Contents 

AdeptSQL Diff Online Help

Options / Comparison / Code Comparison

This page controls how strict the comparison of stored procedures, triggers, functions and views should be. Each such entity has a piece of SQL script as its body and this SQL may (or may not) be equivalent to another piece of SQL regardless of minor changes in spacing, indentation, character case or comments. If Lexical comparison is chosen, the SQL is converted into a stream of lexical tokens and those are compared. This is the preferable way.

In this case all SQL keywords will be case-insensitive and the case-sensitivity of identifiers will be determined by the "Name comparison" settings above. If you don't want to have SQL comments compared, they can be removed from the token stream before the comparison by setting the Ignore comments option. When you uncheck the Lexical comparison option, the two SQL definitions will be compared as monolithic strings, together with spaces, line breaks and all. Still, the case-sensitivity option would apply in this case, as well.

The Smart comparison option works only for expressions, such as those in DEFAULT or CHECK constraints or in computed columns. The server parses such expressions and tends to insert extra parenthesis around numeric literals and generally re-format them before storing in the database. This "smart comparison" option allows to normalize numeric literals before they are compared and removes parenthesis around them, so some of the minor changes introduced by server's parsing and reformatting can be eliminated.  It does not solve the problem completely, as other auto-formatting changes made by the server may involve adding/removing of optional parameters (e.g. in CONVERT(...)) or of parenthesis around sub-expressions.

The "Pair objects with identical code" option allows the Diff to match stored procedures (or other code objects) that have different names, but otherwise identical definitions. Such pairing would only happen when no procedure with matching name is found in the opposite database.  When scripting, the Diff would synchronize these procedures by renaming them ("exec sp_rename ...").

Ignore changes in expressions. This option allows to ignore changes in certain expressions.  You don't normally need this, except for situations when equivalent expressions differ in minor details (e.g. "(2+3)" vs. (3+2)") and the comparison algorithm is too stupid to recognize that they are in fact equivalent.  You should always try to use the "[x] Smart comparison" option before resorting to the "Ignore expressions" one. 

NOTE that although the comparison settings primarily apply to the schema tree comparison, some of them also affect the side-by-side textual comparison. In other words, first of all, they determine if a specific pair of procedures will be marked in the schema tree as identical or as changed. When you double-click on this schema item and the procedures are displayed in the side-by-side comparison window, the visual comparison will use these settings as follows:

It is worth to remember that it is always the schema comparison that matters, not the side-by-side view.  It is possible to encounter situations when two SQL entities (such as procedures) may appear identical in the schema view, but different in the side-by-side comparison, or vice versa.