support many word analyzer

[boliza]

in some situation,we should use different analyzer for different string field.

[mbroecheler]

Can you elaborate on what you mean? Could this be achieved by modifying the analyzer directly in ES?

[boliza]

my es schema like this.
yrp: {
index: not_analyzed
type: string
}
vlx: {
analyzer: whitespace
type: string
}
kjp: {
analyzer: jieba
type: string
}
It can update the schema using es api after titan create the schema.

[boliza]

@mbroecheler this issue should close,it seems the same with #933