Add "ignore_schemas" option

README.md

@@ -40,6 +40,8 @@ end
 
 ## Other options
 
+### indexes_after_tables
+
 You can optionally have indexes following the respective tables setting `indexes_after_tables`:
 
 ```ruby
@@ -61,6 +63,8 @@ CREATE INDEX index_users_on_tentant_id ON public.users USING btree (tenant_id);
 CREATE UNIQUE INDEX index_users_on_email ON public.users USING btree (email);
 ```
 
+### order_column_definitions
+
 To enable sorting the table column definitions alphabetically, discarding the actual order provided by `pg_dump`, set `order_column_definitions`:
 
 ```ruby
@@ -69,6 +73,8 @@ Rails.application.configure do
 end
 ```
 
+### order_schema_migrations_values
+
 You can have the schema_migrations values reorganized to prevent merge conflicts by setting `order_schema_migrations_values`:
 
 ```ruby
@@ -87,14 +93,26 @@ INSERT INTO "schema_migrations" (version) VALUES
 ;
 ```
 
+### keep_extensions
+
 By default the gem will remove [some extensions](https://github.com/ghiculescu/activerecord-clean-db-structure/blob/c9551391476a5e7a08ff314501af89baddcf669a/lib/activerecord-clean-db-structure/clean_dump.rb#L24) that typically aren't needed in structure dumps. You can choose to keep all, or just some, of those extensions:
 
 ```ruby
 Rails.application.configure do
   config.activerecord_clean_db_structure.keep_extensions = :all
 
   # This does the same thing as :all. You can choose which optional extensions to keep.
-  config.activerecord_clean_db_structure.keep_extensions = ["pg_stat_statements", "pg_buffercache"]
+  config.activerecord_clean_db_structure.keep_extensions = ['pg_stat_statements', 'pg_buffercache']
+end
+```
+
+### ignore_schmeas
+
+You can ignore specific schemas, for example when dumping from a database copy that is integrated with pganalyze and has helper functions:
+
+```ruby
+Rails.application.configure do
+  config.activerecord_clean_db_structure.ignore_schemas = ['pganalyze']
 end
 ```
 

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

@@ -10,8 +10,10 @@ def initialize(dump, options = {})
     def run
       clean_partition_tables # Must be first because it makes assumptions about string format
       clean
+      clean_ignored_schemas
       clean_inherited_tables
       clean_options
+      clean_schema_comments
     end
 
     def clean
@@ -44,7 +46,7 @@ def clean
       dump.gsub!(/^COMMENT ON EXTENSION .*/, '')
 
       # Remove useless, version-specific parts of comments
-      dump.gsub!(/^-- (.*); Schema: ([\w\.]+|-); Owner: -.*/, '-- \1')
+      dump.gsub!(/^-- (.*); Owner: -.*/, '-- \1')
 
       # Remove useless comment lines
       dump.gsub!(/^--$/, '')
@@ -61,27 +63,36 @@ def clean
         dump.gsub!(/^ALTER SEQUENCE [\w\.]+_id_seq OWNED BY .*;$/, '')
         dump.gsub!(/^ALTER TABLE ONLY [\w\.]+ ALTER COLUMN id SET DEFAULT nextval\('[\w\.]+_id_seq'::regclass\);$/, '')
         dump.gsub!(/^ALTER TABLE ONLY [\w\.]+\s+ADD CONSTRAINT [\w\.]+_pkey PRIMARY KEY \(id\);$/, '')
-        dump.gsub!(/^-- Name: (\w+\s+)?id; Type: DEFAULT$/, '')
+        dump.gsub!(/^-- Name: (\w+\s+)?id; Type: DEFAULT; Schema: \w+$/, '')
         dump.gsub!(/^-- .*_id_seq; Type: SEQUENCE.*/, '')
-        dump.gsub!(/^-- Name: (\w+\s+)?\w+_pkey; Type: CONSTRAINT$/, '')
+        dump.gsub!(/^-- Name: (\w+\s+)?\w+_pkey; Type: CONSTRAINT; Schema: \w+$/, '')
       end
     end
 
     def clean_inherited_tables
-      inherited_tables_regexp = /-- Name: ([\w\.]+); Type: TABLE\n\n[^;]+?INHERITS \([\w\.]+\);/m
+      inherited_tables_regexp = /-- Name: ([\w\.]+); Type: TABLE; Schema: \w+\n\n[^;]+?INHERITS \([\w\.]+\);/m
       inherited_tables = dump.scan(inherited_tables_regexp).map(&:first)
       dump.gsub!(inherited_tables_regexp, '')
       inherited_tables.each do |inherited_table|
         dump.gsub!(/ALTER TABLE ONLY ([\w_]+\.)?#{inherited_table}[^;]+;/, '')
 
         index_regexp = /CREATE INDEX ([\w_]+) ON ([\w_]+\.)?#{inherited_table}[^;]+;/m
         dump.scan(index_regexp).map(&:first).each do |inherited_table_index|
-          dump.gsub!("-- Name: #{inherited_table_index}; Type: INDEX", '')
+          dump.gsub!(/-- Name: #{inherited_table_index}; Type: INDEX; Schema: \w+/, '')
         end
         dump.gsub!(index_regexp, '')
       end
     end
 
+    def clean_ignored_schemas
+      return if options[:ignore_schemas].nil?
+      options[:ignore_schemas].each do |schema|
+        dump.gsub!(/-- Name: ([^;]+); Type: \w+; Schema: #{schema}\n\n(?:(?!(-- Name:|SET default_tablespace)).)*/m, '')
+        dump.gsub!(/-- Name: #{schema}; Type: SCHEMA; Schema: -\n\n(?:(?!(-- Name:|SET default_tablespace)).)*/m, '')
+        dump.gsub!(/-- Name: \w+; Type: EXTENSION; Schema: -\n\n\nCREATE EXTENSION IF NOT EXISTS \w+ WITH SCHEMA #{schema};(?:(?!(-- Name:|SET default_tablespace)).)*/m, '')
+      end
+    end
+
     def clean_partition_tables
       partitioned_tables = []
 
@@ -99,7 +110,7 @@ def clean_partition_tables
       names = []
       partitioned_tables.each { |table| names << table.split('.', 2)[1] }
       if names.any?
-        dump.scan(/CREATE (UNIQUE )?INDEX ([\w_]+) ON ([\w_]+\.)?(#{names.join('|')})[^;]+;/m).each { |m| names << m[1] }
+        dump.scan(/CREATE (UNIQUE )?INDEX (\w+) ON (\w+\.)?(#{names.join('|')})[^;]+;/m).each { |m| names << m[1] }
       end
       statements.reject! { |stmt| names.any? { |name| stmt.include?(name) } }
       @dump = statements.join("\n\n")
@@ -126,7 +137,7 @@ def clean_options
             .transform_values(&:join)
 
         dump.gsub!(/^CREATE( UNIQUE)? INDEX \w+ ON .+\n+/, '')
-        dump.gsub!(/^-- Name: \w+; Type: INDEX\n+/, '')
+        dump.gsub!(/^-- Name: \w+; Type: INDEX; Schema: \w+\n+/, '')
         indexes.each do |table, indexes_for_table|
           dump.gsub!(/^(CREATE TABLE #{table}\b(:?[^;\n]*\n)+\);*\n(?:.*);*)/) { $1 + "\n\n" + indexes_for_table }
         end
@@ -140,6 +151,11 @@ def clean_options
       end
     end
 
+    def clean_schema_comments
+      # Remove schema in comments for backwards compatibility
+      dump.gsub!(/^-- (.*); Schema: [\w-]+/, '-- \1')
+    end
+
     def order_column_definitions(source)
       result = []
 

test/clean_dump_test.rb

@@ -29,6 +29,11 @@ def test_partitions
     assert_cleans_dump "data/partitions.sql", "expectations/partitions.sql"
   end
 
+  def test_ignored_schemas
+    assert_cleans_dump "data/ignored_schemas.sql", "expectations/ignored_schemas_pganalyze.sql", ignore_schemas: ['pganalyze']
+    assert_cleans_dump "data/ignored_schemas.sql", "expectations/ignored_schemas_myschema.sql", ignore_schemas: ['myschema']
+  end
+
   private
 
   def assert_cleans_dump(input, output, props = {})