The RegExp may have bug to separate the words

[99MengXin]

Describe the bug The RegExp may have bug to separate the words.

To Reproduce Steps to reproduce the behavior:

1. Go to Languages
2. Click on Edit (In this case, English)
3. Scroll down to RegExp Split Sentences, and set .!?:;
4. Scroll down to RegExp Word Characters, and set \-\'a-zA-ZÀ-ÖØ-öø-ȳЀ-ӹ
5. Click on Save
6. Choose any English text and read
7. See error

Expected behavior e.g. 'I should like it.' The word it in this sentence should be recognized to it but it. with a period. In other cases, symbols can be recognized perfectly. e.g. ?!;:

Screenshots

Desktop (please complete the following information):

• OS: MacOS
• Safari 16.2
• Firefox 109.0
• Brave 1.48.164

Smartphone (please complete the following information):

• iPhone 8
• iOS8.1 14.8
• Safari
• 14

Additional context This function works well in original LWT.

[HugoFara]

Hi! Here, I can give you a few suggestions: In step 4 you detail

Scroll down to RegExp Word Characters, and set \-\'a-zA-ZÀ-ÖØ-öø-ȳЀ-ӹ

Base on the provided RegExp, ' should be considered a word, which is useful in some contexts (english genitive such as "Tom's house" → "Tom's + house"). If it's a trouble to you, you can simply remove the apostrophe for words characters (use \-a-zA-ZÀ-ÖØ-öø-ȳЀ-ӹ).

In your context, I would suggest to use " instead of ' whenever possible.

Finally, I can look into the word parser if I have some time, but it a core function so it's really hard to change without breaking something for users.

I hope that helps you :smile: