Sorry, let me clarify: I know what DeRPC is and why we'd want to get rid of it. I'm asking why we're discussing getting rid of it *now*.
Like I said, Google has an internal user for it still, so someone (most likely me) would need to invest time in helping them transition off before we can delete it. I want to make sure that effort is well spent. On Tue, Nov 5, 2013 at 5:00 PM, Ray Cromwell <[email protected]> wrote: > DeRPC was essentially an experiment that didn't work out, it was supposed > to improve the speed of IE6. It added a lot of hacks to the compiler to > make it work, @ArtificialRescue, ability to invoke client-side methods from > server emitted functions, which actually requires clever hacks in some > classes to prevent inlining (see java/lang/Class emulation code) > > It would be good to remove it. If you want really fast RPC, go with > something like JAXRS + JSOs. Parsing is blistering fast with JSON.parse() > on today's browsers, and there's no need to deserialize/copy. RestyGWT is > one possibility. > > > > On Tue, Nov 5, 2013 at 10:07 AM, Matthew Dempsky <[email protected]>wrote: > >> Just curious, is there a particular impetus to removing DeRPC at this >> time? E.g., is it causing code bloat or slowness, or making new >> development more difficult? >> >> I'm fine with marking the class as @Deprecated, but unfortunately we >> [Google] still have an internal customer using DeRPC and so far I haven't >> been successful in convincing them to prioritize moving off of it. (I did >> manage to remove our other internal uses of DeRPC.) >> >> Alternatively, if there's an easy way for us to split DeRPC out of GWT >> core and maintain it separately, I'm happy to do that (e.g., we're already >> doing that for GWTShell). But it sounds like DeRPC is heavily entangled >> into GWT-RPC? >> >> >> On Tue, Nov 5, 2013 at 9:45 AM, Thomas Broyer <[email protected]> wrote: >> >>> The first step would be to delete com.google.gwt.rpc and the >>> HybridServiceServlet, and then remove all the hooks in RPC that were added >>> to make deRPC possible (there's at least one in RemoteServiceServlet), and >>> probably others in the RPC generator. >>> >>> >>> On Tuesday, November 5, 2013 5:26:21 PM UTC+1, Stephen Haberman wrote: >>>> >>>> Hi Naresh, >>>> >>>> > Could you please let me know is there any patch available to check >>>> > the impact of removing de rpc calls. >>>> >>>> I'm glad you're interested in looking in to this. >>>> >>>> Unfortunately, I don't really know much about DeRPC. There are probably >>>> others on the gwt-contrib list who do though...we'll see if anyone can >>>> point you in the right direction. >>>> >>>> - Stephen >>>> >>>> -- >>> http://groups.google.com/group/Google-Web-Toolkit-Contributors >>> --- >>> You received this message because you are subscribed to the Google >>> Groups "GWT Contributors" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to [email protected] >>> . >>> For more options, visit https://groups.google.com/groups/opt_out. >>> >> >> -- >> http://groups.google.com/group/Google-Web-Toolkit-Contributors >> --- >> You received this message because you are subscribed to the Google Groups >> "GWT Contributors" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected]. >> For more options, visit https://groups.google.com/groups/opt_out. >> > > -- > http://groups.google.com/group/Google-Web-Toolkit-Contributors > --- > You received this message because you are subscribed to the Google Groups > "GWT Contributors" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > For more options, visit https://groups.google.com/groups/opt_out. > -- http://groups.google.com/group/Google-Web-Toolkit-Contributors --- You received this message because you are subscribed to the Google Groups "GWT Contributors" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
