Hard-code sample rate based on fractional_hardcoded_sample_rate flag (#4762)

* Hard-code sample rate based on fractional_hardcoded_sample_rate flag

We found cases where using a numeric sample rate would cause issues when
joining two tables due to different _sample_factor. Ref: https://3.basecamp.com/5308029/buckets/26383192/card_tables/cards/7973456592#__recording_7978780711

The proper fix is to use fractional sample rates everywhere, but this is
a whole project due to not wanting to sample small sites. For now, hard-code
sample rate for specific sites having the issue while we work on a
larger fix.

* is_number

extra/lib/plausible/stats/sampling.ex

@@ -17,8 +17,8 @@ defmodule Plausible.Stats.Sampling do
     end
   end
 
-  @spec add_query_hint(Ecto.Query.t(), pos_integer()) :: Ecto.Query.t()
+  @spec add_query_hint(Ecto.Query.t(), pos_integer() | float()) :: Ecto.Query.t()
-  def add_query_hint(%Ecto.Query{} = query, threshold) when is_integer(threshold) do
+  def add_query_hint(%Ecto.Query{} = query, threshold) when is_number(threshold) do
     from(x in query, hints: unsafe_fragment(^"SAMPLE #{threshold}"))
   end
 
@@ -27,15 +27,32 @@ defmodule Plausible.Stats.Sampling do
     add_query_hint(query, @default_sample_threshold)
   end
 
-  @spec put_threshold(Plausible.Stats.Query.t(), map()) :: Plausible.Stats.Query.t()
+  @spec put_threshold(Plausible.Stats.Query.t(), Plausible.Site.t(), map()) ::
-  def put_threshold(query, params) do
+          Plausible.Stats.Query.t()
+  def put_threshold(query, site, params) do
     sample_threshold =
       case params["sample_threshold"] do
-        nil -> @default_sample_threshold
+        nil ->
-        "infinite" -> :infinite
+          site_default_threshold(site)
-        value -> String.to_integer(value)
+
+        "infinite" ->
+          :infinite
+
+        value_string ->
+          {value, _} = Float.parse(value_string)
+          value
       end
 
     Map.put(query, :sample_threshold, sample_threshold)
   end
+
+  defp site_default_threshold(site) do
+    if FunWithFlags.enabled?(:fractional_hardcoded_sample_rate, for: site) do
+      # Hard-coded sample rate to temporarily fix an issue for a client.
+      # To be solved as part of https://3.basecamp.com/5308029/buckets/39750953/messages/7978775089
+      0.1
+    else
+      @default_sample_threshold
+    end
+  end
 end

lib/plausible/stats/legacy/legacy_query_builder.ex

@@ -26,7 +26,7 @@ defmodule Plausible.Stats.Legacy.QueryBuilder do
       |> Query.put_imported_opts(site, params)
 
     on_ee do
-      query = Plausible.Stats.Sampling.put_threshold(query, params)
+      query = Plausible.Stats.Sampling.put_threshold(query, site, params)
     end
 
     query

lib/plausible/stats/query.ex

@@ -37,6 +37,10 @@ defmodule Plausible.Stats.Query do
         |> put_experimental_reduced_joins(site, params)
         |> struct!(v2: true, now: DateTime.utc_now(:second), debug_metadata: debug_metadata)
 
+      on_ee do
+        query = Plausible.Stats.Sampling.put_threshold(query, site, params)
+      end
+
       {:ok, query}
     end
   end