Compare and sort text with accents correctly

Prompt that explains why text with accents does not match or sorts wrongly and sets clear rules. Paste it into an AI tool of your choice, which then sets the text comparison rules.

Topic: Apps & prototypes
Type: Prompt
Level: Intermediate
Time: about 15 min

Results depend on the tool, the model and what you enter.

Result preview

Example: written by our editors, not real AI output.

Example: "Zoë" ≠ "Zoë" Cause: one uses ë as one character, the other e plus accent Rules: - store: normalise to composed form (NFC) - compare: case- and accent-insensitive for search only - sort: German order for German users Tests: ß/ss, é/e, ø, O'Brien

Fill in

What you type stays in this browser window. We store and send nothing.

ProblemWhat goes wrong?

DataWhich data is affected?

StackWhere does the text live?

Your prompt

Help me compare and sort text with special characters correctly.
Problem: [Problem]
Data: [Data]
Stack: [Stack]
Approach: 1. From the examples, explain why two texts that look the same are not equal, for example a combined letter versus letter plus accent. 2. Decide where text should be normalised: on input, on storage or only for comparison. 3. Propose rules for case-insensitive and accent-aware comparison for this use case. 4. Propose a sort order that follows the language of the users instead of character codes. 5. Check database collation, search index and code use the same rules. 6. Write test cases with tricky examples such as ß, é, ø and names with apostrophes.
Output: cause per example, rules for storage, comparison and sorting, and a list of test cases.

Take particular care to agree on one place to normalise and test with real edge cases.

Before you answer, check: Are all instructions met? If key information is missing or a placeholder in square brackets is still in the text, ask instead of guessing. Mark assumptions as assumptions.

Help me compare and sort text with special characters correctly.
Problem: [Problem]
Data: [Data]
Stack: [Stack]
Approach: 1. From the examples, explain why two texts that look the same are not equal, for example a combined letter versus letter plus accent. 2. Decide where text should be normalised: on input, on storage or only for comparison. 3. Propose rules for case-insensitive and accent-aware comparison for this use case. 4. Propose a sort order that follows the language of the users instead of character codes. 5. Check database collation, search index and code use the same rules. 6. Write test cases with tricky examples such as ß, é, ø and names with apostrophes.
Output: cause per example, rules for storage, comparison and sorting, and a list of test cases.

Take particular care to agree on one place to normalise and test with real edge cases.

Before you answer, check: Are all instructions met? If key information is missing or a placeholder in square brackets is still in the text, ask instead of guessing. Mark assumptions as assumptions.

This is an instruction for an AI, not the result itself.

Where this prompt works

These AI tools are a good fit for this prompt:

How to use it

Fill in the fields or tap a suggestion. Leave out names, addresses and other personal data.
Copy the finished prompt into an AI tool of your choice.
Use invented names such as "Zoë Müller-Ødegaard" for the test cases; they cover several special cases at once.

Why does this work?

Text that looks equal can be stored differently. Normalising in one agreed place and testing with real edge cases stops duplicates and failed searches.

Tips

Changing normalisation on existing data can merge or split records; plan it on a copy first.

You may use our prompts freely, including at work. Please do not republish them as your own collection. We give no warranty for AI results; check them yourself. Terms of use

Compare and sort text with accents correctly

Result preview

Fill in

Your prompt

Where this prompt works

How to use it

Why does this work?

Tips

Also a good fit

Choose pagination for an API list New

Write an API reference people can use right away New

Plan a throttled data backfill New