tdigest 0.0.2 → 0.0.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: be74af8cae96393fca31017d57cd99a125c06af6
-  data.tar.gz: 7f17839d5f50ee56071d213c55891f189c99de37
+  metadata.gz: fb2b554c557253ae6826b504943a315bf4f660a0
+  data.tar.gz: 4f86fdef222ec8f3ddd7746ccfb7767cfbd1bd26
 SHA512:
-  metadata.gz: 9aa1a1379afab686402272ae6cd04a147c522d443924436cc599745abe3cf3bdf3bac86e04ef0d47428c1f8329c1a12c9789838a2b1cfae844b4b0cc089489c9
-  data.tar.gz: a7ceaf5fe58f37f18cd4ff5c72a026296637d73e6f7071ac9b8ed98c4c944ecbd36f7fc36d6b927e0b53102fe653688eac4c4baa013208c9e887b0e8bbf09242
+  metadata.gz: c5b5567069b6958c551721f8c6c6d707e9a08f6f20c23473b39e732c853a92e4ef1e20406dba46d67ddf2b496ce94224501f6605def6f36be0b65b7f586aaa01
+  data.tar.gz: c00dfc8c3bb483793c5f5b2c8a4ea8ad9b324c4ce506a7930db362d31f42e2f1759215552d25140ab5667be969e62ee79fcc4e72cf7dcb0ca976d0fbae8b95d3

data/lib/tdigest/tdigest.rb

@@ -72,12 +72,16 @@ module TDigest
     def p_rank(x)
       is_array = x.is_a? Array
       x = [x] unless is_array
+
+      min = @centroids.first
+      max = @centroids.last
+
       x.map! do |item|
         if size == 0
           nil
-        elsif item < @centroids.min[1].mean
+        elsif item < min[1].mean
           0.0
-        elsif item > @centroids.max[1].mean
+        elsif item > max[1].mean
           1.0
         else
           _cumulate(true)
@@ -97,6 +101,9 @@ module TDigest
       is_array = p.is_a? Array
       p = [p] unless is_array
       p.map! do |item|
+        unless (0..1).include? item
+          fail ArgumentError, "p should be in [0,1], got #{item}"
+        end
         if size == 0
           nil
         else
@@ -135,7 +142,7 @@ module TDigest
     end
 
     def size
-      @centroids.count
+      @centroids.size
     end
 
     def to_a
@@ -161,7 +168,8 @@ module TDigest
 
     def _cumulate(exact = false)
       factor = @last_cumulate == 0 ? Float::INFINITY : (@n / @last_cumulate)
-      if @n == @last_cumulate
+      if @n == @last_cumulate ||
+        !exact && @cx && @cx > (factor)
         return
       end
 
@@ -175,8 +183,11 @@ module TDigest
     end
 
     def _digest(x, n)
-      min = @centroids.min
-      max = @centroids.max
+      # Use 'first' and 'last' instead of min/max because of performance reasons
+      # This works because RBTree is sorted
+      min = @centroids.first
+      max = @centroids.last
+
       min = min.nil? ? nil : min[1]
       max = max.nil? ? nil : max[1]
       nearest = find_nearest(x)

data/lib/tdigest/version.rb

@@ -1,3 +1,3 @@
 module TDigest
-  VERSION = "0.0.2"
+  VERSION = "0.0.3"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: tdigest
 version: !ruby/object:Gem::Version
-  version: 0.0.2
+  version: 0.0.3
 platform: ruby
 authors:
 - Sebastian Wallin
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2015-11-22 00:00:00.000000000 Z
+date: 2015-12-07 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rbtree