🐛 [Bug] Wrong German umlaut handling for usernames at account creation

[Tirokk]

Authored by sushidave
Apr 25, 2020

---

🐛 Bugreport

At account creation, usernames are not encoded correctly if the profile name contains one or more German umlauts:
Ä/ä -> a
Ö/ö -> o
Ü/ü -> u

Examples:

• Sandra Märtens -> sandra-martens
• Oliver Stöckel -> oliver-stockel
• Günter Hürst -> gunter-hurst

Steps to reproduce the behavior

1. Create a new account using one or more German umlauts in the profile name.
2. Check the username.

Expected behavior

German umlauts should be encoded like this:
Ä/ä -> ae
Ö/ö -> oe
Ü/ü -> ue

Examples:

• Sandra Märtens -> sandra-maertens
• Oliver Stöckel -> oliver-stoeckel
• Günter Hürst -> guenter-huerst

Version & Environment

Type: [any]

• OS: [any]
• Browser: [any]
• Version [any]
• Device: [any]

[Tirokk]

Authored by ATOMktn
Jun 18, 2020

---

Hi! I'm new to open source and would like to take a crack at this!

[Tirokk]

Authored by sushidave
Jun 18, 2020

---

Hello @ATOMktn and welcome on board! Thank you for your interest in this project. @Tirokk Please provide ATOMktn with further info. Thanks!

[Tirokk]

Authored by Tirokk
Jun 22, 2020

---

Hello @ATOMktn ,

I have invited you to our repository as volunteer via an e-mail invitation you can except.
After you excepted, please push your feature branch directly to this repo and create a PR using master as the base branch.

I assigned you to this issue …

Our conversation runs over our discord channels:
http://hc.world/discord
Another possibility to get to there:
https://discord.com/invite/DFSjPaX

On Discord we have kind of every day a Daily Standup at UTC 09:30 am (CEST 11:30 am) in the Conference Room.

[ulfgebhardt]

#3676

[Brandon-G-Tripp]

@Tirokk Hey Wolle. I want to discuss taking on this issue. In discord you said we should chat about how to go about this. I have the app up and running again. Somehow it got corrupted and I was missing some modules so I cloned the app down again.

[Tirokk]

Hey @Brandon-G-Tripp ,

we have already the PR #3676 for this issue done by @ATOMktn which you should overtake.
But for me this solution was too specific for German.

I would like to have a very general regex solution for all languages and all characters. I mean a regex where every non ANSII characters gets converted to an ANSII string.

Have a quick investigation and If you don’t find anything, please let me know. I think I have seen a solution last year somewhere.

[Brandon-G-Tripp]

@Tirokk I found some ways of replacing all the diacritics in the characters with their standard keyboard equivalent but not the changing of ü -> ue and so on. I will take a look at @ATOMktn pull request and see if I can combine them so we have to correct handling of the umlaut and then just remove the diacritics for other languages.

[Tirokk]

I have investigated a bit and couldn’t find what I meant to have seen.
Yeah, just have a table of umlaut etc. (Unicode) characters to replace and delete the rest or replace them by underlines or dashes. @Brandon-G-Tripp 👍🏼

PS: In case this could be of some help:
https://javascript.info/regexp-unicode

[Brandon-G-Tripp]

@Tirokk Hey I see that the previous work on this issue was just doing a charmap for the characters. Would you prefer this in regex or should I just expand the work there?

[Mogge]

I have this post title Tösting ße ßlüg ççç ñññ ÄÖÜ and I get the slug: tosting-sse-sslug-ccc-nnn-aou