Handle tables with custom fillfactor + add tests + add CI

.github/workflows/ci.yml

@@ -0,0 +1,16 @@
+name: CI
+on:
+  pull_request:
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        ruby: ["3.1", "3.2"]
+    steps:
+      - uses: actions/checkout@v3
+      - uses: ruby/setup-ruby@v1
+        with:
+          ruby-version: ${{ matrix.ruby }}
+          bundler-cache: true
+      - run: bundle exec rake test

Gemfile.lock

@@ -7,30 +7,45 @@ PATH
 GEM
   remote: https://rubygems.org/
   specs:
-    activemodel (7.0.4.3)
-      activesupport (= 7.0.4.3)
-    activerecord (7.0.4.3)
-      activemodel (= 7.0.4.3)
-      activesupport (= 7.0.4.3)
-    activesupport (7.0.4.3)
-      concurrent-ruby (~> 1.0, >= 1.0.2)
+    activemodel (7.2.1)
+      activesupport (= 7.2.1)
+    activerecord (7.2.1)
+      activemodel (= 7.2.1)
+      activesupport (= 7.2.1)
+      timeout (>= 0.4.0)
+    activesupport (7.2.1)
+      base64
+      bigdecimal
+      concurrent-ruby (~> 1.0, >= 1.3.1)
+      connection_pool (>= 2.2.5)
+      drb
       i18n (>= 1.6, < 2)
+      logger (>= 1.4.2)
       minitest (>= 5.1)
-      tzinfo (~> 2.0)
-    concurrent-ruby (1.3.3)
+      securerandom (>= 0.3)
+      tzinfo (~> 2.0, >= 2.0.5)
+    base64 (0.2.0)
+    bigdecimal (3.1.8)
+    concurrent-ruby (1.3.4)
+    connection_pool (2.4.1)
+    drb (2.2.1)
     i18n (1.14.5)
       concurrent-ruby (~> 1.0)
-    minitest (5.24.1)
-    rake (13.0.6)
+    logger (1.6.1)
+    minitest (5.25.1)
+    rake (13.2.1)
+    securerandom (0.3.1)
+    timeout (0.4.1)
     tzinfo (2.0.6)
       concurrent-ruby (~> 1.0)
 
 PLATFORMS
+  arm64-darwin-22
   ruby
 
 DEPENDENCIES
   activerecord-clean-db-structure!
   rake (~> 13)
 
 BUNDLED WITH
-   1.17.3
+   2.5.18

Rakefile

@@ -1,3 +1,12 @@
 require 'rubygems'
 require 'bundler/setup'
 require 'bundler/gem_tasks'
+require "rake/testtask"
+
+Rake::TestTask.new(:test) do |t|
+  t.libs << "test"
+  t.libs << "lib"
+  t.test_files = FileList["test/**/*_test.rb"]
+end
+
+task default: %i[test]

lib/activerecord-clean-db-structure/clean_dump.rb

@@ -84,7 +84,7 @@ def run
       partitioned_tables += dump.scan(partitioned_tables_regexp2).map(&:first)
 
       partitioned_tables.each do |partitioned_table|
-        partitioned_schema_name, partitioned_table_name_only = partitioned_table.split('.', 2)
+        _partitioned_schema_name, partitioned_table_name_only = partitioned_table.split('.', 2)
         dump.gsub!(/-- Name: #{partitioned_table_name_only}; Type: TABLE/, '')
         dump.gsub!(/CREATE TABLE #{partitioned_table} \([^;]+;/m, '')
         dump.gsub!(/ALTER TABLE ONLY ([\w_\.]+) ATTACH PARTITION #{partitioned_table}[^;]+;/m, '')
@@ -124,7 +124,7 @@ def run
         dump.gsub!(/^CREATE( UNIQUE)? INDEX \w+ ON .+\n+/, '')
         dump.gsub!(/^-- Name: \w+; Type: INDEX\n+/, '')
         indexes.each do |table, indexes_for_table|
-          dump.gsub!(/^(CREATE TABLE #{table}\b(:?[^;\n]*\n)+\);\n)/) { $1 + "\n" + indexes_for_table }
+          dump.gsub!(/^(CREATE TABLE #{table}\b(:?[^;\n]*\n)+\);*\n(?:.*);*)/) { $1 + "\n\n" + indexes_for_table }
         end
       end
 
@@ -151,7 +151,7 @@ def order_column_definitions(source)
           inside_table = true
           columns = []
           result << source_line
-        elsif source_line.start_with?(");")
+        elsif source_line.start_with?(")")
           if inside_table
             inside_table = false
             columns.sort_by!(&:first)

test/clean_dump_test.rb

@@ -0,0 +1,35 @@
+require "test_helper"
+
+class CleanDumpTest < Minitest::Test
+  def test_basic_case
+    assert_cleans_dump "expectations/default_props.sql", {}
+  end
+
+  def test_ignore_ids
+    assert_cleans_dump "expectations/ignore_ids.sql", { ignore_ids: true }
+  end
+
+  def test_order_column_definitions
+    assert_cleans_dump "expectations/order_column_definitions.sql", { order_column_definitions: true }
+  end
+
+  def test_order_schema_migrations_values
+    assert_cleans_dump "expectations/order_schema_migrations_values.sql", { order_schema_migrations_values: true }
+  end
+
+  def test_indexes_after_tables
+    assert_cleans_dump "expectations/indexes_after_tables.sql", { indexes_after_tables: true }
+  end
+
+  def test_keep_extensions_all
+    assert_cleans_dump "expectations/keep_extensions_all.sql", { keep_extensions: :all }
+  end
+
+  private
+
+  def assert_cleans_dump(expected_output_path, props)
+    cleaner = ActiveRecordCleanDbStructure::CleanDump.new(File.read(File.join(__dir__, "data/input.sql")), props)
+    cleaner.run
+    assert_equal File.read(File.join(__dir__, expected_output_path)), cleaner.dump
+  end
+end

test/data/input.sql

@@ -0,0 +1,162 @@
+SET statement_timeout = 0;
+SET lock_timeout = 0;
+SET idle_in_transaction_session_timeout = 0;
+SET client_encoding = 'UTF8';
+SET standard_conforming_strings = on;
+SELECT pg_catalog.set_config('search_path', '', false);
+SET check_function_bodies = false;
+SET xmloption = content;
+SET client_min_messages = warning;
+SET row_security = off;
+
+--
+-- Name: pg_stat_statements; Type: EXTENSION; Schema: -; Owner: -
+--
+
+CREATE EXTENSION IF NOT EXISTS pg_stat_statements WITH SCHEMA public;
+
+
+--
+-- Name: EXTENSION pg_stat_statements; Type: COMMENT; Schema: -; Owner: -
+--
+
+COMMENT ON EXTENSION pg_stat_statements IS 'track execution statistics of all SQL statements executed';
+
+
+SET default_tablespace = '';
+
+SET default_table_access_method = heap;
+
+--
+-- Name: ar_internal_metadata; Type: TABLE; Schema: public; Owner: -
+--
+
+CREATE TABLE public.ar_internal_metadata (
+    key character varying NOT NULL,
+    value character varying,
+    created_at timestamp(6) without time zone NOT NULL,
+    updated_at timestamp(6) without time zone NOT NULL
+);
+
+
+--
+-- Name: delayed_jobs; Type: TABLE; Schema: public; Owner: -
+--
+
+CREATE TABLE public.delayed_jobs (
+    id bigint NOT NULL,
+    priority integer DEFAULT 0,
+    attempts integer DEFAULT 0,
+    handler text,
+    last_error text,
+    run_at timestamp without time zone,
+    locked_at timestamp without time zone,
+    failed_at timestamp without time zone,
+    locked_by character varying(255),
+    queue character varying(255),
+    created_at timestamp without time zone NOT NULL,
+    updated_at timestamp without time zone NOT NULL,
+    metadata jsonb DEFAULT '{}'::jsonb,
+    whodunnit text
+)
+WITH (fillfactor='85');
+
+
+--
+-- Name: delayed_jobs_id_seq; Type: SEQUENCE; Schema: public; Owner: -
+--
+
+CREATE SEQUENCE public.delayed_jobs_id_seq
+    START WITH 1
+    INCREMENT BY 1
+    NO MINVALUE
+    NO MAXVALUE
+    CACHE 1;
+
+
+--
+-- Name: delayed_jobs_id_seq; Type: SEQUENCE OWNED BY; Schema: public; Owner: -
+--
+
+ALTER SEQUENCE public.delayed_jobs_id_seq OWNED BY public.delayed_jobs.id;
+
+
+--
+-- Name: schema_migrations; Type: TABLE; Schema: public; Owner: -
+--
+
+CREATE TABLE public.schema_migrations (
+    version character varying NOT NULL
+);
+
+
+--
+-- Name: delayed_jobs id; Type: DEFAULT; Schema: public; Owner: -
+--
+
+ALTER TABLE ONLY public.delayed_jobs ALTER COLUMN id SET DEFAULT nextval('public.delayed_jobs_id_seq'::regclass);
+
+
+--
+-- Name: ar_internal_metadata ar_internal_metadata_pkey; Type: CONSTRAINT; Schema: public; Owner: -
+--
+
+ALTER TABLE ONLY public.ar_internal_metadata
+    ADD CONSTRAINT ar_internal_metadata_pkey PRIMARY KEY (key);
+
+
+--
+-- Name: delayed_jobs delayed_jobs_pkey; Type: CONSTRAINT; Schema: public; Owner: -
+--
+
+ALTER TABLE ONLY public.delayed_jobs
+    ADD CONSTRAINT delayed_jobs_pkey PRIMARY KEY (id);
+
+
+--
+-- Name: schema_migrations schema_migrations_pkey; Type: CONSTRAINT; Schema: public; Owner: -
+--
+
+ALTER TABLE ONLY public.schema_migrations
+    ADD CONSTRAINT schema_migrations_pkey PRIMARY KEY (version);
+
+
+--
+-- Name: index_delayed_jobs_on_locked_by; Type: INDEX; Schema: public; Owner: -
+--
+
+CREATE INDEX index_delayed_jobs_on_locked_by ON public.delayed_jobs USING btree (locked_by);
+
+
+--
+-- Name: index_delayed_jobs_on_queue; Type: INDEX; Schema: public; Owner: -
+--
+
+CREATE INDEX index_delayed_jobs_on_queue ON public.delayed_jobs USING btree (queue);
+
+
+--
+-- Name: index_delayed_jobs_on_run_at; Type: INDEX; Schema: public; Owner: -
+--
+
+CREATE INDEX index_delayed_jobs_on_run_at ON public.delayed_jobs USING btree (run_at) WHERE (locked_at IS NULL);
+
+
+--
+-- PostgreSQL database dump complete
+--
+
+SET search_path TO "$user", public;
+
+INSERT INTO "schema_migrations" (version) VALUES
+('20240822225012'),
+('20240822224954'),
+('20240725043656'),
+('20240621041110'),
+('20240621020038'),
+('20220802204003'),
+('20211125055031'),
+('20211012054749'),
+('20210923052631'),
+('20210903003251');
+

test/expectations/default_props.sql

@@ -0,0 +1,78 @@
+SET statement_timeout = 0;
+SET lock_timeout = 0;
+SET client_encoding = 'UTF8';
+SET standard_conforming_strings = on;
+SELECT pg_catalog.set_config('search_path', '', false);
+SET check_function_bodies = false;
+SET client_min_messages = warning;
+
+SET default_tablespace = '';
+
+-- Name: ar_internal_metadata; Type: TABLE
+
+CREATE TABLE public.ar_internal_metadata (
+    key character varying NOT NULL,
+    value character varying,
+    created_at timestamp(6) without time zone NOT NULL,
+    updated_at timestamp(6) without time zone NOT NULL
+);
+
+-- Name: delayed_jobs; Type: TABLE
+
+CREATE TABLE public.delayed_jobs (
+    id BIGSERIAL PRIMARY KEY,
+    priority integer DEFAULT 0,
+    attempts integer DEFAULT 0,
+    handler text,
+    last_error text,
+    run_at timestamp without time zone,
+    locked_at timestamp without time zone,
+    failed_at timestamp without time zone,
+    locked_by character varying(255),
+    queue character varying(255),
+    created_at timestamp without time zone NOT NULL,
+    updated_at timestamp without time zone NOT NULL,
+    metadata jsonb DEFAULT '{}'::jsonb,
+    whodunnit text
+)
+WITH (fillfactor='85');
+
+-- Name: schema_migrations; Type: TABLE
+
+CREATE TABLE public.schema_migrations (
+    version character varying NOT NULL
+);
+
+ALTER TABLE ONLY public.ar_internal_metadata
+    ADD CONSTRAINT ar_internal_metadata_pkey PRIMARY KEY (key);
+
+ALTER TABLE ONLY public.schema_migrations
+    ADD CONSTRAINT schema_migrations_pkey PRIMARY KEY (version);
+
+-- Name: index_delayed_jobs_on_locked_by; Type: INDEX
+
+CREATE INDEX index_delayed_jobs_on_locked_by ON public.delayed_jobs USING btree (locked_by);
+
+-- Name: index_delayed_jobs_on_queue; Type: INDEX
+
+CREATE INDEX index_delayed_jobs_on_queue ON public.delayed_jobs USING btree (queue);
+
+-- Name: index_delayed_jobs_on_run_at; Type: INDEX
+
+CREATE INDEX index_delayed_jobs_on_run_at ON public.delayed_jobs USING btree (run_at) WHERE (locked_at IS NULL);
+
+-- PostgreSQL database dump complete
+
+SET search_path TO "$user", public;
+
+INSERT INTO "schema_migrations" (version) VALUES
+('20240822225012'),
+('20240822224954'),
+('20240725043656'),
+('20240621041110'),
+('20240621020038'),
+('20220802204003'),
+('20211125055031'),
+('20211012054749'),
+('20210923052631'),
+('20210903003251');