dictionaries not recognizing ligatures and diacritics

[h3ct0r]

I have a problem with some of my texts which contain diacritics (principally apices and diaeresis) and ligatures (principally æ, œ) being encoded as stand alone letters whereas my dictionaries don't include them at all and can't identify the word.

Does anyone know how to format my texts so that:
Æ is always recognized as AE
Πis OE
Ë is E
ÁÉÍÓÚ is AEIOU

I want to do this without replacing the diacritics in the text. Or conversely make the dictionaries themselves recognized those letter as such

Please help me if you have any information
Thank you

[Aleron Ives]

I don't think you can do what you want. You would either have to replace all the letters in the book with the letters that the dictionary recognises, or you would have to add new entries to the dictionary for every word that might be spelled with special characters. Since you don't want to do the former, typing in the words manually to look them up is easier and faster than doing the latter.

If it's possible to use NickelMenu to perform dictionary lookups, you could potentially inject a shell script that would find and replace special characters before passing the word to the dictionary, but I don't know of a way to perform dictionary lookups with a shell script on Kobo.

[JSWolf]

delete post

[h3ct0r]

Originally Posted by Aleron Ives

I don't think you can do what you want. You would either have to replace all the letters in the book with the letters that the dictionary recognises, or you would have to add new entries to the dictionary for every word that might be spelled with special characters. Since you don't want to do the former, typing in the words manually to look them up is easier and faster than doing the latter.

If it's possible to use NickelMenu to perform dictionary lookups, you could potentially inject a shell script that would find and replace special characters before passing the word to the dictionary, but I don't know of a way to perform dictionary lookups with a shell script on Kobo.

In that case do you know if there's a way to automate the process of changing every letter in the book?

[geek1011]

New Kobo dictionaries can contain a `prefix_exceptions` file, which is a marisa-trie of original words to the new one to look up in its place.

Unfortunately, I have not had a chance to implement this in dictutil, and there's no publicly available software for working with the newer dictionary format. It can still be done manually if you care enough.

[h3ct0r]

Originally Posted by geek1011

New Kobo dictionaries can contain a `prefix_exceptions` file, which is a marisa-trie of original words to the new one to look up in its place.

Unfortunately, I have not had a chance to implement this in dictutil, and there's no publicly available software for working with the newer dictionary format. It can still be done manually if you care enough.

They actually aren't official kobo dictionaries, I converted them myself with pyglossary. Maybe someone knows of a way to fix or circumvent this issue. Thanks for your response