From f57d43c1b3f317b66b42b9d748d88dd0227a2079 Mon Sep 17 00:00:00 2001
From: Francis Tyers <ftyers@users.noreply.github.com>
Date: Sat, 14 Jul 2018 14:05:14 +0200
Subject: [PATCH 1/2] tatar.rb

Add an example for Tatar with abbreviations.
---
 lib/pragmatic_segmenter/languages/tatar.rb | 28 ++++++++++++++++++++++
 1 file changed, 28 insertions(+)
 create mode 100644 lib/pragmatic_segmenter/languages/tatar.rb

diff --git a/lib/pragmatic_segmenter/languages/tatar.rb b/lib/pragmatic_segmenter/languages/tatar.rb
new file mode 100644
index 0000000..66c3c82
--- /dev/null
+++ b/lib/pragmatic_segmenter/languages/tatar.rb
@@ -0,0 +1,28 @@
+# frozen_string_literal: true
+
+module PragmaticSegmenter
+  module Languages
+    module Tatar
+      include Languages::Common
+
+      module Abbreviation
+        ABBREVIATIONS = Set.new(["биекл", "биол", "б.э", "б.э.к", "геогр", "геол", "е", "елл", "к", "кг", "км", "км²", "км³", "лат", "м", "м²", "м³", "мәс", "млн", "млрд", "мм", "обл", "окр", "пед", "см", "т", "тех", "тирәнл", "трлн", "һ.б", "хим", "экон", "эл"]).freeze
+        PREPOSITIVE_ABBREVIATIONS = [].freeze
+        NUMBER_ABBREVIATIONS = [].freeze
+      end
+
+      class AbbreviationReplacer < AbbreviationReplacer
+        SENTENCE_STARTERS = [].freeze
+
+        private
+
+        def replace_period_of_abbr(txt, abbr)
+          txt.gsub!(/(?<=\s#{abbr.strip})\./, '∯')
+          txt.gsub!(/(?<=\A#{abbr.strip})\./, '∯')
+          txt.gsub!(/(?<=^#{abbr.strip})\./, '∯')
+          txt
+        end
+      end
+    end
+  end
+end

From 594726d4a0df9c7ece1def1304f5fc5e0422e4b4 Mon Sep 17 00:00:00 2001
From: Francis Tyers <ftyers@users.noreply.github.com>
Date: Sat, 14 Jul 2018 14:14:16 +0200
Subject: [PATCH 2/2] Update tatar.rb

Copy/paste the handling of 1. from Danish
---
 lib/pragmatic_segmenter/languages/tatar.rb | 10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/lib/pragmatic_segmenter/languages/tatar.rb b/lib/pragmatic_segmenter/languages/tatar.rb
index 66c3c82..511ef61 100644
--- a/lib/pragmatic_segmenter/languages/tatar.rb
+++ b/lib/pragmatic_segmenter/languages/tatar.rb
@@ -10,6 +10,16 @@ module Abbreviation
         PREPOSITIVE_ABBREVIATIONS = [].freeze
         NUMBER_ABBREVIATIONS = [].freeze
       end
+      
+      # This handles the case where a dot is used to denote and ordinal (5. Juni)
+      module Numbers
+        NumberPeriodSpaceRule = Rule.new(/(?<=\s[0-9]|\s([1-9][0-9]))\.(?=\s)/, '∯')
+
+        All = [
+          Common::Numbers::All,
+          NumberPeriodSpaceRule
+        ]
+      end
 
       class AbbreviationReplacer < AbbreviationReplacer
         SENTENCE_STARTERS = [].freeze