Create a database of only selected objects with a script

Why not just use the database scripting already provided in Management studio then transfer your data with DTS

in sql management studio
Right click on Database - Taks - Generate scripts

This will let you select what objects you want to script then save your script to a file.
You can the use the import export wizard to transfer your data

I did not know that part of the SSMS existed, thanks.  

I will go with this method but is there any way to make it run based object sets I set saying what to copy and what not to?  The database I am copying has close to 380 tables, 456 stored procedures, and numerous other things.  To go through a wizard may take a while.

For the data transfer I was hoping to keep it with a single script but pick and choose data to transfer.  You can fashion your query with DTS instead of just bulk copying a table correct?

I know the wizards can do a lot but it needs a person unchecking a lot of stuff too.

Well unfortunately what I really needed was a dynamic scripting data tool.  I understand that during complex procedures it is not always easy to explain what you want.  The scripting tool in SSMS is very limited in that it may generate errors when copying massive tables (more than 3 million rows).  I wanted to create a tool to generate scripts.  Eventually what I did was below.  This code is an example of code but it is basically three parts with a fourth left up to the user to generate a list of tables from their

1st part is to get your column values from the sys.columns data table so you can create a 'select *' into a single string.  I chose to make a view to give me something like 'select cast(column1 as varchar(max)) + ', ' + cast(column2 as varchar(max)) ... etc.

2nd part is to dynamically select this data from the view as a procedure.  I select the string to run as a long string then run it dynamically in SQL.

3rd Part 2 is just for a single script so to really get fast on hundreds of tables I need to loop this.  I set up a looping sproc to run this from.

4th Part is the table that part 3 is going to run from.  You need a master list of tables and criteria column to base what tables you want to keep or not on.

There are certain issues this of course will run into:

1. Varchar(max) does not deal well with varbinary.  You could go a level deeper than I did and add case logic to your column listings in the view.  That required too much I thought as my database although massive, did not have but 6 tables that stored in varbinary, I just did those manually.

2. Constraints may upset this process.  I would find someone else online that has a dynamic way to remove all constraints and print them out for reapplying them once data insertion is done.

3. Not all tables store an identity column so I could have set a bit somewhere if the table had an identity column or not.  Did not do that.  It's pretty simple thought to just remove the two lines that reference that though.

4. You should not run this script on tables over 5 million rows except one at a time and you may get an ASP.Net error if you are trying to copy and paste out of SSMS.  I would recommend on the extremely large tables running part 2 by itself on the table listed out explicitly and to file.

This tool actually helps a lot though as the SSMS wizard is slow, clunky, you need to set options.  There may be things I don't have as stated though like RedGate's tools. Or there may be SSIS things to automate the existing wizard faster.  I did not see much though except recreating the wheel each time for a dataset I wanted to copy.  I could have done multiple ones but you would have had to point and click one at a time and save it.  I chose my method because mainly I could run it FAST.  

Sorry I did not reply earlier and I know Redgate probably has a publishing tool that puts this to shame.

-- Part 1 is to create a view that can take the columns apart.  I --- need to create a string to create a script from.
create view {MySchema}.vIdentityInsert as

with obj as
	(
	select 
		s.name + '.' + o.name as TableName
	,	o.name as tbl
	,	o.object_id
	,	o.create_date
	,	o.modify_date
	from {MyDatabase}.sys.objects o
		join Creditnet.sys.schemas s on o.schema_id = s.schema_id
		and o.type_desc = 'USER_TABLE'
	) 
, cols as 
	(
	Select
		TableName
	,	obj.object_id
	,	obj.tbl
	,	create_date
	,	modify_date
	,	(
		select 
			name as n
		from {MyDatabase}.sys.columns c
		where c.object_id = obj.object_id
		for xml auto
		) as cols
	from obj
	where TableName in (select TableName from DataStaging.FTG.CreditnetTables)
	) 
, report as 
	(
	select
		TableName
	,	object_id
	,	tbl
	--,	create_date
	--,	modify_date
	,	replace 
		(
			left(right(cols,len(cols) -6),len(cols)-9)
					, '"/><c n="', '{|}'
		)
					 as ColumnsData
	from cols
	)
select 
	*
,	'Insert into ' + TableName + ' ( ' + replace(ColumnsData, '{|}',',') + ' ) Values' as insertstmt
,	'isnull(cast(' + replace(ColumnsData, '{|}', ' as varchar(max)),''NULL'') + ''{|}'' + isnull(cast(') + 
	' as varchar(max)), ''NULL'') as Data' as TableInsert
from report

-- part 2 is to create a procedure that can script this data but 
-- be dependent on the table you put in to get it's sys object ------ values from

if object_id('{MySchema}.spScripter') is not null
	drop proc {MySchema}.spScripter
GO

create proc Brett.spScripter
	(
		@Table	varchar(128)
	)

as

declare
	@Cols				varchar(max)
,	@Ins				varchar(256)
,	@SQL				varchar(max)


Select 
	@Cols = TableInsert
,	@Ins =	insertstmt
from {MySchema}.vIdentityInsert
where TableName = @Table

Set @SQL = 
'
declare @Temp table ( Data	varchar(max)  )

insert into @Temp values (''Use {NewDatabaseName}'')
;

insert into @Temp values (''set identity_insert ' + @Table + ' on'')
;

insert into @Temp
select ' + @Cols +

' 
from Creditnet.' + @Table + '
;

update @Temp
set Data = '''''''' + replace(Data,''{|}'','''''','''''') + ''''''''
where Data != ''set identity_insert ' + @Table + ' on'' 
and Data != ''Use {NewDatabaseName}''
;

update @Temp
set Data = replace(Data,''''''NULL'''''',''NULL'')
where Data like ''%NULL%''
;

update @Temp
set Data = ''' + @Ins + ' ('' + Data + '')''
where Data != ''set identity_insert ' + @Table + ' on'' 
and Data != ''Use Creditnet_Seed''
;

insert into @Temp values (''set identity_insert ' + @Table + ' off'')
;

insert into @Temp values (''GO'')
;

insert into @Temp values ('''')
;

Select * From @Temp
'

exec (@SQL)

-- Part 3 is to automate this on multiple tables with a query that -- loops the part 2 stored procedure

use DataStaging

if object_ID('{MySchema}.spTablesValidater') is not null
	drop proc {MySchema}.spTablesValidater
GO

create proc {MySchema}.spTablesValidater
	(
		@Type			varchar(16)
	,	@ResultType		varchar(16)
	)

as

declare @Temp table
	(
		TableID		int				identity(1,1)
	,	TableName	varchar(128)
	)

declare 
	@Name		varchar(128)
,	@Start		int

insert into @Temp
select TableName
from {MySchema}.{TablesToRunTable}
where Keep_Remove = @Type

Set @Start = 1

while @Start <= (select count(*) from {MySchema}.{TablesToRunTable} where Keep_Remove = @Type)
begin
	
	Select @Name =  TableName from @Temp where TableID = @Start
	
	
	if @ResultType = 'Script'
	begin
		exec {MySchema}.spScripter @Name
	End
		
	set @Start = @Start + 1
	
End

-- 4th part is a table that is a listing of tables to go off of.
-- I choose to use the sql built in function 'sp_tables' after my
-- 'use {Database to choose from}.  Then I added a column for 
-- identification of what to move or not (Keep_remove) and added
-- another sproc to populate it's criteria.

Select allOpen in new window

>>I know Redgate probably has a publishing tool that puts this to shame.<<
That is correct.

This was a difficult thing to ask for so I did not expect to get a complete answer but a list of possibilities which I did receive.