added setting dictionary_file_max_bytes to config the maximum bytes…

… size

of the yaml file in `dictionary_path` to overcome the 3MB size limit from
SnakeYaml 1.33

Fixed: #96

CHANGELOG.md

@@ -1,3 +1,6 @@
+## 3.4.1
+  - Fix the limitation of the size of yaml file that exceeds 3MB
+
 ## 3.4.0
  - Refactor: leverage scheduler mixin [#93](https://github.com/logstash-plugins/logstash-filter-translate/pull/93)
 

lib/logstash/filters/dictionary/file.rb

@@ -9,9 +9,9 @@ class File
 
     include LogStash::Util::Loggable
 
-    def self.create(path, refresh_interval, refresh_behaviour, exact, regex)
+    def self.create(path, refresh_interval, refresh_behaviour, exact, regex, params)
       if /\.y[a]?ml$/.match(path)
-        instance = YamlFile.new(path, refresh_interval, exact, regex)
+        instance = YamlFile.new(path, refresh_interval, exact, regex, params["dictionary_file_max_bytes"])
       elsif path.end_with?(".json")
         instance = JsonFile.new(path, refresh_interval, exact, regex)
       elsif path.end_with?(".csv")
@@ -31,15 +31,15 @@ def self.create(path, refresh_interval, refresh_behaviour, exact, regex)
 
     attr_reader :dictionary, :fetch_strategy
 
-    def initialize(path, refresh_interval, exact, regex)
+    def initialize(path, refresh_interval, exact, regex, file_max_bytes = nil)
       @dictionary_path = path
       @refresh_interval = refresh_interval
       @short_refresh = @refresh_interval <= 300
       rw_lock = java.util.concurrent.locks.ReentrantReadWriteLock.new
       @write_lock = rw_lock.writeLock
       @dictionary = Hash.new
       @update_method = method(:merge_dictionary)
-      initialize_for_file_type
+      initialize_for_file_type(file_max_bytes)
       args = [@dictionary, rw_lock]
       klass = case
               when exact && regex then FetchStrategy::File::ExactRegex
@@ -68,7 +68,7 @@ def set_update_strategy(method_sym)
 
     protected
 
-    def initialize_for_file_type
+    def initialize_for_file_type(file_max_bytes)
       # sub class specific initializer
     end
 

lib/logstash/filters/dictionary/json_file.rb

@@ -6,9 +6,6 @@ class JsonFile < File
 
     protected
 
-    def initialize_for_file_type
-    end
-
     def read_file_into_dictionary
       content = IO.read(@dictionary_path, :mode => 'r:bom|utf-8')
       @dictionary.update(LogStash::Json.load(content)) unless content.nil? || content.empty?

lib/logstash/filters/dictionary/yaml_file.rb

@@ -7,18 +7,20 @@ class YamlFile < File
 
     protected
 
-    def initialize_for_file_type
+    def initialize_for_file_type(file_max_bytes)
       @visitor = YamlVisitor.create
+
+      @parser = Psych::Parser.new(Psych::TreeBuilder.new)
+      @parser.code_point_limit = file_max_bytes
     end
 
     def read_file_into_dictionary
       # low level YAML read that tries to create as
       # few intermediate objects as possible
       # this overwrites the value at key
-      @visitor.accept_with_dictionary(
-        @dictionary, Psych.parse_stream(
-          IO.read(@dictionary_path, :mode => 'r:bom|utf-8')
-      ))
+      yaml_string = IO.read(@dictionary_path, :mode => 'r:bom|utf-8')
+      @parser.parse(yaml_string, @dictionary_path)
+      @visitor.accept_with_dictionary(@dictionary, @parser.handler.root)
     end
   end
 end end end

lib/logstash/filters/translate.rb

@@ -102,6 +102,11 @@ class Translate < LogStash::Filters::Base
   # as the original text, and the second column as the replacement.
   config :dictionary_path, :validate => :path
 
+  # Setting the maximum bytes size of the file in `dictionary_path`. This setting is effective for YAML file only.
+  # Snakeyaml 1.33 has a default limit 3MB. YAML file over the limit throws exception. JSON and CSV currently do not have such limit.
+  # The limit could be too small in some use cases. Setting a bigger number in `dictionary_file_max_bytes` to relax the restriction.
+  config :dictionary_file_max_bytes, :validate => :number, :default => 3_145_728
+
   # When using a dictionary file, this setting will indicate how frequently
   # (in seconds) logstash will check the dictionary file for updates.
   config :refresh_interval, :validate => :number, :default => 300
@@ -180,8 +185,12 @@ def register
       )
     end
 
+    if @dictionary_path && @dictionary_file_max_bytes <= 0
+      raise LogStash::ConfigurationError, "Please set a positive number in `dictionary_file_max_bytes => #{@dictionary_file_max_bytes}`."
+    end
+
     if @dictionary_path
-      @lookup = Dictionary::File.create(@dictionary_path, @refresh_interval, @refresh_behaviour, @exact, @regex)
+      @lookup = Dictionary::File.create(@dictionary_path, @refresh_interval, @refresh_behaviour, @exact, @regex, params)
     else
       @lookup = Dictionary::Memory.new(@dictionary, @exact, @regex)
     end

logstash-filter-translate.gemspec

@@ -1,7 +1,7 @@
 Gem::Specification.new do |s|
 
   s.name            = 'logstash-filter-translate'
-  s.version         = '3.4.0'
+  s.version         = '3.4.1'
   s.licenses        = ['Apache License (2.0)']
   s.summary         = "Replaces field contents based on a hash or YAML file"
   s.description     = "This gem is a Logstash plugin required to be installed on top of the Logstash core pipeline using $LS_HOME/bin/logstash-plugin install gemname. This gem is not a stand-alone program"
@@ -25,6 +25,7 @@ Gem::Specification.new do |s|
   s.add_runtime_dependency 'logstash-mixin-validator_support', '~> 1.0'
   s.add_runtime_dependency 'logstash-mixin-deprecation_logger_support', '~> 1.0'
   s.add_runtime_dependency "logstash-mixin-scheduler", '~> 1.0'
+  s.add_runtime_dependency "psych", ">= 5.1.0"
 
   s.add_development_dependency 'logstash-devutils'
   s.add_development_dependency 'rspec-sequencing'

spec/filters/translate_spec.rb

@@ -240,6 +240,45 @@ def self.build_fixture_path(filename)
       end
     end
 
+    describe "when using a yml file with size limit" do
+      let(:config) do
+        {
+          "source"      => "status",
+          "target"      => "translation",
+          "dictionary_path"  => dictionary_path,
+          "dictionary_file_max_bytes" => dictionary_size  # the file is 18 bytes
+        }
+      end
+      let(:dictionary_path)  { TranslateUtil.build_fixture_path("dict.yml") }
+      let(:event) { LogStash::Event.new("status" => "a") }
+
+      context "file is over size limit" do
+        let(:dictionary_size) { 17 }
+
+        it "raises exception" do
+          expect { subject.register }.to raise_error(/The incoming YAML document exceeds/)
+        end
+      end
+
+      context "file is within size limit" do
+        let(:dictionary_size) { 18 }
+
+        it "returns the exact translation" do
+          subject.register
+          subject.filter(event)
+          expect(event.get("translation")).to eq(1)
+        end
+      end
+
+      context "file size set to zero" do
+        let(:dictionary_size) { 0 }
+
+        it "raises configuration exception" do
+          expect { subject.register }.to raise_error(LogStash::ConfigurationError, /Please set a positive number/)
+        end
+      end
+    end
+
     context "when using a map tagged yml file" do
       let(:dictionary_path)  { TranslateUtil.build_fixture_path("tag-map-dict.yml") }
       let(:event) { LogStash::Event.new("status" => "six") }