Re: CS question: do private function namechanges count as API changes?

On Nov 16, 2007 12:55 PM, Jorrit Schippers - nCode <jorrit@ncode.nl> wrote: > > This was my first idea, but constructs like those are only allowed in > method bodies, not when declaring the content of a variable directly inside > a class {} > Wow, I missed that completely... 600+ lines before the constructor... phew... Looking more at the context of code around that line, it further looks like it's not really feasible to move that initialization into the constructor. Given no options to solve it, I'd put either a @todo or @internal tag in that variable's docblock to at least acknowledge the PEAR CS violation and the realization that it cannot be solved in the current code. That won't clear the Codesniffer warning, but it will at least show anyone else looking at the code that the warning has been addressed... sort of. I am currently doing this in PhpDocumentor as I do CS cleanup in its code, where such CS cleanups will require true code refactoring... effectively I'm making notes for myself to tackle these CS issues when I later do some code refactoring for later releases. -- CRB Let me introduce you to my very own DMCA-protected encryption key: BC 1B 64 4A 8D DE 49 E8 C3 7D CC EE 1A AD EE F5 (compliments of Freedom-to-Tinker http://www.freedom-to-tinker.com/?p=1155)