1 package DBIx::Class::Schema::Loader::Base;
5 use base qw/Class::Accessor::Grouped Class::C3::Componentised/;
7 use Carp::Clan qw/^DBIx::Class/;
8 use DBIx::Class::Schema::Loader::RelBuilder;
9 use Data::Dump qw/ dump /;
14 use Lingua::EN::Inflect::Number qw//;
15 use Lingua::EN::Inflect::Phrase qw//;
18 use Class::Inspector ();
19 use Scalar::Util 'looks_like_number';
20 use File::Slurp 'slurp';
21 use DBIx::Class::Schema::Loader::Utils qw/split_name dumper_squashed eval_without_redefine_warnings/;
22 use DBIx::Class::Schema::Loader::Optional::Dependencies ();
25 use Class::Load 'load_class';
28 our $VERSION = '0.07007';
30 __PACKAGE__->mk_group_ro_accessors('simple', qw/
37 additional_base_classes
52 default_resultset_class
56 overwrite_modifications
75 __PACKAGE__->mk_group_accessors('simple', qw/
77 schema_version_to_dump
79 _upgrading_from_load_classes
80 _downgrading_to_load_classes
81 _rewriting_result_namespace
86 pod_comment_spillover_length
91 datetime_undef_if_invalid
97 DBIx::Class::Schema::Loader::Base - Base DBIx::Class::Schema::Loader Implementation.
101 See L<DBIx::Class::Schema::Loader>
105 This is the base class for the storage-specific C<DBIx::Class::Schema::*>
106 classes, and implements the common functionality between them.
108 =head1 CONSTRUCTOR OPTIONS
110 These constructor options are the base options for
111 L<DBIx::Class::Schema::Loader/loader_options>. Available constructor options are:
113 =head2 skip_relationships
115 Skip setting up relationships. The default is to attempt the loading
118 =head2 skip_load_external
120 Skip loading of other classes in @INC. The default is to merge all other classes
121 with the same name found in @INC into the schema file we are creating.
125 Static schemas (ones dumped to disk) will, by default, use the new-style
126 relationship names and singularized Results, unless you're overwriting an
127 existing dump made by an older version of L<DBIx::Class::Schema::Loader>, in
128 which case the backward compatible RelBuilder will be activated, and the
129 appropriate monikerization used.
135 will disable the backward-compatible RelBuilder and use
136 the new-style relationship names along with singularized Results, even when
137 overwriting a dump made with an earlier version.
139 The option also takes a hashref:
141 naming => { relationships => 'v7', monikers => 'v7' }
149 How to name relationship accessors.
153 How to name Result classes.
155 =item column_accessors
157 How to name column accessors in Result classes.
167 Latest style, whatever that happens to be.
171 Unsingularlized monikers, C<has_many> only relationships with no _id stripping.
175 Monikers singularized as whole words, C<might_have> relationships for FKs on
176 C<UNIQUE> constraints, C<_id> stripping for belongs_to relationships.
178 Some of the C<_id> stripping edge cases in C<0.05003> have been reverted for
183 All monikers and relationships are inflected using
184 L<Lingua::EN::Inflect::Phrase>, and there is more aggressive C<_id> stripping
185 from relationship names.
187 In general, there is very little difference between v5 and v6 schemas.
191 This mode is identical to C<v6> mode, except that monikerization of CamelCase
192 table names is also done correctly.
194 CamelCase column names in case-preserving mode will also be handled correctly
195 for relationship name inflection. See L</preserve_case>.
197 In this mode, CamelCase L</column_accessors> are normalized based on case
198 transition instead of just being lowercased, so C<FooId> becomes C<foo_id>.
200 If you don't have any CamelCase table or column names, you can upgrade without
201 breaking any of your code.
205 Dynamic schemas will always default to the 0.04XXX relationship names and won't
206 singularize Results for backward compatibility, to activate the new RelBuilder
207 and singularization put this in your C<Schema.pm> file:
209 __PACKAGE__->naming('current');
211 Or if you prefer to use 0.07XXX features but insure that nothing breaks in the
212 next major version upgrade:
214 __PACKAGE__->naming('v7');
218 By default POD will be generated for columns and relationships, using database
219 metadata for the text if available and supported.
221 Reading database metadata (e.g. C<COMMENT ON TABLE some_table ...>) is only
222 supported for Postgres right now.
224 Set this to C<0> to turn off all POD generation.
226 =head2 pod_comment_mode
228 Controls where table comments appear in the generated POD. Smaller table
229 comments are appended to the C<NAME> section of the documentation, and larger
230 ones are inserted into C<DESCRIPTION> instead. You can force a C<DESCRIPTION>
231 section to be generated with the comment always, only use C<NAME>, or choose
232 the length threshold at which the comment is forced into the description.
238 Use C<NAME> section only.
242 Force C<DESCRIPTION> always.
246 Use C<DESCRIPTION> if length > L</pod_comment_spillover_length>, this is the
251 =head2 pod_comment_spillover_length
253 When pod_comment_mode is set to C<auto>, this is the length of the comment at
254 which it will be forced into a separate description section.
258 =head2 relationship_attrs
260 Hashref of attributes to pass to each generated relationship, listed
261 by type. Also supports relationship type 'all', containing options to
262 pass to all generated relationships. Attributes set for more specific
263 relationship types override those set in 'all'.
267 relationship_attrs => {
268 belongs_to => { is_deferrable => 0 },
271 use this to turn off DEFERRABLE on your foreign key constraints.
275 If set to true, each constructive L<DBIx::Class> statement the loader
276 decides to execute will be C<warn>-ed before execution.
280 Set the name of the schema to load (schema in the sense that your database
281 vendor means it). Does not currently support loading more than one schema
286 Only load tables matching regex. Best specified as a qr// regex.
290 Exclude tables matching regex. Best specified as a qr// regex.
294 Overrides the default table name to moniker translation. Can be either
295 a hashref of table keys and moniker values, or a coderef for a translator
296 function taking a single scalar table name argument and returning
297 a scalar moniker. If the hash entry does not exist, or the function
298 returns a false value, the code falls back to default behavior
301 The default behavior is to split on case transition and non-alphanumeric
302 boundaries, singularize the resulting phrase, then join the titlecased words
305 Table Name | Moniker Name
306 ---------------------------------
308 luser_group | LuserGroup
309 luser-opts | LuserOpt
310 stations_visited | StationVisited
311 routeChange | RouteChange
313 =head2 col_accessor_map
315 Same as moniker_map, but for column accessor names. If a coderef is
316 passed, the code is called with arguments of
318 the name of the column in the underlying database,
319 default accessor name that DBICSL would ordinarily give this column,
321 table_class => name of the DBIC class we are building,
322 table_moniker => calculated moniker for this table (after moniker_map if present),
323 table_name => name of the database table,
324 full_table_name => schema-qualified name of the database table (RDBMS specific),
325 schema_class => name of the schema class we are building,
326 column_info => hashref of column info (data_type, is_nullable, etc),
329 =head2 inflect_plural
331 Just like L</moniker_map> above (can be hash/code-ref, falls back to default
332 if hash key does not exist or coderef returns false), but acts as a map
333 for pluralizing relationship names. The default behavior is to utilize
334 L<Lingua::EN::Inflect::Phrase/to_PL>.
336 =head2 inflect_singular
338 As L</inflect_plural> above, but for singularizing relationship names.
339 Default behavior is to utilize L<Lingua::EN::Inflect::Phrase/to_S>.
341 =head2 schema_base_class
343 Base class for your schema classes. Defaults to 'DBIx::Class::Schema'.
345 =head2 result_base_class
347 Base class for your table classes (aka result classes). Defaults to
350 =head2 additional_base_classes
352 List of additional base classes all of your table classes will use.
354 =head2 left_base_classes
356 List of additional base classes all of your table classes will use
357 that need to be leftmost.
359 =head2 additional_classes
361 List of additional classes which all of your table classes will use.
365 List of additional components to be loaded into all of your table
366 classes. A good example would be
367 L<InflateColumn::DateTime|DBIx::Class::InflateColumn::DateTime>
369 =head2 use_namespaces
371 This is now the default, to go back to L<DBIx::Class::Schema/load_classes> pass
374 Generate result class names suitable for
375 L<DBIx::Class::Schema/load_namespaces> and call that instead of
376 L<DBIx::Class::Schema/load_classes>. When using this option you can also
377 specify any of the options for C<load_namespaces> (i.e. C<result_namespace>,
378 C<resultset_namespace>, C<default_resultset_class>), and they will be added
379 to the call (and the generated result class names adjusted appropriately).
381 =head2 dump_directory
383 The value of this option is a perl libdir pathname. Within
384 that directory this module will create a baseline manual
385 L<DBIx::Class::Schema> module set, based on what it creates at runtime.
387 The created schema class will have the same classname as the one on
388 which you are setting this option (and the ResultSource classes will be
389 based on this name as well).
391 Normally you wouldn't hard-code this setting in your schema class, as it
392 is meant for one-time manual usage.
394 See L<DBIx::Class::Schema::Loader/dump_to_dir> for examples of the
395 recommended way to access this functionality.
397 =head2 dump_overwrite
399 Deprecated. See L</really_erase_my_files> below, which does *not* mean
400 the same thing as the old C<dump_overwrite> setting from previous releases.
402 =head2 really_erase_my_files
404 Default false. If true, Loader will unconditionally delete any existing
405 files before creating the new ones from scratch when dumping a schema to disk.
407 The default behavior is instead to only replace the top portion of the
408 file, up to and including the final stanza which contains
409 C<# DO NOT MODIFY THE FIRST PART OF THIS FILE>
410 leaving any customizations you placed after that as they were.
412 When C<really_erase_my_files> is not set, if the output file already exists,
413 but the aforementioned final stanza is not found, or the checksum
414 contained there does not match the generated contents, Loader will
415 croak and not touch the file.
417 You should really be using version control on your schema classes (and all
418 of the rest of your code for that matter). Don't blame me if a bug in this
419 code wipes something out when it shouldn't have, you've been warned.
421 =head2 overwrite_modifications
423 Default false. If false, when updating existing files, Loader will
424 refuse to modify any Loader-generated code that has been modified
425 since its last run (as determined by the checksum Loader put in its
428 If true, Loader will discard any manual modifications that have been
429 made to Loader-generated code.
431 Again, you should be using version control on your schema classes. Be
432 careful with this option.
434 =head2 custom_column_info
436 Hook for adding extra attributes to the
437 L<column_info|DBIx::Class::ResultSource/column_info> for a column.
439 Must be a coderef that returns a hashref with the extra attributes.
441 Receives the table name, column name and column_info.
445 custom_column_info => sub {
446 my ($table_name, $column_name, $column_info) = @_;
448 if ($column_name eq 'dog' && $column_info->{default_value} eq 'snoopy') {
449 return { is_snoopy => 1 };
453 This attribute can also be used to set C<inflate_datetime> on a non-datetime
454 column so it also receives the L</datetime_timezone> and/or L</datetime_locale>.
456 =head2 datetime_timezone
458 Sets the timezone attribute for L<DBIx::Class::InflateColumn::DateTime> for all
459 columns with the DATE/DATETIME/TIMESTAMP data_types.
461 =head2 datetime_locale
463 Sets the locale attribute for L<DBIx::Class::InflateColumn::DateTime> for all
464 columns with the DATE/DATETIME/TIMESTAMP data_types.
466 =head2 datetime_undef_if_invalid
468 Pass a C<0> for this option when using MySQL if you B<DON'T> want C<<
469 datetime_undef_if_invalid => 1 >> in your column info for DATE, DATETIME and
472 The default is recommended to deal with data such as C<00/00/00> which
473 sometimes ends up in such columns in MySQL.
477 File in Perl format, which should return a HASH reference, from which to read
482 Usually column names are lowercased, to make them easier to work with in
483 L<DBIx::Class>. This option lets you turn this behavior off, if the driver
486 Drivers for case sensitive databases like Sybase ASE or MSSQL with a
487 case-sensitive collation will turn this option on unconditionally.
489 Currently the drivers for SQLite, mysql, MSSQL and Firebird/InterBase support
492 =head2 qualify_objects
494 Set to true to prepend the L</db_schema> to table names for C<<
495 __PACKAGE__->table >> calls, and to some other things like Oracle sequences.
499 Creates Schema and Result classes that use L<Moose>, L<MooseX::NonMoose> and
500 L<namespace::autoclean>. The default content after the md5 sum also makes the
503 It is safe to upgrade your existing Schema to this option.
505 =head2 col_collision_map
507 This option controls how accessors for column names which collide with perl
508 methods are named. See L</COLUMN ACCESSOR COLLISIONS> for more information.
510 This option takes either a single L<sprintf|perlfunc/sprintf> format or a hashref of
511 strings which are compiled to regular expressions that map to
512 L<sprintf|perlfunc/sprintf> formats.
516 col_collision_map => 'column_%s'
518 col_collision_map => { '(.*)' => 'column_%s' }
520 col_collision_map => { '(foo).*(bar)' => 'column_%s_%s' }
522 =head2 rel_collision_map
524 Works just like L</col_collision_map>, but for relationship names/accessors
525 rather than column names/accessors.
527 The default is to just append C<_rel> to the relationship name, see
528 L</RELATIONSHIP NAME COLLISIONS>.
532 None of these methods are intended for direct invocation by regular
533 users of L<DBIx::Class::Schema::Loader>. Some are proxied via
534 L<DBIx::Class::Schema::Loader>.
538 my $CURRENT_V = 'v7';
541 schema_base_class result_base_class additional_base_classes
542 left_base_classes additional_classes components
545 # ensure that a peice of object data is a valid arrayref, creating
546 # an empty one or encapsulating whatever's there.
547 sub _ensure_arrayref {
552 $self->{$_} = [ $self->{$_} ]
553 unless ref $self->{$_} eq 'ARRAY';
559 Constructor for L<DBIx::Class::Schema::Loader::Base>, used internally
560 by L<DBIx::Class::Schema::Loader>.
565 my ( $class, %args ) = @_;
567 if (exists $args{column_accessor_map}) {
568 $args{col_accessor_map} = delete $args{column_accessor_map};
571 my $self = { %args };
573 # don't lose undef options
574 for (values %$self) {
575 $_ = 0 unless defined $_;
578 bless $self => $class;
580 if (my $config_file = $self->config_file) {
581 my $config_opts = do $config_file;
583 croak "Error reading config from $config_file: $@" if $@;
585 croak "Config file $config_file must be a hashref" unless ref($config_opts) eq 'HASH';
587 while (my ($k, $v) = each %$config_opts) {
588 $self->{$k} = $v unless exists $self->{$k};
592 $self->_ensure_arrayref(qw/additional_classes
593 additional_base_classes
598 $self->_validate_class_args;
600 if ($self->use_moose) {
601 if (not DBIx::Class::Schema::Loader::Optional::Dependencies->req_ok_for('use_moose')) {
602 die sprintf "You must install the following CPAN modules to enable the use_moose option: %s.\n",
603 DBIx::Class::Schema::Loader::Optional::Dependencies->req_missing_for('use_moose');
607 $self->{monikers} = {};
608 $self->{classes} = {};
609 $self->{_upgrading_classes} = {};
611 $self->{schema_class} ||= ( ref $self->{schema} || $self->{schema} );
612 $self->{schema} ||= $self->{schema_class};
614 croak "dump_overwrite is deprecated. Please read the"
615 . " DBIx::Class::Schema::Loader::Base documentation"
616 if $self->{dump_overwrite};
618 $self->{dynamic} = ! $self->{dump_directory};
619 $self->{temp_directory} ||= File::Temp::tempdir( 'dbicXXXX',
624 $self->{dump_directory} ||= $self->{temp_directory};
626 $self->real_dump_directory($self->{dump_directory});
628 $self->version_to_dump($DBIx::Class::Schema::Loader::VERSION);
629 $self->schema_version_to_dump($DBIx::Class::Schema::Loader::VERSION);
631 if ((not ref $self->naming) && defined $self->naming) {
632 my $naming_ver = $self->naming;
634 relationships => $naming_ver,
635 monikers => $naming_ver,
636 column_accessors => $naming_ver,
641 for (values %{ $self->naming }) {
642 $_ = $CURRENT_V if $_ eq 'current';
645 $self->{naming} ||= {};
647 if ($self->custom_column_info && ref $self->custom_column_info ne 'CODE') {
648 croak 'custom_column_info must be a CODE ref';
651 $self->_check_back_compat;
653 $self->use_namespaces(1) unless defined $self->use_namespaces;
654 $self->generate_pod(1) unless defined $self->generate_pod;
655 $self->pod_comment_mode('auto') unless defined $self->pod_comment_mode;
656 $self->pod_comment_spillover_length(60) unless defined $self->pod_comment_spillover_length;
658 if (my $col_collision_map = $self->col_collision_map) {
659 if (my $reftype = ref $col_collision_map) {
660 if ($reftype ne 'HASH') {
661 croak "Invalid type $reftype for option 'col_collision_map'";
665 $self->col_collision_map({ '(.*)' => $col_collision_map });
672 sub _check_back_compat {
675 # dynamic schemas will always be in 0.04006 mode, unless overridden
676 if ($self->dynamic) {
677 # just in case, though no one is likely to dump a dynamic schema
678 $self->schema_version_to_dump('0.04006');
680 if (not %{ $self->naming }) {
681 warn <<EOF unless $ENV{SCHEMA_LOADER_BACKCOMPAT};
683 Dynamic schema detected, will run in 0.04006 mode.
685 Set the 'naming' attribute or the SCHEMA_LOADER_BACKCOMPAT environment variable
686 to disable this warning.
688 Also consider setting 'use_namespaces => 1' if/when upgrading.
690 See perldoc DBIx::Class::Schema::Loader::Manual::UpgradingFromV4 for more
695 $self->_upgrading_from('v4');
698 $self->naming->{relationships} ||= 'v4';
699 $self->naming->{monikers} ||= 'v4';
701 if ($self->use_namespaces) {
702 $self->_upgrading_from_load_classes(1);
705 $self->use_namespaces(0);
711 # otherwise check if we need backcompat mode for a static schema
712 my $filename = $self->_get_dump_filename($self->schema_class);
713 return unless -e $filename;
715 my ($old_gen, $old_md5, $old_ver, $old_ts, $old_custom) =
716 $self->_parse_generated_file($filename);
718 return unless $old_ver;
720 # determine if the existing schema was dumped with use_moose => 1
721 if (! defined $self->use_moose) {
722 $self->{use_moose} = 1 if $old_gen =~ /^ (?!\s*\#) use \s+ Moose/xm;
725 my $load_classes = ($old_gen =~ /^__PACKAGE__->load_classes;/m) ? 1 : 0;
726 my $result_namespace = do { ($old_gen =~ /result_namespace => '([^']+)'/) ? $1 : '' };
728 if ($load_classes && (not defined $self->use_namespaces)) {
729 warn <<"EOF" unless $ENV{SCHEMA_LOADER_BACKCOMPAT};
731 'load_classes;' static schema detected, turning off 'use_namespaces'.
733 Set the 'use_namespaces' attribute or the SCHEMA_LOADER_BACKCOMPAT environment
734 variable to disable this warning.
736 See perldoc DBIx::Class::Schema::Loader::Manual::UpgradingFromV4 for more
739 $self->use_namespaces(0);
741 elsif ($load_classes && $self->use_namespaces) {
742 $self->_upgrading_from_load_classes(1);
744 elsif ((not $load_classes) && defined $self->use_namespaces && ! $self->use_namespaces) {
745 $self->_downgrading_to_load_classes(
746 $result_namespace || 'Result'
749 elsif ((not defined $self->use_namespaces) || $self->use_namespaces) {
750 if (not $self->result_namespace) {
751 $self->result_namespace($result_namespace || 'Result');
753 elsif ($result_namespace ne $self->result_namespace) {
754 $self->_rewriting_result_namespace(
755 $result_namespace || 'Result'
760 # XXX when we go past .0 this will need fixing
761 my ($v) = $old_ver =~ /([1-9])/;
764 return if ($v eq $CURRENT_V || $old_ver =~ /^0\.\d\d999/);
766 if (not %{ $self->naming }) {
767 warn <<"EOF" unless $ENV{SCHEMA_LOADER_BACKCOMPAT};
769 Version $old_ver static schema detected, turning on backcompat mode.
771 Set the 'naming' attribute or the SCHEMA_LOADER_BACKCOMPAT environment variable
772 to disable this warning.
774 See: 'naming' in perldoc DBIx::Class::Schema::Loader::Base .
776 See perldoc DBIx::Class::Schema::Loader::Manual::UpgradingFromV4 if upgrading
777 from version 0.04006.
780 $self->naming->{relationships} ||= $v;
781 $self->naming->{monikers} ||= $v;
782 $self->naming->{column_accessors} ||= $v;
784 $self->schema_version_to_dump($old_ver);
787 $self->_upgrading_from($v);
791 sub _validate_class_args {
795 foreach my $k (@CLASS_ARGS) {
796 next unless $self->$k;
798 my @classes = ref $self->$k eq 'ARRAY' ? @{ $self->$k } : $self->$k;
799 foreach my $c (@classes) {
800 # components default to being under the DBIx::Class namespace unless they
801 # are preceeded with a '+'
802 if ( $k =~ m/components$/ && $c !~ s/^\+// ) {
803 $c = 'DBIx::Class::' . $c;
806 # 1 == installed, 0 == not installed, undef == invalid classname
807 my $installed = Class::Inspector->installed($c);
808 if ( defined($installed) ) {
809 if ( $installed == 0 ) {
810 croak qq/$c, as specified in the loader option "$k", is not installed/;
813 croak qq/$c, as specified in the loader option "$k", is an invalid class name/;
819 sub _find_file_in_inc {
820 my ($self, $file) = @_;
822 foreach my $prefix (@INC) {
823 my $fullpath = File::Spec->catfile($prefix, $file);
824 return $fullpath if -f $fullpath
825 # abs_path throws on Windows for nonexistant files
826 and (try { Cwd::abs_path($fullpath) }) ne
827 ((try { Cwd::abs_path(File::Spec->catfile($self->dump_directory, $file)) }) || '');
834 my ($self, $class) = @_;
836 my $class_path = $class;
837 $class_path =~ s{::}{/}g;
838 $class_path .= '.pm';
843 sub _find_class_in_inc {
844 my ($self, $class) = @_;
846 return $self->_find_file_in_inc($self->_class_path($class));
852 return $self->_upgrading_from
853 || $self->_upgrading_from_load_classes
854 || $self->_downgrading_to_load_classes
855 || $self->_rewriting_result_namespace
859 sub _rewrite_old_classnames {
860 my ($self, $code) = @_;
862 return $code unless $self->_rewriting;
864 my %old_classes = reverse %{ $self->_upgrading_classes };
866 my $re = join '|', keys %old_classes;
869 $code =~ s/$re/$old_classes{$1} || $1/eg;
875 my ($self, $class) = @_;
877 return if $self->{skip_load_external};
879 # so that we don't load our own classes, under any circumstances
880 local *INC = [ grep $_ ne $self->dump_directory, @INC ];
882 my $real_inc_path = $self->_find_class_in_inc($class);
884 my $old_class = $self->_upgrading_classes->{$class}
885 if $self->_rewriting;
887 my $old_real_inc_path = $self->_find_class_in_inc($old_class)
888 if $old_class && $old_class ne $class;
890 return unless $real_inc_path || $old_real_inc_path;
892 if ($real_inc_path) {
893 # If we make it to here, we loaded an external definition
894 warn qq/# Loaded external class definition for '$class'\n/
897 my $code = $self->_rewrite_old_classnames(scalar slurp $real_inc_path);
899 if ($self->dynamic) { # load the class too
900 eval_without_redefine_warnings($code);
903 $self->_ext_stmt($class,
904 qq|# These lines were loaded from '$real_inc_path' found in \@INC.\n|
905 .qq|# They are now part of the custom portion of this file\n|
906 .qq|# for you to hand-edit. If you do not either delete\n|
907 .qq|# this section or remove that file from \@INC, this section\n|
908 .qq|# will be repeated redundantly when you re-create this\n|
909 .qq|# file again via Loader! See skip_load_external to disable\n|
910 .qq|# this feature.\n|
913 $self->_ext_stmt($class, $code);
914 $self->_ext_stmt($class,
915 qq|# End of lines loaded from '$real_inc_path' |
919 if ($old_real_inc_path) {
920 my $code = slurp $old_real_inc_path;
922 $self->_ext_stmt($class, <<"EOF");
924 # These lines were loaded from '$old_real_inc_path',
925 # based on the Result class name that would have been created by an older
926 # version of the Loader. For a static schema, this happens only once during
927 # upgrade. See skip_load_external to disable this feature.
930 $code = $self->_rewrite_old_classnames($code);
932 if ($self->dynamic) {
935 Detected external content in '$old_real_inc_path', a class name that would have
936 been used by an older version of the Loader.
938 * PLEASE RENAME THIS CLASS: from '$old_class' to '$class', as that is the
939 new name of the Result.
941 eval_without_redefine_warnings($code);
945 $self->_ext_stmt($class, $code);
946 $self->_ext_stmt($class,
947 qq|# End of lines loaded from '$old_real_inc_path' |
954 Does the actual schema-construction work.
962 $self->_tables_list({ constraint => $self->constraint, exclude => $self->exclude })
970 Rescan the database for changes. Returns a list of the newly added table
973 The schema argument should be the schema class or object to be affected. It
974 should probably be derived from the original schema_class used during L</load>.
979 my ($self, $schema) = @_;
981 $self->{schema} = $schema;
982 $self->_relbuilder->{schema} = $schema;
985 my @current = $self->_tables_list({ constraint => $self->constraint, exclude => $self->exclude });
987 foreach my $table (@current) {
988 if(!exists $self->{_tables}->{$table}) {
989 push(@created, $table);
994 @current{@current} = ();
995 foreach my $table (keys %{ $self->{_tables} }) {
996 if (not exists $current{$table}) {
997 $self->_unregister_source_for_table($table);
1001 delete $self->{_dump_storage};
1002 delete $self->{_relations_started};
1004 my $loaded = $self->_load_tables(@current);
1006 return map { $self->monikers->{$_} } @created;
1012 return if $self->{skip_relationships};
1014 return $self->{relbuilder} ||= do {
1016 no warnings 'uninitialized';
1017 my $relbuilder_suff =
1023 ->{ $self->naming->{relationships}};
1025 my $relbuilder_class = 'DBIx::Class::Schema::Loader::RelBuilder'.$relbuilder_suff;
1026 load_class $relbuilder_class;
1027 $relbuilder_class->new( $self );
1033 my ($self, @tables) = @_;
1035 # Save the new tables to the tables list
1037 $self->{_tables}->{$_} = 1;
1040 $self->_make_src_class($_) for @tables;
1042 # sanity-check for moniker clashes
1043 my $inverse_moniker_idx;
1044 for (keys %{$self->monikers}) {
1045 push @{$inverse_moniker_idx->{$self->monikers->{$_}}}, $_;
1049 for (keys %$inverse_moniker_idx) {
1050 my $tables = $inverse_moniker_idx->{$_};
1052 push @clashes, sprintf ("tables %s reduced to the same source moniker '%s'",
1053 join (', ', map { "'$_'" } @$tables),
1060 die 'Unable to load schema - chosen moniker/class naming style results in moniker clashes. '
1061 . 'Either change the naming style, or supply an explicit moniker_map: '
1062 . join ('; ', @clashes)
1068 $self->_setup_src_meta($_) for @tables;
1070 if(!$self->skip_relationships) {
1071 # The relationship loader needs a working schema
1073 local $self->{dump_directory} = $self->{temp_directory};
1074 $self->_reload_classes(\@tables);
1075 $self->_load_relationships($_) for @tables;
1076 $self->_relbuilder->cleanup;
1079 # Remove that temp dir from INC so it doesn't get reloaded
1080 @INC = grep $_ ne $self->dump_directory, @INC;
1083 $self->_load_external($_)
1084 for map { $self->classes->{$_} } @tables;
1086 # Reload without unloading first to preserve any symbols from external
1088 $self->_reload_classes(\@tables, { unload => 0 });
1090 # Drop temporary cache
1091 delete $self->{_cache};
1096 sub _reload_classes {
1097 my ($self, $tables, $opts) = @_;
1099 my @tables = @$tables;
1101 my $unload = $opts->{unload};
1102 $unload = 1 unless defined $unload;
1104 # so that we don't repeat custom sections
1105 @INC = grep $_ ne $self->dump_directory, @INC;
1107 $self->_dump_to_dir(map { $self->classes->{$_} } @tables);
1109 unshift @INC, $self->dump_directory;
1112 my %have_source = map { $_ => $self->schema->source($_) }
1113 $self->schema->sources;
1115 for my $table (@tables) {
1116 my $moniker = $self->monikers->{$table};
1117 my $class = $self->classes->{$table};
1120 no warnings 'redefine';
1121 local *Class::C3::reinitialize = sub {}; # to speed things up, reinitialized below
1124 if (my $mc = $self->_moose_metaclass($class)) {
1127 Class::Unload->unload($class) if $unload;
1128 my ($source, $resultset_class);
1130 ($source = $have_source{$moniker})
1131 && ($resultset_class = $source->resultset_class)
1132 && ($resultset_class ne 'DBIx::Class::ResultSet')
1134 my $has_file = Class::Inspector->loaded_filename($resultset_class);
1135 if (my $mc = $self->_moose_metaclass($resultset_class)) {
1138 Class::Unload->unload($resultset_class) if $unload;
1139 $self->_reload_class($resultset_class) if $has_file;
1141 $self->_reload_class($class);
1143 push @to_register, [$moniker, $class];
1146 Class::C3->reinitialize;
1147 for (@to_register) {
1148 $self->schema->register_class(@$_);
1152 sub _moose_metaclass {
1153 return undef unless $INC{'Class/MOP.pm'}; # if CMOP is not loaded the class could not have loaded in the 1st place
1157 my $mc = try { Class::MOP::class_of($class) }
1160 return $mc->isa('Moose::Meta::Class') ? $mc : undef;
1163 # We use this instead of ensure_class_loaded when there are package symbols we
1166 my ($self, $class) = @_;
1168 my $class_path = $self->_class_path($class);
1169 delete $INC{ $class_path };
1171 # kill redefined warnings
1173 eval_without_redefine_warnings ("require $class");
1176 my $source = slurp $self->_get_dump_filename($class);
1177 die "Failed to reload class $class: $_.\n\nCLASS SOURCE:\n\n$source";
1181 sub _get_dump_filename {
1182 my ($self, $class) = (@_);
1184 $class =~ s{::}{/}g;
1185 return $self->dump_directory . q{/} . $class . q{.pm};
1188 =head2 get_dump_filename
1192 Returns the full path to the file for a class that the class has been or will
1193 be dumped to. This is a file in a temp dir for a dynamic schema.
1197 sub get_dump_filename {
1198 my ($self, $class) = (@_);
1200 local $self->{dump_directory} = $self->real_dump_directory;
1202 return $self->_get_dump_filename($class);
1205 sub _ensure_dump_subdirs {
1206 my ($self, $class) = (@_);
1208 my @name_parts = split(/::/, $class);
1209 pop @name_parts; # we don't care about the very last element,
1210 # which is a filename
1212 my $dir = $self->dump_directory;
1215 mkdir($dir) or croak "mkdir('$dir') failed: $!";
1217 last if !@name_parts;
1218 $dir = File::Spec->catdir($dir, shift @name_parts);
1223 my ($self, @classes) = @_;
1225 my $schema_class = $self->schema_class;
1226 my $schema_base_class = $self->schema_base_class || 'DBIx::Class::Schema';
1228 my $target_dir = $self->dump_directory;
1229 warn "Dumping manual schema for $schema_class to directory $target_dir ...\n"
1230 unless $self->{dynamic} or $self->{quiet};
1233 qq|package $schema_class;\n\n|
1234 . qq|# Created by DBIx::Class::Schema::Loader\n|
1235 . qq|# DO NOT MODIFY THE FIRST PART OF THIS FILE\n\n|;
1237 if ($self->use_moose) {
1238 $schema_text.= qq|use Moose;\nuse namespace::autoclean;\nextends '$schema_base_class';\n\n|;
1241 $schema_text .= qq|use strict;\nuse warnings;\n\nuse base '$schema_base_class';\n\n|;
1244 if ($self->use_namespaces) {
1245 $schema_text .= qq|__PACKAGE__->load_namespaces|;
1246 my $namespace_options;
1248 my @attr = qw/resultset_namespace default_resultset_class/;
1250 unshift @attr, 'result_namespace' unless (not $self->result_namespace) || $self->result_namespace eq 'Result';
1252 for my $attr (@attr) {
1254 $namespace_options .= qq| $attr => '| . $self->$attr . qq|',\n|
1257 $schema_text .= qq|(\n$namespace_options)| if $namespace_options;
1258 $schema_text .= qq|;\n|;
1261 $schema_text .= qq|__PACKAGE__->load_classes;\n|;
1265 local $self->{version_to_dump} = $self->schema_version_to_dump;
1266 $self->_write_classfile($schema_class, $schema_text, 1);
1269 my $result_base_class = $self->result_base_class || 'DBIx::Class::Core';
1271 foreach my $src_class (@classes) {
1273 qq|package $src_class;\n\n|
1274 . qq|# Created by DBIx::Class::Schema::Loader\n|
1275 . qq|# DO NOT MODIFY THE FIRST PART OF THIS FILE\n\n|
1276 . qq|use strict;\nuse warnings;\n\n|;
1277 if ($self->use_moose) {
1278 $src_text.= qq|use Moose;\nuse MooseX::NonMoose;\nuse namespace::autoclean;|;
1280 # these options 'use base' which is compile time
1281 if (@{ $self->left_base_classes } || @{ $self->additional_base_classes }) {
1282 $src_text .= qq|\nBEGIN { extends '$result_base_class' }\n\n|;
1285 $src_text .= qq|\nextends '$result_base_class';\n\n|;
1289 $src_text .= qq|use base '$result_base_class';\n\n|;
1291 $self->_write_classfile($src_class, $src_text);
1294 # remove Result dir if downgrading from use_namespaces, and there are no
1296 if (my $result_ns = $self->_downgrading_to_load_classes
1297 || $self->_rewriting_result_namespace) {
1298 my $result_namespace = $self->_result_namespace(
1303 (my $result_dir = $result_namespace) =~ s{::}{/}g;
1304 $result_dir = $self->dump_directory . '/' . $result_dir;
1306 unless (my @files = glob "$result_dir/*") {
1311 warn "Schema dump completed.\n" unless $self->{dynamic} or $self->{quiet};
1316 my ($self, $version, $ts) = @_;
1317 return qq|\n\n# Created by DBIx::Class::Schema::Loader|
1320 . qq|\n# DO NOT MODIFY THIS OR ANYTHING ABOVE! md5sum:|;
1323 sub _write_classfile {
1324 my ($self, $class, $text, $is_schema) = @_;
1326 my $filename = $self->_get_dump_filename($class);
1327 $self->_ensure_dump_subdirs($class);
1329 if (-f $filename && $self->really_erase_my_files) {
1330 warn "Deleting existing file '$filename' due to "
1331 . "'really_erase_my_files' setting\n" unless $self->{quiet};
1335 my ($old_gen, $old_md5, $old_ver, $old_ts, $old_custom)
1336 = $self->_parse_generated_file($filename);
1338 if (! $old_gen && -f $filename) {
1339 croak "Cannot overwrite '$filename' without 'really_erase_my_files',"
1340 . " it does not appear to have been generated by Loader"
1343 my $custom_content = $old_custom || '';
1345 # prepend extra custom content from a *renamed* class (singularization effect)
1346 if (my $renamed_class = $self->_upgrading_classes->{$class}) {
1347 my $old_filename = $self->_get_dump_filename($renamed_class);
1349 if (-f $old_filename) {
1350 my $extra_custom = ($self->_parse_generated_file ($old_filename))[4];
1352 $extra_custom =~ s/\n\n# You can replace.*\n1;\n//;
1354 $custom_content = join ("\n", '', $extra_custom, $custom_content)
1357 unlink $old_filename;
1361 $custom_content ||= $self->_default_custom_content($is_schema);
1363 # If upgrading to use_moose=1 replace default custom content with default Moose custom content.
1364 # If there is already custom content, which does not have the Moose content, add it.
1365 if ($self->use_moose) {
1367 my $non_moose_custom_content = do {
1368 local $self->{use_moose} = 0;
1369 $self->_default_custom_content;
1372 if ($custom_content eq $non_moose_custom_content) {
1373 $custom_content = $self->_default_custom_content($is_schema);
1375 elsif ($custom_content !~ /\Q@{[$self->_default_moose_custom_content($is_schema)]}\E/) {
1376 $custom_content .= $self->_default_custom_content($is_schema);
1379 elsif (defined $self->use_moose && $old_gen) {
1380 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'
1381 if $old_gen =~ /use \s+ MooseX?\b/x;
1384 $custom_content = $self->_rewrite_old_classnames($custom_content);
1387 for @{$self->{_dump_storage}->{$class} || []};
1389 # Check and see if the dump is infact differnt
1393 $compare_to = $text . $self->_sig_comment($old_ver, $old_ts);
1394 if (Digest::MD5::md5_base64($compare_to) eq $old_md5) {
1395 return unless $self->_upgrading_from && $is_schema;
1399 $text .= $self->_sig_comment(
1400 $self->version_to_dump,
1401 POSIX::strftime('%Y-%m-%d %H:%M:%S', localtime)
1404 open(my $fh, '>', $filename)
1405 or croak "Cannot open '$filename' for writing: $!";
1407 # Write the top half and its MD5 sum
1408 print $fh $text . Digest::MD5::md5_base64($text) . "\n";
1410 # Write out anything loaded via external partial class file in @INC
1412 for @{$self->{_ext_storage}->{$class} || []};
1414 # Write out any custom content the user has added
1415 print $fh $custom_content;
1418 or croak "Error closing '$filename': $!";
1421 sub _default_moose_custom_content {
1422 my ($self, $is_schema) = @_;
1424 if (not $is_schema) {
1425 return qq|\n__PACKAGE__->meta->make_immutable;|;
1428 return qq|\n__PACKAGE__->meta->make_immutable(inline_constructor => 0);|;
1431 sub _default_custom_content {
1432 my ($self, $is_schema) = @_;
1433 my $default = qq|\n\n# You can replace this text with custom|
1434 . qq| code or comments, and it will be preserved on regeneration|;
1435 if ($self->use_moose) {
1436 $default .= $self->_default_moose_custom_content($is_schema);
1438 $default .= qq|\n1;\n|;
1442 sub _parse_generated_file {
1443 my ($self, $fn) = @_;
1445 return unless -f $fn;
1447 open(my $fh, '<', $fn)
1448 or croak "Cannot open '$fn' for reading: $!";
1451 qr{^(# DO NOT MODIFY THIS OR ANYTHING ABOVE! md5sum:)([A-Za-z0-9/+]{22})\n};
1453 my ($md5, $ts, $ver, $gen);
1459 # Pull out the version and timestamp from the line above
1460 ($ver, $ts) = $gen =~ m/^# Created by DBIx::Class::Schema::Loader v(.*?) @ (.*?)\Z/m;
1463 croak "Checksum mismatch in '$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"
1464 if !$self->overwrite_modifications && Digest::MD5::md5_base64($gen) ne $md5;
1473 my $custom = do { local $/; <$fh> }
1478 return ($gen, $md5, $ver, $ts, $custom);
1486 warn "$target: use $_;" if $self->debug;
1487 $self->_raw_stmt($target, "use $_;");
1495 my $blist = join(q{ }, @_);
1497 return unless $blist;
1499 warn "$target: use base qw/$blist/;" if $self->debug;
1500 $self->_raw_stmt($target, "use base qw/$blist/;");
1503 sub _result_namespace {
1504 my ($self, $schema_class, $ns) = @_;
1505 my @result_namespace;
1507 if ($ns =~ /^\+(.*)/) {
1508 # Fully qualified namespace
1509 @result_namespace = ($1)
1512 # Relative namespace
1513 @result_namespace = ($schema_class, $ns);
1516 return wantarray ? @result_namespace : join '::', @result_namespace;
1519 # Create class with applicable bases, setup monikers, etc
1520 sub _make_src_class {
1521 my ($self, $table) = @_;
1523 my $schema = $self->schema;
1524 my $schema_class = $self->schema_class;
1526 my $table_moniker = $self->_table2moniker($table);
1527 my @result_namespace = ($schema_class);
1528 if ($self->use_namespaces) {
1529 my $result_namespace = $self->result_namespace || 'Result';
1530 @result_namespace = $self->_result_namespace(
1535 my $table_class = join(q{::}, @result_namespace, $table_moniker);
1537 if ((my $upgrading_v = $self->_upgrading_from)
1538 || $self->_rewriting) {
1539 local $self->naming->{monikers} = $upgrading_v
1542 my @result_namespace = @result_namespace;
1543 if ($self->_upgrading_from_load_classes) {
1544 @result_namespace = ($schema_class);
1546 elsif (my $ns = $self->_downgrading_to_load_classes) {
1547 @result_namespace = $self->_result_namespace(
1552 elsif ($ns = $self->_rewriting_result_namespace) {
1553 @result_namespace = $self->_result_namespace(
1559 my $old_class = join(q{::}, @result_namespace,
1560 $self->_table2moniker($table));
1562 $self->_upgrading_classes->{$table_class} = $old_class
1563 unless $table_class eq $old_class;
1566 # this was a bad idea, should be ok now without it
1567 # my $table_normalized = lc $table;
1568 # $self->classes->{$table_normalized} = $table_class;
1569 # $self->monikers->{$table_normalized} = $table_moniker;
1571 $self->classes->{$table} = $table_class;
1572 $self->monikers->{$table} = $table_moniker;
1574 $self->_use ($table_class, @{$self->additional_classes});
1575 $self->_inject($table_class, @{$self->left_base_classes});
1577 if (my @components = @{ $self->components }) {
1578 $self->_dbic_stmt($table_class, 'load_components', @components);
1581 $self->_inject($table_class, @{$self->additional_base_classes});
1584 sub _is_result_class_method {
1585 my ($self, $name) = @_;
1587 if (not $self->_result_class_methods) {
1588 my (@methods, %methods);
1589 my $base = $self->result_base_class || 'DBIx::Class::Core';
1590 my @components = map { /^\+/ ? substr($_,1) : "DBIx::Class::$_" } @{ $self->components || [] };
1592 for my $class ($base, @components, $self->use_moose ? 'Moose::Object' : ()) {
1595 push @methods, @{ Class::Inspector->methods($class) || [] };
1598 push @methods, @{ Class::Inspector->methods('UNIVERSAL') };
1600 @methods{@methods} = ();
1603 $methods{meta} = undef;
1605 $self->_result_class_methods(\%methods);
1607 my $result_methods = $self->_result_class_methods;
1609 return exists $result_methods->{$name};
1612 sub _resolve_col_accessor_collisions {
1613 my ($self, $table, $col_info) = @_;
1615 my $table_name = ref $table ? $$table : $table;
1617 while (my ($col, $info) = each %$col_info) {
1618 my $accessor = $info->{accessor} || $col;
1620 next if $accessor eq 'id'; # special case (very common column)
1622 if ($self->_is_result_class_method($accessor)) {
1625 if (my $map = $self->col_collision_map) {
1626 for my $re (keys %$map) {
1627 if (my @matches = $col =~ /$re/) {
1628 $info->{accessor} = sprintf $map->{$re}, @matches;
1636 Column '$col' in table '$table_name' collides with an inherited method.
1637 See "COLUMN ACCESSOR COLLISIONS" in perldoc DBIx::Class::Schema::Loader::Base .
1639 $info->{accessor} = undef;
1645 # use the same logic to run moniker_map, col_accessor_map, and
1646 # relationship_name_map
1648 my ( $self, $map, $default_code, $ident, @extra ) = @_;
1650 my $default_ident = $default_code->( $ident, @extra );
1652 if( $map && ref $map eq 'HASH' ) {
1653 $new_ident = $map->{ $ident };
1655 elsif( $map && ref $map eq 'CODE' ) {
1656 $new_ident = $map->( $ident, $default_ident, @extra );
1659 $new_ident ||= $default_ident;
1664 sub _default_column_accessor_name {
1665 my ( $self, $column_name ) = @_;
1667 my $accessor_name = $column_name;
1668 $accessor_name =~ s/\W+/_/g;
1670 if ((($self->naming->{column_accessors}||'') =~ /(\d+)/ && $1 < 7) || (not $self->preserve_case)) {
1671 # older naming just lc'd the col accessor and that's all.
1672 return lc $accessor_name;
1675 return join '_', map lc, split_name $column_name;
1679 sub _make_column_accessor_name {
1680 my ($self, $column_name, $column_context_info ) = @_;
1682 my $accessor = $self->_run_user_map(
1683 $self->col_accessor_map,
1684 sub { $self->_default_column_accessor_name( shift ) },
1686 $column_context_info,
1692 # Set up metadata (cols, pks, etc)
1693 sub _setup_src_meta {
1694 my ($self, $table) = @_;
1696 my $schema = $self->schema;
1697 my $schema_class = $self->schema_class;
1699 my $table_class = $self->classes->{$table};
1700 my $table_moniker = $self->monikers->{$table};
1702 my $table_name = $table;
1703 my $name_sep = $self->schema->storage->sql_maker->name_sep;
1705 if ($name_sep && $table_name =~ /\Q$name_sep\E/) {
1706 $table_name = \ $self->_quote_table_name($table_name);
1709 my $full_table_name = ($self->qualify_objects ? ($self->db_schema . '.') : '') . (ref $table_name ? $$table_name : $table_name);
1711 # be careful to not create refs Data::Dump can "optimize"
1712 $full_table_name = \do {"".$full_table_name} if ref $table_name;
1714 $self->_dbic_stmt($table_class, 'table', $full_table_name);
1716 my $cols = $self->_table_columns($table);
1717 my $col_info = $self->__columns_info_for($table);
1719 ### generate all the column accessor names
1720 while (my ($col, $info) = each %$col_info) {
1721 # hashref of other info that could be used by
1722 # user-defined accessor map functions
1724 table_class => $table_class,
1725 table_moniker => $table_moniker,
1726 table_name => $table_name,
1727 full_table_name => $full_table_name,
1728 schema_class => $schema_class,
1729 column_info => $info,
1732 $info->{accessor} = $self->_make_column_accessor_name( $col, $context );
1735 $self->_resolve_col_accessor_collisions($full_table_name, $col_info);
1737 # prune any redundant accessor names
1738 while (my ($col, $info) = each %$col_info) {
1739 no warnings 'uninitialized';
1740 delete $info->{accessor} if $info->{accessor} eq $col;
1743 my $fks = $self->_table_fk_info($table);
1745 foreach my $fkdef (@$fks) {
1746 for my $col (@{ $fkdef->{local_columns} }) {
1747 $col_info->{$col}{is_foreign_key} = 1;
1751 my $pks = $self->_table_pk_info($table) || [];
1753 foreach my $pkcol (@$pks) {
1754 $col_info->{$pkcol}{is_nullable} = 0;
1760 map { $_, ($col_info->{$_}||{}) } @$cols
1763 my %uniq_tag; # used to eliminate duplicate uniqs
1765 @$pks ? $self->_dbic_stmt($table_class,'set_primary_key',@$pks)
1766 : carp("$table has no primary key");
1767 $uniq_tag{ join("\0", @$pks) }++ if @$pks; # pk is a uniq
1769 my $uniqs = $self->_table_uniq_info($table) || [];
1771 my ($name, $cols) = @$_;
1772 next if $uniq_tag{ join("\0", @$cols) }++; # skip duplicates
1773 $self->_dbic_stmt($table_class,'add_unique_constraint', $name, $cols);
1778 sub __columns_info_for {
1779 my ($self, $table) = @_;
1781 my $result = $self->_columns_info_for($table);
1783 while (my ($col, $info) = each %$result) {
1784 $info = { %$info, %{ $self->_custom_column_info ($table, $col, $info) } };
1785 $info = { %$info, %{ $self->_datetime_column_info($table, $col, $info) } };
1787 $result->{$col} = $info;
1795 Returns a sorted list of loaded tables, using the original database table
1803 return keys %{$self->_tables};
1806 # Make a moniker from a table
1807 sub _default_table2moniker {
1808 no warnings 'uninitialized';
1809 my ($self, $table) = @_;
1811 if ($self->naming->{monikers} eq 'v4') {
1812 return join '', map ucfirst, split /[\W_]+/, lc $table;
1814 elsif ($self->naming->{monikers} eq 'v5') {
1815 return join '', map ucfirst, split /[\W_]+/,
1816 Lingua::EN::Inflect::Number::to_S(lc $table);
1818 elsif ($self->naming->{monikers} eq 'v6') {
1819 (my $as_phrase = lc $table) =~ s/_+/ /g;
1820 my $inflected = Lingua::EN::Inflect::Phrase::to_S($as_phrase);
1822 return join '', map ucfirst, split /\W+/, $inflected;
1825 my @words = map lc, split_name $table;
1826 my $as_phrase = join ' ', @words;
1828 my $inflected = Lingua::EN::Inflect::Phrase::to_S($as_phrase);
1830 return join '', map ucfirst, split /\W+/, $inflected;
1833 sub _table2moniker {
1834 my ( $self, $table ) = @_;
1836 $self->_run_user_map(
1838 sub { $self->_default_table2moniker( shift ) },
1843 sub _load_relationships {
1844 my ($self, $table) = @_;
1846 my $tbl_fk_info = $self->_table_fk_info($table);
1847 foreach my $fkdef (@$tbl_fk_info) {
1848 $fkdef->{remote_source} =
1849 $self->monikers->{delete $fkdef->{remote_table}};
1851 my $tbl_uniq_info = $self->_table_uniq_info($table);
1853 my $local_moniker = $self->monikers->{$table};
1854 my $rel_stmts = $self->_relbuilder->generate_code($local_moniker, $tbl_fk_info, $tbl_uniq_info);
1856 foreach my $src_class (sort keys %$rel_stmts) {
1857 my $src_stmts = $rel_stmts->{$src_class};
1858 foreach my $stmt (@$src_stmts) {
1859 $self->_dbic_stmt($src_class,$stmt->{method},@{$stmt->{args}});
1864 # Overload these in driver class:
1866 # Returns an arrayref of column names
1867 sub _table_columns { croak "ABSTRACT METHOD" }
1869 # Returns arrayref of pk col names
1870 sub _table_pk_info { croak "ABSTRACT METHOD" }
1872 # Returns an arrayref of uniqs [ [ foo => [ col1, col2 ] ], [ bar => [ ... ] ] ]
1873 sub _table_uniq_info { croak "ABSTRACT METHOD" }
1875 # Returns an arrayref of foreign key constraints, each
1876 # being a hashref with 3 keys:
1877 # local_columns (arrayref), remote_columns (arrayref), remote_table
1878 sub _table_fk_info { croak "ABSTRACT METHOD" }
1880 # Returns an array of lower case table names
1881 sub _tables_list { croak "ABSTRACT METHOD" }
1883 # Execute a constructive DBIC class method, with debug/dump_to_dir hooks.
1889 # generate the pod for this statement, storing it with $self->_pod
1890 $self->_make_pod( $class, $method, @_ ) if $self->generate_pod;
1892 my $args = dump(@_);
1893 $args = '(' . $args . ')' if @_ < 2;
1894 my $stmt = $method . $args . q{;};
1896 warn qq|$class\->$stmt\n| if $self->debug;
1897 $self->_raw_stmt($class, '__PACKAGE__->' . $stmt);
1901 # generates the accompanying pod for a DBIC class method statement,
1902 # storing it with $self->_pod
1908 if ( $method eq 'table' ) {
1910 my $pcm = $self->pod_comment_mode;
1911 my ($comment, $comment_overflows, $comment_in_name, $comment_in_desc);
1912 $comment = $self->__table_comment($table);
1913 $comment_overflows = ($comment and length $comment > $self->pod_comment_spillover_length);
1914 $comment_in_name = ($pcm eq 'name' or ($pcm eq 'auto' and !$comment_overflows));
1915 $comment_in_desc = ($pcm eq 'description' or ($pcm eq 'auto' and $comment_overflows));
1916 $self->_pod( $class, "=head1 NAME" );
1917 my $table_descr = $class;
1918 $table_descr .= " - " . $comment if $comment and $comment_in_name;
1919 $self->{_class2table}{ $class } = $table;
1920 $self->_pod( $class, $table_descr );
1921 if ($comment and $comment_in_desc) {
1922 $self->_pod( $class, "=head1 DESCRIPTION" );
1923 $self->_pod( $class, $comment );
1925 $self->_pod_cut( $class );
1926 } elsif ( $method eq 'add_columns' ) {
1927 $self->_pod( $class, "=head1 ACCESSORS" );
1928 my $col_counter = 0;
1930 while( my ($name,$attrs) = splice @cols,0,2 ) {
1932 $self->_pod( $class, '=head2 ' . $name );
1933 $self->_pod( $class,
1935 my $s = $attrs->{$_};
1936 $s = !defined $s ? 'undef' :
1937 length($s) == 0 ? '(empty string)' :
1938 ref($s) eq 'SCALAR' ? $$s :
1939 ref($s) ? dumper_squashed $s :
1940 looks_like_number($s) ? $s : qq{'$s'};
1943 } sort keys %$attrs,
1945 if (my $comment = $self->__column_comment($self->{_class2table}{$class}, $col_counter, $name)) {
1946 $self->_pod( $class, $comment );
1949 $self->_pod_cut( $class );
1950 } elsif ( $method =~ /^(belongs_to|has_many|might_have)$/ ) {
1951 $self->_pod( $class, "=head1 RELATIONS" ) unless $self->{_relations_started} { $class } ;
1952 my ( $accessor, $rel_class ) = @_;
1953 $self->_pod( $class, "=head2 $accessor" );
1954 $self->_pod( $class, 'Type: ' . $method );
1955 $self->_pod( $class, "Related object: L<$rel_class>" );
1956 $self->_pod_cut( $class );
1957 $self->{_relations_started} { $class } = 1;
1961 sub _filter_comment {
1962 my ($self, $txt) = @_;
1964 $txt = '' if not defined $txt;
1966 $txt =~ s/(?:\015?\012|\015\012?)/\n/g;
1971 sub __table_comment {
1974 if (my $code = $self->can('_table_comment')) {
1975 return $self->_filter_comment($self->$code(@_));
1981 sub __column_comment {
1984 if (my $code = $self->can('_column_comment')) {
1985 return $self->_filter_comment($self->$code(@_));
1991 # Stores a POD documentation
1993 my ($self, $class, $stmt) = @_;
1994 $self->_raw_stmt( $class, "\n" . $stmt );
1998 my ($self, $class ) = @_;
1999 $self->_raw_stmt( $class, "\n=cut\n" );
2002 # Store a raw source line for a class (for dumping purposes)
2004 my ($self, $class, $stmt) = @_;
2005 push(@{$self->{_dump_storage}->{$class}}, $stmt);
2008 # Like above, but separately for the externally loaded stuff
2010 my ($self, $class, $stmt) = @_;
2011 push(@{$self->{_ext_storage}->{$class}}, $stmt);
2014 sub _quote_table_name {
2015 my ($self, $table) = @_;
2017 my $qt = $self->schema->storage->sql_maker->quote_char;
2019 return $table unless $qt;
2022 return $qt->[0] . $table . $qt->[1];
2025 return $qt . $table . $qt;
2028 sub _custom_column_info {
2029 my ( $self, $table_name, $column_name, $column_info ) = @_;
2031 if (my $code = $self->custom_column_info) {
2032 return $code->($table_name, $column_name, $column_info) || {};
2037 sub _datetime_column_info {
2038 my ( $self, $table_name, $column_name, $column_info ) = @_;
2040 my $type = $column_info->{data_type} || '';
2041 if ((grep $_, @{ $column_info }{map "inflate_$_", qw/date datetime timestamp/})
2042 or ($type =~ /date|timestamp/i)) {
2043 $result->{timezone} = $self->datetime_timezone if $self->datetime_timezone;
2044 $result->{locale} = $self->datetime_locale if $self->datetime_locale;
2050 my ($self, $name) = @_;
2052 return $self->preserve_case ? $name : lc($name);
2056 my ($self, $name) = @_;
2058 return $self->preserve_case ? $name : uc($name);
2061 sub _unregister_source_for_table {
2062 my ($self, $table) = @_;
2066 my $schema = $self->schema;
2067 # in older DBIC it's a private method
2068 my $unregister = $schema->can('unregister_source') || $schema->can('_unregister_source');
2069 $schema->$unregister($self->_table2moniker($table));
2070 delete $self->monikers->{$table};
2071 delete $self->classes->{$table};
2072 delete $self->_upgrading_classes->{$table};
2073 delete $self->{_tables}{$table};
2077 # remove the dump dir from @INC on destruction
2081 @INC = grep $_ ne $self->dump_directory, @INC;
2086 Returns a hashref of loaded table to moniker mappings. There will
2087 be two entries for each table, the original name and the "normalized"
2088 name, in the case that the two are different (such as databases
2089 that like uppercase table names, or preserve your original mixed-case
2090 definitions, or what-have-you).
2094 Returns a hashref of table to class mappings. In some cases it will
2095 contain multiple entries per table for the original and normalized table
2096 names, as above in L</monikers>.
2098 =head1 COLUMN ACCESSOR COLLISIONS
2100 Occasionally you may have a column name that collides with a perl method, such
2101 as C<can>. In such cases, the default action is to set the C<accessor> of the
2102 column spec to C<undef>.
2104 You can then name the accessor yourself by placing code such as the following
2107 __PACKAGE__->add_column('+can' => { accessor => 'my_can' });
2109 Another option is to use the L</col_collision_map> option.
2111 =head1 RELATIONSHIP NAME COLLISIONS
2113 In very rare cases, you may get a collision between a generated relationship
2114 name and a method in your Result class, for example if you have a foreign key
2115 called C<belongs_to>.
2117 This is a problem because relationship names are also relationship accessor
2118 methods in L<DBIx::Class>.
2120 The default behavior is to append C<_rel> to the relationship name and print
2121 out a warning that refers to this text.
2123 You can also control the renaming with the L</rel_collision_map> option.
2127 L<DBIx::Class::Schema::Loader>
2131 See L<DBIx::Class::Schema::Loader/AUTHOR> and L<DBIx::Class::Schema::Loader/CONTRIBUTORS>.
2135 This library is free software; you can redistribute it and/or modify it under
2136 the same terms as Perl itself.
2141 # vim:et sts=4 sw=4 tw=0: