traject 2.3.3 → 2.3.4

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 2507bfdf51675c233b64ebbecb3247aaf53281ec
-  data.tar.gz: 0fd4edc089aa6cc09e477d4e7e8538f830d20f1c
+  metadata.gz: 70d8b265e2e00866a63fdc067172ee1174efb068
+  data.tar.gz: 6588b8231b636d268765a5428a607b731a34fe7f
 SHA512:
-  metadata.gz: 1091dbd01a7f2adf5ac7a8e4d09f9aec6c67b162379280442f62677e32dc82ec3d575fee3dc131983e5a23e4680cd92cbe4c739c2daf0f25dbf4db83febf7e95
-  data.tar.gz: 7220b2acb51c9ccb6cbbf8cb7caef8c478beaf16f115cbee7b8f719b4de91261926d7a6f6d059d84d0d4abced778148199864fd754c6abc3c7a54ebe85950d1a
+  metadata.gz: 564a834087b4b5d0b032a9a0797cd1587f241afcbf35bf0c3aa928724f9f26fb8d3e7220d43e909c9bf4c03dd91873113e4d2c081ee68ad12ed151ff3ba34284
+  data.tar.gz: 9cbc506f6f2ab5bcbf77811c5018a91d8b588eff2d0c75960b3aada514756e6d90bce441da3db59d1aaa0a9a0f62b3e0c56d1be2c494aae091b06a61f94de46c

data/.travis.yml

@@ -5,12 +5,8 @@ rvm:
   - jruby-19mode
   - jruby-9.0.4.0
   - 1.9
-  - 2.2
-  - 2.3.3
-  - 2.4.0
-before_install:
-  - gem update --system
-  - gem uninstall bundler
-  - gem update bundler
+  - 2.2.8
+  - 2.3.5
+  - 2.4.2
 jdk:
   - oraclejdk8

data/CHANGES.md

@@ -1,5 +1,8 @@
 # Changes
 
+## 2.3.4
+  * Totally internal change to provide easier hooks into indexing process
+
 ## 2.3.3
   * Further squash use of capture-variabels ('$1', etc.)
     to try to work around the non-thread-safety of

data/lib/traject/indexer.rb

@@ -184,7 +184,7 @@ class Traject::Indexer
     @after_processing_steps = []
   end
 
-  # Pass a string file path, or a File object, for
+  # Pass a string file path, a Pathname, or a File object, for
   # a config file to load into indexer.
   #
   # Can raise:
@@ -195,9 +195,9 @@ class Traject::Indexer
   def load_config_file(file_path)
     File.open(file_path) do |file|
       begin
-        self.instance_eval(file.read, file_path)
+        self.instance_eval(file.read, file_path.to_s)
       rescue ScriptError, StandardError => e
-        raise ConfigLoadError.new(file_path, e)
+        raise ConfigLoadError.new(file_path.to_s, e)
       end
     end
   end
@@ -342,12 +342,10 @@ class Traject::Indexer
 
       # Set the index step for error reporting
       context.index_step = index_step
-      accumulator        = log_mapping_errors(context, index_step) do
+      log_mapping_errors(context, index_step) do
         index_step.execute(context) # will always return [] for an each_record step
       end
 
-      add_accumulator_to_context!(accumulator, context) if index_step.to_field_step?
-
       # And unset the index step now that we're finished
       context.index_step = nil
     end
@@ -355,37 +353,6 @@ class Traject::Indexer
     return context
   end
 
-
-  # Add the accumulator to the context with the correct field name
-  # Do post-processing on the accumulator (remove nil values, allow empty
-  # fields, etc)
-  #
-  # Only get here if we've got a to_field step; otherwise the
-  # call to get a field_name will throw an error
-
-  ALLOW_NIL_VALUES       = "allow_nil_values".freeze
-  ALLOW_EMPTY_FIELDS     = "allow_empty_fields".freeze
-  ALLOW_DUPLICATE_VALUES = "allow_duplicate_values".freeze
-
-  def add_accumulator_to_context!(accumulator, context)
-
-    accumulator.compact! unless settings[ALLOW_NIL_VALUES]
-    return if accumulator.empty? and not (settings[ALLOW_EMPTY_FIELDS])
-
-    field_name                      = context.index_step.field_name
-    context.output_hash[field_name] ||= []
-
-    existing_accumulator = context.output_hash[field_name].concat(accumulator)
-    existing_accumulator.uniq! unless settings[ALLOW_DUPLICATE_VALUES]
-
-  rescue NameError => e
-    msg = "Tried to call add_accumulator_to_context with a non-to_field step"
-    msg += context.index_step.inspect
-    logger.error msg
-    raise ArgumentError.new(msg)
-  end
-
-
   # just a wrapper that captures and records any unexpected
   # errors raised in mapping, along with contextual information
   # on record and location in source file of mapping rule.
@@ -414,7 +381,6 @@ class Traject::Indexer
     end
   end
 
-
   # Processes a stream of records, reading from the configured Reader,
   # mapping according to configured mapping rules, and then writing
   # to configured Writer.

data/lib/traject/indexer/step.rb

@@ -150,9 +150,26 @@ class Traject::Indexer
       end
 
 
+      add_accumulator_to_context!(accumulator, context)
       return accumulator
     end
 
+    # Add the accumulator to the context with the correct field name
+    # Do post-processing on the accumulator (remove nil values, allow empty
+    # fields, etc)
+    ALLOW_NIL_VALUES       = "allow_nil_values".freeze
+    ALLOW_EMPTY_FIELDS     = "allow_empty_fields".freeze
+    ALLOW_DUPLICATE_VALUES = "allow_duplicate_values".freeze
+
+    def add_accumulator_to_context!(accumulator, context)
+      accumulator.compact! unless context.settings[ALLOW_NIL_VALUES]
+      return if accumulator.empty? and not (context.settings[ALLOW_EMPTY_FIELDS])
+
+      context.output_hash[field_name] ||= []
+
+      existing_accumulator = context.output_hash[field_name].concat(accumulator)
+      existing_accumulator.uniq! unless context.settings[ALLOW_DUPLICATE_VALUES]
+    end
   end
 
 # A class representing a block of logic called after

data/lib/traject/version.rb

@@ -1,3 +1,3 @@
 module Traject
-  VERSION = "2.3.3"
+  VERSION = "2.3.4"
 end

data/test/indexer/load_config_file_test.rb

@@ -20,7 +20,7 @@ describe "Traject::Indexer#load_config_path" do
     end
   end
 
-  describe "with good config" do
+  describe "with good config provided" do
     before do
       @config_file = tmp_config_file_with(%Q{
         settings do
@@ -32,11 +32,19 @@ describe "Traject::Indexer#load_config_path" do
     after do
       @config_file.unlink
     end
-    it "loads config file by path" do
+
+
+    it "loads config file by path (as a String)" do
       @indexer.load_config_file(@config_file.path)
 
       assert_equal "our_value", @indexer.settings["our_key"]
     end
+
+    it "loads config file by path (as a Pathname)" do
+      @indexer.load_config_file(Pathname.new(@config_file.path))
+
+      assert_equal "our_value", @indexer.settings["our_key"]
+    end
   end
 
   describe "with error in config" do
@@ -60,7 +68,7 @@ describe "Traject::Indexer#load_config_path" do
       assert_equal 4,  e.config_file_lineno
     end
 
-    it "raises good error on StandardError type" do
+    it "raises good error on StandardError type (when passing String)" do
       @config_file = tmp_config_file_with(%Q{
         # Intentional non-syntax error, bad extract_marc spec
         to_field "foo", extract_marc("#%^%^%^")
@@ -74,6 +82,20 @@ describe "Traject::Indexer#load_config_path" do
       assert_equal @config_file.path, e.config_file
       assert_equal 3,  e.config_file_lineno
     end
+
+    it "raises good error on StandardError type (when passing Pathname)" do
+      @config_file = tmp_config_file_with(%Q{
+        # Intentional non-syntax error, bad extract_marc spec
+        to_field "foo", extract_marc("#%^%^%^")
+      }) 
+
+      e = assert_raises(Traject::Indexer::ConfigLoadError) do
+        @indexer.load_config_file(Pathname.new(@config_file.path))
+      end
+
+      assert_kind_of StandardError, e.original
+      assert_equal @config_file.path, e.config_file
+    end
   end
 
 
@@ -84,6 +106,4 @@ describe "Traject::Indexer#load_config_path" do
 
     return file
   end
-
-
-end
+end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: traject
 version: !ruby/object:Gem::Version
-  version: 2.3.3
+  version: 2.3.4
 platform: ruby
 authors:
 - Jonathan Rochkind
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2017-01-25 00:00:00.000000000 Z
+date: 2017-12-06 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: concurrent-ruby
@@ -331,7 +331,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.6.8
+rubygems_version: 2.6.13
 signing_key: 
 specification_version: 4
 summary: Index MARC to Solr; or generally process source records to hash-like structures