On Tue, Jan 16, 2018 at 3:54 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> FWIW, I think that that represents bad practice in those changes,
> precisely because of the hazard it poses for uncommitted patches.
> If you're changing a function signature, it's usually not that hard
> to make sure that un-updated code will produce a failure or warning,
> and you should generally do so IMO.
I strongly agree. That's an example of the programmer exploiting
mechanical detection of conflicts deliberately, which is great. All of
these things are tools, and like all tools they are generally not
helpful unless used thoughtfully.
--
Peter Geoghegan
