curation 1.9 → 1.11

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: e68898021d1c54927e120e46d2b1282535aec94a4daa0c552d9edeb4d6dc1d88
-  data.tar.gz: d6ba0ccabe71d10efb60fb0a5806f94f3fdd42f7e49fe39f00d64c427455d450
+  metadata.gz: bcbcb0a8ecb81b659e2ae2aa35e27acac6006265cf2743b6394f81203c634425
+  data.tar.gz: 25b25c7f30be8f9b004cec7efbd41caae4e69e87000ff585b92c8377b7855439
 SHA512:
-  metadata.gz: 88310bda3ba8221af689f5848c24af2dfd48668caaf875b55f05626e09dbf072747b877e779b93a91c4e602b39cd23e61dc602662bb2b34e8b0e095c6cd9c488
-  data.tar.gz: c750a811383b541beefd3214accbdacda85ab7552ded3863734101cf1ffa42ec963b455c83f817e6176a2e31dc8d57a07ce14fac0f2567a54fd353c42f26d7fb
+  metadata.gz: 420056127e2c0ca86a4ad2e08fb37aa3da31a394264e4d7807a844dd9db61bc44124d6e978a076d9f50a1df37f7a5f80cc782145f54d11c7c3ce00566feca80d
+  data.tar.gz: a3b387e1c7345968c0eaaf1d2e8fbdbf752aa8b7a8b1e3b3807f206c43e7b096419b61b907622065f690d00a55b933038f0f3ed1b4fefa4d970f272afc1438d5

data/Gemfile.lock

@@ -1,7 +1,7 @@
 PATH
   remote: .
   specs:
-    curation (1.9)
+    curation (1.11)
       htmlentities
       metainspector
       nokogiri
@@ -9,82 +9,69 @@ PATH
 GEM
   remote: https://rubygems.org/
   specs:
-    addressable (2.8.0)
-      public_suffix (>= 2.0.2, < 5.0)
+    addressable (2.8.5)
+      public_suffix (>= 2.0.2, < 6.0)
     ansi (1.5.0)
     builder (3.2.4)
     byebug (11.1.3)
     domain_name (0.5.20190701)
       unf (>= 0.0.5, < 1.0.0)
-    faraday (1.10.0)
-      faraday-em_http (~> 1.0)
-      faraday-em_synchrony (~> 1.0)
-      faraday-excon (~> 1.1)
-      faraday-httpclient (~> 1.0)
-      faraday-multipart (~> 1.0)
-      faraday-net_http (~> 1.0)
-      faraday-net_http_persistent (~> 1.0)
-      faraday-patron (~> 1.0)
-      faraday-rack (~> 1.0)
-      faraday-retry (~> 1.0)
+    faraday (2.7.10)
+      faraday-net_http (>= 2.0, < 3.1)
       ruby2_keywords (>= 0.0.4)
     faraday-cookie_jar (0.0.7)
       faraday (>= 0.8.0)
       http-cookie (~> 1.0.0)
-    faraday-em_http (1.0.0)
-    faraday-em_synchrony (1.0.0)
     faraday-encoding (0.0.5)
       faraday
-    faraday-excon (1.1.0)
-    faraday-http-cache (2.2.0)
+    faraday-follow_redirects (0.3.0)
+      faraday (>= 1, < 3)
+    faraday-gzip (1.0.0)
+      faraday (>= 1.0)
+      zlib (~> 2.1)
+    faraday-http-cache (2.5.0)
       faraday (>= 0.8)
-    faraday-httpclient (1.0.1)
-    faraday-multipart (1.0.3)
-      multipart-post (>= 1.2, < 3)
-    faraday-net_http (1.0.1)
-    faraday-net_http_persistent (1.2.0)
-    faraday-patron (1.0.0)
-    faraday-rack (1.0.0)
-    faraday-retry (1.0.3)
-    faraday_middleware (1.2.0)
-      faraday (~> 1.0)
-    fastimage (2.2.6)
+    faraday-net_http (3.0.2)
+    faraday-retry (2.2.0)
+      faraday (~> 2.0)
+    fastimage (2.2.7)
     htmlentities (4.3.4)
-    http-cookie (1.0.4)
+    http-cookie (1.0.5)
       domain_name (~> 0.5)
-    metainspector (5.12.1)
-      addressable (~> 2.7)
-      faraday (>= 1.4, < 3.0)
+    metainspector (5.15.0)
+      addressable (~> 2.8.4)
+      faraday (~> 2.5)
       faraday-cookie_jar (~> 0.0)
       faraday-encoding (~> 0.0)
-      faraday-http-cache (~> 2.2)
-      faraday_middleware (~> 1.0)
+      faraday-follow_redirects (~> 0.3)
+      faraday-gzip (>= 0.1, < 2.0)
+      faraday-http-cache (~> 2.5)
+      faraday-retry (~> 2.0)
       fastimage (~> 2.2)
       nesty (~> 1.0)
-      nokogiri (~> 1.11)
-    mini_portile2 (2.8.0)
-    minitest (5.15.0)
-    minitest-reporters (1.5.0)
+      nokogiri (~> 1.13)
+    minitest (5.19.0)
+    minitest-reporters (1.6.1)
       ansi
       builder
       minitest (>= 5.0)
       ruby-progressbar
-    multipart-post (2.1.1)
     nesty (1.0.2)
-    nokogiri (1.13.6)
-      mini_portile2 (~> 2.8.0)
+    nokogiri (1.15.4-x86_64-darwin)
       racc (~> 1.4)
-    public_suffix (4.0.7)
-    racc (1.6.0)
+    public_suffix (5.0.3)
+    racc (1.7.1)
     rake (12.3.3)
-    ruby-progressbar (1.11.0)
+    ruby-progressbar (1.13.0)
     ruby2_keywords (0.0.5)
     unf (0.1.4)
       unf_ext
-    unf_ext (0.0.8.1)
+    unf_ext (0.0.8.2)
+    zlib (2.1.1)
 
 PLATFORMS
-  ruby
+  x86_64-darwin-21
+  x86_64-darwin-22
 
 DEPENDENCIES
   byebug
@@ -94,4 +81,4 @@ DEPENDENCIES
   rake (~> 12.0)
 
 BUNDLED WITH
-   2.1.4
+   2.4.6

data/lib/curation/version.rb

@@ -1,3 +1,3 @@
 module Curation
-  VERSION = "1.9"
+  VERSION = "1.11"
 end

data/lib/curation.rb

@@ -8,6 +8,7 @@ module Curation
 
   class Page
     attr_reader :url
+    attr_accessor :verbose
 
     BLACKLIST = [
       'head', 'script', 'style', 'iframe', 'nav', 'noscript', 'header', 'footer', 'aside',
@@ -23,6 +24,7 @@ module Curation
     def initialize(url, html = nil)
       @url = url.to_s.gsub('http://', 'https://')
       @html = html
+      @verbose = false
     end
 
     def title
@@ -66,24 +68,37 @@ module Curation
           return possibility unless possibility.to_s.empty?
         end
       rescue
-        puts 'Curation::Page find_title error'
+        log 'Curation::Page find_title error'
       end
       return ''
     end
 
     def find_image
+      log "Curation::Page find_image #{url}"
       if json_ld.any?
         json_ld.each do |ld|
           ld = ld.first if ld.is_a?(Array)
           if ld.has_key? 'image'
             image_data = ld['image']
-            return image_data if image_data.is_a? String
+            if image_data.is_a? String
+              log "Curation::Page find_image json_ld string"
+              return image_data 
+            end
             if image_data.is_a? Array
               first = image_data.first
-              return first if first.is_a? String
-              return first['url'] if first.is_a? Hash
+              if first.is_a? String
+                log "Curation::Page find_image json_ld array"
+                return first 
+              end
+              if first.is_a? Hash
+                log "Curation::Page find_image json_ld array url"
+                return first['url'] 
+              end
+            end
+            if image_data.is_a? Hash
+              log "Curation::Page find_image json_ld url"
+              return image_data['url'] 
             end
-            return image_data['url'] if image_data.is_a? Hash
           end
         end
       end
@@ -179,7 +194,7 @@ module Curation
           @json_ld.flatten!
           # require 'byebug'; byebug
         rescue
-          puts 'Curation::Page json_ld error'
+          log 'Curation::Page json_ld error'
         end
       end
       @json_ld
@@ -194,7 +209,7 @@ module Curation
     def file
       @file ||= URI.open url, 'User-Agent' => "Mozilla/5.0"
     rescue
-      puts "Curation::Page file error with url #{url}"
+      log "Curation::Page file error with url #{url}"
     end
 
     def html
@@ -205,7 +220,7 @@ module Curation
       end
       @html
     rescue
-      puts "Curation::Page html error"
+      log "Curation::Page html error"
     end
 
     def nokogiri
@@ -220,7 +235,7 @@ module Curation
       end
       @nokogiri
     rescue
-      puts 'Curation::Page nokogiri error'
+      log 'Curation::Page nokogiri error'
     end
 
     def metainspector
@@ -230,13 +245,13 @@ module Curation
       end
       @metainspector
     rescue
-      puts 'Curation::Page metainspector error'
+      log 'Curation::Page metainspector error'
     end
 
     def metatags
       @metatags ||= metainspector.meta_tag['name']
     rescue
-      puts 'Curation::Page metatags error'
+      log 'Curation::Page metatags error'
     end
 
     # réforme -> réforme
@@ -259,5 +274,9 @@ module Curation
         text
       end
     end
+
+    def log(message)
+      puts message if verbose
+    end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: curation
 version: !ruby/object:Gem::Version
-  version: '1.9'
+  version: '1.11'
 platform: ruby
 authors:
 - Arnaud Levy
 autorequire:
 bindir: exe
 cert_chain: []
-date: 2022-05-16 00:00:00.000000000 Z
+date: 2023-08-29 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: metainspector
@@ -94,7 +94,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.1.6
+rubygems_version: 3.4.6
 signing_key:
 specification_version: 4
 summary: Curation of content