We use the ElasticSearch Service on AWS and we want to use the hyphenation decompounder token filter (https://www.elastic.co/guide/en/elasticsearch/reference/6.0/analysis-compound-word-tokenfilter.html#_hyphenation_decompounder).
This filter requires some files (hyphenation_patterns_path and word_list_path, of which the latter can be inline, but with thousands of words that doesn't seem like a smart solution) that need to be present on the ElasticSearch server. Because in the case of ElasticSearch Service we do not control the server, there seems to be no documented way to upload files to the server. How can we still use this token filter?