es_dump_restore 2.2.2 → 3.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: a4965b79edc9ae5cc1f6eaf0f27199792ee991d9
-  data.tar.gz: 1f061a54c152128037b5510e0ec901ebef0d5799
+  metadata.gz: fba1c866a6ebce9f261aaf8415155de3f450d5ef
+  data.tar.gz: 9ebbc5d6ffc601122b84864ce6bb7750e86e4b43
 SHA512:
-  metadata.gz: 5b1609965037f9a4e0acf23fa29e3ebcb7c5756d5098e90edd255de74c353eb9823070922fe2e444537120d832bd43c51923c23297d09f881fdf683551c7b8f9
-  data.tar.gz: aa4cc30bf4c3299a4f388b9103e1d043525f27003a8d72cd2286252fdae1c6497ccb42de2980b618b95229aab55d84b4cad5dcf8937abfd41181a253f7b06cda
+  metadata.gz: 6f5d6f71fc4a58c8989643e1d31aa1fe4952a227d392595d73525ff10ec2b89abe7af1d77281006af986a836aff0e6386a0b3fa5d14e37bb2434744cc663da35
+  data.tar.gz: 4729f3c751b2606aa8a4c714bcc9add0719bd3403a78a1471708d177d4544ce2e3fcec42faf0f2514ccc4cd240a5c244dd991b1f3156a77c4d597f017cb3849e

data/CHANGELOG.md

@@ -1,3 +1,10 @@
+# Version 3.0.0 - December 1, 2017
+
+* Fixes a bug causing document versions not to be recorded
+* BREAKING CHANGE: Exposes a --preserve-versions option to dump and dump_type with a default value
+  of **false** in order to record document versions. Previous intended default behavior was to
+  preserve versions.
+
 # Version 2.2.2 - November 13, 2017
 
 * Fix for incorrect parsing of the retry command-line parameter (thanks to Mat Moore!)

data/lib/es_dump_restore/app.rb

@@ -14,15 +14,17 @@ module EsDumpRestore
     option :verbose, :type => :boolean # add some additional output
 
     desc "dump URL INDEX_NAME FILENAME", "Creates a dumpfile based on the given ElasticSearch index"
+    method_option :preserve_versions, type: :boolean, desc: "Whether to record the current version for each indexed document"
     def dump(url, index_name, filename)
       client = EsClient.new(url, index_name, nil)
-      _dump(client, filename)
+      _dump(client, filename, options[:preserve_versions])
     end
 
     desc "dump_type URL INDEX_NAME TYPE FILENAME", "Creates a dumpfile based on the given ElasticSearch index"
+    method_option :preserve_versions, type: :boolean, desc: "Whether to record the current version for each indexed document"
     def dump_type(url, index_name, type, filename)
       client = EsClient.new(url, index_name, type)
-      _dump(client, filename)
+      _dump(client, filename, options[:preserve_versions])
     end
 
     desc "restore URL INDEX_NAME FILENAME", "Restores a dumpfile into the given ElasticSearch index"
@@ -62,7 +64,7 @@ module EsDumpRestore
 
     private
 
-    def _dump(client, filename)
+    def _dump(client, filename, preserve_versions)
       Dumpfile.write(filename) do |dumpfile|
         dumpfile.index = {
           settings: client.settings,
@@ -77,8 +79,12 @@ module EsDumpRestore
             client.each_scroll_hit(scroll_id) do |hit|
               hit['fields'] ||= {}
               metadata = { index: { _type: hit["_type"], _id: hit["_id"] } }
+              if preserve_versions
+                metadata[:index][:version_type] = :external
+                metadata[:index][:_version] = hit["_version"]
+              end
 
-              %w(_timestamp _version _routing _percolate _parent _ttl).each do |metadata_field|
+              %w(_timestamp _routing _percolate _parent _ttl).each do |metadata_field|
                 metadata[:index][metadata_field] = hit['fields'][metadata_field] if hit['fields'][metadata_field]
               end
 

data/lib/es_dump_restore/es_client.rb

@@ -46,9 +46,9 @@ module EsDumpRestore
 
     def start_scan(&block)
       scroll = request(:get, "#{@path_prefix}/_search",
-        query: { search_type: 'scan', scroll: '10m', size: 500 },
+        query: { search_type: 'scan', scroll: '10m', size: 500, version: true },
         body: MultiJson.dump({
-          fields: ['_source', '_timestamp', '_version', '_routing', '_percolate', '_parent', '_ttl'],
+          fields: ['_source', '_timestamp', '_routing', '_percolate', '_parent', '_ttl'],
           query: { match_all: {} } }
         ))
       total = scroll["hits"]["total"]
@@ -60,9 +60,13 @@ module EsDumpRestore
     def each_scroll_hit(scroll_id, &block)
       done = 0
       loop do
-        batch = request(:get, '_search/scroll', {query: {
-          scroll: '10m', scroll_id: scroll_id
-        }}, [404])
+        batch = request(:get, '_search/scroll', {
+          query: {
+            version: true,
+            scroll: '10m',
+            scroll_id: scroll_id
+          }
+        }, [404])
 
         batch_hits = batch["hits"]
         break if batch_hits.nil?

data/lib/es_dump_restore/version.rb

@@ -1,3 +1,3 @@
 module EsDumpRestore
-  VERSION = "2.2.2"
+  VERSION = "3.0.0"
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: es_dump_restore
 version: !ruby/object:Gem::Version
-  version: 2.2.2
+  version: 3.0.0
 platform: ruby
 authors:
 - Nat Budin
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-11-13 00:00:00.000000000 Z
+date: 2017-12-01 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: multi_json
@@ -135,8 +135,9 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.6.4
+rubygems_version: 2.6.13
 signing_key: 
 specification_version: 4
 summary: Dump ElasticSearch indexes to files and restore them back
 test_files: []
+has_rdoc: