X-Git-Url: http://git.shadowcat.co.uk/gitweb/gitweb.cgi?a=blobdiff_plain;f=lib%2FSQL%2FTranslator.pm;h=232e83aadc72515000ebf87be716059785a982cf;hb=00db1ad50a7fc9696272ca554ea63ac4f03e3b49;hp=02cbc9485a0701ac0ebdffb8c80f2d5a094de409;hpb=c2d3a526263c0948d03e0b2922773eb7db00b0ed;p=dbsrgits%2FSQL-Translator.git diff --git a/lib/SQL/Translator.pm b/lib/SQL/Translator.pm index 02cbc94..232e83a 100644 --- a/lib/SQL/Translator.pm +++ b/lib/SQL/Translator.pm @@ -1,10 +1,7 @@ package SQL::Translator; # ---------------------------------------------------------------------- -# $Id: Translator.pm,v 1.9 2002-07-23 19:21:16 dlc Exp $ -# ---------------------------------------------------------------------- -# Copyright (C) 2002 Ken Y. Clark , -# darren chamberlain +# Copyright (C) 2002-2009 The SQLFairy Authors # # This program is free software; you can redistribute it and/or # modify it under the terms of the GNU General Public License as @@ -21,70 +18,986 @@ package SQL::Translator; # 02111-1307 USA # ------------------------------------------------------------------- +use strict; +use vars qw( $VERSION $DEFAULT_SUB $DEBUG $ERROR ); +use base 'Class::Base'; + +require 5.005; + +$VERSION = '0.09005'; +$DEBUG = 0 unless defined $DEBUG; +$ERROR = ""; + +use Carp qw(carp); + +use Data::Dumper; +use File::Find; +use File::Spec::Functions qw(catfile); +use File::Basename qw(dirname); +use IO::Dir; +use SQL::Translator::Producer; +use SQL::Translator::Schema; + +# ---------------------------------------------------------------------- +# The default behavior is to "pass through" values (note that the +# SQL::Translator instance is the first value ($_[0]), and the stuff +# to be parsed is the second value ($_[1]) +# ---------------------------------------------------------------------- +$DEFAULT_SUB = sub { $_[0]->schema } unless defined $DEFAULT_SUB; + +# ---------------------------------------------------------------------- +# init([ARGS]) +# The constructor. +# +# new takes an optional hash of arguments. These arguments may +# include a parser, specified with the keys "parser" or "from", +# and a producer, specified with the keys "producer" or "to". +# +# The values that can be passed as the parser or producer are +# given directly to the parser or producer methods, respectively. +# See the appropriate method description below for details about +# what each expects/accepts. +# ---------------------------------------------------------------------- +sub init { + my ( $self, $config ) = @_; + # + # Set the parser and producer. + # + # If a 'parser' or 'from' parameter is passed in, use that as the + # parser; if a 'producer' or 'to' parameter is passed in, use that + # as the producer; both default to $DEFAULT_SUB. + # + $self->parser ($config->{'parser'} || $config->{'from'} || $DEFAULT_SUB); + $self->producer($config->{'producer'} || $config->{'to'} || $DEFAULT_SUB); + + # + # Set up callbacks for formatting of pk,fk,table,package names in producer + # MOVED TO PRODUCER ARGS + # + #$self->format_table_name($config->{'format_table_name'}); + #$self->format_package_name($config->{'format_package_name'}); + #$self->format_fk_name($config->{'format_fk_name'}); + #$self->format_pk_name($config->{'format_pk_name'}); + + # + # Set the parser_args and producer_args + # + for my $pargs ( qw[ parser_args producer_args ] ) { + $self->$pargs( $config->{$pargs} ) if defined $config->{ $pargs }; + } + + # + # Initialize the filters. + # + if ( $config->{filters} && ref $config->{filters} eq "ARRAY" ) { + $self->filters( @{$config->{filters}} ) + || return $self->error('Error inititializing filters: '.$self->error); + } + + # + # Set the data source, if 'filename' or 'file' is provided. + # + $config->{'filename'} ||= $config->{'file'} || ""; + $self->filename( $config->{'filename'} ) if $config->{'filename'}; + + # + # Finally, if there is a 'data' parameter, use that in + # preference to filename and file + # + if ( my $data = $config->{'data'} ) { + $self->data( $data ); + } + + # + # Set various other options. + # + $self->{'debug'} = defined $config->{'debug'} ? $config->{'debug'} : $DEBUG; + + $self->add_drop_table( $config->{'add_drop_table'} ); + + $self->no_comments( $config->{'no_comments'} ); + + $self->show_warnings( $config->{'show_warnings'} ); + + $self->trace( $config->{'trace'} ); + + $self->validate( $config->{'validate'} ); + + $self->quote_table_names( (defined $config->{'quote_table_names'} + ? $config->{'quote_table_names'} : 1) ); + $self->quote_field_names( (defined $config->{'quote_field_names'} + ? $config->{'quote_field_names'} : 1) ); + + return $self; +} + +# ---------------------------------------------------------------------- +# add_drop_table([$bool]) +# ---------------------------------------------------------------------- +sub add_drop_table { + my $self = shift; + if ( defined (my $arg = shift) ) { + $self->{'add_drop_table'} = $arg ? 1 : 0; + } + return $self->{'add_drop_table'} || 0; +} + +# ---------------------------------------------------------------------- +# no_comments([$bool]) +# ---------------------------------------------------------------------- +sub no_comments { + my $self = shift; + my $arg = shift; + if ( defined $arg ) { + $self->{'no_comments'} = $arg ? 1 : 0; + } + return $self->{'no_comments'} || 0; +} + + +# ---------------------------------------------------------------------- +# quote_table_names([$bool]) +# ---------------------------------------------------------------------- +sub quote_table_names { + my $self = shift; + if ( defined (my $arg = shift) ) { + $self->{'quote_table_names'} = $arg ? 1 : 0; + } + return $self->{'quote_table_names'} || 0; +} + +# ---------------------------------------------------------------------- +# quote_field_names([$bool]) +# ---------------------------------------------------------------------- +sub quote_field_names { + my $self = shift; + if ( defined (my $arg = shift) ) { + $self->{'quote_field_names'} = $arg ? 1 : 0; + } + return $self->{'quote_field_names'} || 0; +} + +# ---------------------------------------------------------------------- +# producer([$producer_spec]) +# +# Get or set the producer for the current translator. +# ---------------------------------------------------------------------- +sub producer { + shift->_tool({ + name => 'producer', + path => "SQL::Translator::Producer", + default_sub => "produce", + }, @_); +} + +# ---------------------------------------------------------------------- +# producer_type() +# +# producer_type is an accessor that allows producer subs to get +# information about their origin. This is poptentially important; +# since all producer subs are called as subroutine references, there is +# no way for a producer to find out which package the sub lives in +# originally, for example. +# ---------------------------------------------------------------------- +sub producer_type { $_[0]->{'producer_type'} } + +# ---------------------------------------------------------------------- +# producer_args([\%args]) +# +# Arbitrary name => value pairs of paramters can be passed to a +# producer using this method. +# +# If the first argument passed in is undef, then the hash of arguments +# is cleared; all subsequent elements are added to the hash of name, +# value pairs stored as producer_args. +# ---------------------------------------------------------------------- +sub producer_args { shift->_args("producer", @_); } + +# ---------------------------------------------------------------------- +# parser([$parser_spec]) +# ---------------------------------------------------------------------- +sub parser { + shift->_tool({ + name => 'parser', + path => "SQL::Translator::Parser", + default_sub => "parse", + }, @_); +} + +sub parser_type { $_[0]->{'parser_type'}; } + +sub parser_args { shift->_args("parser", @_); } + +# ---------------------------------------------------------------------- +# e.g. +# $sqlt->filters => [ +# sub { }, +# [ "NormalizeNames", field => "lc", tabel => "ucfirst" ], +# [ +# "DataTypeMap", +# "TEXT" => "BIGTEXT", +# ], +# ], +# ---------------------------------------------------------------------- +sub filters { + my $self = shift; + my $filters = $self->{filters} ||= []; + return @$filters unless @_; + + # Set. Convert args to list of [\&code,@args] + foreach (@_) { + my ($filt,@args) = ref($_) eq "ARRAY" ? @$_ : $_; + if ( isa($filt,"CODE") ) { + push @$filters, [$filt,@args]; + next; + } + else { + $self->debug("Adding $filt filter. Args:".Dumper(\@args)."\n"); + $filt = _load_sub("$filt\::filter", "SQL::Translator::Filter") + || return $self->error(__PACKAGE__->error); + push @$filters, [$filt,@args]; + } + } + return @$filters; +} + +# ---------------------------------------------------------------------- +sub show_warnings { + my $self = shift; + my $arg = shift; + if ( defined $arg ) { + $self->{'show_warnings'} = $arg ? 1 : 0; + } + return $self->{'show_warnings'} || 0; +} + + +# filename - get or set the filename +sub filename { + my $self = shift; + if (@_) { + my $filename = shift; + if (-d $filename) { + my $msg = "Cannot use directory '$filename' as input source"; + return $self->error($msg); + } elsif (ref($filename) eq 'ARRAY') { + $self->{'filename'} = $filename; + $self->debug("Got array of files: ".join(', ',@$filename)."\n"); + } elsif (-f _ && -r _) { + $self->{'filename'} = $filename; + $self->debug("Got filename: '$self->{'filename'}'\n"); + } else { + my $msg = "Cannot use '$filename' as input source: ". + "file does not exist or is not readable."; + return $self->error($msg); + } + } + + $self->{'filename'}; +} + +# ---------------------------------------------------------------------- +# data([$data]) +# +# if $self->{'data'} is not set, but $self->{'filename'} is, then +# $self->{'filename'} is opened and read, with the results put into +# $self->{'data'}. +# ---------------------------------------------------------------------- +sub data { + my $self = shift; + + # Set $self->{'data'} based on what was passed in. We will + # accept a number of things; do our best to get it right. + if (@_) { + my $data = shift; + if (isa($data, "SCALAR")) { + $self->{'data'} = $data; + } + else { + if (isa($data, 'ARRAY')) { + $data = join '', @$data; + } + elsif (isa($data, 'GLOB')) { + seek ($data, 0, 0) if eof ($data); + local $/; + $data = <$data>; + } + elsif (! ref $data && @_) { + $data = join '', $data, @_; + } + $self->{'data'} = \$data; + } + } + + # If we have a filename but no data yet, populate. + if (not $self->{'data'} and my $filename = $self->filename) { + $self->debug("Opening '$filename' to get contents.\n"); + local *FH; + local $/; + my $data; + + my @files = ref($filename) eq 'ARRAY' ? @$filename : ($filename); + + foreach my $file (@files) { + unless (open FH, $file) { + return $self->error("Can't read file '$file': $!"); + } + + $data .= ; + + unless (close FH) { + return $self->error("Can't close file '$file': $!"); + } + } + + $self->{'data'} = \$data; + } + + return $self->{'data'}; +} + +# ---------------------------------------------------------------------- +sub reset { +# +# Deletes the existing Schema object so that future calls to translate +# don't append to the existing. +# + my $self = shift; + $self->{'schema'} = undef; + return 1; +} + +# ---------------------------------------------------------------------- +sub schema { +# +# Returns the SQL::Translator::Schema object +# + my $self = shift; + + unless ( defined $self->{'schema'} ) { + $self->{'schema'} = SQL::Translator::Schema->new( + translator => $self, + ); + } + + return $self->{'schema'}; +} + +# ---------------------------------------------------------------------- +sub trace { + my $self = shift; + my $arg = shift; + if ( defined $arg ) { + $self->{'trace'} = $arg ? 1 : 0; + } + return $self->{'trace'} || 0; +} + +# ---------------------------------------------------------------------- +# translate([source], [\%args]) +# +# translate does the actual translation. The main argument is the +# source of the data to be translated, which can be a filename, scalar +# reference, or glob reference. +# +# Alternatively, translate takes optional arguements, which are passed +# to the appropriate places. Most notable of these arguments are +# parser and producer, which can be used to set the parser and +# producer, respectively. This is the applications last chance to set +# these. +# +# translate returns a string. +# ---------------------------------------------------------------------- +sub translate { + my $self = shift; + my ($args, $parser, $parser_type, $producer, $producer_type); + my ($parser_output, $producer_output, @producer_output); + + # Parse arguments + if (@_ == 1) { + # Passed a reference to a hash? + if (isa($_[0], 'HASH')) { + # yep, a hashref + $self->debug("translate: Got a hashref\n"); + $args = $_[0]; + } + + # Passed a GLOB reference, i.e., filehandle + elsif (isa($_[0], 'GLOB')) { + $self->debug("translate: Got a GLOB reference\n"); + $self->data($_[0]); + } + + # Passed a reference to a string containing the data + elsif (isa($_[0], 'SCALAR')) { + # passed a ref to a string + $self->debug("translate: Got a SCALAR reference (string)\n"); + $self->data($_[0]); + } + + # Not a reference; treat it as a filename + elsif (! ref $_[0]) { + # Not a ref, it's a filename + $self->debug("translate: Got a filename\n"); + $self->filename($_[0]); + } + + # Passed something else entirely. + else { + # We're not impressed. Take your empty string and leave. + # return ""; + + # Actually, if data, parser, and producer are set, then we + # can continue. Too bad, because I like my comment + # (above)... + return "" unless ($self->data && + $self->producer && + $self->parser); + } + } + else { + # You must pass in a hash, or you get nothing. + return "" if @_ % 2; + $args = { @_ }; + } + + # ---------------------------------------------------------------------- + # Can specify the data to be transformed using "filename", "file", + # "data", or "datasource". + # ---------------------------------------------------------------------- + if (my $filename = ($args->{'filename'} || $args->{'file'})) { + $self->filename($filename); + } + + if (my $data = ($args->{'data'} || $args->{'datasource'})) { + $self->data($data); + } + + # ---------------------------------------------------------------- + # Get the data. + # ---------------------------------------------------------------- + my $data = $self->data; + + # ---------------------------------------------------------------- + # Local reference to the parser subroutine + # ---------------------------------------------------------------- + if ($parser = ($args->{'parser'} || $args->{'from'})) { + $self->parser($parser); + } + $parser = $self->parser; + $parser_type = $self->parser_type; + + # ---------------------------------------------------------------- + # Local reference to the producer subroutine + # ---------------------------------------------------------------- + if ($producer = ($args->{'producer'} || $args->{'to'})) { + $self->producer($producer); + } + $producer = $self->producer; + $producer_type = $self->producer_type; + + # ---------------------------------------------------------------- + # Execute the parser, the filters and then execute the producer. + # Allowances are made for each piece to die, or fail to compile, + # since the referenced subroutines could be almost anything. In + # the future, each of these might happen in a Safe environment, + # depending on how paranoid we want to be. + # ---------------------------------------------------------------- + + # Run parser + unless ( defined $self->{'schema'} ) { + eval { $parser_output = $parser->($self, $$data) }; + if ($@ || ! $parser_output) { + my $msg = sprintf "translate: Error with parser '%s': %s", + $parser_type, ($@) ? $@ : " no results"; + return $self->error($msg); + } + } + $self->debug("Schema =\n", Dumper($self->schema), "\n"); + + # Validate the schema if asked to. + if ($self->validate) { + my $schema = $self->schema; + return $self->error('Invalid schema') unless $schema->is_valid; + } + + # Run filters + my $filt_num = 0; + foreach ($self->filters) { + $filt_num++; + my ($code,@args) = @$_; + eval { $code->($self->schema, @args) }; + my $err = $@ || $self->error || 0; + return $self->error("Error with filter $filt_num : $err") if $err; + } + + # Run producer + # Calling wantarray in the eval no work, wrong scope. + my $wantarray = wantarray ? 1 : 0; + eval { + if ($wantarray) { + @producer_output = $producer->($self); + } else { + $producer_output = $producer->($self); + } + }; + if ($@ || !( $producer_output || @producer_output)) { + my $err = $@ || $self->error || "no results"; + my $msg = "translate: Error with producer '$producer_type': $err"; + return $self->error($msg); + } + + return wantarray ? @producer_output : $producer_output; +} + +# ---------------------------------------------------------------------- +# list_parsers() +# +# Hacky sort of method to list all available parsers. This has +# several problems: +# +# - Only finds things in the SQL::Translator::Parser namespace +# +# - Only finds things that are located in the same directory +# as SQL::Translator::Parser. Yeck. +# +# This method will fail in several very likely cases: +# +# - Parser modules in different namespaces +# +# - Parser modules in the SQL::Translator::Parser namespace that +# have any XS componenets will be installed in +# arch_lib/SQL/Translator. +# +# ---------------------------------------------------------------------- +sub list_parsers { + return shift->_list("parser"); +} + +# ---------------------------------------------------------------------- +# list_producers() +# +# See notes for list_parsers(), above; all the problems apply to +# list_producers as well. +# ---------------------------------------------------------------------- +sub list_producers { + return shift->_list("producer"); +} + + +# ====================================================================== +# Private Methods +# ====================================================================== + +# ---------------------------------------------------------------------- +# _args($type, \%args); +# +# Gets or sets ${type}_args. Called by parser_args and producer_args. +# ---------------------------------------------------------------------- +sub _args { + my $self = shift; + my $type = shift; + $type = "${type}_args" unless $type =~ /_args$/; + + unless (defined $self->{$type} && isa($self->{$type}, 'HASH')) { + $self->{$type} = { }; + } + + if (@_) { + # If the first argument is an explicit undef (remember, we + # don't get here unless there is stuff in @_), then we clear + # out the producer_args hash. + if (! defined $_[0]) { + shift @_; + %{$self->{$type}} = (); + } + + my $args = isa($_[0], 'HASH') ? shift : { @_ }; + %{$self->{$type}} = (%{$self->{$type}}, %$args); + } + + $self->{$type}; +} + +# ---------------------------------------------------------------------- +# Does the get/set work for parser and producer. e.g. +# return $self->_tool({ +# name => 'producer', +# path => "SQL::Translator::Producer", +# default_sub => "produce", +# }, @_); +# ---------------------------------------------------------------------- +sub _tool { + my ($self,$args) = (shift, shift); + my $name = $args->{name}; + return $self->{$name} unless @_; # get accessor + + my $path = $args->{path}; + my $default_sub = $args->{default_sub}; + my $tool = shift; + + # passed an anonymous subroutine reference + if (isa($tool, 'CODE')) { + $self->{$name} = $tool; + $self->{"$name\_type"} = "CODE"; + $self->debug("Got $name: code ref\n"); + } + + # Module name was passed directly + # We try to load the name; if it doesn't load, there's a + # possibility that it has a function name attached to it, + # so we give it a go. + else { + $tool =~ s/-/::/g if $tool !~ /::/; + my ($code,$sub); + ($code,$sub) = _load_sub("$tool\::$default_sub", $path); + unless ($code) { + if ( __PACKAGE__->error =~ m/Can't find module/ ) { + # Mod not found so try sub + ($code,$sub) = _load_sub("$tool", $path) unless $code; + die "Can't load $name subroutine '$tool' : ".__PACKAGE__->error + unless $code; + } + else { + die "Can't load $name '$tool' : ".__PACKAGE__->error; + } + } + + # get code reference and assign + my (undef,$module,undef) = $sub =~ m/((.*)::)?(\w+)$/; + $self->{$name} = $code; + $self->{"$name\_type"} = $sub eq "CODE" ? "CODE" : $module; + $self->debug("Got $name: $sub\n"); + } + + # At this point, $self->{$name} contains a subroutine + # reference that is ready to run + + # Anything left? If so, it's args + my $meth = "$name\_args"; + $self->$meth(@_) if (@_); + + return $self->{$name}; +} + +# ---------------------------------------------------------------------- +# _list($type) +# ---------------------------------------------------------------------- +sub _list { + my $self = shift; + my $type = shift || return (); + my $uctype = ucfirst lc $type; + + # + # First find all the directories where SQL::Translator + # parsers or producers (the "type") appear to live. + # + load("SQL::Translator::$uctype") or return (); + my $path = catfile "SQL", "Translator", $uctype; + my @dirs; + for (@INC) { + my $dir = catfile $_, $path; + $self->debug("_list_${type}s searching $dir\n"); + next unless -d $dir; + push @dirs, $dir; + } + + # + # Now use File::File::find to look recursively in those + # directories for all the *.pm files, then present them + # with the slashes turned into dashes. + # + my %found; + find( + sub { + if ( -f && m/\.pm$/ ) { + my $mod = $_; + $mod =~ s/\.pm$//; + my $cur_dir = $File::Find::dir; + my $base_dir = quotemeta catfile 'SQL', 'Translator', $uctype; + + # + # See if the current directory is below the base directory. + # + if ( $cur_dir =~ m/$base_dir(.*)/ ) { + $cur_dir = $1; + $cur_dir =~ s!^/!!; # kill leading slash + $cur_dir =~ s!/!-!g; # turn other slashes into dashes + } + else { + $cur_dir = ''; + } + + $found{ join '-', map { $_ || () } $cur_dir, $mod } = 1; + } + }, + @dirs + ); + + return sort { lc $a cmp lc $b } keys %found; +} + +# ---------------------------------------------------------------------- +# load(MODULE [,PATH[,PATH]...]) +# +# Loads a Perl module. Short circuits if a module is already loaded. +# +# MODULE - is the name of the module to load. +# +# PATH - optional list of 'package paths' to look for the module in. e.g +# If you called load('Super::Foo' => 'My', 'Other') it will +# try to load the mod Super::Foo then My::Super::Foo then Other::Super::Foo. +# +# Returns package name of the module actually loaded or false and sets error. +# +# Note, you can't load a name from the root namespace (ie one without '::' in +# it), therefore a single word name without a path fails. +# ---------------------------------------------------------------------- +sub load { + my $name = shift; + my @path; + push @path, "" if $name =~ /::/; # Empty path to check name on its own first + push @path, @_ if @_; + + foreach (@path) { + my $module = $_ ? "$_\::$name" : $name; + my $file = $module; $file =~ s[::][/]g; $file .= ".pm"; + __PACKAGE__->debug("Loading $name as $file\n"); + return $module if $INC{$file}; # Already loaded + + eval { require $file }; + next if $@ =~ /Can't locate $file in \@INC/; + eval { $module->import() } unless $@; + return __PACKAGE__->error("Error loading $name as $module : $@") + if $@ && $@ !~ /"SQL::Translator::Producer" is not exported/; + + return $module; # Module loaded ok + } + + return __PACKAGE__->error("Can't find module $name. Path:".join(",",@path)); +} + +# ---------------------------------------------------------------------- +# Load the sub name given (including package), optionally using a base package +# path. Returns code ref and name of sub loaded, including its package. +# (\&code, $sub) = load_sub( 'MySQL::produce', "SQL::Translator::Producer" ); +# (\&code, $sub) = load_sub( 'MySQL::produce', @path ); +# ---------------------------------------------------------------------- +sub _load_sub { + my ($tool, @path) = @_; + + my (undef,$module,$func_name) = $tool =~ m/((.*)::)?(\w+)$/; + if ( my $module = load($module => @path) ) { + my $sub = "$module\::$func_name"; + return wantarray ? ( \&{ $sub }, $sub ) : \&$sub; + } + return undef; +} + +# ---------------------------------------------------------------------- +sub format_table_name { + return shift->_format_name('_format_table_name', @_); +} + +# ---------------------------------------------------------------------- +sub format_package_name { + return shift->_format_name('_format_package_name', @_); +} + +# ---------------------------------------------------------------------- +sub format_fk_name { + return shift->_format_name('_format_fk_name', @_); +} + +# ---------------------------------------------------------------------- +sub format_pk_name { + return shift->_format_name('_format_pk_name', @_); +} + +# ---------------------------------------------------------------------- +# The other format_*_name methods rely on this one. It optionally +# accepts a subroutine ref as the first argument (or uses an identity +# sub if one isn't provided or it doesn't already exist), and applies +# it to the rest of the arguments (if any). +# ---------------------------------------------------------------------- +sub _format_name { + my $self = shift; + my $field = shift; + my @args = @_; + + if (ref($args[0]) eq 'CODE') { + $self->{$field} = shift @args; + } + elsif (! exists $self->{$field}) { + $self->{$field} = sub { return shift }; + } + + return @args ? $self->{$field}->(@args) : $self->{$field}; +} + +# ---------------------------------------------------------------------- +# isa($ref, $type) +# +# Calls UNIVERSAL::isa($ref, $type). I think UNIVERSAL::isa is ugly, +# but I like function overhead. +# ---------------------------------------------------------------------- +sub isa($$) { + my ($ref, $type) = @_; + return UNIVERSAL::isa($ref, $type); +} + +# ---------------------------------------------------------------------- +# version +# +# Returns the $VERSION of the main SQL::Translator package. +# ---------------------------------------------------------------------- +sub version { + my $self = shift; + return $VERSION; +} + +# ---------------------------------------------------------------------- +sub validate { + my ( $self, $arg ) = @_; + if ( defined $arg ) { + $self->{'validate'} = $arg ? 1 : 0; + } + return $self->{'validate'} || 0; +} + +1; + +# ---------------------------------------------------------------------- +# Who killed the pork chops? +# What price bananas? +# Are you my Angel? +# Allen Ginsberg +# ---------------------------------------------------------------------- + +=pod + =head1 NAME -SQL::Translator - convert schema from one database to another +SQL::Translator - manipulate structured data definitions (SQL and more) =head1 SYNOPSIS use SQL::Translator; - my $translator = SQL::Translator->new; - my $output = $translator->translate( - from => "MySQL", - to => "Oracle", - filename => $file, - ) or die $translator->error; + my $translator = SQL::Translator->new( + # Print debug info + debug => 1, + # Print Parse::RecDescent trace + trace => 0, + # Don't include comments in output + no_comments => 0, + # Print name mutations, conflicts + show_warnings => 0, + # Add "drop table" statements + add_drop_table => 1, + # to quote or not to quote, thats the question + quote_table_names => 1, + quote_field_names => 1, + # Validate schema object + validate => 1, + # Make all table names CAPS in producers which support this option + format_table_name => sub {my $tablename = shift; return uc($tablename)}, + # Null-op formatting, only here for documentation's sake + format_package_name => sub {return shift}, + format_fk_name => sub {return shift}, + format_pk_name => sub {return shift}, + ); + + my $output = $translator->translate( + from => 'MySQL', + to => 'Oracle', + # Or an arrayref of filenames, i.e. [ $file1, $file2, $file3 ] + filename => $file, + ) or die $translator->error; + print $output; =head1 DESCRIPTION -This module attempts to simplify the task of converting one database -create syntax to another through the use of Parsers (which understand -the sourced format) and Producers (which understand the destination -format). The idea is that any Parser can be used with any Producer in -the conversion process. So, if you wanted PostgreSQL-to-Oracle, you -would use the PostgreSQL parser and the Oracle producer. +This documentation covers the API for SQL::Translator. For a more general +discussion of how to use the modules and scripts, please see +L. + +SQL::Translator is a group of Perl modules that converts +vendor-specific SQL table definitions into other formats, such as +other vendor-specific SQL, ER diagrams, documentation (POD and HTML), +XML, and Class::DBI classes. The main focus of SQL::Translator is +SQL, but parsers exist for other structured data formats, including +Excel spreadsheets and arbitrarily delimited text files. Through the +separation of the code into parsers and producers with an object model +in between, it's possible to combine any parser with any producer, to +plug in custom parsers or producers, or to manipulate the parsed data +via the built-in object model. Presently only the definition parts of +SQL are handled (CREATE, ALTER), not the manipulation of data (INSERT, +UPDATE, DELETE). + +=head1 CONSTRUCTOR -=cut +The constructor is called C, and accepts a optional hash of options. +Valid options are: -use strict; -use vars qw($VERSION $DEFAULT_SUB $DEBUG $ERROR); -use base qw(Class::Base); +=over 4 -$VERSION = sprintf "%d.%02d", q$Revision: 1.9 $ =~ /(\d+)\.(\d+)/; -$DEBUG = 1 unless defined $DEBUG; -$ERROR = ""; +=item * -use Carp qw(carp); +parser / from -# ---------------------------------------------------------------------- -# The default behavior is to "pass through" values (note that the -# SQL::Translator instance is the first value ($_[0]), and the stuff -# to be parsed is the second value ($_[1]) -# ---------------------------------------------------------------------- -$DEFAULT_SUB = sub { $_[1] } unless defined $DEFAULT_SUB; +=item * + +parser_args + +=item * + +producer / to + +=item * + +producer_args + +=item * + +filters + +=item * + +filename / file + +=item * + +data + +=item * -=head1 CONSTRUCTOR +debug -The constructor is called B, and accepts a optional hash of options. -Valid options are: +=item * -=over 4 +add_drop_table + +=item * + +quote_table_names + +=item * -=item parser (aka from) +quote_field_names -=item parser_args +=item * -=item producer (aka to) +no_comments -=item producer_args +=item * -=item filename (aka file) +trace -=item data +=item * -=item debug +validate =back @@ -92,88 +1005,54 @@ All options are, well, optional; these attributes can be set via instance methods. Internally, they are; no (non-syntactical) advantage is gained by passing options to the constructor. -=cut +=head1 METHODS -# ---------------------------------------------------------------------- -# init([ARGS]) -# The constructor. -# -# new takes an optional hash of arguments. These arguments may -# include a parser, specified with the keys "parser" or "from", -# and a producer, specified with the keys "producer" or "to". -# -# The values that can be passed as the parser or producer are -# given directly to the parser or producer methods, respectively. -# See the appropriate method description below for details about -# what each expects/accepts. -# ---------------------------------------------------------------------- -sub init { - my ($self, $config) = @_; +=head2 add_drop_table - # ------------------------------------------------------------------ - # Set the parser and producer. - # - # If a 'parser' or 'from' parameter is passed in, use that as the - # parser; if a 'producer' or 'to' parameter is passed in, use that - # as the producer; both default to $DEFAULT_SUB. - # ------------------------------------------------------------------ - $self->parser( $config->{'parser'} || $config->{'from'} || $DEFAULT_SUB); - $self->producer($config->{'producer'} || $config->{'to'} || $DEFAULT_SUB); +Toggles whether or not to add "DROP TABLE" statements just before the +create definitions. - # ------------------------------------------------------------------ - # Set the parser_args and producer_args - # ------------------------------------------------------------------ - for my $pargs (qw(parser_args producer_args)) { - $self->$pargs($config->{$pargs}) if defined $config->{$pargs}; - } +=head2 quote_table_names - # ------------------------------------------------------------------ - # Set the data source, if 'filename' or 'file' is provided. - # ------------------------------------------------------------------ - $config->{'filename'} ||= $config->{'file'} || ""; - $self->filename($config->{'filename'}) if $config->{'filename'}; +Toggles whether or not to quote table names with " in DROP and CREATE +statements. The default (true) is to quote them. - # ------------------------------------------------------------------ - # Finally, if there is a 'data' parameter, use that in preference - # to filename and file - # ------------------------------------------------------------------ - if (my $data = $config->{'data'}) { - $self->data($data); - } +=head2 quote_field_names - $self->{'debug'} = $DEBUG; - $self->{'debug'} = $config->{'debug'} if (defined $config->{'debug'}); +Toggles whether or not to quote field names with " in most +statements. The default (true), is to quote them. - return $self; -} +=head2 no_comments -=head1 METHODS +Toggles whether to print comments in the output. Accepts a true or false +value, returns the current value. -=head2 B +=head2 producer -The B method is an accessor/mutator, used to retrieve or +The C method is an accessor/mutator, used to retrieve or define what subroutine is called to produce the output. A subroutine -defined as a producer will be invoked as a function (not a method) and -passed 2 parameters: its container SQL::Translator instance and a -data structure. It is expected that the function transform the data -structure to a string. The SQL::Transformer instance is provided for -informational purposes; for example, the type of the parser can be -retrieved using the B method, and the B and -B methods can be called when needed. - -When defining a producer, one of several things can be passed -in: A module name (e.g., My::Groovy::Producer), a module name -relative to the SQL::Translator::Producer namespace (e.g., MySQL), a -module name and function combination (My::Groovy::Producer::transmogrify), +defined as a producer will be invoked as a function (I) +and passed its container C instance, which it should +call the C method on, to get the C +generated by the parser. It is expected that the function transform the +schema structure to a string. The C instance is also useful +for informational purposes; for example, the type of the parser can be +retrieved using the C method, and the C and +C methods can be called when needed. + +When defining a producer, one of several things can be passed in: A +module name (e.g., C), a module name relative to +the C namespace (e.g., C), a module +name and function combination (C), or a reference to an anonymous subroutine. If a full module name is passed in (for the purposes of this method, a string containing "::" is considered to be a module name), it is treated as a package, and a -function called "produce" will be invoked: $modulename::produce. If -$modulename cannot be loaded, the final portion is stripped off and +function called "produce" will be invoked: C<$modulename::produce>. +If $modulename cannot be loaded, the final portion is stripped off and treated as a function. In other words, if there is no file named -My/Groovy/Producer/transmogrify.pm, SQL::Translator will attempt to load -My/Groovy/Producer.pm and use transmogrify as the name of the function, -instead of the default "produce". +F, C will attempt +to load F and use C as the name of +the function, instead of the default C. my $tr = SQL::Translator->new; @@ -192,12 +1071,12 @@ instead of the default "produce". # $subref->($tr, $data); $tr->producer(\&my_producer); -There is also a method named B, which is a string -containing the classname to which the above B function +There is also a method named C, which is a string +containing the classname to which the above C function belongs. In the case of anonymous subroutines, this method returns the string "CODE". -Finally, there is a method named B, which is both an +Finally, there is a method named C, which is both an accessor and a mutator. Arbitrary data may be stored in name => value pairs for the producer subroutine to access: @@ -207,8 +1086,8 @@ pairs for the producer subroutine to access: # $pr_args is a hashref. -Extra data passed to the B method is passed to -B: +Extra data passed to the C method is passed to +C: $tr->producer("xSV", delimiter => ',\s*'); @@ -216,106 +1095,14 @@ B: my $args = $tr->producer_args; my $delimiter = $args->{'delimiter'}; # value is ,\s* -=cut - -# producer and producer_type -sub producer { - my $self = shift; - - # producer as a mutator - if (@_) { - my $producer = shift; - - # Passed a module name (string containing "::") - if ($producer =~ /::/) { - my $func_name; - - # Module name was passed directly - # We try to load the name; if it doesn't load, there's - # a possibility that it has a function name attached to - # it. - if (load($producer)) { - $func_name = "produce"; - } - - # Module::function was passed - else { - # Passed Module::Name::function; try to recover - my @func_parts = split /::/, $producer; - $func_name = pop @func_parts; - $producer = join "::", @func_parts; - - # If this doesn't work, then we have a legitimate - # problem. - load($producer) or die "Can't load $producer: $@"; - } - - # get code reference and assign - $self->{'producer'} = \&{ "$producer\::$func_name" }; - $self->{'producer_type'} = $producer; - $self->debug("Got producer: $producer\::$func_name"); - } - - # passed an anonymous subroutine reference - elsif (isa($producer, 'CODE')) { - $self->{'producer'} = $producer; - $self->{'producer_type'} = "CODE"; - $self->debug("Got producer: code ref"); - } - - # passed a string containing no "::"; relative package name - else { - my $Pp = sprintf "SQL::Translator::Producer::$producer"; - load($Pp) or die "Can't load $Pp: $@"; - $self->{'producer'} = \&{ "$Pp\::produce" }; - $self->{'producer_type'} = $Pp; - $self->debug("Got producer: $Pp"); - } - - # At this point, $self->{'producer'} contains a subroutine - # reference that is ready to run - - # Anything left? If so, it's producer_args - $self->producer_args(@_) if (@_); - } - - return $self->{'producer'}; -}; - -# ---------------------------------------------------------------------- -# producer_type -# -# producer_type is an accessor that allows producer subs to get -# information about their origin. This is poptentially important; -# since all producer subs are called as subroutine refernces, there is -# no way for a producer to find out which package the sub lives in -# originally, for example. -# ---------------------------------------------------------------------- -sub producer_type { $_[0]->{'producer_type'} } - -# ---------------------------------------------------------------------- -# producer_args -# -# Arbitrary name => value pairs of paramters can be passed to a -# producer using this method. -# ---------------------------------------------------------------------- -sub producer_args { - my $self = shift; - if (@_) { - my $args = isa($_[0], 'HASH') ? shift : { @_ }; - $self->{'producer_args'} = $args; - } - $self->{'producer_args'}; -} - -=head2 B +=head2 parser -The B method defines or retrieves a subroutine that will be +The C method defines or retrieves a subroutine that will be called to perform the parsing. The basic idea is the same as that of -B (see above), except the default subroutine name is -"parse", and will be invoked as $module_name::parse($tr, $data). +C (see above), except the default subroutine name is +"parse", and will be invoked as C<$module_name::parse($tr, $data)>. Also, the parser subroutine will be passed a string containing the -entirety of the data to be parsed (or possibly a reference to a string?). +entirety of the data to be parsed. # Invokes SQL::Translator::Parser::MySQL::parse() $tr->parser("MySQL"); @@ -330,101 +1117,84 @@ entirety of the data to be parsed (or possibly a reference to a string?). return $dumper->Dump; }); -There is also B and B, which perform -analogously to B and B - -=cut - -sub parser { - my $self = shift; - - # parser as a mutator - if (@_) { - my $parser = shift; - - # Passed a module name (string containing "::") - if ($parser =~ /::/) { - my $func_name; - - # Module name was passed directly - # We try to load the name; if it doesn't load, there's - # a possibility that it has a function name attached to - # it. - if (load($parser)) { - $func_name = "parse"; - } - - # Module::function was passed - else { - # Passed Module::Name::function; try to recover - my @func_parts = split /::/, $parser; - $func_name = pop @func_parts; - $parser = join "::", @func_parts; - - # If this doesn't work, then we have a legitimate - # problem. - load($parser) or die "Can't load $parser: $@"; - } - - # get code reference and assign - $self->{'parser'} = \&{ "$parser\::$func_name" }; - $self->{'parser_type'} = $parser; - $self->debug("Got parser: $parser\::$func_name"); - } - - # passed an anonymous subroutine reference - elsif (isa($parser, 'CODE')) { - $self->{'parser'} = $parser; - $self->{'parser_type'} = "CODE"; - $self->debug("Got parser: code ref"); - } - - # passed a string containing no "::"; relative package name - else { - my $Pp = sprintf "SQL::Translator::Parser::$parser"; - load($Pp) or die "Can't load $Pp: $@"; - $self->{'parser'} = \&{ "$Pp\::parse" }; - $self->{'parser_type'} = $Pp; - $self->debug("Got parser: $Pp"); - } - - # At this point, $self->{'parser'} contains a subroutine - # reference that is ready to run - - $self->parser_args(@_) if (@_); - } - - return $self->{'parser'}; -} - -sub parser_type { $_[0]->{'parser_type'} } - -# parser_args -sub parser_args { - my $self = shift; - if (@_) { - my $args = isa($_[0], 'HASH') ? shift : { @_ }; - $self->{'parser_args'} = $args; - } - $self->{'parser_args'}; -} - -=head2 B - -The B method calls the subroutines referenced by the -B and B data members (described above). It accepts -as arguments a number of things, in key => value format, including -(potentially) a parser and a producer (they are passed directly to the -B and B methods). - -Here is how the parameter list to B is parsed: +There is also C and C, which perform +analogously to C and C + +=head2 filters + +Set or retreive the filters to run over the schema during the +translation, before the producer creates its output. Filters are sub +routines called, in order, with the schema object to filter as the 1st +arg and a hash of options (passed as a list) for the rest of the args. +They are free to do whatever they want to the schema object, which will be +handed to any following filters, then used by the producer. + +Filters are set as an array, which gives the order they run in. +Like parsers and producers, they can be defined by a module name, a +module name relative to the SQL::Translator::Filter namespace, a module +name and function name together or a reference to an anonymous subroutine. +When using a module name a function called C will be invoked in +that package to do the work. + +To pass args to the filter set it as an array ref with the 1st value giving +the filter (name or sub) and the rest its args. e.g. + + $tr->filters( + sub { + my $schema = shift; + # Do stuff to schema here! + }, + DropFKeys, + [ "Names", table => 'lc' ], + [ "Foo", foo => "bar", hello => "world" ], + [ "Filter5" ], + ); + +Although you normally set them in the constructor, which calls +through to filters. i.e. + + my $translator = SQL::Translator->new( + ... + filters => [ + sub { ... }, + [ "Names", table => 'lc' ], + ], + ... + ); + +See F for more examples. + +Multiple set calls to filters are cumulative with new filters added to +the end of the current list. + +Returns the filters as a list of array refs, the 1st value being a +reference to the filter sub and the rest its args. + +=head2 show_warnings + +Toggles whether to print warnings of name conflicts, identifier +mutations, etc. Probably only generated by producers to let the user +know when something won't translate very smoothly (e.g., MySQL "enum" +fields into Oracle). Accepts a true or false value, returns the +current value. + +=head2 translate + +The C method calls the subroutine referenced by the +C data member, then calls any C and finally calls +the C sub routine (these members are described above). +It accepts as arguments a number of things, in key => value format, +including (potentially) a parser and a producer (they are passed +directly to the C and C methods). + +Here is how the parameter list to C is parsed: =over =item * 1 argument means it's the data to be parsed; which could be a string -(filename) or a refernce to a scalar (a string stored in memory), or a +(filename) or a reference to a scalar (a string stored in memory), or a reference to a hash, which is parsed as being more than one argument (see next section). @@ -455,12 +1225,12 @@ You get the idea. =back -=head2 B, B +=head2 filename, data -Using the B method, the filename of the data to be parsed -can be set. This method can be used in conjunction with the B -method, below. If both the B and B methods are -invoked as mutators, the data set in the B method is used. +Using the C method, the filename of the data to be parsed +can be set. This method can be used in conjunction with the C +method, below. If both the C and C methods are +invoked as mutators, the data set in the C method is used. $tr->filename("/my/data/files/create.sql"); @@ -473,209 +1243,42 @@ or: }; $tr->data(\$create_script); -B takes a string, which is interpreted as a filename. -B takes a reference to a string, which is used as the data to be +C takes a string, which is interpreted as a filename. +C takes a reference to a string, which is used as the data to be parsed. If a filename is set, then that file is opened and read when -the B method is called, as long as the data instance +the C method is called, as long as the data instance variable is not set. -=cut - -# filename - get or set the filename -sub filename { - my $self = shift; - if (@_) { - my $filename = shift; - if (-d $filename) { - my $msg = "Cannot use directory '$filename' as input source"; - return $self->error($msg); - } elsif (-f _ && -r _) { - $self->{'filename'} = $filename; - $self->debug("Got filename: $self->{'filename'}"); - } else { - my $msg = "Cannot use '$filename' as input source: ". - "file does not exist or is not readable."; - return $self->error($msg); - } - } - - $self->{'filename'}; -} - -# data - get or set the data -# if $self->{'data'} is not set, but $self->{'filename'} is, then -# $self->{'filename'} is opened and read, whith the results put into -# $self->{'data'}. -sub data { - my $self = shift; - - # Set $self->{'data'} to $_[0], if it is provided. - if (@_) { - my $data = shift; - if (isa($data, "SCALAR")) { - $self->{'data'} = $data; - } - elsif (! ref $data) { - $self->{'data'} = \$data; - } - } - - # If we have a filename but no data yet, populate. - if (not $self->{'data'} and my $filename = $self->filename) { - $self->debug("Opening '$filename' to get contents..."); - local *FH; - local $/; - my $data; - - unless (open FH, $filename) { - return $self->error("Can't open $filename for reading: $!"); - } - - $data = ; - $self->{'data'} = \$data; - - unless (close FH) { - return $self->error("Can't close $filename: $!"); - } - } - - return $self->{'data'}; -} - -# translate -sub translate { - my $self = shift; - my ($args, $parser, $parser_type, $producer, $producer_type); - my ($parser_output, $producer_output); - - # Parse arguments - if (@_ == 1) { - # Passed a reference to a hash? - if (isa($_[0], 'HASH')) { - # yep, a hashref - $self->debug("translate: Got a hashref"); - $args = $_[0]; - } - - # Passed a reference to a string containing the data - elsif (isa($_[0], 'SCALAR')) { - # passed a ref to a string - $self->debug("translate: Got a SCALAR reference (string)"); - $self->data($_[0]); - } - - # Not a reference; treat it as a filename - elsif (! ref $_[0]) { - # Not a ref, it's a filename - $self->debug("translate: Got a filename"); - $self->filename($_[0]); - } - - # Passed something else entirely. - else { - # We're not impressed. Take your empty string and leave. - # return ""; - - # Actually, if data, parser, and producer are set, then we - # can continue. Too bad, because I like my comment - # (above)... - return "" unless ($self->data && - $self->producer && - $self->parser); - } - } - else { - # You must pass in a hash, or you get nothing. - return "" if @_ % 2; - $args = { @_ }; - } - - # ---------------------------------------------------------------------- - # Can specify the data to be transformed using "filename", "file", - # "data", or "datasource". - # ---------------------------------------------------------------------- - if (my $filename = ($args->{'filename'} || $args->{'file'})) { - $self->filename($filename); - } +=head2 schema - if (my $data = ($self->{'data'} || $self->{'datasource'})) { - $self->data($data); - } +Returns the SQL::Translator::Schema object. - # ---------------------------------------------------------------- - # Get the data. - # ---------------------------------------------------------------- - my $data = $self->data; - unless (length $$data) { - return $self->error("Empty data file!"); - } +=head2 trace - # ---------------------------------------------------------------- - # Local reference to the parser subroutine - # ---------------------------------------------------------------- - if ($parser = ($args->{'parser'} || $args->{'from'})) { - $self->parser($parser); - } - $parser = $self->parser; - $parser_type = $self->parser_type; +Turns on/off the tracing option of Parse::RecDescent. - # ---------------------------------------------------------------- - # Local reference to the producer subroutine - # ---------------------------------------------------------------- - if ($producer = ($args->{'producer'} || $args->{'to'})) { - $self->producer($producer); - } - $producer = $self->producer; - $producer_type = $self->producer_type; +=head2 validate - # ---------------------------------------------------------------- - # Execute the parser, then execute the producer with that output. - # Allowances are made for each piece to die, or fail to compile, - # since the referenced subroutines could be almost anything. In - # the future, each of these might happen in a Safe environment, - # depending on how paranoid we want to be. - # ---------------------------------------------------------------- - eval { $parser_output = $parser->($self, $$data) }; - if ($@ || ! $parser_output) { - my $msg = sprintf "translate: Error with parser '%s': %s", - $parser_type, ($@) ? $@ : " no results"; - return $self->error($msg); - } +Whether or not to validate the schema object after parsing and before +producing. - eval { $producer_output = $producer->($self, $parser_output) }; - if ($@ || ! $producer_output) { - my $msg = sprintf "translate: Error with producer '%s': %s", - $producer_type, ($@) ? $@ : " no results"; - return $self->error($msg); - } +=head2 version - return $producer_output; -} +Returns the version of the SQL::Translator release. -sub load { - my $module = do { my $m = shift; $m =~ s[::][/]g; "$m.pm" }; - return 1 if $INC{$module}; - - eval { require $module }; - - return if ($@); - return 1; -} +=head1 AUTHORS -sub isa { UNIVERSAL::isa($_[0], $_[1]) } +See the included AUTHORS file: +L -1; +If you would like to contribute to the project, you can send patches +to the developers mailing list: -__END__ -#----------------------------------------------------- -# Rescue the drowning and tie your shoestrings. -# Henry David Thoreau -#----------------------------------------------------- + sqlfairy-developers@lists.sourceforge.net -=head1 AUTHORS +Or send us a message (with your Sourceforge username) asking to be +added to the project and what you'd like to contribute. -Ken Y. Clark, Ekclark@logsoft.comE, -darren chamberlain Edarren@cpan.orgE =head1 COPYRIGHT @@ -693,8 +1296,23 @@ along with this program; if not, write to the Free Software Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA -=head1 SEE ALSO +=head1 BUGS -L, L +Please use L for reporting bugs. + +=head1 PRAISE + +If you find this module useful, please use +L to rate it. + +=head1 SEE ALSO -=cut +L, +L, +L, +L, +L, +L, +L, +L, +L.