Allowing hypen based words to be tokenized in elasticsearch - elasticsearch

Question

I have the following mapping for a field name that will hold products name for ecommerce.
'properties': {
'name': {
'type': 'text',
'analyzer': 'standard',
'fields': {
'english': {
'type': 'text',
'analyzer': 'english'
},
}
},
Assuming that I have the following string to be indexed/searched.
A pack of 3 T-shirts
Both of the analyerzs are producing terms [t, shirts], [t, shirt] respectively.
This gives me the problem of not getting any result when a user types "mens tshirts"
How can i get the term in the inverted index like [t, shirts, shirt, tshirt', tshirts]
I tried to look into Stemmers exclusions but I couldn't find any thing to deal with hyphens. Also i will be helpful if a more generic solution is found rather than doing exclusions manually. Because there could be many possiblities which i don't know now e.g emails, e-mails

Answer 1

whitespace tokenizer could do the job
https://www.elastic.co/guide/en/elasticsearch/reference/current/analysis-whitespace-tokenizer.html
POST _analyze
{
"tokenizer": "whitespace",
"text": "The 2 QUICK Brown-Foxes jumped over the lazy dog's bone."
}
will produce
[ The, 2, QUICK, Brown-Foxes, jumped, over, the, lazy, dog's, bone. ]

Answer 2

I found one solution which i guess could help me achieving the desired results. However, I would still like to see if there is some good and recommended approach for this problem.
Basically I will use multi fields for this problem where the first analyzer will be standard and the second will be my custom.
According to Elasticsearch documentation, chars_filters happens before tokenizer. So the idea is to remove - with a empty character which will make t-shirts to tshirt. Hence the tokenizer will token the whole term as tshirts in invertded index.
GET _analyze
{
"tokenizer": "standard",
"filter": [
"lowercase",
{"type": "stop", "stopwords": "_english_"}
],
"char_filter" : [
"html_strip",
{"type": "mapping", "mappings": ["- => "]}
],
"text": "these are t-shirts <table>"
}
will give the following tokens
{
"tokens": [
{
"token": "tshirts",
"start_offset": 10,
"end_offset": 18,
"type": "<ALPHANUM>",
"position": 2
}
]
}