Applied patch from Ryan to uniqify index names sanely for the mysql producer

[dbsrgits/SQL-Translator.git] / lib / SQL / Translator / Utils.pm

diff --git a/lib/SQL/Translator/Utils.pm b/lib/SQL/Translator/Utils.pm
index 88fe399..e85000a 100644 (file)
--- a/lib/SQL/Translator/Utils.pm
+++ b/lib/SQL/Translator/Utils.pm
@@ -1,9 +1,9 @@
 package SQL::Translator::Utils;
 
 # ----------------------------------------------------------------------
-# $Id: Utils.pm,v 1.4 2003-05-09 16:54:03 kycl4rk Exp $
+# $Id: Utils.pm,v 1.12 2004-02-09 23:04:26 kycl4rk Exp $
 # ----------------------------------------------------------------------
-# Copyright (C) 2003 darren chamberlain <darren@cpan.org>
+# Copyright (C) 2002-4 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
@@ -24,12 +24,14 @@ use strict;
 use base qw(Exporter);
 use vars qw($VERSION $DEFAULT_COMMENT @EXPORT_OK);
 
+use Digest::SHA1 qw( sha1_hex );
+
 use Exporter;
 
-$VERSION = 1.00;
+$VERSION = sprintf "%d.%02d", q$Revision: 1.12 $ =~ /(\d+)\.(\d+)/;
 $DEFAULT_COMMENT = '-- ';
 @EXPORT_OK = qw(
-    debug normalize_name header_comment parse_list_arg $DEFAULT_COMMENT
+    debug normalize_name header_comment parse_list_arg truncate_id_uniquely $DEFAULT_COMMENT
 );
 
 # ----------------------------------------------------------------------
@@ -72,7 +74,7 @@ sub debug {
 
 # ----------------------------------------------------------------------
 sub normalize_name {
-    my $name = shift;
+    my $name = shift or return '';
 
     # The name can only begin with a-zA-Z_; if there's anything
     # else, prefix with _
@@ -116,11 +118,57 @@ HEADER_COMMENT
 }
 
 # ----------------------------------------------------------------------
+# parse_list_arg
+#
+# Meant to accept a list, an array reference, or a string of 
+# comma-separated values.  Retuns an array reference of the 
+# arguments.  Modified to also handle a list of references.
+# ----------------------------------------------------------------------
 sub parse_list_arg {
-    return UNIVERSAL::isa( $_[0], 'ARRAY' ) 
-        ? shift 
-        : [ map { s/^\s+|\s+$//g; $_ } map { split /,/ } @_ ]
-    ;
+    my $list = UNIVERSAL::isa( $_[0], 'ARRAY' ) ? shift : [ @_ ];
+
+    #
+    # This protects stringification of references.
+    #
+    if ( @$list && ref $list->[0] ) {
+        return $list;
+    }
+    #
+    # This processes string-like arguments.
+    #
+    else {
+        return [ 
+            map { s/^\s+|\s+$//g; $_ }
+            map { split /,/ }
+            grep { defined && length } @$list
+        ];
+    }
+}
+
+# ----------------------------------------------------------------------
+# truncate_id_uniquely( $desired_name, $max_symbol_length )
+#
+# Truncates the name $desired_name to the $max_symbol_length by
+# including part of the hash of the full name at the end of the
+# truncated name, giving a high probability that the symbol will be
+# unique.
+# ----------------------------------------------------------------------
+my $COLLISION_TAG_LENGTH = 8;
+sub truncate_id_uniquely {
+    my ( $desired_name, $max_symbol_length ) = @_;
+
+    return $desired_name unless defined $desired_name && length $desired_name > $max_symbol_length;
+
+    my $truncated_name = substr $desired_name, 0, $max_symbol_length - $COLLISION_TAG_LENGTH - 1;
+
+    # Hex isn't the most space-efficient, but it skirts around allowed
+    # charset issues
+    my $digest = sha1_hex($desired_name);
+    my $collision_tag = substr $digest, 0, $COLLISION_TAG_LENGTH;
+
+    return $truncated_name
+         . '_'
+         . $collision_tag;
 }
 
 1;
@@ -152,7 +200,7 @@ Nothing is exported by default.
 C<debug> takes 0 or more messages, which will be sent to STDERR using
 C<warn>.  Occurances of the strings I<PKG>, I<SUB>, and I<LINE>
 will be replaced by the calling package, subroutine, and line number,
-respectively, as reported by C<caller(1)>.  
+respectively, as reported by C<caller(1)>.
 
 For example, from within C<foo> in F<SQL/Translator.pm>, at line 666:
 
@@ -204,17 +252,17 @@ this use).  For example, this:
   use SQL::Translator::Utils qw(header_comment $DEFAULT_COMMENT);
 
   print header_comment(__PACKAGE__,
-                       $DEFAULT_COMMENT, 
+                       $DEFAULT_COMMENT,
                        "Hi mom!");
 
 produces:
 
-  -- 
+  --
   -- Created by My::Prodcuer
   -- Created on Fri Apr 25 06:56:02 2003
-  -- 
+  --
   -- Hi mom!
-  -- 
+  --
 
 Note the gratuitous spacing.
 
@@ -230,6 +278,23 @@ All of the following will return equivalent values:
   parse_list_arg( [ 'id', 'name' ] );
   parse_list_arg( qw[ id name ] );
 
+=head2 truncate_id_uniquely
+
+Takes a string ($desired_name) and int ($max_symbol_length). Truncates
+$desired_name to $max_symbol_length by including part of the hash of
+the full name at the end of the truncated name, giving a high
+probability that the symbol will be unique. For example,
+
+  truncate_id_uniquely( 'a' x 100, 64 )
+  truncate_id_uniquely( 'a' x 99 . 'b', 64 );
+  truncate_id_uniquely( 'a' x 99,  64 )
+
+Will give three different results; specifically:
+
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa_7f900025
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa_6191e39a
+  aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa_8cd96af2
+
 =head2 $DEFAULT_COMMENT
 
 This is the default comment string, '-- ' by default.  Useful for