search_solr_tools 3.3.1 → 3.3.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: 3a9a32ebeed875799854bd53c0b5013604fa9b71
-  data.tar.gz: 755fe9ebb1fb6af9cb6ad4c813c6c288bcca9a65
+  metadata.gz: 7ede056ad3f242ee3181914e874335216920d6fb
+  data.tar.gz: 778b003b040565581be421b43cccfb49f4086952
 SHA512:
-  metadata.gz: fe7cb79dd54c457bbb6ab5d863c14e2a0a4497fce9bf8e5b33da6a9376217f6b06fab12aecbd184d60435c3675dc7d4aac6d0f00f6ead2e411029f98f4f83ce6
-  data.tar.gz: e059a89eb73aa392ec660ad28492dc79fb5c27c7591ba2851ac1dc116fff0191922544d0619a8cc3c79f3cfe9487af650d92de056e7c845283ac94a64ab377cb
+  metadata.gz: 3e9665f5671a2b1f195ec51902d396dd302b262dd848776259db3ba176e6eea53f716f4d1557ca6e8cc1d3954dc5034ae7d66664faa25db112b0a78c12f9733f
+  data.tar.gz: 3a632303193cec2b4784dce4cf0311ba69826b6d1b00c8cbe8da43748bce11434299cc378fdff507411b5003f943da37326b880d860118c51319d849c6f7a90a

data/CHANGELOG.md

@@ -1,3 +1,9 @@
+## v3.3.2
+
+Bugfix
+
+  - Added quote checking for cisl offset parsing check
+
 ## v3.3.1
 
 Bugfix

data/lib/search_solr_tools/harvesters/base.rb

@@ -90,7 +90,7 @@ module SearchSolrTools
 
         # Some of the docs will cause Solr to crash - CPU goes to 195% with `top` and it
         # doesn't seem to recover.
-        return success unless doc_valid?(doc) if content_type == XML_CONTENT_TYPE
+        return success if content_type == XML_CONTENT_TYPE && !doc_valid?(doc)
 
         doc_serialized = get_serialized_doc(doc, content_type)
 

data/lib/search_solr_tools/harvesters/cisl.rb

@@ -42,12 +42,14 @@ module SearchSolrTools
       # To get around this I'd prefer to make assumptions about the token and let it break if
       # they change the formatting.  For now, all fields other than offset should be able to be
       # assumed to remain constant.
+      # glewis 2016-01-15: It broke, offset has quotes around it, so I updated the regex to account for
+      # the possibility, including '"' or '"'
       # If the input is empty, then we are done - return an empty string, which is checked for
       # in the harvest loop.
       def format_resumption_token(resumption_token)
         return '' if resumption_token.empty?
 
-        resumption_token =~ /offset:(\d+)/
+        resumption_token =~ /offset(?:"|")?:(\d+)/
         offset = Regexp.last_match(1)
 
         {

data/lib/search_solr_tools/version.rb

@@ -1,3 +1,3 @@
 module SearchSolrTools
-  VERSION = '3.3.1'
+  VERSION = '3.3.3'
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: search_solr_tools
 version: !ruby/object:Gem::Version
-  version: 3.3.1
+  version: 3.3.3
 platform: ruby
 authors:
 - Chris Chalstrom
@@ -12,7 +12,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2015-09-25 00:00:00.000000000 Z
+date: 2016-01-14 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: iso8601
@@ -347,7 +347,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.4.8
+rubygems_version: 2.4.5.1
 signing_key: 
 specification_version: 4
 summary: Tools to harvest and manage various scientific dataset feeds in a Solr instance.