RE: CVS update: subversion/subversion/libsvn_fs txn-table.c

Yeah, I know, but the discussion seemed to be related to noticing from
the function prototype section of the code what the in/out semantics of
each of the parameters was. As opposed to having a very detailed
exposition in a large comment preceding a large chunk of code.

Having the distinction listed in the prototype area is very useful,
since const isn't useful enough at first glance to give you all of the
details.

That's more of what I was trying to get at. It's been one of those days
at work, you'll have to forgive me for seeming a little rough today.

Bill

 -----Original Message-----
From: Karl Fogel [mailto:kfogel@galois.collab.net]
Sent: Thursday, December 14, 2000 1:17 PM
To: Bill Tutt
Cc: Greg Stein; Jim Blandy; dev@subversion.tigris.org
Subject: Re: CVS update: subversion/subversion/libsvn_fs
txn-table.c

Bill, I don't think Jim was saying there's no way available to state
his intentions regarding `skel_t *' arguments. He's saying that
`const' isn't it, because const doesn't technically mean what he needs
to assert about the parameter.

If you look at his code, he uses comments all the time to say what
needs to be said.

-Karl