Initial commit

.gitignore

@@ -0,0 +1,3 @@
+config.json
+tmp/
+dump.db

Gemfile

@@ -0,0 +1,6 @@
+# frozen_string_literal: true
+
+source "https://rubygems.org"
+
+gem "sqlite3"
+gem "mini_sql"

Gemfile.lock

@@ -0,0 +1,15 @@
+GEM
+  remote: https://rubygems.org/
+  specs:
+    mini_sql (1.4.0)
+    sqlite3 (1.6.2-x86_64-linux)
+
+PLATFORMS
+  x86_64-linux
+
+DEPENDENCIES
+  mini_sql
+  sqlite3
+
+BUNDLED WITH
+   2.4.7

README.md

@@ -0,0 +1,31 @@
+### Public Data Dump for you forum
+
+This repo attempts to establish a pattern for a public data dump. It includes 2 data explorer queries you can use to export all your public data.
+
+Public data is defined as forum topics and posts that anonymous users can access.
+
+### How to use this?
+
+First you need to define 2 queries using data explorer:
+
+1. Topic query: [here](topic_query.sql)
+2. Post query: [here](post_query.sql)
+
+Once defined note the data explorer query ids as specified in the URL
+
+Next, define an API key with rights to run the 2 queries.
+
+### config.json
+
+Create a [config.json](config.json.sample) specifying the domain of your discourse site, api key and data explorer query ids.
+
+### Importing the site into Sqlite
+
+The first phase of the import is importing the site into a sqlite3 db. This intermediary db stores all the content.
+
+Run: `ruby download_topics.rb`
+
+### Importing the Sqlite db into Discourse
+
+1. Start with a blank DB
+2. ... (in progress)

config.json.sample

@@ -0,0 +1,7 @@
+{
+  "api_username": "USER",
+  "api_key": "API_KEY",
+  "topics_query_id": "QUERY_ID",
+  "posts_query_id": "QUERY_ID",
+  "domain": "YOUR_DISCOURSE.com"
+}

download_topics.rb

@@ -0,0 +1,168 @@
+require "json"
+require "net/http"
+require "uri"
+require "sqlite3"
+require "mini_sql"
+require "cgi"
+
+begin
+  config = JSON.parse(File.read("config.json"))
+rescue StandardError
+  puts "Please create a file called .creds with your API KEY and USERNAME"
+end
+
+# Replace these values with your Discourse instance details
+DISCOURSE_DOMAIN = config["domain"]
+API_KEY = config["api_key"]
+API_USERNAME = config["api_username"]
+TOPIC_QUERY_ID = config["topics_query_id"]
+POST_QUERY_ID = config["posts_query_id"]
+
+sqlite_conn = SQLite3::Database.new("dump.db")
+conn = MiniSql::Connection.get(sqlite_conn)
+
+def run_report(query_id:, min_id: 0, limit:)
+  params = CGI.escape({ min_id: min_id.to_s }.to_json)
+
+  uri =
+    URI(
+      "https://#{DISCOURSE_DOMAIN}/admin/plugins/explorer/queries/#{query_id}/run?limit=#{limit}&params=#{params}"
+    )
+  http = Net::HTTP.new(uri.host, uri.port)
+  http.use_ssl = true
+
+  request = Net::HTTP::Post.new(uri.request_uri)
+  request["Content-Type"] = "application/json"
+  request["Api-Key"] = API_KEY
+  request["Api-Username"] = API_USERNAME
+
+  response = http.request(request)
+  JSON.parse(response.body)
+end
+
+def create_schema(conn)
+  conn.exec <<-SQL
+    CREATE TABLE IF NOT EXISTS topics (
+      id INTEGER PRIMARY KEY,
+      category,
+      title,
+      created_at,
+      user_id,
+      tags
+    )
+  SQL
+
+  conn.exec <<-SQL
+    CREATE TABLE IF NOT EXISTS users(
+      id INTEGER PRIMARY KEY,
+      username,
+      name
+    )
+  SQL
+
+  conn.exec <<-SQL
+    CREATE TABLE IF NOT EXISTS posts(
+      id INTEGER PRIMARY KEY,
+      raw,
+      post_number,
+      topic_id,
+      user_id,
+      created_at
+    )
+  SQL
+end
+
+def load_posts(conn, rows)
+  highest_id = 0
+  posts_loaded = 0
+
+  conn.exec "BEGIN TRANSACTION"
+
+  rows.each do |row|
+    conn.exec <<~SQL, *row
+    INSERT OR IGNORE INTO posts (id, raw, post_number, topic_id, user_id, created_at)
+    VALUES (?, ?, ?, ?, ?, ?)
+  SQL
+    posts_loaded += 1
+    highest_id = row[0] if row[0] > highest_id
+  end
+
+  conn.exec "COMMIT TRANSACTION"
+
+  { highest_id: highest_id, posts_loaded: posts_loaded }
+end
+
+def load_topics(conn, rows)
+  highest_id = 0
+  topics_loaded = 0
+
+  conn.exec "BEGIN TRANSACTION"
+
+  rows.each do |row|
+    conn.exec <<~SQL, *row
+    INSERT OR IGNORE INTO topics (id, category, title, created_at, user_id, tags)
+    VALUES (?, ?, ?, ?, ?, ?)
+  SQL
+    topics_loaded += 1
+    highest_id = row[0] if row[0] > highest_id
+  end
+
+  conn.exec "COMMIT TRANSACTION"
+
+  { highest_id: highest_id, topics_loaded: topics_loaded }
+end
+
+def load_users(conn, rows)
+  conn.exec "BEGIN TRANSACTION"
+  loaded = 0
+
+  rows.each do |row|
+    conn.exec <<~SQL, *row
+    INSERT OR IGNORE INTO users(id, username, name)
+    VALUES (?, ?, ?)
+  SQL
+    loaded += 1
+  end
+
+  conn.exec "COMMIT TRANSACTION"
+  loaded
+end
+
+def load_users_from_json(conn, json)
+  users = json.dig("relations", "user")
+  if users
+    users = users.map { |user| [user["id"], user["username"], user["name"]] }
+    loaded = load_users(conn, users)
+    puts "Loaded #{loaded} users"
+  end
+end
+
+create_schema(conn)
+
+min_id = 0
+while true
+  response_data =
+    run_report(query_id: TOPIC_QUERY_ID, min_id: min_id, limit: 10_000)
+
+  load_users_from_json(conn, response_data)
+
+  result = load_topics(conn, response_data["rows"])
+  puts "Loaded #{result[:topics_loaded]} topics (highest id is #{result[:highest_id]})"
+
+  min_id = result[:highest_id]
+  break if result[:topics_loaded] == 0
+end
+
+min_id = 0
+while true
+  response_data =
+    run_report(query_id: POST_QUERY_ID, min_id: min_id, limit: 10_000)
+
+  load_users_from_json(conn, response_data)
+
+  result = load_posts(conn, response_data["rows"])
+  puts "Loaded #{result[:posts_loaded]} posts (highest id is #{result[:highest_id]})"
+
+  min_id = result[:highest_id]
+  break if result[:posts_loaded] == 0
+end

import_db.rb

@@ -0,0 +1,28 @@
+require "sqlite3"
+require "mini_sql"
+
+sqlite_conn = SQLite3::Database.new("dump.db")
+conn = MiniSql::Connection.get(sqlite_conn)
+
+Dir.chdir("/home/sam/Source/discourse")
+require "/home/sam/Source/discourse/config/environment"
+
+puts "Importing users..."
+
+created = 0
+conn
+  .query("SELECT * FROM users")
+  .each do |row|
+    if !User.exists?(row.id)
+      User.create(
+        id: row.id,
+        username: row.username,
+        name: row.username,
+        password: SecureRandom.hex,
+        email: "#{SecureRandom.hex}@email.com"
+      )
+    end
+    print "."
+    created += 1
+    puts "#{created} users created" if created % 500 == 0
+  end

post_query.sql

@@ -0,0 +1,21 @@
+-- [params]
+-- int :min_id = 0
+
+SELECT
+   p.id,
+   p.raw,
+   p.post_number,
+   p.topic_id,
+   p.user_id,
+   p.created_at
+FROM topics t
+JOIN posts p ON p.topic_id = t.id
+JOIN categories c ON c.id = t.category_id
+WHERE NOT c.read_restricted 
+  AND t.deleted_at IS NULL
+  AND p.deleted_at IS NULL
+  AND p.post_type = 1
+  AND NOT p.hidden
+AND p.id > :min_id
+ORDER BY p.id ASC
+

topic_query.sql

@@ -0,0 +1,15 @@
+-- [params]
+-- int :min_id = 0
+
+SELECT
+    t.id,
+    c.name,
+    t.title,
+    t.created_at,
+    t.user_id,
+    (SELECT STRING_AGG(tag.name, ', ') FROM topic_tags tt JOIN tags tag ON tag.id = tt.tag_id WHERE tt.topic_id = t.id) AS all_tags
+FROM topics t
+JOIN categories c ON c.id = t.category_id
+WHERE NOT c.read_restricted AND t.deleted_at IS NULL
+AND t.id > :min_id
+ORDER BY t.id ASC