package SQL::Translator::Parser::MySQL;
# -------------------------------------------------------------------
-# $Id: MySQL.pm,v 1.5 2002-11-20 04:03:04 kycl4rk Exp $
+# $Id: MySQL.pm,v 1.17 2003-05-09 16:55:07 kycl4rk Exp $
# -------------------------------------------------------------------
-# Copyright (C) 2002 Ken Y. Clark <kycl4rk@users.sourceforge.net>,
-# darren chamberlain <darren@cpan.org>
+# Copyright (C) 2003 Ken Y. Clark <kclark@cpan.org>,
+# darren chamberlain <darren@cpan.org>,
+# Chris Mungall <cjm@fruitfly.org>
#
# This program is free software; you can redistribute it and/or
# modify it under the terms of the GNU General Public License as
# 02111-1307 USA
# -------------------------------------------------------------------
+=head1 NAME
+
+SQL::Translator::Parser::MySQL - parser for MySQL
+
+=head1 SYNOPSIS
+
+ use SQL::Translator;
+ use SQL::Translator::Parser::MySQL;
+
+ my $translator = SQL::Translator->new;
+ $translator->parser("SQL::Translator::Parser::MySQL");
+
+=head1 DESCRIPTION
+
+The grammar is influenced heavily by Tim Bunce's "mysql2ora" grammar.
+
+Here's the word from the MySQL site
+(http://www.mysql.com/doc/en/CREATE_TABLE.html):
+
+ CREATE [TEMPORARY] TABLE [IF NOT EXISTS] tbl_name [(create_definition,...)]
+ [table_options] [select_statement]
+
+ or
+
+ CREATE [TEMPORARY] TABLE [IF NOT EXISTS] tbl_name LIKE old_table_name;
+
+ create_definition:
+ col_name type [NOT NULL | NULL] [DEFAULT default_value] [AUTO_INCREMENT]
+ [PRIMARY KEY] [reference_definition]
+ or PRIMARY KEY (index_col_name,...)
+ or KEY [index_name] (index_col_name,...)
+ or INDEX [index_name] (index_col_name,...)
+ or UNIQUE [INDEX] [index_name] (index_col_name,...)
+ or FULLTEXT [INDEX] [index_name] (index_col_name,...)
+ or [CONSTRAINT symbol] FOREIGN KEY [index_name] (index_col_name,...)
+ [reference_definition]
+ or CHECK (expr)
+
+ type:
+ TINYINT[(length)] [UNSIGNED] [ZEROFILL]
+ or SMALLINT[(length)] [UNSIGNED] [ZEROFILL]
+ or MEDIUMINT[(length)] [UNSIGNED] [ZEROFILL]
+ or INT[(length)] [UNSIGNED] [ZEROFILL]
+ or INTEGER[(length)] [UNSIGNED] [ZEROFILL]
+ or BIGINT[(length)] [UNSIGNED] [ZEROFILL]
+ or REAL[(length,decimals)] [UNSIGNED] [ZEROFILL]
+ or DOUBLE[(length,decimals)] [UNSIGNED] [ZEROFILL]
+ or FLOAT[(length,decimals)] [UNSIGNED] [ZEROFILL]
+ or DECIMAL(length,decimals) [UNSIGNED] [ZEROFILL]
+ or NUMERIC(length,decimals) [UNSIGNED] [ZEROFILL]
+ or CHAR(length) [BINARY]
+ or VARCHAR(length) [BINARY]
+ or DATE
+ or TIME
+ or TIMESTAMP
+ or DATETIME
+ or TINYBLOB
+ or BLOB
+ or MEDIUMBLOB
+ or LONGBLOB
+ or TINYTEXT
+ or TEXT
+ or MEDIUMTEXT
+ or LONGTEXT
+ or ENUM(value1,value2,value3,...)
+ or SET(value1,value2,value3,...)
+
+ index_col_name:
+ col_name [(length)]
+
+ reference_definition:
+ REFERENCES tbl_name [(index_col_name,...)]
+ [MATCH FULL | MATCH PARTIAL]
+ [ON DELETE reference_option]
+ [ON UPDATE reference_option]
+
+ reference_option:
+ RESTRICT | CASCADE | SET NULL | NO ACTION | SET DEFAULT
+
+ table_options:
+ TYPE = {BDB | HEAP | ISAM | InnoDB | MERGE | MRG_MYISAM | MYISAM }
+ or AUTO_INCREMENT = #
+ or AVG_ROW_LENGTH = #
+ or CHECKSUM = {0 | 1}
+ or COMMENT = "string"
+ or MAX_ROWS = #
+ or MIN_ROWS = #
+ or PACK_KEYS = {0 | 1 | DEFAULT}
+ or PASSWORD = "string"
+ or DELAY_KEY_WRITE = {0 | 1}
+ or ROW_FORMAT= { default | dynamic | fixed | compressed }
+ or RAID_TYPE= {1 | STRIPED | RAID0 } RAID_CHUNKS=# RAID_CHUNKSIZE=#
+ or UNION = (table_name,[table_name...])
+ or INSERT_METHOD= {NO | FIRST | LAST }
+ or DATA DIRECTORY="absolute path to directory"
+ or INDEX DIRECTORY="absolute path to directory"
+
+=cut
+
use strict;
-use vars qw($VERSION $GRAMMAR @EXPORT_OK);
-$VERSION = sprintf "%d.%02d", q$Revision: 1.5 $ =~ /(\d+)\.(\d+)/;
+use vars qw[ $DEBUG $VERSION $GRAMMAR @EXPORT_OK ];
+$VERSION = sprintf "%d.%02d", q$Revision: 1.17 $ =~ /(\d+)\.(\d+)/;
+$DEBUG = 0 unless defined $DEBUG;
-#use SQL::Translator::Parser; # This is not necessary!
+use Data::Dumper;
use Parse::RecDescent;
use Exporter;
use base qw(Exporter);
@EXPORT_OK = qw(parse);
+# Enable warnings within the Parse::RecDescent module.
+$::RD_ERRORS = 1; # Make sure the parser dies when it encounters an error
+$::RD_WARN = 1; # Enable warnings. This will warn on unused rules &c.
+$::RD_HINT = 1; # Give out hints to help fix problems.
+
my $parser; # should we do this? There's no programmic way to
# change the grammar, so I think this is safe.
-sub parse {
- my ( $translator, $data ) = @_;
- $parser ||= Parse::RecDescent->new($GRAMMAR);
- unless (defined $parser) {
- return $translator->error("Error instantiating Parse::RecDescent ".
- "instance: Bad grammer");
- }
+$GRAMMAR = q!
- # Is this right? It was $parser->parse before, but that didn't
- # work; Parse::RecDescent appears to need the name of a rule
- # with which to begin, so I chose the first rule in the grammar.
- return $parser->file($data);
+{
+ our ( %tables, $table_order );
}
-$GRAMMAR =
- q!
- { our ( %tables ) }
-
- file : statement(s) { \%tables }
-
- statement : comment
- | create
- | <error>
-
- create : create_table table_name '(' line(s /,/) ')' table_type(?) ';'
- {
- my $i = 0;
- for my $line ( @{ $item[4] } ) {
- if ( $line->{'type'} eq 'field' ) {
- my $field_name = $line->{'name'};
- $tables{ $item{'table_name'} }
- {'fields'}{$field_name} =
- { %$line, order => $i };
- $i++;
-
- if ( $line->{'is_primary_key'} ) {
- push
- @{ $tables{ $item{'table_name'} }{'indices'} },
- {
- type => 'primary_key',
- fields => [ $field_name ],
- };
- }
- }
- else {
- push @{ $tables{ $item{'table_name'} }{'indices'} },
- $line;
- }
- $tables{ $item{'table_name'} }{'type'} =
- $item{'table_type'}[0];
+#
+# The "eofile" rule makes the parser fail if any "statement" rule
+# fails. Otherwise, the first successful match by a "statement"
+# won't cause the failure needed to know that the parse, as a whole,
+# failed. -ky
+#
+startrule : statement(s) eofile { \%tables }
+
+eofile : /^\Z/
+
+statement : comment
+ | drop
+ | create
+ | <error>
+
+drop : /drop/i WORD(s) ';'
+
+create : create_table table_name '(' create_definition(s /,/) ')' table_option(s?) ';'
+ {
+ my $table_name = $item{'table_name'};
+ $tables{ $table_name }{'order'} = ++$table_order;
+ $tables{ $table_name }{'table_name'} = $table_name;
+
+ my $i = 1;
+ for my $definition ( @{ $item[4] } ) {
+ if ( $definition->{'type'} eq 'field' ) {
+ my $field_name = $definition->{'name'};
+ $tables{ $table_name }{'fields'}{ $field_name } =
+ { %$definition, order => $i };
+ $i++;
+
+ if ( $definition->{'is_primary_key'} ) {
+ push @{ $tables{ $table_name }{'indices'} },
+ {
+ type => 'primary_key',
+ fields => [ $field_name ],
}
- }
- | <error>
-
- create : create_index index_name /on/i table_name '(' field_name(s /,/) ')' ';'
-# create : create_index index_name keyword_on table_name '(' field_name ')' ';'
- {
- # do nothing just now
- my $dummy = 0;
- }
- | <error>
-
- keyword_on : /on/i
-
- line : index
- | field
- | <error>
-
- comment : /^\s*[#-]+.*\n/
-
- blank : /\s*/
-
-
- field : field_name data_type field_qualifier(s?)
- {
- my %qualifier_h =
- map {%$_} @{$item{'field_qualifier'} || []};
- my $null = defined $item{'not_null'}
- ? $item{'not_null'} : 1 ;
- delete $qualifier_h{'not_null'};
- $return = {
- type => 'field',
- name => $item{'field_name'},
- data_type => $item{'data_type'}{'type'},
- null => $null,
- %qualifier_h,
- }
- }
- | <error>
-
- field_qualifier : not_null
- {
- $return = {
- null => $item{'not_null'},
- }
+ ;
+ }
}
-
- field_qualifier : default_val
- {
- $return = {
- default => $item{default_val},
- }
+ else {
+ push @{ $tables{ $table_name }{'indices'} },
+ $definition;
}
+ }
- field_qualifier : auto_inc
- {
- $return = {
- is_auto_inc => $item{auto_inc},
- }
+ for my $opt ( @{ $item{'table_option(s?)'} } ) {
+ if ( my ( $key, $val ) = each %$opt ) {
+ $tables{ $table_name }{'table_options'}{ $key } = $val;
}
+ }
- field_qualifier : primary_key
- {
- $return = {
- is_primary_key => $item{primary_key},
- }
- }
+ 1;
+ }
- field_qualifier : unsigned
- {
- $return = {
- is_unsigned => $item{unsigned},
- }
+create : /CREATE/i unique(?) /(INDEX|KEY)/i index_name /on/i table_name '(' field_name(s /,/) ')' ';'
+ {
+ push @{ $tables{ $item{'table_name'} }{'indices'} },
+ {
+ name => $item[4],
+ type => $item[2] ? 'unique' : 'normal',
+ fields => $item[8],
}
+ ;
+ }
- index : primary_key_index
- | unique_index
- | normal_index
+create_definition : index
+ | field
+ | <error>
+
+comment : /^\s*(?:#|-{2}).*\n/
+
+blank : /\s*/
+
+field : field_name data_type field_qualifier(s?) reference_definition(?)
+ {
+ my %qualifiers = map { %$_ } @{ $item{'field_qualifier(s?)'} || [] };
+ my $null = defined $item{'not_null'} ? $item{'not_null'} : 1;
+ delete $qualifiers{'not_null'};
+ if ( my @type_quals = @{ $item{'data_type'}{'qualifiers'} || [] } ) {
+ $qualifiers{ $_ } = 1 for @type_quals;
+ }
+
+ $return = {
+ type => 'field',
+ name => $item{'field_name'},
+ data_type => $item{'data_type'}{'type'},
+ size => $item{'data_type'}{'size'},
+ list => $item{'data_type'}{'list'},
+ null => $null,
+ constraints => $item{'reference_definition(?)'},
+ %qualifiers,
+ }
+ }
+ | <error>
- table_name : WORD
+field_qualifier : not_null
+ {
+ $return = {
+ null => $item{'not_null'},
+ }
+ }
- field_name : WORD
+field_qualifier : default_val
+ {
+ $return = {
+ default => $item{'default_val'},
+ }
+ }
- index_name : WORD
+field_qualifier : auto_inc
+ {
+ $return = {
+ is_auto_inc => $item{'auto_inc'},
+ }
+ }
- data_type : WORD field_size(?)
- {
- $return = {
- type => $item[1],
- size => $item[2][0]
- }
- }
+field_qualifier : primary_key
+ {
+ $return = {
+ is_primary_key => $item{'primary_key'},
+ }
+ }
- field_type : WORD
+field_qualifier : unsigned
+ {
+ $return = {
+ is_unsigned => $item{'unsigned'},
+ }
+ }
- field_size : '(' num_range ')' { $item{'num_range'} }
+reference_definition : /references/i table_name parens_field_list(?) match_type(?) on_delete_do(?) on_update_do(?)
+ {
+ $return = {
+ type => 'foreign_key',
+ reference_table => $item[2],
+ reference_fields => $item[3][0],
+ match_type => $item[4][0],
+ on_delete_do => $item[5][0],
+ on_update_do => $item[6][0],
+ }
+ }
- num_range : DIGITS ',' DIGITS
- { $return = $item[1].','.$item[3] }
- | DIGITS
- { $return = $item[1] }
+match_type : /match full/i { 'match_full' }
+ |
+ /match partial/i { 'match_partial' }
+
+on_delete_do : /on delete/i reference_option
+ { $item[2] }
+
+on_update_do : /on update/i reference_option
+ { $item[2] }
+
+reference_option: /restrict/i |
+ /cascade/i |
+ /set null/i |
+ /no action/i |
+ /set default/i
+ { $item[1] }
+
+index : primary_key_index
+ | unique_index
+ | fulltext_index
+ | normal_index
+ | <error>
+
+table_name : WORD
+
+field_name : WORD
+
+index_name : WORD
+
+data_type : WORD parens_value_list(s?) type_qualifier(s?)
+ {
+ my $type = $item[1];
+ my $size; # field size, applicable only to non-set fields
+ my $list; # set list, applicable only to sets (duh)
+
+ if ( uc($type) =~ /^(SET|ENUM)$/ ) {
+ $size = undef;
+ $list = $item[2][0];
+ }
+ else {
+ $size = $item[2][0];
+ $list = [];
+ }
+
+ $return = {
+ type => $type,
+ size => $size,
+ list => $list,
+ qualifiers => $item[3],
+ }
+ }
+parens_field_list : '(' field_name(s /,/) ')'
+ { $item[2] }
- create_table : /create/i /table/i
+parens_value_list : '(' VALUE(s /,/) ')'
+ { $item[2] }
- create_index : /create/i /index/i
+type_qualifier : /(BINARY|UNSIGNED|ZEROFILL)/i
+ { lc $item[1] }
- not_null : /not/i /null/i { $return = 0 }
+field_type : WORD
- unsigned : /unsigned/i { $return = 0 }
+field_size : '(' num_range ')' { $item{'num_range'} }
- default_val : /default/i /(?:')?[\w\d.-]*(?:')?/ { $item[2]=~s/'//g; $return=$item[2] }
+num_range : DIGITS ',' DIGITS
+ { $return = $item[1].','.$item[3] }
+ | DIGITS
+ { $return = $item[1] }
- auto_inc : /auto_increment/i { 1 }
+create_table : /create/i /table/i
- primary_key : /primary/i /key/i { 1 }
+create_index : /create/i /index/i
- primary_key_index : primary_key index_name(?) '(' field_name(s /,/) ')'
- {
- $return = {
- name => $item{'index_name'}[0],
- type => 'primary_key',
- fields => $item[4],
- }
- }
+not_null : /not/i /null/i { $return = 0 }
- normal_index : key index_name(?) '(' field_name(s /,/) ')'
- {
- $return = {
- name => $item{'index_name'}[0],
- type => 'normal',
- fields => $item[4],
- }
- }
+unsigned : /unsigned/i { $return = 0 }
- unique_index : /unique/i key(?) index_name(?) '(' field_name(s /,/) ')'
- {
- $return = {
- name => $item{'index_name'}[0],
- type => 'unique',
- fields => $item[5],
- }
- }
+default_val : /default/i /(?:')?[\w\d:.-]*(?:')?/
+ {
+ $item[2] =~ s/'//g;
+ $return = $item[2];
+ }
- key : /key/i
- | /index/i
+auto_inc : /auto_increment/i { 1 }
- table_type : /TYPE=/i /\w+/ { $item[2] }
+primary_key : /primary/i /key/i { 1 }
- WORD : /\w+/
+primary_key_index : primary_key index_name(?) '(' field_name(s /,/) ')'
+ {
+ $return = {
+ name => $item{'index_name(?)'}[0],
+ type => 'primary_key',
+ fields => $item[4],
+ };
+ }
- DIGITS : /\d+/
+normal_index : key index_name(?) '(' name_with_opt_paren(s /,/) ')'
+ {
+ $return = {
+ name => $item{'index_name(?)'}[0],
+ type => 'normal',
+ fields => $item[4],
+ }
+ }
- COMMA : ','
+unique_index : unique key(?) index_name(?) '(' name_with_opt_paren(s /,/) ')'
+ {
+ $return = {
+ name => $item{'index_name(?)'}[0],
+ type => 'unique',
+ fields => $item[5],
+ }
+ }
- !;
+fulltext_index : fulltext key(?) index_name(?) '(' name_with_opt_paren(s /,/) ')'
+ {
+ $return = {
+ name => $item{'index_name(?)'}[0],
+ type => 'fulltext',
+ fields => $item[5],
+ }
+ }
-1;
+name_with_opt_paren : NAME parens_value_list(s?)
+ { $item[2][0] ? "$item[1]($item[2][0][0])" : $item[1] }
-#-----------------------------------------------------
-# Where man is not nature is barren.
-# William Blake
-#-----------------------------------------------------
+fulltext : /fulltext/i { 1 }
-=head1 NAME
+unique : /unique/i { 1 }
-SQL::Translator::Parser::MySQL - parser for MySQL
+key : /key/i | /index/i
-=head1 SYNOPSIS
+table_option : /[^\s;]*/
+ {
+ $return = { split /=/, $item[1] }
+ }
- use SQL::Translator;
- use SQL::Translator::Parser::MySQL;
+WORD : /\w+/
- my $translator = SQL::Translator->new;
- $translator->parser("SQL::Translator::Parser::MySQL");
+DIGITS : /\d+/
-=head1 DESCRIPTION
+COMMA : ','
+
+NAME : "`" /\w+/ "`"
+ { $item[2] }
+ | /\w+/
+ { $item[1] }
+
+VALUE : /[-+]?\.?\d+(?:[eE]\d+)?/
+ { $item[1] }
+ | /'.*?'/ # XXX doesn't handle embedded quotes
+ { $item[1] }
+ | /NULL/
+ { 'NULL' }
+# {
+# {
+# value => $item[1],
+# attribute => $item[2]
+# }
+# }
+
+!;
+
+# -------------------------------------------------------------------
+sub parse {
+ my ( $translator, $data, $schema ) = @_;
+ $parser ||= Parse::RecDescent->new($GRAMMAR);
+
+ local $::RD_TRACE = $translator->trace ? 1 : undef;
+ local $DEBUG = $translator->debug;
+
+ unless (defined $parser) {
+ return $translator->error("Error instantiating Parse::RecDescent ".
+ "instance: Bad grammer");
+ }
+
+ my $result = $parser->startrule($data);
+ die "Parse failed.\n" unless defined $result;
+ warn Dumper( $result ) if $DEBUG;
+
+ for my $table_name ( keys %{ $result } ) {
+ my $tdata = $result->{ $table_name };
+ my $table = $schema->add_table(
+ name => $tdata->{'table_name'},
+ );
+
+ my @fields = sort {
+ $tdata->{'fields'}->{$a}->{'order'}
+ <=>
+ $tdata->{'fields'}->{$b}->{'order'}
+ } keys %{ $tdata->{'fields'} };
+
+ for my $fname ( @fields ) {
+ my $fdata = $tdata->{'fields'}{ $fname };
+ my $field = $table->add_field(
+ name => $fdata->{'name'},
+ data_type => $fdata->{'data_type'},
+ size => $fdata->{'size'},
+ default_value => $fdata->{'default'},
+ is_auto_increment => $fdata->{'is_auto_inc'},
+ is_nullable => $fdata->{'null'},
+ );
+ }
+ }
+
+ return $result;
+}
+
+1;
+
+# ----------------------------------------------------
+# Where man is not nature is barren.
+# William Blake
+# ----------------------------------------------------
-Blah blah blah.
+=pod
=head1 AUTHOR
-Ken Y. Clark, kclark@logsoft.com
+Ken Y. Clark E<lt>kclark@cpan.orgE<gt>,
+Chris Mungall E<lt>cjm@fruitfly.orgE<gt>.
=head1 SEE ALSO
-perl(1).
+perl(1), Parse::RecDescent, SQL::Translator::Schema.
=cut