Hello,
The default behaviour of ES to tokenize on slash does not suit me, because my data is really full of slashes.
Tokenizing on space is more relevent in my case, but how to copy the index with the good tokinizing settings?
You can use whitespace tokenizer in custom analyzer.