youtube_transcript2020 0.3.2 → 0.3.3

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 9462fe85e21ee061f717e564efb7b66bec2e62b1c454228d6fb8f31633f7363d
-  data.tar.gz: bb04a03be0cb61058682ce4d9c1159666e41feb2a794fe65d07fbae418412056
+  metadata.gz: 10d150f7396c9e0e18db51381fc5f262eb83595b4f3d5aad313192d86041071c
+  data.tar.gz: 607ae0291272ff40d08db5398f465d99872747e300500c214602b754c43811bb
 SHA512:
-  metadata.gz: d8d4f57bb617e07f647727eaf56c81f1a7efc5eaccc71c323b54e887a270691f01f1d58cc64fe505a40ad6bd5ebd2cea3ea5fc87b7950abe773697acb5e7cc4b
-  data.tar.gz: 9a9a4e72f99f70aca736171490bba54d563aa722576ce6d91435177ed9f0fb327080d84ead963474ab8e46cd1c2dfc4709d2ab93c20853c2a376b091399ed6ad
+  metadata.gz: 244d0242415308c65b07d9bfb09f6776fe3f67817dadbeb5189b4bb5756a4733668fde8b139f636c361598cb6e623ffbdfbad40c3675fd78b74df17701514183
+  data.tar.gz: '07884cdc2b52ad2ac5cf86814f8ae88aa864109d24584e2b256b65b0b8c8168ac7d08f71ce5f2343751b4bbcfe6d81960d4b81b50d4c49109d5b0855198761b1'

data/lib/youtube_transcript2020.rb

@@ -3,10 +3,12 @@
 # file: youtube_transcript2020.rb
 
 require 'yawc'
+require 'json'
 require 'subunit'
 require 'youtube_id'
 require 'simple-config'
 
+# https://github.com/jdepoix/youtube-transcript-api
 
 class YoutubeTranscript2020
 
@@ -54,14 +56,26 @@ class YoutubeTranscript2020
     s = RXFReader.read(obj).first
 
     if s =~ /------+/ then
+
       header, body = s.split(/-----+/,2)
 
       h = SimpleConfig.new(header).to_h
       @id, @author, @title = h[:id], h[:author], h[:title]
       @s = body
+
+    elsif File.extname(obj) == '.json'
+
+      r = JSON.parse(s)
+      @a = r.map {|x| [x['start'], x['text']]}
+      @s = join_sentences(@a)
+
+      return
+
     else
+
       body = obj
       raw_transcript = true
+
     end
 
     puts 'body: ' + body[0..400] if @debug
@@ -193,7 +207,14 @@ EOF
       elsif s[/^"/]
         a2[-1][-1] = a2[-1][-1].chomp + ' ' + s
       elsif s[/^So,? /]
-        a2[-1][-1] += ' ' + s.sub(/^So,? /,'').capitalize
+
+        puts 'so? a2[-1]' + a2[-1].inspect if @debug
+
+        if a2.empty? then
+          a2 << [time, s.sub(/^So,? /,'').capitalize]
+        else
+          a2[-1][-1] += ' ' + s.sub(/^So,? /,'').capitalize
+        end
       elsif s[/^\[(?:Music|Applause)\]/i]
 
         # ignore it
@@ -220,7 +241,17 @@ EOF
 
     # formats the paragraph with the timestamp appearing above
     @a = a2
-    a2.map {|time, s| "\n%s\n\n%s" % [time, s]}.join("\n")
+
+    a2.map do |rawtime, s|
+
+      time = if rawtime.is_a? Float then
+        Subunit.seconds(rawtime).strfunit("%sc")
+      else
+        time
+      end
+
+      "\n%s\n\n%s" % [time, s]
+    end.join("\n")
 
   end
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: youtube_transcript2020
 version: !ruby/object:Gem::Version
-  version: 0.3.2
+  version: 0.3.3
 platform: ruby
 authors:
 - James Robertson
@@ -35,7 +35,7 @@ cert_chain:
   Eh31Azzsjb9JoMQLQliugChaXNzGUL7z5A4jmxeBd91yoD6odSGqLbGuUwjMfyd/
   bYe6x24BppPTKnvGv7iKJQHe
   -----END CERTIFICATE-----
-date: 2022-03-21 00:00:00.000000000 Z
+date: 2022-03-22 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: yawc
@@ -66,7 +66,7 @@ dependencies:
         version: '0.8'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.8.5
+        version: 0.8.7
   type: :runtime
   prerelease: false
   version_requirements: !ruby/object:Gem::Requirement
@@ -76,7 +76,7 @@ dependencies:
         version: '0.8'
     - - ">="
       - !ruby/object:Gem::Version
-        version: 0.8.5
+        version: 0.8.7
 - !ruby/object:Gem::Dependency
   name: simple-config
   requirement: !ruby/object:Gem::Requirement