ruby)

I know that if i add string.gsub(/[^[:upper:][+|^ =>]]/, "") into my regex, the last = in my string passed in parameter will be selected too

`(=>)|[^[:upper:]+|^ ]` replace with `$1`? – ctwheels Apr 09 '18 at 15:46 — ctwheels, Apr 09 '18 at 15:46
Why does your string contain those extra characters? – Stefan Apr 09 '18 at 15:56 — Stefan, Apr 09 '18 at 15:56

Sweeper · Accepted Answer · 2018-04-09T20:24:10.040

5

You can try an alternative approach: matching everything you want to keep then joining the result.

You can use this regex to match everything you want to keep:

[A-Z\d+| ^]|<?=>

As you can see this is just a using | and [] to create a list of strings that you want to keep: uppercase, numbers, +, |, space, ^, => and <=>.

Example:

"aA azee + B => C=".scan(/[A-Z\d+| ^]|<?=>/).join()

Output:

"A  + B => C"

Note that there are 2 consecutive spaces between "A" and "+". If you don't want that you can call String#squeeze.

edited Apr 09 '18 at 20:24

answered Apr 09 '18 at 15:52

Sweeper

213,210
22
193
313

`join` does not require a default argument to be explicitly passed and _squeezing_ would probably make sense afterwards. – Aleksei Matiushkin Apr 09 '18 at 15:53
1

this is nevertheless the best approach AFAICT. – Aleksei Matiushkin Apr 09 '18 at 15:57
Thank you, i think that it's the best approach to solve my problem ! – Vincent Cheloudiakoff Apr 09 '18 at 17:12
1

I would think the regex should also include `\d`. – Cary Swoveland Apr 09 '18 at 20:18
`/[A-Z\d+| ^]|=>/` is faster and also includes the `^` character that you forgot :) – ctwheels Apr 09 '18 at 20:23

score 1 · Answer 2 · answered Apr 09 '18 at 15:58

1

See regex in use here

(<?=>)|[^[:upper:]+|^ ]

(<?=>) Captures <=> or => into capture group 1
[^[:upper:]+|^ ] Matches any character that is not an uppercase letter (same as [A-Z]) or +, |, ^ or a space

See code in use here

p "aA azee + B => C=".gsub(/(<?=>)|[^[:upper:]+|^ ]/, '\1')

Result: A + B => C

answered Apr 09 '18 at 15:58

ctwheels

21,901
9
42
77

I prefer this solution because it explicitly excludes characters, as opposed to including what is inferred to be the strings to be kept. Also, the POSIX expression for uppercase letters has wider applicability than `A-Z`. – Cary Swoveland Apr 09 '18 at 20:31

score 0 · Answer 3 · answered Nov 02 '19 at 04:55

r = /[a-z\s[:punct:]&&[^+ |^]]/

"The cat, 'Boots', had 9+5=4 ^lIVEs^ leF|t.".gsub(r,'')
  #=> "T  B  9+54 ^IVE^ F|"

The regular expression reads, "Match lowercase letters, whitespace and punctuation that are not the characters '+', ' ', '|' and '^'. && within a character class is the set intersection operator. Here it intersects the set of characters that match a-z\s[:punct:] with those that match [^+ |^]. (Note that this includes whitespaces other than spaces.) For more information search for "character classes also support the && operator" in Regexp.

I have not included '=>' and '<=>' as those, unlike '+', ' ', '|' and '^', are multi-character strings and therefore require a different approach than simply removing certain characters.

Select a string in regex with ruby

3 Answers3