enumark 0.1.2 → 1.0.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 5ff7445e38a40b2f5e2e7196c9873833d1853c8994cf0ac13699c322174658ee
-  data.tar.gz: 2d679d07bed90e0973701f6fd01a5a6b2c0c86bf61529a97d3ade1d57d68ca4c
+  metadata.gz: fa3a4cc7fa496b1863e6de8c004144d85d983e1b8886f252c909c1031b215f66
+  data.tar.gz: a462ef2d2b0ee84b4330e6daa2b528bbf1e643181715ff8bfb0030c2fabc6207
 SHA512:
-  metadata.gz: 6ab4b15cf7de8e56c3af1e681f369258e9806b34736e2ffb5df85c4195a9be354c272d46e2627c5f1bf8bd69b26f9144ac1f46736835926f481c2a5bc0204a92
-  data.tar.gz: c86177d92d36168653a669d5f11ed2136549026760c042a89ad385aece50bcc1faf69e269c30fcf2ae57d53b9df839db200fe3a0a7cd5f335b28463fdfc2a255
+  metadata.gz: 0712afefeb2f3950b7763cb4d2246e39034a393bf8beb75bc88e10c22f6d45122fcf19f38cda647f4f3ec441705648b46d88fc729f3162cc2cfc50bb66bbd1a3
+  data.tar.gz: 482ce353a24835363a27a64b15e0d74e08e352ab57cc5e19b0008f3bfa0dabbeeed195b669f006d212bc76fcd8a604d8e16bb077d9e1fb85fcf012eb52ba348c

data/.ruby-version

@@ -0,0 +1 @@
+3.1.1

data/README.md

@@ -27,6 +27,8 @@ enum = Enumark.new('/path/to/bookmark_dump_file')
 
 enum.each do |item|
   item.name
+  item.dump_date
+  item.add_date
   item.href
   item.host
   item.categories
@@ -46,6 +48,29 @@ enum.each_dup_href do |href|
   href.name
   href.items
 end
+
+enum.each_category do |cate|
+  cate.name
+  cate.items
+end
+```
+
+Explore trends of your dump files:
+
+```ruby
+dir = Enumark::Dir.new('/path/to/directory_with_bookmark_dump_files_more_than_one')
+
+dir.added # select items in last file but not in second to last
+dir.deleted # reject items in last file
+dir.uniq # union all items
+dir.static # select items appear in all files
+dir.all # enumerator of all items in all files
+```
+
+Set config
+
+```ruby
+Enumark::Config.set(logger: STDOUT)
 ```
 
 ## Development
@@ -56,7 +81,7 @@ To install this gem onto your local machine, run `bundle exec rake install`. To
 
 ## Contributing
 
-Bug reports and pull requests are welcome on GitHub at https://github.com/[USERNAME]/enumark. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [code of conduct](https://github.com/[USERNAME]/enumark/blob/master/CODE_OF_CONDUCT.md).
+Bug reports and pull requests are welcome on GitHub at https://github.com/turnon/enumark. This project is intended to be a safe, welcoming space for collaboration, and contributors are expected to adhere to the [code of conduct](https://github.com/[USERNAME]/enumark/blob/master/CODE_OF_CONDUCT.md).
 
 ## License
 

data/enumark.gemspec

@@ -11,7 +11,7 @@ Gem::Specification.new do |spec|
   spec.summary       = "Enumerate chrome bookmark dump file"
   spec.homepage      = "https://github.com/turnon/enumark"
   spec.license       = "MIT"
-  spec.required_ruby_version = Gem::Requirement.new(">= 2.4.0")
+  spec.required_ruby_version = Gem::Requirement.new(">= 3.1.0")
 
   # Specify which files should be added to the gem when it is released.
   # The `git ls-files -z` loads the files in the RubyGem that have been added into git.

data/lib/enumark/category.rb

@@ -0,0 +1,21 @@
+# frozen_string_literal: true
+
+class Enumark
+  class Category
+
+    START = /^\s.*<DT><H3/
+    ENDIND = /^\s.*<\/DL><p>/
+    PATTERN = /ADD_DATE="(.*?)".*LAST_MODIFIED="(.*?)".*>(.*)<\/H3/
+
+    attr_reader :name
+    alias_method :inspect, :name
+    alias_method :to_s, :name
+
+    def initialize(line)
+      m = line.match(PATTERN)
+      @add_date = m[1]
+      @last_mod = m[2]
+      @name = m[3]
+    end
+  end
+end

data/lib/enumark/config.rb

@@ -0,0 +1,17 @@
+# frozen_string_literal: true
+
+class Enumark
+  module Config
+    class << self
+      def set(**cfg)
+        (@cfg ||= {}).merge!(cfg)
+      end
+
+      def get(key)
+        @cfg[key]
+      end
+    end
+
+    set
+  end
+end

data/lib/enumark/dir.rb

@@ -0,0 +1,42 @@
+# frozen_string_literal: true
+
+class Enumark
+  class Dir
+    def initialize(dir)
+      @enumarks = ::Dir.glob(dir).map{ |f| ::Enumark.new(f) }
+      raise 'Not enough to process' if @enumarks.count <= 1
+    end
+
+    def added
+      @added ||= (@enumarks[-1] - @enumarks[-2])
+    end
+
+    def deleted
+      @deleted ||= @enumarks[0..-2].reverse_each.reduce(&:|) - @enumarks[-1]
+    end
+
+    def uniq
+      @uniq ||= @enumarks.reverse_each.reduce(&:|)
+    end
+
+    def static
+      @static ||= @enumarks.reverse_each.reduce(&:&)
+    end
+
+    def all
+      Enumerator.new do |yielder|
+        logger = Config.get(:logger)
+        file_count = @enumarks.count
+
+        @enumarks.each_with_index do |enum, idx|
+          enum.each do |item|
+            yielder << item
+            logger.printf("--> %6d/%-6d = %3f \r", idx + 1, file_count, ((idx + 1).to_f / file_count * 100).round(2)) if logger
+          end
+        end
+
+        logger.puts if logger
+      end
+    end
+  end
+end

data/lib/enumark/grouping.rb

@@ -0,0 +1,29 @@
+# frozen_string_literal: true
+
+class Enumark
+
+  class Grouping
+    Group = Struct.new(:name, :items)
+
+    def initialize(enumark, key, &post)
+      @lock = Mutex.new
+      @collection = nil
+
+      @enumark = enumark
+      @key = key
+      @post = post
+    end
+
+    def each(&block)
+      unless @collection
+        @lock.synchronize do
+          @collection = @enumark.group_by(&@key)
+          @collection = @post.call(@collection) if @post
+          @collection = @collection.map{ |k, items| Group.new(k, items) }
+        end
+      end
+
+      @collection.each(&block)
+    end
+  end
+end

data/lib/enumark/item.rb

@@ -0,0 +1,45 @@
+# frozen_string_literal: true
+
+class Enumark
+  class Item
+
+    PREFIX = /^\s.*<DT><A/
+    PATTERN = /HREF="(.*?)".*ADD_DATE="(.*?)".*>(.*)<\/A>/
+    USELESS_SHARP = /\#.*$/
+
+    attr_reader :dump_date, :name, :href, :add_date, :categories
+
+    def initialize(dump_date, line, categories)
+      @dump_date = dump_date
+      m = line.match(PATTERN)
+      @href = m[1].gsub(USELESS_SHARP, '')
+      @add_date = Time.at(m[2].to_i)
+      @name = m[3]
+      @categories = categories
+    end
+
+    def inspect
+      @inspect ||= "#{add_date.strftime('%F %T')} #{categories_str}> #{name}"
+    end
+
+    def categories_str
+      @categories_str ||= "/#{categories.join('/')}"
+    end
+
+    def to_s
+      inspect
+    end
+
+    def hash
+      href.hash
+    end
+
+    def eql?(another)
+      href.eql?(another.href)
+    end
+
+    def host
+      @host ||= (URI.parse(href).host rescue 'unknown')
+    end
+  end
+end

data/lib/enumark/version.rb

@@ -1,5 +1,5 @@
 # frozen_string_literal: true
 
 class Enumark
-  VERSION = "0.1.2"
+  VERSION = "1.0.0"
 end

data/lib/enumark.rb

@@ -1,115 +1,29 @@
 # frozen_string_literal: true
 
 require_relative "enumark/version"
+require 'enumark/item'
+require 'enumark/category'
+require 'enumark/grouping'
+require 'enumark/dir'
+require 'enumark/config'
 
 class Enumark
   include Enumerable
 
-  CATEGORY_START = /^\s.*<DT><H3/
-  CATEGORY_END = /^\s.*<\/DL><p>/
-  CATEGORY_NAME = /ADD_DATE="(.*?)".*LAST_MODIFIED="(.*?)".*>(.*)<\/H3/
-
-  ITEM_PREFIX = /^\s.*<DT><A/
-  ITEM_NAME = /HREF="(.*?)".*ADD_DATE="(.*?)".*>(.*)<\/A>/
-
-  class Category
-    attr_reader :name
-    alias_method :inspect, :name
-    alias_method :to_s, :name
-
-    def initialize(line)
-      m = line.match(CATEGORY_NAME)
-      @add_date = m[1]
-      @last_mod = m[2]
-      @name = m[3]
-    end
-  end
-
-  class Item
-    attr_reader :name, :href, :categories
-
-    USELESS_SHARP = /\#.*$/
-
-    def initialize(line, categories)
-      m = line.match(ITEM_NAME)
-      @href = m[1].gsub(USELESS_SHARP, '')
-      @add_date = m[2]
-      @name = m[3]
-      @categories = categories
-    end
-
-    def inspect
-      @inspect ||= "/#{categories.join('/')}> #{name}"
-    end
-
-    def to_s
-      inspect
-    end
-
-    def host
-      @host ||= (URI.parse(href).host rescue 'unknown')
-    end
-  end
-
-  class Hostname
-    attr_reader :name, :items
-
-    def initialize(name)
-      @name = name
-      @items = []
-    end
-
-    def add(item)
-      @items << item
-    end
-
-    def inspect
-      @name
-    end
-
-    def to
-      inspect
-    end
-  end
-
-  class Grouping
-    Group = Struct.new(:name, :items)
-
-    def initialize(enumark, key, &post)
-      @lock = Mutex.new
-      @collection = nil
-
-      @enumark = enumark
-      @key = key
-      @post = post
-    end
-
-    def each(&block)
-      unless @collection
-        @lock.synchronize do
-          @collection = @enumark.group_by(&@key)
-          @collection = @post.call(@collection) if @post
-          @collection = @collection.map{ |k, items| Group.new(k, items) }
-        end
-      end
-
-      @collection.each(&block)
-    end
-  end
-
-  def initialize(file)
+  def initialize(file, items: nil)
     @file = file
     @lock = Mutex.new
-    @read = false
-    @items = []
+    @items = items
 
     @hosts = Grouping.new(self, :host)
     @dup_titles = Grouping.new(self, :name){ |groups| groups.select{ |_, items| items.count > 1 } }
     @dup_hrefs = Grouping.new(self, :href){ |groups| groups.select{ |_, items| items.count > 1 } }
+    @cates = Grouping.new(self, :categories_str)
   end
 
   def each(&block)
     read_all_lines
+    sort_by_add_date!
     @items.each(&block)
   end
 
@@ -125,32 +39,55 @@ class Enumark
     @dup_hrefs.each(&block)
   end
 
+  def each_category(&block)
+    @cates.each(&block)
+  end
+
+  [:+ ,:-, :&, :|].each do |op|
+    class_eval <<-EOM
+      def #{op}(another)
+        new_items = self.to_a #{op} another.to_a
+        Enumark.new(nil, items: new_items)
+      end
+    EOM
+  end
+
   private
 
   def read_all_lines
-    return if @read
+    return if @items
 
     @lock.synchronize do
-      next if @read
-
-      _read_all_lines
-      @read = true
+      _read_all_lines unless @items
     end
   end
 
   def _read_all_lines
+    dump_date = Pathname.new(@file).basename('.html').to_s
     categories = []
+    @items = []
 
     File.new(@file).each do |line|
       case line
-      when ITEM_PREFIX
-        item = Item.new(line, categories.dup)
+      when Item::PREFIX
+        item = Item.new(dump_date, line, categories.dup)
         @items.push(item)
-      when CATEGORY_START
+      when Category::START
         categories.push(Category.new(line))
-      when CATEGORY_END
+      when Category::ENDIND
         categories.pop
       end
     end
   end
+
+  def sort_by_add_date!
+    return if @sorted
+
+    @lock.synchronize do
+      next if @sorted
+
+      @items.sort!{ |i1, i2| i2.add_date <=> i1.add_date }
+      @sorted = true
+    end
+  end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: enumark
 version: !ruby/object:Gem::Version
-  version: 0.1.2
+  version: 1.0.0
 platform: ruby
 authors:
 - ken
 autorequire: 
 bindir: exe
 cert_chain: []
-date: 2021-02-13 00:00:00.000000000 Z
+date: 2022-03-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: pry
@@ -32,6 +32,7 @@ extensions: []
 extra_rdoc_files: []
 files:
 - ".gitignore"
+- ".ruby-version"
 - CODE_OF_CONDUCT.md
 - Gemfile
 - LICENSE.txt
@@ -41,6 +42,11 @@ files:
 - bin/setup
 - enumark.gemspec
 - lib/enumark.rb
+- lib/enumark/category.rb
+- lib/enumark/config.rb
+- lib/enumark/dir.rb
+- lib/enumark/grouping.rb
+- lib/enumark/item.rb
 - lib/enumark/version.rb
 homepage: https://github.com/turnon/enumark
 licenses:
@@ -54,14 +60,14 @@ required_ruby_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
-      version: 2.4.0
+      version: 3.1.0
 required_rubygems_version: !ruby/object:Gem::Requirement
   requirements:
   - - ">="
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.2.3
+rubygems_version: 3.3.7
 signing_key: 
 specification_version: 4
 summary: Enumerate chrome bookmark dump file