On 01/06/15 12:49, Simon Lees wrote: > > > On 06/01/2015 06:42 PM, Tom Hacohen wrote: >> On 20/05/15 17:55, Tom Hacohen wrote: >>> Hey, >>> >>> It is finally time to release Eo. >>> >>> Since the next EFL release is in August, this will only become official >>> EFL API starting from then, however, I'm announcing this now, so >>> developers who track EFL git can already start using Eo and be assured >>> the API will remain the same. >>> >>> This includes all of the APIs of Eo and Eo.Base, essentially everything >>> in Eo.h (unless marked otherwise). Please note, this means that all the >>> other EFL/Elementary Eo functions, for example Evas.Object, are still >>> *not* stable. >>> >>> >>> I plan on doing it in three stages: >>> 1. "Everything goes": nothing is going to change until the 27th. Until >>> then Eo is still considered unstable and things may change. This should >>> give people enough time for a last review before we decide to go forward >>> with this. >>> 2. "Very stable": after the 27th we are not going to break Eo API unless >>> there's a VERY good reason to. This means that starting from the 27th, >>> Eo is essentially "released". >>> 3. "Released": Starting from 1.15, Eo is stable API. Same as all the >>> other released EFL APIs. >>> >>> Again, this does *not* apply to APIs outside of Eo.h like Evas.Object. >>> Essentially, this means you can create your own classes (as long as they >>> inherit from Eo.Base), but not the EFL/Elementary classes. >>> This is still very useful, and we already have users using Eo this way. >>> This also means people writing new libraries, can create them using >>> Eolian, and create Eo based API, which users of that API can safely use >>> (we also have users using Eo this way). >>> >>> >>> Eolian: while we are still making no promises about Eolian, I believe >>> the Eolian syntax won't break (too much) until the 1.15 release. We also >>> hope to declare the Eolian syntax as stable at the 1.15 release, so we >>> can only add to it, but not break existing syntax. This however depends >>> on more testing within the EFL and generated bindings. >>> You can however already start using Eolian, just make sure to ship the >>> generated .eo.[ch] in your tarball. This will just work regardless of >>> any Eolian syntax breaks. >>> As always, all the info about Eolian, including reference .eo file, can >>> be found on the wiki: https://phab.enlightenment.org/w/eolian/ >>> >>> >>> As for when you'll be able to use the EFL/Elementary API with Eo? No >>> plans at the moment. I plan on making it mostly good for 1.15, however >>> there'll still be a lot of reviewing and changes that need to be done, >>> and that will take time. >>> >>> I'd also like to take this opportunity to thank the 28 different >>> contributors who contributed to Eo since its inception, everyone who >>> reported bugs, and everyone else who helped in getting it to this state. >>> >>> If you have any questions, thoughts or suggestions, please let me know. >> Btw, Eo should now be considered "very stable". >> >> -- >> Tom. >> >> > Cool, now someone just needs to fix the python bindings for this new > "very stable" eo > > [ 78s] efl/eo/efl.eo.c: In function '__pyx_f_3efl_2eo_2Eo__set_obj': > [ 78s] efl/eo/efl.eo.c:3023:3: error: too many arguments to function > 'eo_key_data_set' > [ 78s] eo_do(__pyx_v_self->obj, eo_key_data_set(__pyx_k_python_eo, > ((void *)__pyx_v_self), NULL)); > [ 78s] >
Hey hey, Please remove the quotes. :) As I promised, this is not going to change anymore, that change was within the mercy week. :P Not going to happen again (unless we find something very wrong before the 1.15 release). -- Tom. ------------------------------------------------------------------------------ _______________________________________________ enlightenment-devel mailing list enlightenment-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
