1 package DBIx::Class::Schema::Loader::Base;
5 use base qw/Class::Accessor::Grouped Class::C3::Componentised/;
8 use Carp::Clan qw/^DBIx::Class/;
9 use DBIx::Class::Schema::Loader::RelBuilder ();
10 use Data::Dump 'dump';
15 use Lingua::EN::Inflect::Number ();
16 use Lingua::EN::Inflect::Phrase ();
17 use String::ToIdentifier::EN ();
18 use String::ToIdentifier::EN::Unicode ();
21 use Class::Inspector ();
22 use Scalar::Util 'looks_like_number';
23 use DBIx::Class::Schema::Loader::Column;
24 use DBIx::Class::Schema::Loader::Utils qw/split_name dumper_squashed eval_package_without_redefine_warnings class_path slurp_file sigwarn_silencer firstidx uniq/;
25 use DBIx::Class::Schema::Loader::Optional::Dependencies ();
28 use Encode qw/encode decode/;
29 use List::Util qw/all any none/;
30 use File::Temp 'tempfile';
33 our $VERSION = '0.07047';
35 __PACKAGE__->mk_group_ro_accessors('simple', qw/
42 additional_base_classes
58 default_resultset_class
65 overwrite_modifications
93 __PACKAGE__->mk_group_accessors('simple', qw/
95 schema_version_to_dump
97 _upgrading_from_load_classes
98 _downgrading_to_load_classes
99 _rewriting_result_namespace
104 pod_comment_spillover_length
110 result_components_map
112 datetime_undef_if_invalid
113 _result_class_methods
115 filter_generated_code
119 moniker_part_separator
123 my $CURRENT_V = 'v7';
126 schema_components schema_base_class result_base_class
127 additional_base_classes left_base_classes additional_classes components
133 my $CRLF = "\x0d\x0a";
137 DBIx::Class::Schema::Loader::Base - Base DBIx::Class::Schema::Loader Implementation.
141 See L<DBIx::Class::Schema::Loader>.
145 This is the base class for the storage-specific C<DBIx::Class::Schema::*>
146 classes, and implements the common functionality between them.
148 =head1 CONSTRUCTOR OPTIONS
150 These constructor options are the base options for
151 L<DBIx::Class::Schema::Loader/loader_options>. Available constructor options are:
153 =head2 skip_relationships
155 Skip setting up relationships. The default is to attempt the loading
158 =head2 skip_load_external
160 Skip loading of other classes in @INC. The default is to merge all other classes
161 with the same name found in @INC into the schema file we are creating.
165 Static schemas (ones dumped to disk) will, by default, use the new-style
166 relationship names and singularized Results, unless you're overwriting an
167 existing dump made by an older version of L<DBIx::Class::Schema::Loader>, in
168 which case the backward compatible RelBuilder will be activated, and the
169 appropriate monikerization used.
175 will disable the backward-compatible RelBuilder and use
176 the new-style relationship names along with singularized Results, even when
177 overwriting a dump made with an earlier version.
179 The option also takes a hashref:
182 relationships => 'v8',
184 column_accessors => 'v8',
190 naming => { ALL => 'v8', force_ascii => 1 }
198 Set L</relationships>, L</monikers> and L</column_accessors> to the specified
203 How to name relationship accessors.
207 How to name Result classes.
209 =item column_accessors
211 How to name column accessors in Result classes.
215 For L</v8> mode and later, uses L<String::ToIdentifier::EN> instead of
216 L<String::ToIdentifier::EN::Unicode> to force monikers and other identifiers to
227 Latest style, whatever that happens to be.
231 Unsingularlized monikers, C<has_many> only relationships with no _id stripping.
235 Monikers singularized as whole words, C<might_have> relationships for FKs on
236 C<UNIQUE> constraints, C<_id> stripping for belongs_to relationships.
238 Some of the C<_id> stripping edge cases in C<0.05003> have been reverted for
243 All monikers and relationships are inflected using
244 L<Lingua::EN::Inflect::Phrase>, and there is more aggressive C<_id> stripping
245 from relationship names.
247 In general, there is very little difference between v5 and v6 schemas.
251 This mode is identical to C<v6> mode, except that monikerization of CamelCase
252 table names is also done better (but best in v8.)
254 CamelCase column names in case-preserving mode will also be handled better
255 for relationship name inflection (but best in v8.) See L</preserve_case>.
257 In this mode, CamelCase L</column_accessors> are normalized based on case
258 transition instead of just being lowercased, so C<FooId> becomes C<foo_id>.
264 The default mode is L</v7>, to get L</v8> mode, you have to specify it in
265 L</naming> explicitly until C<0.08> comes out.
267 L</monikers> and L</column_accessors> are created using
268 L<String::ToIdentifier::EN::Unicode> or L<String::ToIdentifier::EN> if
269 L</force_ascii> is set; this is only significant for names with non-C<\w>
270 characters such as C<.>.
272 CamelCase identifiers with words in all caps, e.g. C<VLANValidID> are supported
273 correctly in this mode.
275 For relationships, belongs_to accessors are made from column names by stripping
276 postfixes other than C<_id> as well, for example just C<Id>, C<_?ref>, C<_?cd>,
277 C<_?code> and C<_?num>, case insensitively.
281 For L</monikers>, this option does not inflect the table names but makes
282 monikers based on the actual name. For L</column_accessors> this option does
283 not normalize CamelCase column names to lowercase column accessors, but makes
284 accessors that are the same names as the columns (with any non-\w chars
285 replaced with underscores.)
289 For L</monikers>, singularizes the names using the most current inflector. This
290 is the same as setting the option to L</current>.
294 For L</monikers>, pluralizes the names, using the most current inflector.
298 Dynamic schemas will always default to the 0.04XXX relationship names and won't
299 singularize Results for backward compatibility, to activate the new RelBuilder
300 and singularization put this in your C<Schema.pm> file:
302 __PACKAGE__->naming('current');
304 Or if you prefer to use 0.07XXX features but insure that nothing breaks in the
305 next major version upgrade:
307 __PACKAGE__->naming('v7');
311 If true, will not print the usual C<Dumping manual schema ... Schema dump
312 completed.> messages. Does not affect warnings (except for warnings related to
313 L</really_erase_my_files>.)
317 If true, don't actually write out the generated files. This can only be
318 used with static schema generation.
322 By default POD will be generated for columns and relationships, using database
323 metadata for the text if available and supported.
325 Comment metadata can be stored in two ways.
327 The first is that you can create two tables named C<table_comments> and
328 C<column_comments> respectively. These tables must exist in the same database
329 and schema as the tables they describe. They both need to have columns named
330 C<table_name> and C<comment_text>. The second one needs to have a column named
331 C<column_name>. Then data stored in these tables will be used as a source of
332 metadata about tables and comments.
334 (If you wish you can change the name of these tables with the parameters
335 L</table_comments_table> and L</column_comments_table>.)
337 As a fallback you can use built-in commenting mechanisms. Currently this is
338 only supported for PostgreSQL, Oracle and MySQL. To create comments in
339 PostgreSQL you add statements of the form C<COMMENT ON TABLE some_table IS
340 '...'>, the same syntax is used in Oracle. To create comments in MySQL you add
341 C<COMMENT '...'> to the end of the column or table definition. Note that MySQL
342 restricts the length of comments, and also does not handle complex Unicode
345 Set this to C<0> to turn off all POD generation.
347 =head2 pod_comment_mode
349 Controls where table comments appear in the generated POD. Smaller table
350 comments are appended to the C<NAME> section of the documentation, and larger
351 ones are inserted into C<DESCRIPTION> instead. You can force a C<DESCRIPTION>
352 section to be generated with the comment always, only use C<NAME>, or choose
353 the length threshold at which the comment is forced into the description.
359 Use C<NAME> section only.
363 Force C<DESCRIPTION> always.
367 Use C<DESCRIPTION> if length > L</pod_comment_spillover_length>, this is the
372 =head2 pod_comment_spillover_length
374 When pod_comment_mode is set to C<auto>, this is the length of the comment at
375 which it will be forced into a separate description section.
379 =head2 table_comments_table
381 The table to look for comments about tables in. By default C<table_comments>.
382 See L</generate_pod> for details.
384 This must not be a fully qualified name, the table will be looked for in the
385 same database and schema as the table whose comment is being retrieved.
387 =head2 column_comments_table
389 The table to look for comments about columns in. By default C<column_comments>.
390 See L</generate_pod> for details.
392 This must not be a fully qualified name, the table will be looked for in the
393 same database and schema as the table/column whose comment is being retrieved.
395 =head2 relationship_attrs
397 Hashref of attributes to pass to each generated relationship, listed by type.
398 Also supports relationship type 'all', containing options to pass to all
399 generated relationships. Attributes set for more specific relationship types
400 override those set in 'all', and any attributes specified by this option
401 override the introspected attributes of the foreign key if any.
405 relationship_attrs => {
406 has_many => { cascade_delete => 1, cascade_copy => 1 },
407 might_have => { cascade_delete => 1, cascade_copy => 1 },
410 use this to turn L<DBIx::Class> cascades to on on your
411 L<has_many|DBIx::Class::Relationship/has_many> and
412 L<might_have|DBIx::Class::Relationship/might_have> relationships, they default
415 Can also be a coderef, for more precise control, in which case the coderef gets
416 this hash of parameters (as a list):
418 rel_name # the name of the relationship
419 rel_type # the type of the relationship: 'belongs_to', 'has_many' or 'might_have'
420 local_source # the DBIx::Class::ResultSource object for the source the rel is *from*
421 remote_source # the DBIx::Class::ResultSource object for the source the rel is *to*
422 local_table # the DBIx::Class::Schema::Loader::Table object for the table of the source the rel is from
423 local_cols # an arrayref of column names of columns used in the rel in the source it is from
424 remote_table # the DBIx::Class::Schema::Loader::Table object for the table of the source the rel is to
425 remote_cols # an arrayref of column names of columns used in the rel in the source it is to
426 attrs # the attributes that would be set
428 it should return the new hashref of attributes, or nothing for no changes.
432 relationship_attrs => sub {
435 say "the relationship name is: $p{rel_name}";
436 say "the relationship is a: $p{rel_type}";
437 say "the local class is: ", $p{local_source}->result_class;
438 say "the remote class is: ", $p{remote_source}->result_class;
439 say "the local table is: ", $p{local_table}->sql_name;
440 say "the rel columns in the local table are: ", (join ", ", @{$p{local_cols}});
441 say "the remote table is: ", $p{remote_table}->sql_name;
442 say "the rel columns in the remote table are: ", (join ", ", @{$p{remote_cols}});
444 if ($p{local_table} eq 'dogs' && @{$p{local_cols}} == 1 && $p{local_cols}[0] eq 'name') {
445 $p{attrs}{could_be_snoopy} = 1;
451 These are the default attributes:
462 on_delete => 'CASCADE',
463 on_update => 'CASCADE',
467 For L<belongs_to|DBIx::Class::Relationship/belongs_to> relationships, these
468 defaults are overridden by the attributes introspected from the foreign key in
469 the database, if this information is available (and the driver is capable of
472 This information overrides the defaults mentioned above, and is then itself
473 overridden by the user's L</relationship_attrs> for C<belongs_to> if any are
476 In general, for most databases, for a plain foreign key with no rules, the
477 values for a L<belongs_to|DBIx::Class::Relationship/belongs_to> relationship
480 on_delete => 'NO ACTION',
481 on_update => 'NO ACTION',
484 In the cases where an attribute is not supported by the DB, a value matching
485 the actual behavior is used, for example Oracle does not support C<ON UPDATE>
486 rules, so C<on_update> is set to C<NO ACTION>. This is done so that the
487 behavior of the schema is preserved when cross deploying to a different RDBMS
488 such as SQLite for testing.
490 In the cases where the DB does not support C<DEFERRABLE> foreign keys, the
491 value is set to C<1> if L<DBIx::Class> has a working C<<
492 $storage->with_deferred_fk_checks >>. This is done so that the same
493 L<DBIx::Class> code can be used, and cross deployed from and to such databases.
497 If set to true, each constructive L<DBIx::Class> statement the loader
498 decides to execute will be C<warn>-ed before execution.
502 Set the name of the schema to load (schema in the sense that your database
505 Can be set to an arrayref of schema names for multiple schemas, or the special
506 value C<%> for all schemas.
508 For MSSQL, Sybase ASE, and Informix can be set to a hashref of databases as
509 keys and arrays of owners as values, set to the value:
513 for all owners in all databases.
515 Name clashes resulting from the same table name in different databases/schemas
516 will be resolved automatically by prefixing the moniker with the database
519 To prefix/suffix all monikers with the database and/or schema, see
524 The database table names are represented by the
525 L<DBIx::Class::Schema::Loader::Table> class in the loader, the
526 L<DBIx::Class::Schema::Loader::Table::Sybase> class for Sybase ASE and
527 L<DBIx::Class::Schema::Loader::Table::Informix> for Informix.
529 Monikers are created normally based on just the
530 L<name|DBIx::Class::Schema::Loader::DBObject/name> property, corresponding to
531 the table name, but can consist of other parts of the fully qualified name of
534 The L</moniker_parts> option is an arrayref of methods on the table class
535 corresponding to parts of the fully qualified table name, defaulting to
536 C<['name']>, in the order those parts are used to create the moniker name.
537 The parts are joined together using L</moniker_part_separator>.
539 The C<'name'> entry B<must> be present.
541 Below is a table of supported databases and possible L</moniker_parts>.
545 =item * DB2, Firebird, mysql, Oracle, Pg, SQLAnywhere, SQLite, MS Access
549 =item * Informix, MSSQL, Sybase ASE
551 C<database>, C<schema>, C<name>
555 =head2 moniker_part_separator
557 String used to join L</moniker_parts> when creating the moniker.
558 Defaults to the empty string. Use C<::> to get a separate namespace per
559 database and/or schema.
563 Only load matching tables.
565 These can be specified either as a regex (preferably on the C<qr//>
566 form), or as an arrayref of arrayrefs. Regexes are matched against
567 the (unqualified) table name, while arrayrefs are matched according to
572 db_schema => [qw(some_schema other_schema)],
573 moniker_parts => [qw(schema name)],
575 [ qr/\Asome_schema\z/ => qr/\A(?:foo|bar)\z/ ],
576 [ qr/\Aother_schema\z/ => qr/\Abaz\z/ ],
579 In this case only the tables C<foo> and C<bar> in C<some_schema> and
580 C<baz> in C<other_schema> will be dumped.
584 Exclude matching tables.
586 The tables to exclude are specified in the same way as for the
587 L</constraint> option.
591 Overrides the default table name to moniker translation. Either
597 a nested hashref, which will be traversed according to L</moniker_parts>
601 moniker_parts => [qw(schema name)],
608 In which case the table C<bar> in the C<foo> schema would get the moniker
613 a hashref of unqualified table name keys and moniker values
617 a coderef that returns the moniker, which is called with the following
624 the L<DBIx::Class::Schema::Loader::Table> object for the table
628 the default moniker that DBIC would ordinarily give this table
632 a coderef that can be called with either of the hashref forms to get
633 the moniker mapped accordingly. This is useful if you need to handle
634 some monikers specially, but want to use the hashref form for the
641 If the hash entry does not exist, or the function returns a false
642 value, the code falls back to default behavior for that table name.
644 The default behavior is to split on case transition and non-alphanumeric
645 boundaries, singularize the resulting phrase, then join the titlecased words
648 Table Name | Moniker Name
649 ---------------------------------
651 luser_group | LuserGroup
652 luser-opts | LuserOpt
653 stations_visited | StationVisited
654 routeChange | RouteChange
656 =head2 moniker_part_map
658 Map for overriding the monikerization of individual L</moniker_parts>.
659 The keys are the moniker part to override, the value is either a
660 hashref or coderef for mapping the corresponding part of the
661 moniker. If a coderef is used, it gets called with the moniker part
662 and the hash key the code ref was found under.
666 moniker_part_map => {
667 schema => sub { ... },
670 Given the table C<foo.bar>, the code ref would be called with the
671 arguments C<foo> and C<schema>, plus a coderef similar to the one
672 described in L</moniker_map>.
674 L</moniker_map> takes precedence over this.
676 =head2 col_accessor_map
678 Same as moniker_map, but for column accessor names. The nested
679 hashref form is traversed according to L</moniker_parts>, with an
680 extra level at the bottom for the column name. If a coderef is
681 passed, the code is called with the following arguments:
687 the L<DBIx::Class::Schema::Loader::Column> object for the column
691 the default accessor name that DBICSL would ordinarily give this column
695 a hashref of this form:
698 table_class => name of the DBIC class we are building,
699 table_moniker => calculated moniker for this table (after moniker_map if present),
700 table => the DBIx::Class::Schema::Loader::Table object for the table,
701 full_table_name => schema-qualified name of the database table (RDBMS specific),
702 schema_class => name of the schema class we are building,
703 column_info => hashref of column info (data_type, is_nullable, etc),
708 a coderef that can be called with a hashref map
714 Similar in idea to moniker_map, but different in the details. It can be
715 a hashref or a code ref.
717 If it is a hashref, keys can be either the default relationship name, or the
718 moniker. The keys that are the default relationship name should map to the
719 name you want to change the relationship to. Keys that are monikers should map
720 to hashes mapping relationship names to their translation. You can do both at
721 once, and the more specific moniker version will be picked up first. So, for
722 instance, you could have
731 and relationships that would have been named C<bar> will now be named C<baz>
732 except that in the table whose moniker is C<Foo> it will be named C<blat>.
734 If it is a coderef, it will be passed a hashref of this form:
737 name => default relationship name,
738 type => the relationship type eg: C<has_many>,
739 local_class => name of the DBIC class we are building,
740 local_moniker => moniker of the DBIC class we are building,
741 local_columns => columns in this table in the relationship,
742 remote_class => name of the DBIC class we are related to,
743 remote_moniker => moniker of the DBIC class we are related to,
744 remote_columns => columns in the other table in the relationship,
745 # for type => "many_to_many" only:
746 link_class => name of the DBIC class for the link table,
747 link_moniker => moniker of the DBIC class for the link table,
748 link_rel_name => name of the relationship to the link table,
751 In addition it is passed a coderef that can be called with a hashref map.
753 DBICSL will try to use the value returned as the relationship name.
755 =head2 inflect_plural
757 Just like L</moniker_map> above (can be hash/code-ref, falls back to default
758 if hash key does not exist or coderef returns false), but acts as a map
759 for pluralizing relationship names. The default behavior is to utilize
760 L<Lingua::EN::Inflect::Phrase/to_PL>.
762 =head2 inflect_singular
764 As L</inflect_plural> above, but for singularizing relationship names.
765 Default behavior is to utilize L<Lingua::EN::Inflect::Phrase/to_S>.
767 =head2 schema_base_class
769 Base class for your schema classes. Defaults to 'DBIx::Class::Schema'.
771 =head2 schema_components
773 List of components to load into the Schema class.
775 =head2 result_base_class
777 Base class for your table classes (aka result classes). Defaults to
780 =head2 additional_base_classes
782 List of additional base classes all of your table classes will use.
784 =head2 left_base_classes
786 List of additional base classes all of your table classes will use
787 that need to be leftmost.
789 =head2 additional_classes
791 List of additional classes which all of your table classes will use.
795 List of additional components to be loaded into all of your Result
796 classes. A good example would be
797 L<InflateColumn::DateTime|DBIx::Class::InflateColumn::DateTime>
799 =head2 result_components_map
801 A hashref of moniker keys and component values. Unlike L</components>, which
802 loads the given components into every Result class, this option allows you to
803 load certain components for specified Result classes. For example:
805 result_components_map => {
806 StationVisited => '+YourApp::Schema::Component::StationVisited',
808 '+YourApp::Schema::Component::RouteChange',
809 'InflateColumn::DateTime',
813 You may use this in conjunction with L</components>.
817 List of L<Moose> roles to be applied to all of your Result classes.
819 =head2 result_roles_map
821 A hashref of moniker keys and role values. Unlike L</result_roles>, which
822 applies the given roles to every Result class, this option allows you to apply
823 certain roles for specified Result classes. For example:
825 result_roles_map => {
827 'YourApp::Role::Building',
828 'YourApp::Role::Destination',
830 RouteChange => 'YourApp::Role::TripEvent',
833 You may use this in conjunction with L</result_roles>.
835 =head2 use_namespaces
837 This is now the default, to go back to L<DBIx::Class::Schema/load_classes> pass
840 Generate result class names suitable for
841 L<DBIx::Class::Schema/load_namespaces> and call that instead of
842 L<DBIx::Class::Schema/load_classes>. When using this option you can also
843 specify any of the options for C<load_namespaces> (i.e. C<result_namespace>,
844 C<resultset_namespace>, C<default_resultset_class>), and they will be added
845 to the call (and the generated result class names adjusted appropriately).
847 =head2 dump_directory
849 The value of this option is a perl libdir pathname. Within
850 that directory this module will create a baseline manual
851 L<DBIx::Class::Schema> module set, based on what it creates at runtime.
853 The created schema class will have the same classname as the one on
854 which you are setting this option (and the ResultSource classes will be
855 based on this name as well).
857 Normally you wouldn't hard-code this setting in your schema class, as it
858 is meant for one-time manual usage.
860 See L<DBIx::Class::Schema::Loader/dump_to_dir> for examples of the
861 recommended way to access this functionality.
863 =head2 dump_overwrite
865 Deprecated. See L</really_erase_my_files> below, which does *not* mean
866 the same thing as the old C<dump_overwrite> setting from previous releases.
868 =head2 really_erase_my_files
870 Default false. If true, Loader will unconditionally delete any existing
871 files before creating the new ones from scratch when dumping a schema to disk.
873 The default behavior is instead to only replace the top portion of the
874 file, up to and including the final stanza which contains
875 C<# DO NOT MODIFY THE FIRST PART OF THIS FILE>
876 leaving any customizations you placed after that as they were.
878 When C<really_erase_my_files> is not set, if the output file already exists,
879 but the aforementioned final stanza is not found, or the checksum
880 contained there does not match the generated contents, Loader will
881 croak and not touch the file.
883 You should really be using version control on your schema classes (and all
884 of the rest of your code for that matter). Don't blame me if a bug in this
885 code wipes something out when it shouldn't have, you've been warned.
887 =head2 overwrite_modifications
889 Default false. If false, when updating existing files, Loader will
890 refuse to modify any Loader-generated code that has been modified
891 since its last run (as determined by the checksum Loader put in its
894 If true, Loader will discard any manual modifications that have been
895 made to Loader-generated code.
897 Again, you should be using version control on your schema classes. Be
898 careful with this option.
902 Omit the package version from the signature comment.
904 =head2 omit_timestamp
906 Omit the creation timestamp from the signature comment.
908 =head2 custom_column_info
910 Hook for adding extra attributes to the
911 L<column_info|DBIx::Class::ResultSource/column_info> for a column.
913 Must be a coderef that returns a hashref with the extra attributes.
915 Receives the L<DBIx::Class::Schema::Loader::Table> object, column name
920 custom_column_info => sub {
921 my ($table, $column_name, $column_info) = @_;
923 if ($column_name eq 'dog' && $column_info->{default_value} eq 'snoopy') {
924 return { is_snoopy => 1 };
928 This attribute can also be used to set C<inflate_datetime> on a non-datetime
929 column so it also receives the L</datetime_timezone> and/or L</datetime_locale>.
931 =head2 datetime_timezone
933 Sets the timezone attribute for L<DBIx::Class::InflateColumn::DateTime> for all
934 columns with the DATE/DATETIME/TIMESTAMP data_types.
936 =head2 datetime_locale
938 Sets the locale attribute for L<DBIx::Class::InflateColumn::DateTime> for all
939 columns with the DATE/DATETIME/TIMESTAMP data_types.
941 =head2 datetime_undef_if_invalid
943 Pass a C<0> for this option when using MySQL if you B<DON'T> want C<<
944 datetime_undef_if_invalid => 1 >> in your column info for DATE, DATETIME and
947 The default is recommended to deal with data such as C<00/00/00> which
948 sometimes ends up in such columns in MySQL.
952 File in Perl format, which should return a HASH reference, from which to read
957 Normally database names are lowercased and split by underscore, use this option
958 if you have CamelCase database names.
960 Drivers for case sensitive databases like Sybase ASE or MSSQL with a
961 case-sensitive collation will turn this option on unconditionally.
963 B<NOTE:> L</naming> = C<v8> is highly recommended with this option as the
964 semantics of this mode are much improved for CamelCase database names.
966 L</naming> = C<v7> or greater is required with this option.
968 =head2 qualify_objects
970 Set to true to prepend the L</db_schema> to table names for C<<
971 __PACKAGE__->table >> calls, and to some other things like Oracle sequences.
973 This attribute is automatically set to true for multi db_schema configurations,
974 unless explicitly set to false by the user.
978 Creates Schema and Result classes that use L<Moose>, L<MooseX::NonMoose> and
979 L<MooseX::MarkAsMethods> (or L<namespace::autoclean>, see below). The default
980 content after the md5 sum also makes the classes immutable.
982 It is safe to upgrade your existing Schema to this option.
986 Creates Schema and Result classes that use L<Moo> and
987 L<namespace::autoclean>.
989 It is safe to upgrade your existing Schema to this option.
991 =head2 only_autoclean
993 By default, we use L<MooseX::MarkAsMethods> to remove imported functions from
994 your generated classes. It uses L<namespace::autoclean> to do this, after
995 telling your object's metaclass that any operator L<overload>s in your class
996 are methods, which will cause namespace::autoclean to spare them from removal.
998 This prevents the "Hey, where'd my overloads go?!" effect.
1000 If you don't care about operator overloads, enabling this option falls back to
1001 just using L<namespace::autoclean> itself.
1003 If none of the above made any sense, or you don't have some pressing need to
1004 only use L<namespace::autoclean>, leaving this set to the default is
1007 =head2 col_collision_map
1009 This option controls how accessors for column names which collide with perl
1010 methods are named. See L</COLUMN ACCESSOR COLLISIONS> for more information.
1012 This option takes either a single L<sprintf|perlfunc/sprintf> format or a hashref of
1013 strings which are compiled to regular expressions that map to
1014 L<sprintf|perlfunc/sprintf> formats.
1018 col_collision_map => 'column_%s'
1020 col_collision_map => { '(.*)' => 'column_%s' }
1022 col_collision_map => { '(foo).*(bar)' => 'column_%s_%s' }
1024 =head2 rel_collision_map
1026 Works just like L</col_collision_map>, but for relationship names/accessors
1027 rather than column names/accessors.
1029 The default is to just append C<_rel> to the relationship name, see
1030 L</RELATIONSHIP NAME COLLISIONS>.
1032 =head2 uniq_to_primary
1034 Automatically promotes the largest unique constraints with non-nullable columns
1035 on tables to primary keys, assuming there is only one largest unique
1038 =head2 allow_extra_m2m_cols
1040 Generate C<many_to_many> relationship bridges even if the link table has
1041 extra columns other than the foreign keys. The primary key must still
1042 equal the union of the foreign keys.
1045 =head2 filter_generated_code
1047 An optional hook that lets you filter the generated text for various classes
1048 through a function that change it in any way that you want. The function will
1049 receive the type of file, C<schema> or C<result>, class and code; and returns
1050 the new code to use instead. For instance you could add custom comments, or do
1051 anything else that you want.
1053 The option can also be set to a string, which is then used as a filter program,
1056 If this exists but fails to return text matching C</\bpackage\b/>, no file will
1059 filter_generated_code => sub {
1060 my ($type, $class, $text) = @_;
1065 You can also use this option to set L<perltidy markers|perltidy/Skipping
1066 Selected Sections of Code> in your generated classes. This will leave
1067 the generated code in the default format, but will allow you to tidy
1068 your classes at any point in future, without worrying about changing the
1069 portions of the file which are checksummed, since C<perltidy> will just
1070 ignore all text between the markers.
1072 filter_generated_code => sub {
1073 return "#<<<\n$_[2]\n#>>>";
1078 None of these methods are intended for direct invocation by regular
1079 users of L<DBIx::Class::Schema::Loader>. Some are proxied via
1080 L<DBIx::Class::Schema::Loader>.
1084 # ensure that a piece of object data is a valid arrayref, creating
1085 # an empty one or encapsulating whatever's there.
1086 sub _ensure_arrayref {
1091 $self->{$_} = [ $self->{$_} ]
1092 unless ref $self->{$_} eq 'ARRAY';
1098 Constructor for L<DBIx::Class::Schema::Loader::Base>, used internally
1099 by L<DBIx::Class::Schema::Loader>.
1104 my ( $class, %args ) = @_;
1106 if (exists $args{column_accessor_map}) {
1107 $args{col_accessor_map} = delete $args{column_accessor_map};
1110 my $self = { %args };
1112 # don't lose undef options
1113 for (values %$self) {
1114 $_ = 0 unless defined $_;
1117 bless $self => $class;
1119 if (my $config_file = $self->config_file) {
1120 my $config_opts = do $config_file;
1122 croak "Error reading config from $config_file: $@" if $@;
1124 croak "Config file $config_file must be a hashref" unless ref($config_opts) eq 'HASH';
1126 while (my ($k, $v) = each %$config_opts) {
1127 $self->{$k} = $v unless exists $self->{$k};
1131 if (defined $self->{result_component_map}) {
1132 if (defined $self->result_components_map) {
1133 croak "Specify only one of result_components_map or result_component_map";
1135 $self->result_components_map($self->{result_component_map})
1138 if (defined $self->{result_role_map}) {
1139 if (defined $self->result_roles_map) {
1140 croak "Specify only one of result_roles_map or result_role_map";
1142 $self->result_roles_map($self->{result_role_map})
1145 croak "Specify only one of use_moose or use_moo"
1146 if $self->use_moose and $self->use_moo;
1148 croak "the result_roles and result_roles_map options may only be used in conjunction with use_moose=1 or use_moo=1"
1149 if ((not $self->use_moose) && (not $self->use_moo))
1150 && ((defined $self->result_roles) || (defined $self->result_roles_map));
1152 $self->_ensure_arrayref(qw/schema_components
1154 additional_base_classes
1160 $self->_validate_class_args;
1162 croak "result_components_map must be a hash"
1163 if defined $self->result_components_map
1164 && ref $self->result_components_map ne 'HASH';
1166 if ($self->result_components_map) {
1167 my %rc_map = %{ $self->result_components_map };
1168 foreach my $moniker (keys %rc_map) {
1169 $rc_map{$moniker} = [ $rc_map{$moniker} ] unless ref $rc_map{$moniker};
1171 $self->result_components_map(\%rc_map);
1174 $self->result_components_map({});
1176 $self->_validate_result_components_map;
1178 croak "result_roles_map must be a hash"
1179 if defined $self->result_roles_map
1180 && ref $self->result_roles_map ne 'HASH';
1182 if ($self->result_roles_map) {
1183 my %rr_map = %{ $self->result_roles_map };
1184 foreach my $moniker (keys %rr_map) {
1185 $rr_map{$moniker} = [ $rr_map{$moniker} ] unless ref $rr_map{$moniker};
1187 $self->result_roles_map(\%rr_map);
1189 $self->result_roles_map({});
1191 $self->_validate_result_roles_map;
1193 for my $use_oo (qw(use_moose use_moo)) {
1194 if ($self->$use_oo) {
1195 if (not DBIx::Class::Schema::Loader::Optional::Dependencies->req_ok_for($use_oo)) {
1196 die sprintf "You must install the following CPAN modules to enable the $use_oo option: %s.\n",
1197 DBIx::Class::Schema::Loader::Optional::Dependencies->req_missing_for($use_oo);
1202 $self->{_tables} = {};
1203 $self->{monikers} = {};
1204 $self->{moniker_to_table} = {};
1205 $self->{class_to_table} = {};
1206 $self->{classes} = {};
1207 $self->{_upgrading_classes} = {};
1208 $self->{generated_classes} = [];
1210 $self->{schema_class} ||= ( ref $self->{schema} || $self->{schema} );
1211 $self->{schema} ||= $self->{schema_class};
1212 $self->{table_comments_table} ||= 'table_comments';
1213 $self->{column_comments_table} ||= 'column_comments';
1215 croak "dump_overwrite is deprecated. Please read the"
1216 . " DBIx::Class::Schema::Loader::Base documentation"
1217 if $self->{dump_overwrite};
1219 $self->{dynamic} = ! $self->{dump_directory};
1221 croak "dry_run can only be used with static schema generation"
1222 if $self->dynamic and $self->dry_run;
1224 $self->{temp_directory} ||= File::Temp::tempdir( 'dbicXXXX',
1229 $self->{dump_directory} ||= $self->{temp_directory};
1231 $self->real_dump_directory($self->{dump_directory});
1233 $self->version_to_dump($DBIx::Class::Schema::Loader::VERSION);
1234 $self->schema_version_to_dump($DBIx::Class::Schema::Loader::VERSION);
1236 if (not defined $self->naming) {
1237 $self->naming_set(0);
1240 $self->naming_set(1);
1243 if ((not ref $self->naming) && defined $self->naming) {
1244 my $naming_ver = $self->naming;
1246 relationships => $naming_ver,
1247 monikers => $naming_ver,
1248 column_accessors => $naming_ver,
1251 elsif (ref $self->naming eq 'HASH' && exists $self->naming->{ALL}) {
1252 my $val = delete $self->naming->{ALL};
1254 $self->naming->{$_} = $val
1255 foreach qw/relationships monikers column_accessors/;
1258 if ($self->naming) {
1259 foreach my $key (qw/relationships monikers column_accessors/) {
1260 $self->naming->{$key} = $CURRENT_V if ($self->naming->{$key}||'') eq 'current';
1263 $self->{naming} ||= {};
1265 if ($self->custom_column_info && ref $self->custom_column_info ne 'CODE') {
1266 croak 'custom_column_info must be a CODE ref';
1269 $self->_check_back_compat;
1271 $self->use_namespaces(1) unless defined $self->use_namespaces;
1272 $self->generate_pod(1) unless defined $self->generate_pod;
1273 $self->pod_comment_mode('auto') unless defined $self->pod_comment_mode;
1274 $self->pod_comment_spillover_length(60) unless defined $self->pod_comment_spillover_length;
1276 if (my $col_collision_map = $self->col_collision_map) {
1277 if (my $reftype = ref $col_collision_map) {
1278 if ($reftype ne 'HASH') {
1279 croak "Invalid type $reftype for option 'col_collision_map'";
1283 $self->col_collision_map({ '(.*)' => $col_collision_map });
1287 if (my $rel_collision_map = $self->rel_collision_map) {
1288 if (my $reftype = ref $rel_collision_map) {
1289 if ($reftype ne 'HASH') {
1290 croak "Invalid type $reftype for option 'rel_collision_map'";
1294 $self->rel_collision_map({ '(.*)' => $rel_collision_map });
1298 if (defined(my $rel_name_map = $self->rel_name_map)) {
1299 my $reftype = ref $rel_name_map;
1300 if ($reftype ne 'HASH' && $reftype ne 'CODE') {
1301 croak "Invalid type $reftype for option 'rel_name_map', must be HASH or CODE";
1305 if (defined(my $filter = $self->filter_generated_code)) {
1306 my $reftype = ref $filter;
1307 if ($reftype && $reftype ne 'CODE') {
1308 croak "Invalid type $reftype for option 'filter_generated_code, must be a scalar or a CODE reference";
1312 if (defined $self->db_schema) {
1313 if (ref $self->db_schema eq 'ARRAY') {
1314 if (@{ $self->db_schema } > 1 && not defined $self->{qualify_objects}) {
1315 $self->{qualify_objects} = 1;
1317 elsif (@{ $self->db_schema } == 0) {
1318 $self->{db_schema} = undef;
1321 elsif (not ref $self->db_schema) {
1322 if ($self->db_schema eq '%' && not defined $self->{qualify_objects}) {
1323 $self->{qualify_objects} = 1;
1326 $self->{db_schema} = [ $self->db_schema ];
1330 if (not $self->moniker_parts) {
1331 $self->moniker_parts(['name']);
1334 if (not ref $self->moniker_parts) {
1335 $self->moniker_parts([ $self->moniker_parts ]);
1337 if (ref $self->moniker_parts ne 'ARRAY') {
1338 croak 'moniker_parts must be an arrayref';
1340 if (none { $_ eq 'name' } @{ $self->moniker_parts }) {
1341 croak "moniker_parts option *must* contain 'name'";
1345 if (not defined $self->moniker_part_separator) {
1346 $self->moniker_part_separator('');
1348 if (not defined $self->moniker_part_map) {
1349 $self->moniker_part_map({}),
1355 sub _check_back_compat {
1358 # dynamic schemas will always be in 0.04006 mode, unless overridden
1359 if ($self->dynamic) {
1360 # just in case, though no one is likely to dump a dynamic schema
1361 $self->schema_version_to_dump('0.04006');
1363 if (not $self->naming_set) {
1364 warn <<EOF unless $ENV{SCHEMA_LOADER_BACKCOMPAT};
1366 Dynamic schema detected, will run in 0.04006 mode.
1368 Set the 'naming' attribute or the SCHEMA_LOADER_BACKCOMPAT environment variable
1369 to disable this warning.
1371 See perldoc DBIx::Class::Schema::Loader::Manual::UpgradingFromV4 for more
1376 $self->_upgrading_from('v4');
1379 if ((not defined $self->use_namespaces) && ($self->naming_set)) {
1380 $self->use_namespaces(1);
1383 $self->naming->{relationships} ||= 'v4';
1384 $self->naming->{monikers} ||= 'v4';
1386 if ($self->use_namespaces) {
1387 $self->_upgrading_from_load_classes(1);
1390 $self->use_namespaces(0);
1396 # otherwise check if we need backcompat mode for a static schema
1397 my $filename = $self->get_dump_filename($self->schema_class);
1398 return unless -e $filename;
1400 my ($old_gen, $old_md5, $old_ver, $old_ts, $old_custom) =
1401 $self->_parse_generated_file($filename);
1403 return unless $old_ver;
1405 # determine if the existing schema was dumped with use_moo(se) => 1
1406 for my $oo (qw(Moose Moo)) {
1407 my $use_oo = "use_".lc($oo);
1408 if (! defined $self->$use_oo) {
1409 $self->{$use_oo} = 1 if $old_gen =~ /^ (?!\s*\#) use \s+ \Q$oo\E\b/xm;
1413 my $load_classes = ($old_gen =~ /^__PACKAGE__->load_classes;/m) ? 1 : 0;
1415 my $result_namespace = do { ($old_gen =~ /result_namespace => (.+)/) ? $1 : '' };
1416 my $ds = eval $result_namespace;
1418 Could not eval expression '$result_namespace' for result_namespace from
1421 $result_namespace = $ds || '';
1423 if ($load_classes && (not defined $self->use_namespaces)) {
1424 warn <<"EOF" unless $ENV{SCHEMA_LOADER_BACKCOMPAT};
1426 'load_classes;' static schema detected, turning off 'use_namespaces'.
1428 Set the 'use_namespaces' attribute or the SCHEMA_LOADER_BACKCOMPAT environment
1429 variable to disable this warning.
1431 See perldoc DBIx::Class::Schema::Loader::Manual::UpgradingFromV4 for more
1434 $self->use_namespaces(0);
1436 elsif ($load_classes && $self->use_namespaces) {
1437 $self->_upgrading_from_load_classes(1);
1439 elsif ((not $load_classes) && defined $self->use_namespaces && ! $self->use_namespaces) {
1440 $self->_downgrading_to_load_classes(
1441 $result_namespace || 'Result'
1444 elsif ((not defined $self->use_namespaces) || $self->use_namespaces) {
1445 if (not $self->result_namespace) {
1446 $self->result_namespace($result_namespace || 'Result');
1448 elsif ($result_namespace ne $self->result_namespace) {
1449 $self->_rewriting_result_namespace(
1450 $result_namespace || 'Result'
1455 # XXX when we go past .0 this will need fixing
1456 my ($v) = $old_ver =~ /([1-9])/;
1459 return if ($v eq $CURRENT_V || $old_ver =~ /^0\.\d\d999/);
1461 if (not %{ $self->naming }) {
1462 warn <<"EOF" unless $ENV{SCHEMA_LOADER_BACKCOMPAT};
1464 Version $old_ver static schema detected, turning on backcompat mode.
1466 Set the 'naming' attribute or the SCHEMA_LOADER_BACKCOMPAT environment variable
1467 to disable this warning.
1469 See: 'naming' in perldoc DBIx::Class::Schema::Loader::Base .
1471 See perldoc DBIx::Class::Schema::Loader::Manual::UpgradingFromV4 if upgrading
1472 from version 0.04006.
1475 $self->naming->{relationships} ||= $v;
1476 $self->naming->{monikers} ||= $v;
1477 $self->naming->{column_accessors} ||= $v;
1479 $self->schema_version_to_dump($old_ver);
1482 $self->_upgrading_from($v);
1486 sub _validate_class_args {
1489 foreach my $k (@CLASS_ARGS) {
1490 next unless $self->$k;
1492 my @classes = ref $self->$k eq 'ARRAY' ? @{ $self->$k } : $self->$k;
1493 $self->_validate_classes($k, \@classes);
1497 sub _validate_result_components_map {
1500 foreach my $classes (values %{ $self->result_components_map }) {
1501 $self->_validate_classes('result_components_map', $classes);
1505 sub _validate_result_roles_map {
1508 foreach my $classes (values %{ $self->result_roles_map }) {
1509 $self->_validate_classes('result_roles_map', $classes);
1513 sub _validate_classes {
1516 my $classes = shift;
1518 # make a copy to not destroy original
1519 my @classes = @$classes;
1521 foreach my $c (@classes) {
1522 # components default to being under the DBIx::Class namespace unless they
1523 # are preceded with a '+'
1524 if ( $key =~ m/component/ && $c !~ s/^\+// ) {
1525 $c = 'DBIx::Class::' . $c;
1528 # 1 == installed, 0 == not installed, undef == invalid classname
1529 my $installed = Class::Inspector->installed($c);
1530 if ( defined($installed) ) {
1531 if ( $installed == 0 ) {
1532 croak qq/$c, as specified in the loader option "$key", is not installed/;
1535 croak qq/$c, as specified in the loader option "$key", is an invalid class name/;
1541 sub _find_file_in_inc {
1542 my ($self, $file) = @_;
1544 foreach my $prefix (@INC) {
1545 my $fullpath = File::Spec->catfile($prefix, $file);
1546 # abs_path pure-perl fallback warns for non-existent files
1547 local $SIG{__WARN__} = sigwarn_silencer(qr/^stat\(.*\Q$file\E\)/);
1548 return $fullpath if -f $fullpath
1549 # abs_path throws on Windows for nonexistent files
1550 and (try { Cwd::abs_path($fullpath) }) ne
1551 ((try { Cwd::abs_path(File::Spec->catfile($self->dump_directory, $file)) }) || '');
1557 sub _find_class_in_inc {
1558 my ($self, $class) = @_;
1560 return $self->_find_file_in_inc(class_path($class));
1566 return $self->_upgrading_from
1567 || $self->_upgrading_from_load_classes
1568 || $self->_downgrading_to_load_classes
1569 || $self->_rewriting_result_namespace
1573 sub _rewrite_old_classnames {
1574 my ($self, $code) = @_;
1576 return $code unless $self->_rewriting;
1578 my %old_classes = reverse %{ $self->_upgrading_classes };
1580 my $re = join '|', keys %old_classes;
1581 $re = qr/\b($re)\b/;
1583 $code =~ s/$re/$old_classes{$1} || $1/eg;
1588 sub _load_external {
1589 my ($self, $class) = @_;
1591 return if $self->{skip_load_external};
1593 # so that we don't load our own classes, under any circumstances
1594 local *INC = [ grep $_ ne $self->dump_directory, @INC ];
1596 my $real_inc_path = $self->_find_class_in_inc($class);
1598 my $old_class = $self->_upgrading_classes->{$class}
1599 if $self->_rewriting;
1601 my $old_real_inc_path = $self->_find_class_in_inc($old_class)
1602 if $old_class && $old_class ne $class;
1604 return unless $real_inc_path || $old_real_inc_path;
1606 if ($real_inc_path) {
1607 # If we make it to here, we loaded an external definition
1608 warn qq/# Loaded external class definition for '$class'\n/
1611 my $code = $self->_rewrite_old_classnames(slurp_file $real_inc_path);
1613 if ($self->dynamic) { # load the class too
1614 eval_package_without_redefine_warnings($class, $code);
1617 $self->_ext_stmt($class,
1618 qq|# These lines were loaded from '$real_inc_path' found in \@INC.\n|
1619 .qq|# They are now part of the custom portion of this file\n|
1620 .qq|# for you to hand-edit. If you do not either delete\n|
1621 .qq|# this section or remove that file from \@INC, this section\n|
1622 .qq|# will be repeated redundantly when you re-create this\n|
1623 .qq|# file again via Loader! See skip_load_external to disable\n|
1624 .qq|# this feature.\n|
1627 $self->_ext_stmt($class, $code);
1628 $self->_ext_stmt($class,
1629 qq|# End of lines loaded from '$real_inc_path'|
1633 if ($old_real_inc_path) {
1634 my $code = slurp_file $old_real_inc_path;
1636 $self->_ext_stmt($class, <<"EOF");
1638 # These lines were loaded from '$old_real_inc_path',
1639 # based on the Result class name that would have been created by an older
1640 # version of the Loader. For a static schema, this happens only once during
1641 # upgrade. See skip_load_external to disable this feature.
1644 $code = $self->_rewrite_old_classnames($code);
1646 if ($self->dynamic) {
1649 Detected external content in '$old_real_inc_path', a class name that would have
1650 been used by an older version of the Loader.
1652 * PLEASE RENAME THIS CLASS: from '$old_class' to '$class', as that is the
1653 new name of the Result.
1655 eval_package_without_redefine_warnings($class, $code);
1659 $self->_ext_stmt($class, $code);
1660 $self->_ext_stmt($class,
1661 qq|# End of lines loaded from '$old_real_inc_path'|
1668 Does the actual schema-construction work.
1675 $self->_load_tables($self->_tables_list);
1682 Rescan the database for changes. Returns a list of the newly added table
1685 The schema argument should be the schema class or object to be affected. It
1686 should probably be derived from the original schema_class used during L</load>.
1691 my ($self, $schema) = @_;
1693 $self->{schema} = $schema;
1694 $self->_relbuilder->{schema} = $schema;
1697 my @current = $self->_tables_list;
1699 foreach my $table (@current) {
1700 if(!exists $self->_tables->{$table->sql_name}) {
1701 push(@created, $table);
1706 @current{map $_->sql_name, @current} = ();
1707 foreach my $table (values %{ $self->_tables }) {
1708 if (not exists $current{$table->sql_name}) {
1709 $self->_remove_table($table);
1713 delete @$self{qw/_dump_storage _relations_started _uniqs_started/};
1715 my $loaded = $self->_load_tables(@current);
1717 foreach my $table (@created) {
1718 $self->monikers->{$table->sql_name} = $self->_table2moniker($table);
1721 return map { $self->monikers->{$_->sql_name} } @created;
1727 return if $self->{skip_relationships};
1729 return $self->{relbuilder} ||= do {
1730 my $relbuilder_suff =
1737 ->{$self->naming->{relationships}||$CURRENT_V} || '';
1739 my $relbuilder_class = 'DBIx::Class::Schema::Loader::RelBuilder'.$relbuilder_suff;
1740 $self->ensure_class_loaded($relbuilder_class);
1741 $relbuilder_class->new($self);
1746 my ($self, @tables) = @_;
1748 # Save the new tables to the tables list and compute monikers
1750 $self->_tables->{$_->sql_name} = $_;
1751 $self->monikers->{$_->sql_name} = $self->_table2moniker($_);
1754 # check for moniker clashes
1755 my $inverse_moniker_idx;
1756 foreach my $imtable (values %{ $self->_tables }) {
1757 push @{ $inverse_moniker_idx->{$self->monikers->{$imtable->sql_name}} }, $imtable;
1761 foreach my $moniker (keys %$inverse_moniker_idx) {
1762 my $imtables = $inverse_moniker_idx->{$moniker};
1763 if (@$imtables > 1) {
1764 my $different_databases =
1765 $imtables->[0]->can('database') && (uniq map $_->database||'', @$imtables) > 1;
1767 my $different_schemas =
1768 (uniq map $_->schema||'', @$imtables) > 1;
1770 if ($different_databases || $different_schemas) {
1771 my ($use_schema, $use_database) = (1, 0);
1773 if ($different_databases) {
1776 # If any monikers are in the same database, we have to distinguish by
1777 # both schema and database.
1779 $db_counts{$_}++ for map $_->database, @$imtables;
1780 $use_schema = any { $_ > 1 } values %db_counts;
1783 foreach my $tbl (@$imtables) { delete $self->monikers->{$tbl->sql_name}; }
1785 my $moniker_parts = [ @{ $self->moniker_parts } ];
1787 my $have_schema = any { $_ eq 'schema' } @{ $self->moniker_parts };
1788 my $have_database = any { $_ eq 'database' } @{ $self->moniker_parts };
1790 unshift @$moniker_parts, 'schema' if $use_schema && !$have_schema;
1791 unshift @$moniker_parts, 'database' if $use_database && !$have_database;
1793 local $self->{moniker_parts} = $moniker_parts;
1797 foreach my $tbl (@$imtables) { $new_monikers{$tbl->sql_name} = $self->_table2moniker($tbl); }
1798 foreach my $name (map $_->sql_name, @$imtables) { $self->monikers->{$name} = $new_monikers{$name}; }
1800 # check if there are still clashes
1803 while (my ($t, $m) = each %new_monikers) {
1804 push @{ $by_moniker{$m} }, $t;
1807 foreach my $m (grep @{ $by_moniker{$_} } > 1, keys %by_moniker) {
1808 push @clashes, sprintf ("tried disambiguating by moniker_parts, but tables %s still reduced to the same source moniker '%s'",
1809 join (', ', @{ $by_moniker{$m} }),
1815 push @clashes, sprintf ("tables %s reduced to the same source moniker '%s'",
1816 join (', ', map $_->sql_name, @$imtables),
1824 die 'Unable to load schema - chosen moniker/class naming style results in moniker clashes. '
1825 . 'Change the naming style, or supply an explicit moniker_map: '
1826 . join ('; ', @clashes)
1831 foreach my $tbl (@tables) { $self->_make_src_class($tbl); }
1832 foreach my $tbl (@tables) { $self->_setup_src_meta($tbl); }
1834 if(!$self->skip_relationships) {
1835 # The relationship loader needs a working schema
1836 local $self->{quiet} = 1;
1837 local $self->{dump_directory} = $self->{temp_directory};
1838 local $self->{generated_classes} = [];
1839 local $self->{dry_run} = 0;
1840 $self->_reload_classes(\@tables);
1841 $self->_load_relationships(\@tables);
1843 # Remove that temp dir from INC so it doesn't get reloaded
1844 @INC = grep $_ ne $self->dump_directory, @INC;
1847 foreach my $tbl (@tables) { $self->_load_roles($tbl); }
1848 foreach my $tbl (map { $self->classes->{$_->sql_name} } @tables) { $self->_load_external($tbl); }
1850 # Reload without unloading first to preserve any symbols from external
1852 $self->_reload_classes(\@tables, { unload => 0 });
1854 # Drop temporary cache
1855 delete $self->{_cache};
1860 sub _reload_classes {
1861 my ($self, $tables, $opts) = @_;
1863 my @tables = @$tables;
1865 my $unload = $opts->{unload};
1866 $unload = 1 unless defined $unload;
1868 # so that we don't repeat custom sections
1869 @INC = grep $_ ne $self->dump_directory, @INC;
1871 $self->_dump_to_dir(map { $self->classes->{$_->sql_name} } @tables);
1873 unshift @INC, $self->dump_directory;
1875 return if $self->dry_run;
1878 my %have_source = map { $_ => $self->schema->source($_) }
1879 $self->schema->sources;
1881 for my $table (@tables) {
1882 my $moniker = $self->monikers->{$table->sql_name};
1883 my $class = $self->classes->{$table->sql_name};
1886 no warnings 'redefine';
1887 local *Class::C3::reinitialize = sub {}; # to speed things up, reinitialized below
1890 if (my $mc = $self->_moose_metaclass($class)) {
1893 Class::Unload->unload($class) if $unload;
1894 my ($source, $resultset_class);
1896 ($source = $have_source{$moniker})
1897 && ($resultset_class = $source->resultset_class)
1898 && ($resultset_class ne 'DBIx::Class::ResultSet')
1900 my $has_file = Class::Inspector->loaded_filename($resultset_class);
1901 if (my $mc = $self->_moose_metaclass($resultset_class)) {
1904 Class::Unload->unload($resultset_class) if $unload;
1905 $self->_reload_class($resultset_class) if $has_file;
1907 $self->_reload_class($class);
1909 push @to_register, [$moniker, $class];
1912 Class::C3->reinitialize;
1913 for (@to_register) {
1914 $self->schema->register_class(@$_);
1918 sub _moose_metaclass {
1919 return undef unless $INC{'Class/MOP.pm'}; # if CMOP is not loaded the class could not have loaded in the 1st place
1923 my $mc = try { Class::MOP::class_of($class) }
1926 return $mc->isa('Moose::Meta::Class') ? $mc : undef;
1929 # We use this instead of ensure_class_loaded when there are package symbols we
1932 my ($self, $class) = @_;
1934 delete $INC{ +class_path($class) };
1937 eval_package_without_redefine_warnings ($class, "require $class");
1940 my $source = slurp_file $self->_get_dump_filename($class);
1941 die "Failed to reload class $class: $_.\n\nCLASS SOURCE:\n\n$source";
1945 sub _get_dump_filename {
1946 my ($self, $class) = (@_);
1948 $class =~ s{::}{/}g;
1949 return $self->dump_directory . q{/} . $class . q{.pm};
1952 =head2 get_dump_filename
1956 Returns the full path to the file for a class that the class has been or will
1957 be dumped to. This is a file in a temp dir for a dynamic schema.
1961 sub get_dump_filename {
1962 my ($self, $class) = (@_);
1964 local $self->{dump_directory} = $self->real_dump_directory;
1966 return $self->_get_dump_filename($class);
1969 sub _ensure_dump_subdirs {
1970 my ($self, $class) = (@_);
1972 return if $self->dry_run;
1974 my @name_parts = split(/::/, $class);
1975 pop @name_parts; # we don't care about the very last element,
1976 # which is a filename
1978 my $dir = $self->dump_directory;
1981 mkdir($dir) or croak "mkdir('$dir') failed: $!";
1983 last if !@name_parts;
1984 $dir = File::Spec->catdir($dir, shift @name_parts);
1989 my ($self, @classes) = @_;
1991 my $schema_class = $self->schema_class;
1992 my $schema_base_class = $self->schema_base_class || 'DBIx::Class::Schema';
1994 my $target_dir = $self->dump_directory;
1995 warn "Dumping manual schema for $schema_class to directory $target_dir ...\n"
1996 unless $self->dynamic or $self->quiet;
2000 . qq|package $schema_class;\n\n|
2001 . qq|# Created by DBIx::Class::Schema::Loader\n|
2002 . qq|# DO NOT MODIFY THE FIRST PART OF THIS FILE\n\n|;
2005 = $self->only_autoclean
2006 ? 'namespace::autoclean'
2007 : 'MooseX::MarkAsMethods autoclean => 1'
2010 if ($self->use_moose) {
2011 $schema_text.= qq|use Moose;\nuse $autoclean;\nextends '$schema_base_class';\n\n|;
2013 elsif ($self->use_moo) {
2014 $schema_text .= qq|use Moo;\nuse namespace::autoclean;\nextends '$schema_base_class';\n\n|;
2017 $schema_text .= qq|use strict;\nuse warnings;\n\nuse base '$schema_base_class';\n\n|;
2020 my @schema_components = @{ $self->schema_components || [] };
2022 if (@schema_components) {
2023 my $schema_components = dump @schema_components;
2024 $schema_components = "($schema_components)" if @schema_components == 1;
2026 $schema_text .= "__PACKAGE__->load_components${schema_components};\n\n";
2029 if ($self->use_namespaces) {
2030 $schema_text .= qq|__PACKAGE__->load_namespaces|;
2031 my $namespace_options;
2033 my @attr = qw/resultset_namespace default_resultset_class/;
2035 unshift @attr, 'result_namespace'
2036 if $self->result_namespace && $self->result_namespace ne 'Result';
2038 for my $attr (@attr) {
2040 my $code = dumper_squashed $self->$attr;
2041 $namespace_options .= qq| $attr => $code,\n|
2044 $schema_text .= qq|(\n$namespace_options)| if $namespace_options;
2045 $schema_text .= qq|;\n|;
2048 $schema_text .= qq|__PACKAGE__->load_classes;\n|;
2052 local $self->{version_to_dump} = $self->schema_version_to_dump;
2053 $self->_write_classfile($schema_class, $schema_text, 1);
2056 my $result_base_class = $self->result_base_class || 'DBIx::Class::Core';
2058 foreach my $src_class (@classes) {
2061 . qq|package $src_class;\n\n|
2062 . qq|# Created by DBIx::Class::Schema::Loader\n|
2063 . qq|# DO NOT MODIFY THE FIRST PART OF THIS FILE\n\n|;
2065 $src_text .= $self->_make_pod_heading($src_class);
2067 $src_text .= qq|use strict;\nuse warnings;\n\n|;
2069 $src_text .= $self->_base_class_pod($result_base_class)
2070 unless $result_base_class eq 'DBIx::Class::Core';
2072 if ($self->use_moose || $self->use_moo) {
2073 $src_text.= $self->use_moose
2074 ? qq|use Moose;\nuse MooseX::NonMoose;\nuse $autoclean;|
2075 : qq|use Moo;\nuse namespace::autoclean;|
2078 # these options 'use base' which is compile time
2079 if (@{ $self->left_base_classes } || @{ $self->additional_base_classes }) {
2080 $src_text .= qq|\nBEGIN { extends '$result_base_class' }\n|;
2083 $src_text .= qq|\nextends '$result_base_class';\n|;
2087 $src_text .= qq|use base '$result_base_class';\n|;
2090 $self->_write_classfile($src_class, $src_text);
2093 # remove Result dir if downgrading from use_namespaces, and there are no
2095 if (my $result_ns = $self->_downgrading_to_load_classes
2096 || $self->_rewriting_result_namespace) {
2097 my $result_namespace = $self->_result_namespace(
2102 (my $result_dir = $result_namespace) =~ s{::}{/}g;
2103 $result_dir = $self->dump_directory . '/' . $result_dir;
2105 unless (my @files = glob "$result_dir/*") {
2110 warn "Schema dump completed.\n" unless $self->dynamic or $self->quiet;
2114 my ($self, $version, $ts) = @_;
2115 return qq|\n\n# Created by DBIx::Class::Schema::Loader|
2116 . (defined($version) ? q| v| . $version : '')
2117 . (defined($ts) ? q| @ | . $ts : '')
2118 . qq|\n# DO NOT MODIFY THIS OR ANYTHING ABOVE! md5sum:|;
2121 sub _write_classfile {
2122 my ($self, $class, $text, $is_schema) = @_;
2124 my $filename = $self->_get_dump_filename($class);
2125 $self->_ensure_dump_subdirs($class);
2127 if (-f $filename && $self->really_erase_my_files && !$self->dry_run) {
2128 warn "Deleting existing file '$filename' due to "
2129 . "'really_erase_my_files' setting\n" unless $self->quiet;
2133 my ($old_gen, $old_md5, $old_ver, $old_ts, $old_custom)
2134 = $self->_parse_generated_file($filename);
2136 if (! $old_gen && -f $filename) {
2137 croak "Cannot overwrite '$filename' without 'really_erase_my_files',"
2138 . " it does not appear to have been generated by Loader"
2141 my $custom_content = $old_custom || '';
2143 # Use custom content from a renamed class, the class names in it are
2145 if (my $renamed_class = $self->_upgrading_classes->{$class}) {
2146 my $old_filename = $self->_get_dump_filename($renamed_class);
2148 if (-f $old_filename) {
2149 $custom_content = ($self->_parse_generated_file ($old_filename))[4];
2151 unlink $old_filename unless $self->dry_run;
2155 $custom_content ||= $self->_default_custom_content($is_schema);
2157 # If upgrading to use_moose=1 replace default custom content with default Moose custom content.
2158 # If there is already custom content, which does not have the Moose content, add it.
2159 if ($self->use_moose) {
2161 my $non_moose_custom_content = do {
2162 local $self->{use_moose} = 0;
2163 $self->_default_custom_content;
2166 if ($custom_content eq $non_moose_custom_content) {
2167 $custom_content = $self->_default_custom_content($is_schema);
2169 elsif ($custom_content !~ /\Q@{[$self->_default_moose_custom_content($is_schema)]}\E/) {
2170 $custom_content .= $self->_default_custom_content($is_schema);
2173 elsif (defined $self->use_moose && $old_gen) {
2174 croak 'It is not possible to "downgrade" a schema that was loaded with use_moose => 1 to use_moose => 0, due to differing custom content'
2175 if $old_gen =~ /use \s+ MooseX?\b/x;
2178 $custom_content = $self->_rewrite_old_classnames($custom_content);
2181 for @{$self->{_dump_storage}->{$class} || []};
2183 if ($self->filter_generated_code) {
2184 my $filter = $self->filter_generated_code;
2186 if (ref $filter eq 'CODE') {
2188 ($is_schema ? 'schema' : 'result'),
2194 my ($fh, $temp_file) = tempfile();
2196 binmode $fh, ':encoding(UTF-8)';
2200 open my $out, qq{$filter < "$temp_file"|}
2201 or croak "Could not open pipe to $filter: $!";
2203 $text = decode('UTF-8', do { local $/; <$out> });
2205 $text =~ s/$CR?$LF/\n/g;
2209 my $exit_code = $? >> 8;
2212 or croak "Could not remove temporary file '$temp_file': $!";
2214 if ($exit_code != 0) {
2215 croak "filter '$filter' exited non-zero: $exit_code";
2218 if (not $text or not $text =~ /\bpackage\b/) {
2219 warn("$class skipped due to filter") if $self->debug;
2224 # Check and see if the dump is in fact different
2228 $compare_to = $text . $self->_sig_comment($old_ver, $old_ts);
2229 if (Digest::MD5::md5_base64(encode 'UTF-8', $compare_to) eq $old_md5) {
2230 return unless $self->_upgrading_from && $is_schema;
2234 push @{$self->generated_classes}, $class;
2236 return if $self->dry_run;
2238 $text .= $self->_sig_comment(
2239 $self->omit_version ? undef : $self->version_to_dump,
2240 $self->omit_timestamp ? undef : POSIX::strftime('%Y-%m-%d %H:%M:%S', localtime)
2243 open(my $fh, '>:raw:encoding(UTF-8)', $filename)
2244 or croak "Cannot open '$filename' for writing: $!";
2246 # Write the top half and its MD5 sum
2247 print $fh $text . Digest::MD5::md5_base64(encode 'UTF-8', $text) . "\n";
2249 # Write out anything loaded via external partial class file in @INC
2251 for @{$self->{_ext_storage}->{$class} || []};
2253 # Write out any custom content the user has added
2254 print $fh $custom_content;
2257 or croak "Error closing '$filename': $!";
2260 sub _default_moose_custom_content {
2261 my ($self, $is_schema) = @_;
2263 if (not $is_schema) {
2264 return qq|\n__PACKAGE__->meta->make_immutable;|;
2267 return qq|\n__PACKAGE__->meta->make_immutable(inline_constructor => 0);|;
2270 sub _default_custom_content {
2271 my ($self, $is_schema) = @_;
2272 my $default = qq|\n\n# You can replace this text with custom|
2273 . qq| code or comments, and it will be preserved on regeneration|;
2274 if ($self->use_moose) {
2275 $default .= $self->_default_moose_custom_content($is_schema);
2277 $default .= qq|\n1;\n|;
2281 sub _parse_generated_file {
2282 my ($self, $fn) = @_;
2284 return unless -f $fn;
2286 open(my $fh, '<:encoding(UTF-8)', $fn)
2287 or croak "Cannot open '$fn' for reading: $!";
2290 qr{^(# DO NOT MODIFY THIS OR ANYTHING ABOVE! md5sum:)([A-Za-z0-9/+]{22})\r?\n};
2292 my ($real_md5, $ts, $ver, $gen);
2299 # Pull out the version and timestamp from the line above
2300 ($ver, $ts) = $gen =~ m/^# Created by DBIx::Class::Schema::Loader( v[\d.]+)?( @ [\d-]+ [\d:]+)?\r?\Z/m;
2301 $ver =~ s/^ v// if $ver;
2302 $ts =~ s/^ @ // if $ts;
2305 $real_md5 = Digest::MD5::md5_base64(encode 'UTF-8', $gen);
2306 if ($real_md5 ne $mark_md5) {
2307 if ($self->overwrite_modifications) {
2308 # Setting this to something that is not a valid MD5 forces
2309 # the file to be rewritten.
2310 $real_md5 = 'not an MD5';
2313 croak "Checksum mismatch in '$fn', the auto-generated part of the file has been modified outside of this loader. Aborting.\nIf you want to overwrite these modifications, set the 'overwrite_modifications' loader option.\n";
2323 my $custom = do { local $/; <$fh> }
2327 $custom =~ s/$CRLF|$LF/\n/g;
2331 return ($gen, $real_md5, $ver, $ts, $custom);
2339 warn "$target: use $_;" if $self->debug;
2340 $self->_raw_stmt($target, "use $_;");
2348 my $blist = join(q{ }, @_);
2350 return unless $blist;
2352 warn "$target: use base qw/$blist/;" if $self->debug;
2353 $self->_raw_stmt($target, "use base qw/$blist/;");
2360 my $rlist = join(q{, }, map { qq{'$_'} } @_);
2362 return unless $rlist;
2364 warn "$target: with $rlist;" if $self->debug;
2365 $self->_raw_stmt($target, "\nwith $rlist;");
2368 sub _result_namespace {
2369 my ($self, $schema_class, $ns) = @_;
2370 my @result_namespace;
2372 $ns = $ns->[0] if ref $ns;
2374 if ($ns =~ /^\+(.*)/) {
2375 # Fully qualified namespace
2376 @result_namespace = ($1)
2379 # Relative namespace
2380 @result_namespace = ($schema_class, $ns);
2383 return wantarray ? @result_namespace : join '::', @result_namespace;
2386 # Create class with applicable bases, setup monikers, etc
2387 sub _make_src_class {
2388 my ($self, $table) = @_;
2390 my $schema = $self->schema;
2391 my $schema_class = $self->schema_class;
2393 my $table_moniker = $self->monikers->{$table->sql_name};
2394 my @result_namespace = ($schema_class);
2395 if ($self->use_namespaces) {
2396 my $result_namespace = $self->result_namespace || 'Result';
2397 @result_namespace = $self->_result_namespace(
2402 my $table_class = join(q{::}, @result_namespace, $table_moniker);
2404 if ((my $upgrading_v = $self->_upgrading_from)
2405 || $self->_rewriting) {
2406 local $self->naming->{monikers} = $upgrading_v
2409 my @result_namespace = @result_namespace;
2410 if ($self->_upgrading_from_load_classes) {
2411 @result_namespace = ($schema_class);
2413 elsif (my $ns = $self->_downgrading_to_load_classes) {
2414 @result_namespace = $self->_result_namespace(
2419 elsif ($ns = $self->_rewriting_result_namespace) {
2420 @result_namespace = $self->_result_namespace(
2426 my $old_table_moniker = do {
2427 local $self->naming->{monikers} = $upgrading_v;
2428 $self->_table2moniker($table);
2431 my $old_class = join(q{::}, @result_namespace, $old_table_moniker);
2433 $self->_upgrading_classes->{$table_class} = $old_class
2434 unless $table_class eq $old_class;
2437 $self->classes->{$table->sql_name} = $table_class;
2438 $self->moniker_to_table->{$table_moniker} = $table;
2439 $self->class_to_table->{$table_class} = $table;
2441 $self->_pod_class_list($table_class, 'ADDITIONAL CLASSES USED', @{$self->additional_classes});
2443 $self->_use ($table_class, @{$self->additional_classes});
2445 $self->_pod_class_list($table_class, 'LEFT BASE CLASSES', @{$self->left_base_classes});
2447 $self->_inject($table_class, @{$self->left_base_classes});
2449 my @components = @{ $self->components || [] };
2451 push @components, @{ $self->result_components_map->{$table_moniker} }
2452 if exists $self->result_components_map->{$table_moniker};
2454 my @fq_components = @components;
2455 foreach my $component (@fq_components) {
2456 if ($component !~ s/^\+//) {
2457 $component = "DBIx::Class::$component";
2461 $self->_pod_class_list($table_class, 'COMPONENTS LOADED', @fq_components);
2463 $self->_dbic_stmt($table_class, 'load_components', @components) if @components;
2465 $self->_pod_class_list($table_class, 'ADDITIONAL BASE CLASSES', @{$self->additional_base_classes});
2467 $self->_inject($table_class, @{$self->additional_base_classes});
2470 sub _is_result_class_method {
2471 my ($self, $name, $table) = @_;
2473 my $table_moniker = $table ? $self->monikers->{$table->sql_name} : '';
2475 $self->_result_class_methods({})
2476 if not defined $self->_result_class_methods;
2478 if (not exists $self->_result_class_methods->{$table_moniker}) {
2479 my (@methods, %methods);
2480 my $base = $self->result_base_class || 'DBIx::Class::Core';
2482 my @components = @{ $self->components || [] };
2484 push @components, @{ $self->result_components_map->{$table_moniker} }
2485 if exists $self->result_components_map->{$table_moniker};
2487 for my $c (@components) {
2488 $c = $c =~ /^\+/ ? substr($c,1) : "DBIx::Class::$c";
2491 my @roles = @{ $self->result_roles || [] };
2493 push @roles, @{ $self->result_roles_map->{$table_moniker} }
2494 if exists $self->result_roles_map->{$table_moniker};
2497 $base, @components, @roles,
2498 ($self->use_moose ? 'Moose::Object' : ()),
2499 ($self->use_moo ? 'Moo::Object' : ()),
2501 $self->ensure_class_loaded($class);
2503 push @methods, @{ Class::Inspector->methods($class) || [] };
2506 push @methods, @{ Class::Inspector->methods('UNIVERSAL') };
2508 @methods{@methods} = ();
2510 $self->_result_class_methods->{$table_moniker} = \%methods;
2512 my $result_methods = $self->_result_class_methods->{$table_moniker};
2514 return exists $result_methods->{$name};
2517 sub _resolve_col_accessor_collisions {
2518 my ($self, $table, $col_info) = @_;
2520 while (my ($col, $info) = each %$col_info) {
2521 my $accessor = $info->{accessor} || $col;
2523 next if $accessor eq 'id'; # special case (very common column)
2525 if ($self->_is_result_class_method($accessor, $table)) {
2528 if (my $map = $self->col_collision_map) {
2529 for my $re (keys %$map) {
2530 if (my @matches = $col =~ /$re/) {
2531 $info->{accessor} = sprintf $map->{$re}, @matches;
2539 Column '$col' in table '$table' collides with an inherited method.
2540 See "COLUMN ACCESSOR COLLISIONS" in perldoc DBIx::Class::Schema::Loader::Base .
2542 $info->{accessor} = undef;
2548 # use the same logic to run moniker_map, col_accessor_map
2550 my ( $self, $map, $default_code, $ident, @extra ) = @_;
2552 my $default_ident = $default_code->( $ident, @extra );
2554 if( $map && ref $map eq 'HASH' ) {
2555 if (my @parts = try { @{ $ident } }) {
2556 my $part_map = $map;
2558 my $part = shift @parts;
2559 last unless exists $part_map->{ $part };
2560 if ( !ref $part_map->{ $part } && !@parts ) {
2561 $new_ident = $part_map->{ $part };
2564 elsif ( ref $part_map->{ $part } eq 'HASH' ) {
2565 $part_map = $part_map->{ $part };
2569 if( !$new_ident && !ref $map->{ $ident } ) {
2570 $new_ident = $map->{ $ident };
2573 elsif( $map && ref $map eq 'CODE' ) {
2576 croak "reentered map must be a hashref"
2577 unless 'HASH' eq ref($cb_map);
2578 return $self->_run_user_map($cb_map, $default_code, $ident, @extra);
2580 $new_ident = $map->( $ident, $default_ident, @extra, $cb );
2583 $new_ident ||= $default_ident;
2588 sub _default_column_accessor_name {
2589 my ( $self, $column_name ) = @_;
2591 my $preserve = ($self->naming->{column_accessors}||'') eq 'preserve';
2593 my $v = $self->_get_naming_v('column_accessors');
2595 my $accessor_name = $preserve ?
2596 $self->_to_identifier('column_accessors', $column_name) # assume CamelCase
2598 $self->_to_identifier('column_accessors', $column_name, '_');
2600 $accessor_name =~ s/\W+/_/g; # only if naming < v8, otherwise to_identifier
2604 return $accessor_name;
2606 elsif ($v < 7 || (not $self->preserve_case)) {
2607 # older naming just lc'd the col accessor and that's all.
2608 return lc $accessor_name;
2611 return join '_', map lc, split_name $column_name, $v;
2614 sub _make_column_accessor_name {
2615 my ($self, $column_name, $column_context_info ) = @_;
2617 my $accessor = $self->_run_user_map(
2618 $self->col_accessor_map,
2619 sub { $self->_default_column_accessor_name( shift ) },
2621 $column_context_info,
2627 sub _table_is_view {
2628 #my ($self, $table) = @_;
2632 sub _view_definition { undef }
2634 # Set up metadata (cols, pks, etc)
2635 sub _setup_src_meta {
2636 my ($self, $table) = @_;
2638 my $schema = $self->schema;
2639 my $schema_class = $self->schema_class;
2641 my $table_class = $self->classes->{$table->sql_name};
2642 my $table_moniker = $self->monikers->{$table->sql_name};
2644 # Must come before ->table
2645 $self->_dbic_stmt($table_class, 'table_class', 'DBIx::Class::ResultSource::View')
2646 if my $is_view = $self->_table_is_view($table);
2648 $self->_dbic_stmt($table_class, 'table', $table->dbic_name);
2650 # Must come after ->table
2651 if ($is_view and my $view_def = $self->_view_definition($table)) {
2652 $self->_dbic_stmt($table_class, 'result_source_instance->view_definition', $view_def);
2655 my $cols = $self->_table_columns($table);
2656 my $col_info = $self->__columns_info_for($table);
2658 ### generate all the column accessor names
2659 while (my ($col, $info) = each %$col_info) {
2660 # hashref of other info that could be used by
2661 # user-defined accessor map functions
2663 table_class => $table_class,
2664 table_moniker => $table_moniker,
2665 table_name => $table, # bugwards compatibility, RT#84050
2667 full_table_name => $table->dbic_name,
2668 schema_class => $schema_class,
2669 column_info => $info,
2671 my $col_obj = DBIx::Class::Schema::Loader::Column->new(
2676 $info->{accessor} = $self->_make_column_accessor_name( $col_obj, $context );
2679 $self->_resolve_col_accessor_collisions($table, $col_info);
2681 # prune any redundant accessor names
2682 while (my ($col, $info) = each %$col_info) {
2683 no warnings 'uninitialized';
2684 delete $info->{accessor} if $info->{accessor} eq $col;
2687 my $fks = $self->_table_fk_info($table);
2689 foreach my $fkdef (@$fks) {
2690 for my $col (@{ $fkdef->{local_columns} }) {
2691 $col_info->{$col}{is_foreign_key} = 1;
2695 my $pks = $self->_table_pk_info($table) || [];
2697 my %uniq_tag; # used to eliminate duplicate uniqs
2699 $uniq_tag{ join("\0", @$pks) }++ if @$pks; # pk is a uniq
2701 my $uniqs = $self->_table_uniq_info($table) || [];
2704 foreach my $uniq (@$uniqs) {
2705 my ($name, $cols) = @$uniq;
2706 next if $uniq_tag{ join("\0", @$cols) }++; # skip duplicates
2707 push @uniqs, [$name, $cols];
2710 my @non_nullable_uniqs = grep {
2711 all { $col_info->{$_}{is_nullable} == 0 } @{ $_->[1] }
2714 if ($self->uniq_to_primary && (not @$pks) && @non_nullable_uniqs) {
2715 my @by_colnum = sort { $b->[0] <=> $a->[0] }
2716 map [ scalar @{ $_->[1] }, $_ ], @non_nullable_uniqs;
2718 if (not (@by_colnum > 1 && $by_colnum[0][0] == $by_colnum[1][0])) {
2719 my @keys = map $_->[1], @by_colnum;
2723 # remove the uniq from list
2724 @uniqs = grep { $_->[0] ne $pk->[0] } @uniqs;
2730 foreach my $pkcol (@$pks) {
2731 $col_info->{$pkcol}{is_nullable} = 0;
2737 map { $_, ($col_info->{$_}||{}) } @$cols
2740 $self->_dbic_stmt($table_class, 'set_primary_key', @$pks)
2743 # Sort unique constraints by constraint name for repeatable results (rels
2744 # are sorted as well elsewhere.)
2745 @uniqs = sort { $a->[0] cmp $b->[0] } @uniqs;
2747 foreach my $uniq (@uniqs) {
2748 my ($name, $cols) = @$uniq;
2749 $self->_dbic_stmt($table_class,'add_unique_constraint', $name, $cols);
2753 sub __columns_info_for {
2754 my ($self, $table) = @_;
2756 my $result = $self->_columns_info_for($table);
2758 while (my ($col, $info) = each %$result) {
2759 $info = { %$info, %{ $self->_custom_column_info ($table, $col, $info) } };
2760 $info = { %$info, %{ $self->_datetime_column_info($table, $col, $info) } };
2762 $result->{$col} = $info;
2770 Returns a sorted list of loaded tables, using the original database table
2778 return values %{$self->_tables};
2782 my ($self, $naming_key) = @_;
2786 if (($self->naming->{$naming_key}||'') =~ /^v(\d+)\z/) {
2790 ($v) = $CURRENT_V =~ /^v(\d+)\z/;
2796 sub _to_identifier {
2797 my ($self, $naming_key, $name, $sep_char, $force) = @_;
2799 my $v = $self->_get_naming_v($naming_key);
2801 my $to_identifier = $self->naming->{force_ascii} ?
2802 \&String::ToIdentifier::EN::to_identifier
2803 : \&String::ToIdentifier::EN::Unicode::to_identifier;
2805 return $v >= 8 || $force ? $to_identifier->($name, $sep_char) : $name;
2808 # Make a moniker from a table
2809 sub _default_table2moniker {
2810 my ($self, $table) = @_;
2812 my $v = $self->_get_naming_v('monikers');
2814 my @moniker_parts = @{ $self->moniker_parts };
2815 my @name_parts = map $table->$_, @moniker_parts;
2817 my $name_idx = firstidx { $_ eq 'name' } @{ $self->moniker_parts };
2821 foreach my $i (0 .. $#name_parts) {
2822 my $part = $name_parts[$i];
2824 my $moniker_part = $self->_run_user_map(
2825 $self->moniker_part_map->{$moniker_parts[$i]},
2827 $part, $moniker_parts[$i],
2829 if (length $moniker_part) {
2830 push @all_parts, $moniker_part;
2834 if ($i != $name_idx || $v >= 8) {
2835 $part = $self->_to_identifier('monikers', $part, '_', 1);
2838 if ($i == $name_idx && $v == 5) {
2839 $part = Lingua::EN::Inflect::Number::to_S($part);
2842 my @part_parts = map lc, $v > 6 ?
2843 # use v8 semantics for all moniker parts except name
2844 ($i == $name_idx ? split_name $part, $v : split_name $part)
2845 : split /[\W_]+/, $part;
2847 if ($i == $name_idx && $v >= 6) {
2848 my $as_phrase = join ' ', @part_parts;
2850 my $inflected = ($self->naming->{monikers}||'') eq 'plural' ?
2851 Lingua::EN::Inflect::Phrase::to_PL($as_phrase)
2853 ($self->naming->{monikers}||'') eq 'preserve' ?
2856 Lingua::EN::Inflect::Phrase::to_S($as_phrase);
2858 @part_parts = split /\s+/, $inflected;
2861 push @all_parts, join '', map ucfirst, @part_parts;
2864 return join $self->moniker_part_separator, @all_parts;
2867 sub _table2moniker {
2868 my ( $self, $table ) = @_;
2870 $self->_run_user_map(
2872 sub { $self->_default_table2moniker( shift ) },
2877 sub _load_relationships {
2878 my ($self, $tables) = @_;
2882 foreach my $table (@$tables) {
2883 my $local_moniker = $self->monikers->{$table->sql_name};
2885 my $tbl_fk_info = $self->_table_fk_info($table);
2887 foreach my $fkdef (@$tbl_fk_info) {
2888 $fkdef->{local_table} = $table;
2889 $fkdef->{local_moniker} = $local_moniker;
2890 $fkdef->{remote_source} =
2891 $self->monikers->{$fkdef->{remote_table}->sql_name};
2893 my $tbl_uniq_info = $self->_table_uniq_info($table);
2895 push @tables, [ $local_moniker, $tbl_fk_info, $tbl_uniq_info ];
2898 my $rel_stmts = $self->_relbuilder->generate_code(\@tables);
2900 foreach my $src_class (sort keys %$rel_stmts) {
2902 my @src_stmts = map $_->[2],
2908 ($_->{method} eq 'many_to_many' ? 1 : 0),
2911 ], @{ $rel_stmts->{$src_class} };
2913 foreach my $stmt (@src_stmts) {
2914 $self->_dbic_stmt($src_class,$stmt->{method}, @{$stmt->{args}});
2920 my ($self, $table) = @_;
2922 my $table_moniker = $self->monikers->{$table->sql_name};
2923 my $table_class = $self->classes->{$table->sql_name};
2925 my @roles = @{ $self->result_roles || [] };
2926 push @roles, @{ $self->result_roles_map->{$table_moniker} }
2927 if exists $self->result_roles_map->{$table_moniker};
2930 my $class = $self->use_moose ? 'Moose' : 'Moo';
2931 $self->_pod_class_list($table_class, "L<$class> ROLES APPLIED", @roles);
2933 $self->_with($table_class, @roles);
2937 # Overload these in driver class:
2939 # Returns an arrayref of column names
2940 sub _table_columns { croak "ABSTRACT METHOD" }
2942 # Returns arrayref of pk col names
2943 sub _table_pk_info { croak "ABSTRACT METHOD" }
2945 # Returns an arrayref of uniqs [ [ foo => [ col1, col2 ] ], [ bar => [ ... ] ] ]
2946 sub _table_uniq_info { croak "ABSTRACT METHOD" }
2948 # Returns an arrayref of foreign key constraints, each
2949 # being a hashref with 3 keys:
2950 # local_columns (arrayref), remote_columns (arrayref), remote_table
2951 sub _table_fk_info { croak "ABSTRACT METHOD" }
2953 # Returns an array of lower case table names
2954 sub _tables_list { croak "ABSTRACT METHOD" }
2956 # Execute a constructive DBIC class method, with debug/dump_to_dir hooks.
2962 # generate the pod for this statement, storing it with $self->_pod
2963 $self->_make_pod( $class, $method, @_ ) if $self->generate_pod;
2965 my $args = dump(@_);
2966 $args = '(' . $args . ')' if @_ < 2;
2967 my $stmt = $method . $args . q{;};
2969 warn qq|$class\->$stmt\n| if $self->debug;
2970 $self->_raw_stmt($class, '__PACKAGE__->' . $stmt);
2974 sub _make_pod_heading {
2975 my ($self, $class) = @_;
2977 return '' if not $self->generate_pod;
2979 my $table = $self->class_to_table->{$class};
2982 my $pcm = $self->pod_comment_mode;
2983 my ($comment, $comment_overflows, $comment_in_name, $comment_in_desc);
2984 $comment = $self->__table_comment($table);
2985 $comment_overflows = ($comment and length $comment > $self->pod_comment_spillover_length);
2986 $comment_in_name = ($pcm eq 'name' or ($pcm eq 'auto' and !$comment_overflows));
2987 $comment_in_desc = ($pcm eq 'description' or ($pcm eq 'auto' and $comment_overflows));
2989 $pod .= "=head1 NAME\n\n";
2991 my $table_descr = $class;
2992 $table_descr .= " - " . $comment if $comment and $comment_in_name;
2994 $pod .= "$table_descr\n\n";
2996 if ($comment and $comment_in_desc) {
2997 $pod .= "=head1 DESCRIPTION\n\n${comment}\n\n";
3004 # generates the accompanying pod for a DBIC class method statement,
3005 # storing it with $self->_pod
3011 if ($method eq 'table') {
3013 $table = $$table if ref $table eq 'SCALAR';
3014 $self->_pod($class, "=head1 TABLE: C<$table>");
3015 $self->_pod_cut($class);
3017 elsif ( $method eq 'add_columns' ) {
3018 $self->_pod( $class, "=head1 ACCESSORS" );
3019 my $col_counter = 0;
3021 while( my ($name,$attrs) = splice @cols,0,2 ) {
3023 $self->_pod( $class, '=head2 ' . $name );
3024 $self->_pod( $class,
3026 my $s = $attrs->{$_};
3027 $s = !defined $s ? 'undef' :
3028 length($s) == 0 ? '(empty string)' :
3029 ref($s) eq 'SCALAR' ? $$s :
3030 ref($s) ? dumper_squashed $s :
3031 looks_like_number($s) ? $s : qq{'$s'};
3034 } sort keys %$attrs,
3036 if (my $comment = $self->__column_comment($self->class_to_table->{$class}, $col_counter, $name)) {
3037 $self->_pod( $class, $comment );
3040 $self->_pod_cut( $class );
3041 } elsif ( $method =~ /^(?:belongs_to|has_many|might_have)\z/ ) {
3042 $self->_pod( $class, "=head1 RELATIONS" ) unless $self->{_relations_started} { $class } ;
3043 my ( $accessor, $rel_class ) = @_;
3044 $self->_pod( $class, "=head2 $accessor" );
3045 $self->_pod( $class, 'Type: ' . $method );
3046 $self->_pod( $class, "Related object: L<$rel_class>" );
3047 $self->_pod_cut( $class );
3048 $self->{_relations_started} { $class } = 1;
3049 } elsif ( $method eq 'many_to_many' ) {
3050 $self->_pod( $class, "=head1 RELATIONS" ) unless $self->{_relations_started} { $class } ;
3051 my ( $accessor, $rel1, $rel2 ) = @_;
3052 $self->_pod( $class, "=head2 $accessor" );
3053 $self->_pod( $class, 'Type: many_to_many' );
3054 $self->_pod( $class, "Composing rels: L</$rel1> -> $rel2" );
3055 $self->_pod_cut( $class );
3056 $self->{_relations_started} { $class } = 1;
3058 elsif ($method eq 'add_unique_constraint') {
3059 $self->_pod($class, '=head1 UNIQUE CONSTRAINTS')
3060 unless $self->{_uniqs_started}{$class};
3062 my ($name, $cols) = @_;
3064 $self->_pod($class, "=head2 C<$name>");
3065 $self->_pod($class, '=over 4');
3067 foreach my $col (@$cols) {
3068 $self->_pod($class, "=item \* L</$col>");
3071 $self->_pod($class, '=back');
3072 $self->_pod_cut($class);
3074 $self->{_uniqs_started}{$class} = 1;
3076 elsif ($method eq 'set_primary_key') {
3077 $self->_pod($class, "=head1 PRIMARY KEY");
3078 $self->_pod($class, '=over 4');
3080 foreach my $col (@_) {
3081 $self->_pod($class, "=item \* L</$col>");
3084 $self->_pod($class, '=back');
3085 $self->_pod_cut($class);
3089 sub _pod_class_list {
3090 my ($self, $class, $title, @classes) = @_;
3092 return unless @classes && $self->generate_pod;
3094 $self->_pod($class, "=head1 $title");
3095 $self->_pod($class, '=over 4');
3097 foreach my $link (@classes) {
3098 $self->_pod($class, "=item * L<$link>");
3101 $self->_pod($class, '=back');
3102 $self->_pod_cut($class);
3105 sub _base_class_pod {
3106 my ($self, $base_class) = @_;
3108 return '' unless $self->generate_pod;
3110 return "\n=head1 BASE CLASS: L<$base_class>\n\n=cut\n\n";
3113 sub _filter_comment {
3114 my ($self, $txt) = @_;
3116 $txt = '' if not defined $txt;
3118 $txt =~ s/(?:\015?\012|\015\012?)/\n/g;
3123 sub __table_comment {
3126 if (my $code = $self->can('_table_comment')) {
3127 return $self->_filter_comment($self->$code(@_));
3133 sub __column_comment {
3136 if (my $code = $self->can('_column_comment')) {
3137 return $self->_filter_comment($self->$code(@_));
3143 # Stores a POD documentation
3145 my ($self, $class, $stmt) = @_;
3146 $self->_raw_stmt( $class, "\n" . $stmt );
3150 my ($self, $class ) = @_;
3151 $self->_raw_stmt( $class, "\n=cut\n" );
3154 # Store a raw source line for a class (for dumping purposes)
3156 my ($self, $class, $stmt) = @_;
3157 push(@{$self->{_dump_storage}->{$class}}, $stmt);
3160 # Like above, but separately for the externally loaded stuff
3162 my ($self, $class, $stmt) = @_;
3163 push(@{$self->{_ext_storage}->{$class}}, $stmt);
3166 sub _custom_column_info {
3167 my ( $self, $table_name, $column_name, $column_info ) = @_;
3169 if (my $code = $self->custom_column_info) {
3170 return $code->($table_name, $column_name, $column_info) || {};
3175 sub _datetime_column_info {
3176 my ( $self, $table_name, $column_name, $column_info ) = @_;
3178 my $type = $column_info->{data_type} || '';
3179 if ((grep $_, @{ $column_info }{map "inflate_$_", qw/date datetime timestamp/})
3180 or ($type =~ /date|timestamp/i)) {
3181 $result->{timezone} = $self->datetime_timezone if $self->datetime_timezone;
3182 $result->{locale} = $self->datetime_locale if $self->datetime_locale;
3188 my ($self, $name) = @_;
3190 return $self->preserve_case ? $name : lc($name);
3194 my ($self, $name) = @_;
3196 return $self->preserve_case ? $name : uc($name);
3200 my ($self, $table) = @_;
3203 my $schema = $self->schema;
3204 # in older DBIC it's a private method
3205 my $unregister = $schema->can('unregister_source') || $schema->can('_unregister_source');
3206 $schema->$unregister(delete $self->monikers->{$table->sql_name});
3207 delete $self->_upgrading_classes->{delete $self->classes->{$table->sql_name}};
3208 delete $self->_tables->{$table->sql_name};
3212 # remove the dump dir from @INC on destruction
3216 @INC = grep $_ ne $self->dump_directory, @INC;
3221 Returns a hashref of loaded table to moniker mappings. There will
3222 be two entries for each table, the original name and the "normalized"
3223 name, in the case that the two are different (such as databases
3224 that like uppercase table names, or preserve your original mixed-case
3225 definitions, or what-have-you).
3229 Returns a hashref of table to class mappings. In some cases it will
3230 contain multiple entries per table for the original and normalized table
3231 names, as above in L</monikers>.
3233 =head2 generated_classes
3235 Returns an arrayref of classes that were actually generated (i.e. not
3236 skipped because there were no changes).
3238 =head1 NON-ENGLISH DATABASES
3240 If you use the loader on a database with table and column names in a language
3241 other than English, you will want to turn off the English language specific
3244 To do so, use something like this in your loader options:
3246 naming => { monikers => 'v4' },
3247 inflect_singular => sub { "$_[0]_rel" },
3248 inflect_plural => sub { "$_[0]_rel" },
3250 =head1 COLUMN ACCESSOR COLLISIONS
3252 Occasionally you may have a column name that collides with a perl method, such
3253 as C<can>. In such cases, the default action is to set the C<accessor> of the
3254 column spec to C<undef>.
3256 You can then name the accessor yourself by placing code such as the following
3259 __PACKAGE__->add_column('+can' => { accessor => 'my_can' });
3261 Another option is to use the L</col_collision_map> option.
3263 =head1 RELATIONSHIP NAME COLLISIONS
3265 In very rare cases, you may get a collision between a generated relationship
3266 name and a method in your Result class, for example if you have a foreign key
3267 called C<belongs_to>.
3269 This is a problem because relationship names are also relationship accessor
3270 methods in L<DBIx::Class>.
3272 The default behavior is to append C<_rel> to the relationship name and print
3273 out a warning that refers to this text.
3275 You can also control the renaming with the L</rel_collision_map> option.
3279 L<DBIx::Class::Schema::Loader>, L<dbicdump>
3283 See L<DBIx::Class::Schema::Loader/AUTHORS>.
3287 This library is free software; you can redistribute it and/or modify it under
3288 the same terms as Perl itself.
3293 # vim:et sts=4 sw=4 tw=0: