On Sun, 15 Apr 2012 01:19:12 -0400 Michael Blumenkrantz <michael.blumenkra...@gmail.com> wrote:
> On Sun, 15 Apr 2012 14:15:15 +0900 > Carsten Haitzler (The Rasterman) <ras...@rasterman.com> wrote: > > > On Mon, 2 Apr 2012 19:53:28 -0300 Gustavo Sverzut Barbieri > > <barbi...@profusion.mobi> said: > > > > here's my take - and i'm fast-forwarding including more comments > > from tom later. > > > > FIRST: > > > > 1. just because it isn't used elsewhere in efl or e atm should NOT > > be a reason for exclusion. if it has a good use case somewhere > > that's more than a one-off, it is a candidate to go in a release. > > 2. we have lots of stuff that is barely tested, so testing is very > > good and needed, but it has never been a necessity. we get to test > > and fix eventually. 3. what matters is the IDEA is sane and the API > > is sane. it won't conflict with other parts of efl (nor or in the > > forseeable future). it can be expanded in future. if we have > > missing things or bugs they can be fixed without creating havoc > > with existing apps using the api. the usual thing here. > > > > NOW: > > > > eina model/value. the reason really tom has is that here are 2 new > > fairly hefty api's with really no "use cases" floating about to see > > other than example files or tests. the rest of the reasoning is > > really just justifying this base. so lets focus on this. > > > > i haven't looked beyond a cursory glance thus i have no deep > > meaningful insights, but let me make an attempt. :) > > > > 1. eina value > > > > * it is heftily documented, so i can't complain. > > * it's pretty darned big. > > * it seems very heavyweight in a c world for holding and dealing > > with "values". it smells like the kind of logic u find in bash or > > perl or other scripting langs where variables are dynamically typed > > - they may be a string, a number, a float etc. and they keep > > shifting type based on how you use them = eg a = b + c ... will > > store the numeric result of b + c... as long as both b and c are > > sanely interpretable numbers (even if both strings- of both are > > strings of numbers etc.). so that's what this looks like to me. > > find - if u are writing a scripting engine this can be useful. > > * could you please go into what eina_value is useful for? what did > > you develop it for? what was the driving reason? > > * other than the above questions it seems pretty sane and > > reasonable. i don't see any problem shipping with it as-is > > (API-wise). i will trust that it has been actually used and tested > > and deployed - we just can't/don't see it (thus why i'm asking you > > to tell us about it :)) > > not to weigh in on either side because I don't really care, but > eina_value has been deployed in esql since january. nobody that I > know of uses esql, but it does work. sometimes. Which might also be used in part of the project I'm writing my scripting engine for. So sometime later this year, I expect there will be two users of esql. That changes my "is it too big" decision though, if it's already being used elsewhere in the project via esql. We shall see when I get around to that part of my project. -- A big old stinking pile of genius that no one wants coz there are too many silver coated monkeys in the world.
signature.asc
Description: PGP signature
------------------------------------------------------------------------------ For Developers, A Lot Can Happen In A Second. Boundary is the first to Know...and Tell You. Monitor Your Applications in Ultra-Fine Resolution. Try it FREE! http://p.sf.net/sfu/Boundary-d2dvs2
_______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
