cul_hydra 1.9.4 → 1.9.5

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
checksums.yaml CHANGED
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  SHA256:
3
- metadata.gz: c425738e5a5c4634305f4dd908e2c2448db50cc7194c88743a560fd1b2ee8db8
4
- data.tar.gz: 0a69f6de80f11f56812f83ebccd1f0610b4db6e447a8b0c39188449135a830b5
3
+ metadata.gz: 9cbe0237339bc24069451b8fae3742430d8ba2b9e2d102ee7d2cebcf0d6905d3
4
+ data.tar.gz: 9ee494352f8aeb9be9f597a6ec2b24e358a89193e5a4e69e0bfd89113ede246a
5
5
  SHA512:
6
- metadata.gz: b8c9b87f2e4dc966c221d7b32e8cc4e7389b40d5db802e6feb7b2209127464bfd3ae616cc369abe80606fb84b85384edc7f2a59fb2da587b94f90fe0352489c5
7
- data.tar.gz: a09fb529dcb1ff328e20e7d8f67afb020767d0c48d4ef29aef77db38283d3923eb63191e0e4e799338a0f52896f7a67dbd3912b4d2689262dda7bc3a723fef54
6
+ metadata.gz: d94d52aa21a5739cce5277e60f72980d7d542ae28bf6e19491ec9baa6cef141147e10cceae46ae94192952b748dc46ea357de4133d12f2e0677f69eb3873af6c
7
+ data.tar.gz: c95522636da024ea1eb7f34fe11fbfed9f2de266c5f0eba22bbfa06ea4084e1ddaaab9b8f466812b2f3ad9de6d5410935be3d4b451568558a5baaf96d38850a9
@@ -0,0 +1,2 @@
1
+ <?xml version='1.0' encoding='UTF-8'?>
2
+ <mods xmlns='http://www.loc.gov/mods/v3' xmlns:xsi='http://www.w3.org/2001/XMLSchema-instance' xsi:schemaLocation='http://www.loc.gov/mods/v3 http://www.loc.gov/standards/mods/v3/mods-3-4.xsd'><titleInfo><nonSort>The </nonSort><title xml:lang="eng">(ġhotogråphs) ゐ </title></titleInfo></mods>
@@ -1 +1 @@
1
- <doc xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:mods="http://www.loc.gov/mods/v3" xmlns="http://solr.apache.org" xmlns:index="http://repository.cul.columbia.edu/namespace/index/1_0/" xmlns:local="http://local.org/custom"><field name="format">multipartitem</field><field name="collection_h">Columbia University Libraries Content/Online Projects/PRD Customer Orders/</field><field name="internal_h">ldpd:91655/ldpd:91656/ldpd:101479/ldpd:102275/</field><field name="format_h">multipartitem/</field><field name="id">ldpd:102275@ldpd:103434</field><field name="pid_ssi">ldpd:102275</field><field name="descriptor">mods</field><field name="object_ssm">https://repository2.cul.columbia.edu:8443/fedora/get/ldpd:102275</field><field name="identifier_ssi">prd.custord.070017</field><field name="all_text_teim">AN070017</field><field name="all_text_teim">prd.custord.070017</field><field name="lib_collection_sim">Pulitzer Prizes</field><field name="lib_collection_ssm">Pulitzer Prizes</field><field name="lib_project_sim">Pres Orders</field><field name="lib_project_ssm">Customer Order Collection</field><field name="lib_project_teim">Customer Order Collection</field><field name="title_sort">Photographs</field><field name="title_ssm">Photographs</field><field name="title_teim">Photographs</field><field name="lib_name_sim">Kennerly, David</field><field name="lib_name_teim">Kennerly, David</field><field name="lib_name_ssm">Kennerly, David</field><field name="lib_repo_long_ssim">Rare Book and Manuscript Library</field><field name="lib_repo_sim">RBML</field><field name="lib_repo_teim">RBML</field><field name="lib_format_ssm">photographs</field><field name="lib_format_sim">photographs</field><field name="all_text_teim">Original PRD customer order number: 070017</field><field name="extent_teim">2 items</field><field name="access_condition_ssi">Columbia Libraries Staff Use Only</field><field name="all_text_teim">Box 8, JP2 Feature Photography</field></doc>
1
+ <doc xmlns:xs="http://www.w3.org/2001/XMLSchema" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:oai="http://www.openarchives.org/OAI/2.0/oai_dc/" xmlns:mods="http://www.loc.gov/mods/v3" xmlns="http://solr.apache.org" xmlns:index="http://repository.cul.columbia.edu/namespace/index/1_0/" xmlns:local="http://local.org/custom"><field name="format">multipartitem</field><field name="collection_h">Columbia University Libraries Content/Online Projects/PRD Customer Orders/</field><field name="internal_h">ldpd:91655/ldpd:91656/ldpd:101479/ldpd:102275/</field><field name="format_h">multipartitem/</field><field name="id">ldpd:102275@ldpd:103434</field><field name="pid_ssi">ldpd:102275</field><field name="descriptor">mods</field><field name="object_ssm">https://repository2.cul.columbia.edu:8443/fedora/get/ldpd:102275</field><field name="identifier_ssi">prd.custord.070017</field><field name="all_text_teim">AN070017</field><field name="all_text_teim">prd.custord.070017</field><field name="lib_collection_sim">Pulitzer Prizes</field><field name="lib_collection_ssm">Pulitzer Prizes</field><field name="lib_project_sim">Pres Orders</field><field name="lib_project_ssm">Customer Order Collection</field><field name="lib_project_teim">Customer Order Collection</field><field name="title_sort">PHOTOGRAPHS</field><field name="title_ssm">Photographs</field><field name="title_teim">Photographs</field><field name="lib_name_sim">Kennerly, David</field><field name="lib_name_teim">Kennerly, David</field><field name="lib_name_ssm">Kennerly, David</field><field name="lib_repo_long_ssim">Rare Book and Manuscript Library</field><field name="lib_repo_sim">RBML</field><field name="lib_repo_teim">RBML</field><field name="lib_format_ssm">photographs</field><field name="lib_format_sim">photographs</field><field name="all_text_teim">Original PRD customer order number: 070017</field><field name="extent_teim">2 items</field><field name="access_condition_ssi">Columbia Libraries Staff Use Only</field><field name="all_text_teim">Box 8, JP2 Feature Photography</field></doc>
@@ -19,7 +19,7 @@ lib_collection_sim: ['Pulitzer Prizes']
19
19
  lib_collection_ssm: ['Pulitzer Prizes']
20
20
  lib_project_short_ssim: ['Pres Orders']
21
21
  lib_project_full_ssim: ['Customer Order Collection']
22
- title_si: 'Photographs'
22
+ title_si: 'PHOTOGRAPHS'
23
23
  title_display_ssm: ['The Photographs']
24
24
  lib_format_sim: ['photographs']
25
25
  lib_format_ssm: ['photographs']
@@ -39,7 +39,11 @@ module Cul::Hydra::Solrizer
39
39
  n_t = n_t.sub(/^"(.*)"$/, "\\1")
40
40
  n_t = n_t.sub(/^'(.*)'$/, "\\1")
41
41
  is_negative_number = n_t =~ /^-\d+$/
42
- n_t = n_t.sub(/^[[:punct:]]+/, '')
42
+ if strip_punctuation == :all
43
+ n_t = n_t.gsub(/[[:punct:]]/, '')
44
+ else
45
+ n_t = n_t.sub(/^[[:punct:]]+/, '')
46
+ end
43
47
  # this may have 'created' leading/trailing space, so strip
44
48
  n_t.strip!
45
49
  n_t = '-' + n_t if is_negative_number
@@ -92,7 +96,13 @@ module Cul::Hydra::Solrizer
92
96
  base_text << child.text unless child.name == 'nonSort'
93
97
  end
94
98
  end
95
- base_text = ModsFieldable.normalize(base_text, true)
99
+ base_text = ModsFieldable.normalize(base_text, :all)
100
+ # decompose and strip unicode combining characters
101
+ base_text = base_text.unicode_normalize(:nfd)
102
+ base_text.gsub!(/[\u0300-\u036F]/,'')
103
+ # uppercase per Unicode, for ASCII/Latin
104
+ # TODO: decide whether to use full Unicode case, other language options (Turkish, Lithuanian, etc.)
105
+ base_text = base_text.upcase(:ascii)
96
106
  base_text = nil if base_text.empty?
97
107
  base_text
98
108
  end
@@ -1,6 +1,6 @@
1
1
  module Cul
2
2
  module Hydra
3
- VERSION = '1.9.4'
3
+ VERSION = '1.9.5'
4
4
  def self.version
5
5
  VERSION
6
6
  end
metadata CHANGED
@@ -1,15 +1,15 @@
1
1
  --- !ruby/object:Gem::Specification
2
2
  name: cul_hydra
3
3
  version: !ruby/object:Gem::Version
4
- version: 1.9.4
4
+ version: 1.9.5
5
5
  platform: ruby
6
6
  authors:
7
7
  - Benjamin Armintor
8
8
  - Eric O'Hanlon
9
- autorequire:
9
+ autorequire:
10
10
  bindir: bin
11
11
  cert_chain: []
12
- date: 2020-04-15 00:00:00.000000000 Z
12
+ date: 2020-08-14 00:00:00.000000000 Z
13
13
  dependencies:
14
14
  - !ruby/object:Gem::Dependency
15
15
  name: rails
@@ -492,6 +492,7 @@ files:
492
492
  - fixtures/spec/CUL_MODS/mods-subjects.xml
493
493
  - fixtures/spec/CUL_MODS/mods-textual-date.xml
494
494
  - fixtures/spec/CUL_MODS/mods-textual-dates-with-unusual-chars.xml
495
+ - fixtures/spec/CUL_MODS/mods-titles-extended.xml
495
496
  - fixtures/spec/CUL_MODS/mods-titles.xml
496
497
  - fixtures/spec/CUL_MODS/mods-top-level-location-vs-relateditem-location.xml
497
498
  - fixtures/spec/CUL_MODS/mods-unmapped-project.xml
@@ -547,7 +548,7 @@ files:
547
548
  homepage: https://github.com/cul/cul_hydra
548
549
  licenses: []
549
550
  metadata: {}
550
- post_install_message:
551
+ post_install_message:
551
552
  rdoc_options: []
552
553
  require_paths:
553
554
  - app
@@ -565,8 +566,8 @@ required_rubygems_version: !ruby/object:Gem::Requirement
565
566
  - !ruby/object:Gem::Version
566
567
  version: '0'
567
568
  requirements: []
568
- rubygems_version: 3.1.2
569
- signing_key:
569
+ rubygems_version: 3.0.8
570
+ signing_key:
570
571
  specification_version: 4
571
572
  summary: ActiveFedora, OM, and Solrizer implementations for CUL repository apps
572
573
  test_files: []