Rewrote post-filter.asciidoc

Closes #5166

docs/reference/search/request/post-filter.asciidoc

@@ -1,90 +1,114 @@
 [[search-request-post-filter]]
 === Post filter
 
-The `post_filter` allows any filter that it holds to be executed as last filter, because
+The `post_filter` is applied to the search `hits` at the very end of a search
-of this the `post_filter` only has affect on the search hits and not facets.
+request,  after aggregations have already been calculated. It's purpose is
+best explained by example:
 
-There are several reasons why to specify filters as `post_filter`. One reason is to force
+Imagine that you are selling shirts, and the user has specified two filters:
-expensive filters to be executed as last filter, so that these filters only operate on the
+`color:red` and `brand:gucci`.  You only want to show them red shirts made by
-docs that match with the rest of the query. An example of for what filter a post_filter
+Gucci in the search results.  Normally you would do this with a 
-should be used for this reason is the `geo_distance` filter. The `geo_distance` filter is
+<<query-dsl-filtered-query,`filtered` query>>:
-in general an expensive filter to execute and to reduce the execution time for this filter,
-one can choose to specify it as `post_filter`, so it runs on documents that are very likely
-to end up as matching documents.
 
-Another important reason is when doing things like facet navigation,
+[source,json]
-sometimes only the hits are needed to be filtered by the chosen facet,
-and all the facets should continue to be calculated based on the original query.
-The `post_filter` element within the search request can be used to accomplish it.
-
-Note, this is different compared to creating a `filtered` query with the
-filter, since this will cause the facets to only process the filtered
-results.
-
-For example, let's create two tweets, with two different tags:
-
-[source,js]
 --------------------------------------------------
-curl -XPUT 'localhost:9200/twitter/tweet/1' -d '
+curl -XGET localhost:9200/shirts/_search -d '
 {
-    "message" : "something blue",
+  "query": {
-    "tag" : "blue"
+    "filtered": {
-}
+      "filter": {
-'
+        "bool": {
-
+          "must": [
-curl -XPUT 'localhost:9200/twitter/tweet/2' -d '
+            { "term": { "color": "red"   }},
-{
+            { "term": { "brand": "gucci" }}
-    "message" : "something green",
+          ]
-    "tag" : "green"
-}
-'
-
-curl -XPOST 'localhost:9200/_refresh'
---------------------------------------------------
-
-We can now search for something, and have a terms facet.
-
-[source,js]
---------------------------------------------------
-curl -XPOST 'localhost:9200/twitter/_search?pretty=true' -d '
-{
-    "query" : {
-        "term" : { "message" : "something" }
-    },
-    "facets" : {
-        "tag" : {
-            "terms" : { "field" : "tag" }
         }
+      }
     }
+  }
 }
 '
 --------------------------------------------------
 
-We get two hits, and the relevant facets with a count of 1 for both
+However, you would also like to use _faceted navigation_ to display a list of
-`green` and `blue`. Now, let's say the `green` facet is chosen, we can
+other options that the user could click on.  Perhaps you have a `model` field
-simply add a filter for it:
+that would allow the user to limit their search results to red Gucci
+`t-shirts` or `dress-shirts`.
 
-[source,js]
+This can be done with a 
+<<search-aggregations-bucket-terms-aggregation,`terms` aggregation>>:
+
+[source,json]
 --------------------------------------------------
-curl -XPOST 'localhost:9200/twitter/_search?pretty=true' -d '
+curl -XGET localhost:9200/shirts/_search -d '
 {
-    "query" : {
+  "query": {
-        "term" : { "message" : "something" }
+    "filtered": {
-    },
+      "filter": {
-    "post_filter" : {
+        "bool": {
-        "term" : { "tag" : "green" }
+          "must": [
-    },
+            { "term": { "color": "red"   }},
-    "facets" : {
+            { "term": { "brand": "gucci" }}
-        "tag" : {
+          ]
-            "terms" : { "field" : "tag" }
         }
+      }
     }
+  },
+  "aggs": {
+    "models": {
+      "terms": { "field": "model" } <1>
+    }
+  }
 }
 '
 --------------------------------------------------
+<1> Returns the most popular models of red shirts by Gucci.
 
-And now, we get only 1 hit back, but the facets remain the same.
+But perhaps you would also like to tell the user how many Gucci shirts are
+available in *other colors*. If you just add a `terms` aggregation on the
+`color` field, you will only get back the color `red`, because your query
+returns only red shirts by Gucci.
 
-Note, if additional filters are required on specific facets, they can be
+Instead, you want to include shirts of all colors during aggregation, then
-added as a `facet_filter` to the relevant facets.
+apply the `colors` filter only to the search results.  This is the purpose of
+the `post_filter`:
+
+[source,json]
+--------------------------------------------------
+curl -XGET localhost:9200/shirts/_search -d '
+{
+  "query": {
+    "filtered": {
+      "filter": {
+        { "term": { "brand": "gucci" }} <1>
+      }
+    }
+  },
+  "aggs": {
+    "colors": {
+      "terms": { "field": "color" }, <2>
+    },
+    "color_red": {
+      "filter": {
+        "term": { "color": "red" } <3>
+      },
+      "aggs": {
+        "models": {
+          "terms": { "field": "model" } <3>
+        }
+      }
+    }
+  },
+  "post_filter": { <4>
+    "term": { "color": "red" },
+  }
+}
+'
+--------------------------------------------------
+<1> The main query now finds all shirts by Gucci, regardless of color.
+<2> The `colors` agg returns popular colors for shirts by Gucci.
+<3> The `color_red` agg limits the `models` sub-aggregation 
+    to *red* Gucci shirts.
+<4> Finally, the `post_filter` removes colors other than red
+    from the search `hits`.