path_hierarchy tokenizer examples

gistfile1.txt

### Path Hierarchy Examples

DELETE /file-path-test/


# create mapping with file_path text field and two subfields for file path tree to be analyzed in two different ways using custom analyzers.
# field <- customer analyzer <- customer tokenizer <- path_hierarchy tokenizer
# 
PUT /file-path-test/
{
  "settings": {
    "analysis": {
      "analyzer": {
        "custom_path_tree": {
          "tokenizer": "custom_hierarchy"
        },
        "custom_path_tree_reversed": {
          "tokenizer": "custom_hierarchy_reversed"
        }
      },
      "tokenizer": {
        "custom_hierarchy": {
          "type": "path_hierarchy",
          "delimiter": "/",
          "search_analyzer": "keyword"
        },
        "custom_hierarchy_reversed": {
          "type": "path_hierarchy",
          "delimiter": "/",
          "search_analyzer": "keyword",
          "reverse": "true"
        }
      }
    },
    "number_of_replicas": 0,
    "number_of_shards": 1
  },
  "mappings": {
    "doc": {
      "properties": {
        "file_path": {
          "type": "text",
          "fields": {
            "tree": {
              "type": "text",
              "analyzer": "custom_path_tree"
            },
            "tree_reversed": {
              "type": "text",
              "analyzer": "custom_path_tree_reversed"
            }
          }
        }
      }
    }
  }
}

GET /file-path-test/_mapping

POST /file-path-test/doc/
{
  "file_path": "/User/harry/photos/2017/05/16/my_photo1.jpg"
}

POST /file-path-test/doc/
{
  "file_path": "/User/harry/photos/2017/05/16/my_photo2.jpg"
}

POST /file-path-test/doc/
{
  "file_path": "/User/harry/photos/2017/05/16/my_photo3.jpg"
}

POST /file-path-test/doc/
{
  "file_path": "/User/harry/photos/2017/05/15/my_photo1.jpg"
}

POST /file-path-test/doc/
{
  "file_path": "/User/wally/photos/2017/05/16/my_photo1.jpg"
}

GET /file-path-test/_search

# this will match all of our example docs, but wally's will rank highest, since more matching keywords add to relevance.
GET file-path-test/_search
{
  "query": {
    "match": {
      "file_path": "/User/wally/photos/2017/05"
    }
  }
}

# this will essentially only match docs with this exact keyword in the file_path terms list, it's also (still doing scoring)
GET file-path-test/_search
{
  "query": {
    "term": {
      "file_path.tree": "/User/wally/photos/2017/05"
    }
  }
}

# another example of a keyword term match, 3 docs match
GET file-path-test/_search
{
  "query": {
    "term": {
      "file_path.tree": "/User/harry/photos/2017/05/16"
    }
  }
}

# What about reversed tokenizer?
# This will do the same type of matching/terms creation but starting from the other end of the string
GET file-path-test/_search
{
  "query": {
    "term": {
      "file_path.tree_reversed": {
        "value": "my_photo1.jpg"
      }
    }
  }
}

# so you can do all the usual things to the top level text field, then use the tree/tree_reversed subfields to do filtering
GET file-path-test/_search
{
  "query": {
    "bool" : {
      "must" : {
        "match" : { "file_path" : "16" }
      },
      "filter": {
        "term" : { "file_path.tree" : "/User/harry" }
      }
    }
  }
}