Do most of you guys register via the Class and Typelib tables where possible?

Can anyone shed some light on the advantages and disadvantages of this vs writing the information via the registry table?

I understand that adverstisement/install on demand is one reason to use the tables, but are there any others?
0 Comments   [ + ] Show Comments

Comments

Please log in to comment

Rating comments in this legacy AppDeploy message board thread won't reorder them,
so that the conversation will remain readable.

Answers

0
The only other reason I've heard of is purely for 'neatness' i.e. it can make it easier to manage the data if it is split up into categories instead of all being lumped together in the one table. I think it's a lot of work to shift information from the registry table to the class tables etc manually so I just rely on my repackager tool to do the work, which it does well for the most part. Occasionally it may omit to delete duplicate information from the registry table during conversion, which can lead to the classic ICE33 warnings.

Alan
Answered 03/27/2006 by: ab2cv
Orange Senior Belt

Please log in to comment
0
Cheers mate.

Further to my question, can anyone explain exactly how self healing could be triggered via COM registry entries generated by the Class table?

Is it as simple as populating the table, and if an application calls the DLL etc it acts as an entry point?
Answered 03/28/2006 by: OldWiseMan
Senior Yellow Belt

Please log in to comment
0
ORIGINAL: OldWiseMan
Is it as simple as populating the table, and if an application calls the DLL etc it acts as an entry point?

That's how I understand it. As the class table references the component that installs the file instead of the actual file itself, it acts as an entry point for the advertising.

Alan
Answered 03/28/2006 by: ab2cv
Orange Senior Belt

Please log in to comment
Answer this question or Comment on this question for clarity