huginn_instagram_agent 0.4.4

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: 645f7e41f1ac78754c549774a4fda0aaee43e8ab8de1e549fba9dc16dde64ebf
+  data.tar.gz: abcb3c6662524c4d74b3d7a3e8a5e627a64c794f89f44547d9223934b385e770
+SHA512:
+  metadata.gz: 4d52637e2616dccb2f056c7a980103b1a5311d769273cc5abf3223d826fb5fa88cedd02608d470568577f69abbd6c7f002d476018f7ad2eb2cbe254a3bea7c8f
+  data.tar.gz: e6287bf0fca1623de4147a10114fefbdadbf3ca6a17d6663bfaec5c8f4d3a8c2607752741effd116115e687ad760d6204f8d71c5f855c1390b0a9845fd1d8b90

data/lib/huginn_instagram_agent/instagram_agent.rb

@@ -0,0 +1,152 @@
+module Agents
+  class InstagramAgent < Agent
+    can_dry_run!
+
+    default_schedule 'every_1h'
+
+    description <<-MD
+      Monitor public Instagram accounts and creates an event for each post.
+
+      It can be scheduled to hit Instagram as much as you want but will obey
+        the `wait_between_refresh` for each account to avoid being banned.
+        If set to `0` it will refresh all accounts at every run.
+
+      Links generally expire after 24 hours but this agent will try to keep the
+        corresponding events updated so they can be used in a feed.
+    MD
+
+
+    def default_options
+      {
+        :wait_between_refresh => 86400,
+        :accounts_to_monitor => []
+      }
+    end
+
+
+    def validate_options
+      options['wait_between_refresh'] ||= 86400
+      errors.add(:base, "`wait_between_refresh` must be an integer >=0") unless (options['wait_between_refresh'].to_i >= 0)
+
+      errors.add(:base, "`accounts_to_monitor` must be an array of strings") unless options['accounts_to_monitor'].is_a?(Array)
+      options['accounts_to_monitor'].each{|v| v.sub!(/^@+/,'')}
+    end
+
+
+    def working?
+      memory['error'] != true
+    end
+
+
+    def check
+      memory['error'] = nil
+
+      accounts_to_refresh.each do |account|
+        remember_fetching(account)
+
+        posts = get_posts(account) or next
+
+        posts.each do |post|
+          if seen_before?(post)
+            update_existing_event(post)
+          else
+            create_event :payload => post
+          end
+        end
+      end
+
+    end
+
+
+    def get_posts(account)
+      url = "https://www.instagram.com/#{account}/?__a=1&__d=di"
+
+      response = HTTParty.get(url,
+        :headers => {
+          'User-Agent' => 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/83.0.4103.97 Safari/537.36'
+        }
+      )
+
+      unless response.success?
+        error("[#{account}] Could not fetch #{url} - error #{response.code} | headers #{response.headers}")
+        memory['error'] = true
+        return nil
+      end
+
+      json = response.parsed_response
+
+      unless json
+        error("[#{account}] Could not extract JSON from #{url} - raw #{response.parsed_response} | headers #{response.headers}")
+        memory['error'] = true
+        return nil
+      end
+
+      posts = extract_posts(json)
+
+      unless posts.any?
+        error("[#{account}] Could not find any posts, strange - raw #{response.parsed_response} | headers #{response.headers} | json #{json}")
+        memory['error'] = true
+        return nil
+      end
+
+      return Array(posts).compact
+    end
+
+    def extract_posts(json)
+      json['graphql']['user']['edge_owner_to_timeline_media']['edges'].map do |edge|
+        edge['node']
+      end
+
+    rescue => e
+      return []
+    end
+
+
+    private
+
+
+    def instagramid_to_eventid
+      @instagramid_to_eventid ||= events.all.map{|e| [e.payload['id'], e.id]}.to_h
+    end
+
+
+    def seen_before?(post)
+      instagramid_to_eventid.has_key?(post['id'])
+    end
+
+
+    def update_existing_event(post)
+      event_id = instagramid_to_eventid.fetch(post['id'])
+      event = events.find(event_id)
+      event.payload = post
+      create_event(event)
+    end
+
+
+    def all_accounts
+      interpolated['accounts_to_monitor']
+    end
+
+
+    def stale_accounts(refresh_every)
+      all_accounts.select do |account|
+        last_fetched_at = memory.dig('last_fetched_at', account)
+        last_fetched_at.nil? || last_fetched_at < refresh_every.seconds.ago.to_i
+      end
+    end
+
+
+    def accounts_to_refresh
+      refresh_every = interpolated['wait_between_refresh'].to_i
+
+      return Array(refresh_every ? stale_accounts(refresh_every).sample : all_accounts)
+    end
+
+
+    def remember_fetching(account)
+      memory['last_fetched_at'] ||= {}
+      memory['last_fetched_at'][account] = Time.now.to_i
+    end
+
+  end
+end

data/lib/huginn_instagram_agent.rb

@@ -0,0 +1,4 @@
+require 'huginn_agent'
+
+#HuginnAgent.load 'huginn_instagram_agent/concerns/my_agent_concern'
+HuginnAgent.register 'huginn_instagram_agent/instagram_agent'

data/spec/instagram_agent_spec.rb

@@ -0,0 +1,13 @@
+require 'rails_helper'
+require 'huginn_agent/spec_helper'
+
+describe Agents::InstagramAgent do
+  before(:each) do
+    @valid_options = Agents::InstagramAgent.new.default_options
+    @checker = Agents::InstagramAgent.new(:name => "InstagramAgent", :options => @valid_options)
+    @checker.user = users(:bob)
+    @checker.save!
+  end
+
+  pending "add specs here"
+end

metadata

@@ -0,0 +1,105 @@
+--- !ruby/object:Gem::Specification
+name: huginn_instagram_agent
+version: !ruby/object:Gem::Version
+  version: 0.4.4
+platform: ruby
+authors:
+- Alessio Signorini
+- Víctor A. Rodríguez
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2022-12-19 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.7'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - "~>"
+      - !ruby/object:Gem::Version
+        version: '1.7'
+- !ruby/object:Gem::Dependency
+  name: rake
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '12.3'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '12.3'
+- !ruby/object:Gem::Dependency
+  name: huginn_agent
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0.6'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0.6'
+- !ruby/object:Gem::Dependency
+  name: httparty
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0.7'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0.7'
+description: 
+email:
+- alessio@signorini.us
+- victor@bit-man.guru
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- lib/huginn_instagram_agent.rb
+- lib/huginn_instagram_agent/instagram_agent.rb
+- spec/instagram_agent_spec.rb
+homepage: https://github.com/alessio-signorini/huginn-instagram-agent
+licenses:
+- MIT
+metadata: {}
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ">="
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubygems_version: 3.1.4
+signing_key: 
+specification_version: 4
+summary: Huginn Agent that monitors public Instagram accounts
+test_files:
+- spec/instagram_agent_spec.rb