Hive Developers logo

Hive Developer Portal

Get Account Replies

Fetching the replies written to a particular account.

Full, runnable src of Get Account Replies can be downloaded as part of: tutorials/ruby (or download just this tutorial: devportal-master-tutorials-ruby-08_get_account_replies.zip).

Historically, applications that wanted to retrieve replies written to a particular account would use get_state. But this method has been scheduled for deprecation. So we’ll use a more supported approach in this tutorial using get_account_history.

Also see:

Sections

  1. Making the api call - Requesting account history
    1. Example api call - make the call in code
    2. Example api call using script - using our tutorial script
    3. Example Output - output from a successful call
  2. Comment Fields - Getting more detail than provided by account history.
  3. To Run - Running the example.

Making the api call

get_account_replies.rb

To request the latest replies to a particular author, we can use the get_account_history method:

api = Radiator::Api.new

options = []
options << account_name
options << -1 # start
options << 1000 # limit

# This is optional, we can mask out all operations other than comment_operation.
operation_mask = 0x02 # comment_operation
options << (operation_mask & 0xFFFFFFFF) # operation_filter_low
options << ((operation_mask & 0xFFFFFFFF00000000) >> 32) # operation_filter_high

api.get_account_history(*options) do |history|
  history.each do |index, item|
    type, op = item.op
    
    next unless type == 'comment'
    next if op.parent_author.empty? # skip posts
    next unless op.parent_author == account_name # skip comments by account

    # .
    # ... your code here
    # .
  end
end

Notice, the above example request up to 1,000 operations from history, starting from the oldest. From these results, we iterate on each item in history to locate a) type of comment, and b) parent_author that match the account_name.

This example also shows how to mask out all but comment_operation. This is optional, but by providing this mask, the api response will only include posts and replies, reducing the bandwidth required to execute this api call.

Example api call

If we want to get the replies to user @lordvader …

api.get_account_history("lordvader") do |content| ...

Example api call using script

And to do the same with our tutorial script

ruby get_account_replies.rb lordvader

Example Output

From the example we get the following output from our script:

Reply by @hivebuzz in discussion: "Join the 501st Legion and Rule Drug Wars!"
	body_length: 1103 (172 words)
	replied at: 2020-06-06T22:36:33
	net_votes: 0
	https://hive.blog/@hivebuzz/hivebuzz-notify-lordvader-20200606t223635000z
Reply by @obrisgold1 in discussion: "Join the 501st Legion and Rule Drug Wars!"
	body_length: 51 (10 words)
	replied at: 2020-07-21T14:25:00
	net_votes: 0
	https://hive.blog/@obrisgold1/re-lordvader-qdtpdo

Comment fields

Replies in the results of get_account_history will only return the following fields:

In our example script, we want more detail than this, so for every comment, we call get_content to retrieve more detail. For a full explanation of the results provided by get_content, please refer to the tutorial: Get Post Details

Final code:

require 'rubygems'
require 'bundler/setup'

Bundler.require

account_name = ARGV[0]
api = Radiator::Api.new

options = []
options << account_name
options << -1 # start
options << 1000 # limit 

# This is optional, we can mask out all operations other than comment_operation.
operation_mask = 0x02 # comment_operation
options << (operation_mask & 0xFFFFFFFF) # operation_filter_low 
options << ((operation_mask & 0xFFFFFFFF00000000) >> 32) # operation_filter_high 

api.get_account_history(*options) do |history|
  history.each do |index, item|
    type, op = item.op
    
    next unless type == 'comment'
    next if op.parent_author.empty? # skip posts
    next unless op.parent_author == account_name # skip comments by account
    
    url = "https://hive.blog/@#{op.author}/#{op.permlink}"
    api.get_content(op.author, op.permlink) do |reply|
      puts "Reply by @#{op.author} in discussion: \"#{reply.root_title}\""
      
      puts "\tbody_length: #{reply.body.size} (#{reply.body.split(/\W+/).size} words)"
      
      # The date and time this reply was created.
      print "\treplied at: #{reply.created}"
      
      if reply.last_update != reply.created
        # The date and time of the last update to this reply.
        print ", updated at: #{reply.last_update}"
      end
      
      if reply.last_update != reply.created
        # The last time this reply was "touched" by voting or reply.
        print ", active at: #{reply.active}"
      end
      
      print "\n"
      
      # Net positive votes
      puts "\tnet_votes: #{reply.net_votes}"
      
      # Link directly to reply.
      puts "\t#{url}"
    end
  end
end

To Run

First, set up your workstation using the steps provided in Getting Started. Then you can create and execute the script (or clone from this repository):

git clone https://gitlab.syncad.com/hive/devportal.git
cd devportal/tutorials/ruby/08_get_account_replies
bundle install
ruby get_account_replies.rb <account-name>