DataSuite 2 Documentation
English Русский
On this page

Regex reference

The regex replace rule in data transformation lets you search for patterns in text data and replace them. If you’ve never used regular expressions before, don’t worry – this page covers the essentials and the most common use cases. For a deeper dive, regexone.com is a good interactive tutorial.

The regex editor

When you create a regex replace rule, the editor shows:

  • Search pattern (RegEx) – the regular expression to match
  • Replacement text – what to substitute in place of each match
  • Options – global (replace all matches), case sensitive, multiline mode
  • Preview – shows the first non-missing value of the first selected variable, with matches highlighted and capture groups in distinct colors

The preview updates as you type, so you can experiment safely before saving the rule.

Pattern basics

Pattern Matches Example match
abc literal text “abc” “abcdef”
. any single character “abc”
\d any digit (0–9) “age 42”
\D any non-digit “42!”
\w any word character (Latin letter, digit, underscore) “hello”
\W any non-word character (a non-Latin letter included) “hello !”
\s any whitespace (space, tab, newline) “hello world”
\S any non-whitespace “hello”

Quantifiers

Quantifier Meaning Example
* zero or more \d* matches “”, “5”, “42”, “123”
+ one or more \d+ matches “5”, “42”, “123” but not “”
? zero or one (optional) colou?r matches “color” and “colour”
{3} exactly 3 \d{3} matches “123” but not “12”
{2,4} between 2 and 4 \d{2,4} matches “12”, “123”, “1234”

Anchors

Anchor Meaning
^ start of string (or line, with multiline mode)
$ end of string (or line, with multiline mode)

Character classes

Square brackets define a set of characters to match:

Class Matches
[abc] “a”, “b”, or “c”
[a-z] any lowercase Latin letter
[A-Za-z] any Latin letter
[0-9] any digit (same as \d)
[^abc] any character except “a”, “b”, or “c”

Groups and alternatives

Syntax Meaning
(abc) capture group – matches “abc” and remembers it for replacement
(?:abc) non-capturing group – groups without remembering
a|b alternation – matches “a” or “b”

Replacement text

The replacement string can reference captured groups:

Token Inserts
$1 first capture group
$2 second capture group
$& the entire match

Options

  • Global – replace all matches in each cell, not just the first. Usually you want this on.
  • Case sensitive – when off, hello also matches “Hello”, “HELLO”, etc.
  • Multiline mode – makes ^ and $ match the start/end of each line within a cell, not just the start/end of the entire cell value.

Practical examples

Remove extra whitespace

Collapse each run of whitespace into a single space.
Pattern \s+
Replacement (single space)
Options global
Before " John Smith "
After " John Smith "

Tip: to also trim leading and trailing spaces, add a second regex rule with the pattern ^\s+|\s+$ and an empty replacement (global mode). Or use a formula with trim(v1).

Extract numbers from text

Keep only the first number of a mixed-text cell.
Pattern ^\D*(\d+).*$
Replacement $1
Options global
Before "Age: 42 years"
After "42"

The pattern spans the whole cell, so the replacement keeps the captured number alone – a bare (\d+) would replace the number with itself and leave the text around it. For decimal numbers, use ^\D*(\d+\.?\d*).*$; a cell with no digits is left as it was.

Standardize date formats

Convert “DD/MM/YYYY” to “YYYY-MM-DD”.
Pattern (\d{2})/(\d{2})/(\d{4})
Replacement $3-$2-$1
Options global
Before "15/03/2024"
After "2024-03-15"

The three capture groups (\d{2}), (\d{2}), (\d{4}) grab day, month, and year. The replacement $3-$2-$1 rearranges them.

Remove everything after a delimiter

Keep only the part before a dash – put a comma or another separator in the dash’s place for those.
Pattern \s*-.*$
Replacement (empty)
Options global
Before "A1 - some notes"
After "A1"

Clean up Likert scale labels

Strip the numeric prefix from Google Forms–style labels like “1 - Strongly disagree”.
Pattern ^\d+\s*[-–]\s*
Replacement (empty)
Options global
Before "1 - Strongly disagree"
After "Strongly disagree"

Replace multiple spellings at once

Standardize different spellings of the same response.
Pattern ^(yes|yep|yeah|y)$
Replacement Yes
Options global, case insensitive
Before "yep", "YEAH", "Y"
After "Yes", "Yes", "Yes"

The ^ and $ anchors ensure the entire cell matches, not just a substring.

Extract a coded segment

Pull a specific part from a structured code like “DEPT-EMP-001”.
Pattern ^(\w+)-(\w+)-(\d+)$
Replacement $2
Options global
Before "DEPT-EMP-001"
After "EMP"

Change $2 to $1 or $3 to extract a different segment.

When to use regex vs. other rules

  • Exact value mapping (e.g. “Male” → “M”) – use value recode instead. It’s simpler and less error-prone.
  • Numeric transformations – use a formula. Regex operates on text, not numbers.
  • Simple find-and-replace with no pattern logic – regex works, but value recode with a single mapping row is more readable.

Regex shines when the text has structure you need to match flexibly – varying formats, optional parts, or segments to rearrange.