From: René Scharfe Date: Sat, 08 Jul 2017 15:15:30 GMT Subject: Re: [PATCH] urlmatch: use hex2chr() in append_normalized_escapes() Message-ID: <46294b4f-5e6d-bbf3-8b93-9d23df5ce07b@web.de> In-Reply-To: Am 08.07.2017 um 16:28 schrieb Kyle J. McKay: > On Jul 8, 2017, at 01:59, René Scharfe wrote: > >> Simplify the code by using hex2chr() to convert and check for invalid >> characters at the same time instead of doing that sequentially with >> one table lookup for each. > > I think that comment may be a bit misleading as the changes are just > switching from one set of inlines to another. Essentially the same > sequential check takes place in the hex2chr inlined function which is > being used to replace the "one table lookup for each". An optimizing > compiler will likely eliminate any difference between the before and > after patch versions. Nothing immediately comes to mind as an alternate > comment though, so I'm not proposing any changes to the comment. Right, the table lookups for isxdigit and hexval are not duplicated when compiling with -O2. René