Clean And Extract Text
These transformation types are useful when the value needs text cleanup rather than numeric or date parsing.
Text Case
Section titled “Text Case”Use this when the text is correct but the letter casing is not.
Field used:
Text Case
Choices:
UppercaseLowercaseTitle Case
Examples:
| Before | Rule | After |
|---|---|---|
invoice total | Uppercase | INVOICE TOTAL |
CUSTOMER CODE | Lowercase | customer code |
main office west | Title Case | Main Office West |
Text Replace
Section titled “Text Replace”Use this when you want a simple find-and-replace.
Fields used:
Find TextReplace With
What it does:
- Replaces every matching occurrence.
- Matches text exactly.
- Is case-sensitive.
- If
Replace Withis blank, the found text is removed.
Examples:
| Before | Rule | After |
|---|---|---|
ORD-10452 | Find ORD-, Replace with blank | 10452 |
Total: $1,500.00 | Find $, Replace with blank | Total: 1,500.00 |
Ref_2024_01 | Find _, Replace with - | Ref-2024-01 |
abc ABC abc | Find abc, Replace with X | X ABC X |
Regex Extract
Section titled “Regex Extract”Use this when the value contains extra text and you only want the part that matches a pattern.
Fields used:
Regex PatternRegex Group
What it does:
- Searches the value using the pattern you enter.
- Returns the first text match it finds.
- If nothing matches, the original value stays unchanged.
Think of this as: “pull out the first piece of text that matches my rule.”
Examples:
| Before | Rule | After |
|---|---|---|
Reference: ORD-10452 | Pattern ORD-\\d+ | ORD-10452 |
Invoice 8841 dated 2024-12-25 | Pattern \\d{4}-\\d{2}-\\d{2} | 2024-12-25 |
Amount due: 1,234.56 USD | Pattern [\\d,.]+ | 1,234.56 |
Customer code C-10025 / batch 77 | Pattern C-\\d+ | C-10025 |
Pattern ideas:
| Goal | Pattern | What it matches |
|---|---|---|
| Pull an order code | ORD-\\d+ | ORD-10452 |
| Pull an invoice date | \\d{4}-\\d{2}-\\d{2} | 2024-12-25 |
| Pull the first number with separators | [\\d,.]+ | 1,234.56 |
| Pull a code with a letter prefix | [A-Z]-\\d+ | C-10025 |
Good fit:
- The value contains the part you want plus extra words.
- You need the first matching code, date, or amount.
Poor fit:
- You need several different clean-up steps in a row.
- You need capture-group style extraction behavior.
Current regex limit
Section titled “Current regex limit”The transformation card shows a Regex Group field, but the regex transformation returns the first overall match. The group number is not currently used when the value is transformed.