elastic (version 0.7.8)

scroll: Scroll search function

Description

Scroll search function

Usage

scroll(scroll_id, scroll = "1m", raw = FALSE, ...)
scroll_clear(scroll_id = NULL, all = FALSE, ...)

Arguments

scroll_id

(character) For scroll, a single scroll id; for scroll_clear, one or more scroll id's

scroll

(character) Specify how long a consistent view of the index should be maintained for scrolled search, e.g., "30s", "1m". See units-time.

raw

(logical) If TRUE (default), data is parsed to list. If FALSE, then raw JSON.

...

Curl args passed on to POST

all

(logical) If TRUE (default) then all search contexts cleared. If FALSE, scroll id's must be passed to scroll_id

Clear scroll

Search context are automatically removed when the scroll timeout has been exceeded. Keeping scrolls open has a cost, so scrolls should be explicitly cleared as soon as the scroll is not being used anymore using scroll_clear

Sliced scrolling

See the example in this man file.

Details

Scores will be zero for all documents that are returned from a scroll request. Dems da rules.

References

https://www.elastic.co/guide/en/elasticsearch/reference/current/search-request-scroll.html

See Also

Search

Examples

# NOT RUN {
# Get a scroll_id
res <- Search(index = 'shakespeare', q="a*", scroll="1m")
res$`_scroll_id`

# Setting search_type=scan turns off sorting of results, is faster
res <- Search(index = 'shakespeare', q="a*", scroll="1m", search_type = "scan")
res$`_scroll_id`

# Pass scroll_id to scroll function
scroll(scroll_id = res$`_scroll_id`)

# Get all results - one approach is to use a while loop
res <- Search(index = 'shakespeare', q="a*", scroll="5m", search_type = "scan")
out <- list()
hits <- 1
while(hits != 0){
  res <- scroll(scroll_id = res$`_scroll_id`)
  hits <- length(res$hits$hits)
  if(hits > 0)
    out <- c(out, res$hits$hits)
}
length(out)
out[[1]]

# clear scroll
## individual scroll id
res <- Search(index = 'shakespeare', q="a*", scroll="5m", search_type = "scan")
scroll_clear(scroll_id = res$`_scroll_id`)

## many scroll ids
res1 <- Search(index = 'shakespeare', q="c*", scroll="5m", search_type = "scan")
res2 <- Search(index = 'shakespeare', q="d*", scroll="5m", search_type = "scan")
nodes_stats(metric = "indices")$nodes[[1]]$indices$search$open_contexts
scroll_clear(scroll_id = c(res1$`_scroll_id`, res2$`_scroll_id`))
nodes_stats(metric = "indices")$nodes[[1]]$indices$search$open_contexts

## all scroll ids
res1 <- Search(index = 'shakespeare', q="f*", scroll="1m", search_type = "scan")
res2 <- Search(index = 'shakespeare', q="g*", scroll="1m", search_type = "scan")
res3 <- Search(index = 'shakespeare', q="k*", scroll="1m", search_type = "scan")
scroll_clear(all = TRUE)

## sliced scrolling
body1 <- '{
  "slice": {
    "id": 0, 
    "max": 2 
  },
  "query": {
    "match" : {
      "text_entry" : "a*"
    }
  }
}'

body2 <- '{
  "slice": {
    "id": 1, 
    "max": 2 
  },
  "query": {
    "match" : {
      "text_entry" : "a*"
    }
  }
}'

res1 <- Search(index = 'shakespeare', scroll="1m", body = body1)
res2 <- Search(index = 'shakespeare', scroll="1m", body = body2)
scroll(scroll_id = res1$`_scroll_id`)
scroll(scroll_id = res2$`_scroll_id`)

out1 <- list()
hits <- 1
while(hits != 0){
  tmp1 <- scroll(scroll_id = res1$`_scroll_id`)
  hits <- length(tmp1$hits$hits)
  if(hits > 0)
    out1 <- c(out1, tmp1$hits$hits)
}

out2 <- list()
hits <- 1
while(hits != 0){
  tmp2 <- scroll(scroll_id = res2$`_scroll_id`)
  hits <- length(tmp2$hits$hits)
  if(hits > 0)
    out2 <- c(out2, tmp2$hits$hits)
}

c(
 lapply(out1, "[[", "_source"),
 lapply(out2, "[[", "_source")
) 
# }