logstash-filter-kv 4.2.0 → 4.2.1

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 60449ebe2807038eecb8421ac199171dbdb021c5ed8207d5fa37ffa5904c27eb
-  data.tar.gz: 9f9b20456f61eae245a69cdcad9adc86f917711d14cd96cc1e8f8d3317caf038
+  metadata.gz: b34ac3de7b0a2195dd497f68fe24b52acf22469467cd4fdce5052fc3c893b703
+  data.tar.gz: c8979c98ca325f26d09722000ada76d205f1a5499fcaf77e0ee5b03fd4f00088
 SHA512:
-  metadata.gz: ee5f0e5a66cf9a88d70d757e7b729636a4f600d9db8022e0eeabd1220ff80065e62221dca5941efe38fdf9cb059eef026fba663a1c045ff24008c98bf985dfd2
-  data.tar.gz: 5894f8dfe7231d2a77e6a262f0422be1768613147d8a27cb565a08474dab95ec9026405f4fdbd76683ac7011271464361f98e7f3e65b65e358b81a97e9c4ba7d
+  metadata.gz: 91ba5899a87d934d7bd0b19a2fc18f0c6ea8dabd01bb093e558536afcf53a1568cbe72035b1c61e4cb071c7cdd4c2b9d8195fa7fe4d2d3f8bec91aae2ec99979
+  data.tar.gz: feef4f88bd22e8e3486e583fdab40d1b3e5e70590f704436579790f464afcf76de183ba49a57502d915fab2cc1358f0faaf215c3bf5759deb882879aad9b7b09

data/CHANGELOG.md

@@ -1,3 +1,6 @@
+## 4.2.1
+ - Fixes performance regression introduced in 4.1.0 ([#70](https://github.com/logstash-plugins/logstash-filter-kv/issues/70))
+
 ## 4.2.0
  - Added `whitespace => strict` mode, which allows the parser to behave more predictably when input is known to avoid unnecessary whitespace.
  - Added error handling, which tags the event with `_kv_filter_error` if an exception is raised while handling an event instead of allowing the plugin to crash.

data/lib/logstash/filters/kv.rb

@@ -364,7 +364,9 @@ class LogStash::Filters::KV < LogStash::Filters::Base
 
     # a key is a _captured_ sequence of characters or escaped spaces before optional whitespace
     # and followed by either a `value_split`, a `field_split`, or EOF.
-    key_pattern = unquoted_capture(value_split_pattern, field_split_pattern, eof)
+    key_pattern = (original_params.include?('value_split_pattern') || original_params.include?('field_split_pattern')) ?
+                      unquoted_capture_until_pattern(value_split_pattern, field_split_pattern) :
+                      unquoted_capture_until_charclass(@value_split + @field_split)
 
     value_pattern = begin
       # each component expression within value_pattern _must_ capture exactly once.
@@ -379,12 +381,14 @@ class LogStash::Filters::KV < LogStash::Filters::Base
       end
 
       # an unquoted value is a _captured_ sequence of characters or escaped spaces before a `field_split` or EOF.
-      value_patterns << unquoted_capture(field_split_pattern, eof)
+      value_patterns << (original_params.include?('field_split_pattern') ?
+                             unquoted_capture_until_pattern(field_split_pattern) :
+                             unquoted_capture_until_charclass(@field_split))
 
       Regexp.union(value_patterns)
     end
 
-    @scan_re = /#{key_pattern}#{value_split_pattern}#{value_pattern}#{Regexp::union(field_split_pattern, eof)}/
+    @scan_re = /#{key_pattern}#{value_split_pattern}#{value_pattern}?#{Regexp::union(field_split_pattern, eof)}/
     @value_split_re = value_split_pattern
 
     @logger.debug? && @logger.debug("KV scan regex", :regex => @scan_re.inspect)
@@ -446,7 +450,7 @@ class LogStash::Filters::KV < LogStash::Filters::Base
     close_pattern = /#{Regexp.quote(close_quote_sequence)}/
 
     # matches a sequence of zero or more characters are _not_ the `close_quote_sequence`
-    quoted_value_pattern = unquoted_capture(close_pattern)
+    quoted_value_pattern = unquoted_capture_until_charclass(Regexp.quote(close_quote_sequence))
 
     /#{open_pattern}#{quoted_value_pattern}?#{close_pattern}/
   end
@@ -457,8 +461,19 @@ class LogStash::Filters::KV < LogStash::Filters::Base
   # @api private
   # @param *until_lookahead_patterns [Regexp]
   # @return [Regexp]
-  def unquoted_capture(*until_lookahead_patterns)
-    /((?:\\.|(?!#{Regexp::union(until_lookahead_patterns)}).)+)/
+  def unquoted_capture_until_pattern(*patterns)
+    pattern = patterns.size > 1 ? Regexp.union(patterns) : patterns.first
+    /((?:\\.|(?!#{pattern}).)+)/
+  end
+
+  # Helper function for generating *capturing* `Regexp` that will _efficiently_ match any sequence of characters
+  # that are either backslash-escaped or do _not_ belong to the given charclass.
+  #
+  # @api private
+  # @param charclass [String] characters to be injected directly into a regexp charclass; special characters must be pre-escaped.
+  # @return [Regexp]
+  def unquoted_capture_until_charclass(charclass)
+    /((?:\\.|[^#{charclass}])+)/
   end
 
   def transform(text, method)

data/logstash-filter-kv.gemspec

@@ -1,7 +1,7 @@
 Gem::Specification.new do |s|
 
   s.name            = 'logstash-filter-kv'
-  s.version         = '4.2.0'
+  s.version         = '4.2.1'
   s.licenses        = ['Apache License (2.0)']
   s.summary         = "Parses key-value pairs"
   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"

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: logstash-filter-kv
 version: !ruby/object:Gem::Version
-  version: 4.2.0
+  version: 4.2.1
 platform: ruby
 authors:
 - Elastic
 autorequire:
 bindir: bin
 cert_chain: []
-date: 2018-07-23 00:00:00.000000000 Z
+date: 2018-09-04 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   requirement: !ruby/object:Gem::Requirement