I didn't decipher the subtlety you're talking about, but I just wanted to point out that we may be doing this manually moving forward, and it would be best if it wasn't complicated, confusing, or overly subtle.
Gabe nathan binkert wrote: >> Wow, I forgot I had written that... it was only seven years ago. >> > Yeah. Me too. I only noticed it when I wrote the new version and > noticed that tab completion didn't work trivially. > > >> I guess it does make some sense to have the includes with paths before the >> ones that are in the local directory, though applying that recursively may >> be overkill. (Or maybe it's inconsistent not to apply that recursively... I >> don't know.) >> > Consistency is why I applied it recursively, but any of these things > are trivial. > > >> Of course, if your new script is sufficiently robust, maybe we don't worry >> too much about the mental effort of implementing the right kind of sort >> manually and instead just run your script occasionally to get things right. >> > I think it is pretty robust. I've run it on the tree many times. (I > did after each one of the patches in my queue to make sure I didn't > screw anything up and the patch in the queue was completely done by > the script. I could make it part of the style hook so that it detects > the error and offers to fix it up. > > By the way, I've got a patch that allows the style hook to be a > pre-qrefresh hook so that it will be triggered on a "hg qref" (it > always annoyed me that it only worked on push/commit). > > Nate > _______________________________________________ > m5-dev mailing list > [email protected] > http://m5sim.org/mailman/listinfo/m5-dev > _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
