When sp_depends fails

  • Comments posted to this topic are about the item When sp_depends fails

  • Thanks.

    You might want to do something to avoid also listing references to tables that include the target name with a prefix or suffix; e.g. 'MyTable' will also return references to 'MyTableOld', 'IndexedMyTable', etc.

  • Jim Russell (5/27/2008)


    Thanks.

    You might want to do something to avoid also listing references to tables that include the target name with a prefix or suffix; e.g. 'MyTable' will also return references to 'MyTableOld', 'IndexedMyTable', etc.

    Not completely to your point, but I found that adding a space to the end of the table name ('MyTable ') eliminated the suffixes. As to prefixes, if I always fully qualified my tables ([database].[schema].[tablename], I suppose putting a period in the front would work...but I don't.

    Obviously, naming convention comes to play in a hurry too.

    Nate

  • An alternative I recently discovered is to use sp_refreshsqlmodule, which updates the dependency and other metadata for a procedure, function, or view. The following generates EXEC statements:

    SELECT

    'EXEC sp_refreshsqlmodule N''' + s.[name] + '.' + o.name + ''';' AS [stmt]

    , s.[name] AS [schema_name]-- schema name

    , o.[name] AS [object_name]-- procedure, function, or view name

    , o.[type] AS [object_type]-- type (P, FN, IF, TF, or V)

    FROM sys.objects o

    INNER JOIN sys.schemas s ON o.[schema_id] = s.[schema_id]

    WHERE o.[type] IN ('FN','IF','TF','P','V')

    ORDER BY

    o.[type]

    , s.[name]

    , o.[name]

    Copy and paste the first column into the editor and execute them, or you could use a cursor instead of the copy/paste method. sp_refreshsqlmodule raises an error if the object is no longer valid (for example a view referencing a column that has been dropped from a table).

    After executing sp_refreshsqlmodule, you can then use sys.sql_dependencies and sp_depends.

  • Not trusting sp_depends or information_schema.routines when they get updated I took the idea of the first code and wrote a queries to scan syscomments Still need to clean it up but here is the first pass.

    --gets the list of the table.column that are used in a sp

    select o.name+'.'+sc.name

    from sys.objects o

    join sys.columns sc

    on sc.object_id = o.object_id

    join syscomments c

    on 1 = 1 --don't want to join

    where charindex (' '+o.name , c.text)> 0 --not sure if the space is needed.

    and charindex (sc.name , c.text)> 0

    and c.id = object_id('stored procedure name')

    and o.type = 'u' --might what this to be o.type in ('u','p')

    group by o.name, sc.name

    order by 1 ;

    FYI, sorry about not putting the [] on the object_id columns but it runs fine and who ever made the columns a reserved name should be

    Thanks for posting the original.

  • Does not work

    SELECT * FROM information_schema.routines r

    WHERE charindex('Dbo.People2', r.ROUTINE_DEFINITION)>0

    Does work

    SELECT * FROM information_schema.routines r

    WHERE charindex('People2', r.ROUTINE_DEFINITION)>0

    Nate Schmidt

    fully qualified my tables ([database].[schema].[tablename], I suppose putting a period in the front would work...but I don't

    Tried that and it does NOT work

    My conclusion is it will not work with a fully qualified name, and/or the inclusion of the schema name.

    Suggest that a note accompany your script to that effect.

    Now knowing that ,it is an EXCELLENT script and will prove to be very useful... THANKS for contributing.

    If everything seems to be going well, you have obviously overlooked something.

    Ron

    Please help us, help you -before posting a question please read[/url]
    Before posting a performance problem please read[/url]

  • One more thing to keep in mind, if the proc has more than 4000 characters and that the objects used is splitted between 2 rows in syscomments, that still won't work (quite rare, but it does happen).

    The solution is to left join on syscomments on c1.id = c2.id and c1.colid = c2.colid + 1

    then do the search on c1.text + isnull(c2.text,'') like '%Whatever%'

    That was true in 2000, maybe it's been fixed in 2005, but I'd be surprised.

Viewing 7 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic. Login to reply