My objection to this change is only that there is little general agreement on this sort of thing, so there is no hope of maintaining this style throughout the code base. As such, this type of patch just churns the code base.
Before doing this level of patch, I would suggest writing a coging standards document for the code base. Once you get agreement on the coding standards, then you have a basis for making these cosmetic cleanups.
The coding standards issues has been raised before[1], but I don't think there was any resolution or output document.
Ted anderson
[1] https://lists.openafs.org/pipermail/openafs-devel/2003-January/003812.html
The benefit to placing the semi-colon on a separate line in my opinion is for debuggers. If you want to be able to place a break point on the execution of each step of the loop there must be a unique line number for the statement.
However, I completely agree that most of the flexelint patches are superficial in nature. They are useful to the extent that they point out locations where there are recognizable problems. On the other hand, there number of changes are so significant that the process of applying them can easily mask new errors.
I have applied the patches for aggregate initialization; printfs using %p in place of %x in the Windows only code; and bugs found.
A couple of comments about the patches themselves. It is exceedingly difficult to apply the supplied patches in any automated way. The patches are not between an original tree and a modified tree but between new and old versions of specific files in a common directory. In addition, due to the desire to break the patches up based upon their category of issues, multiple patches apply to the same source files producing conflicts.
On the other hand, these problems are somewhat beneficial in that the vast number of changes and the subtle questions many of them raise really do require significant attention to the purpose of the code as it was originally written. There is more than 400K of additional patches which will take significant effort to review and apply (assuming there is agreement that they should be applied.)
Jeffrey Altman
smime.p7s
Description: S/MIME Cryptographic Signature
