The more bits you use, the more likely that rounding errors will adversely affect you, such as two reals that you consider equal not providing the same hash code (caused by rounding errors). I think that you would be better served by a different structures, such as a RB tree, where a node is

Dear Alexander, I hadn't thought of the storable aspect, but there must be some way to get both the characteristic and the mantissa of the floating values as integers, and then add them together and take the absolute value as the hash. Probably it could be an internal routine. regards Finnian

Is there a way to have a separate object to be checked as ensure with the separate obj as l_obj statement?

I guess having a boolean function should work.

Any reason for that I don't understand?

set_position (a_pos: like position)
    do
        position := a_pos
        separate status_keeper as l_status_keeper_sep do
            l_status_keeper_sep.set_position (position)
        end
    ensure
        position = a_pos
        separate status_keeper as l_status_keeper_sep do -- the compiler doesn't agree with separate keyword here
            l_status_keeper_sep.position = a_pos
        end
    end

Hi Luca, The confirmation email should come from the Eiffel.com support web site. The subject: "Eiffel Support Site: Account Activation" Maybe check your spam folder. You can also try to register directly via https://cloud.eiffel.com/esa/register , or https://support.eiffel.com/ Regards, --

Actually as far as I understood SCOOP is implemented with THREADS so PROCESS is the keyword for "SYSTEM" which is the executable instance of your app, I also have a ONCE in my example within STATUS_KEEPER class On Tue, Feb 25, 2020 at 1:39 PM Anders Persson wrote: > Hi > > Try

Hi all, going through my SCOOP basics I'm actually trying to SCOOP my application which seems to be a big work. Into the steps of SCOOPING my app, I have `file_processors` which are going to be `separate file_processor` 1) creating them with `create l_file_processor.make (a_sep_arg)` as far as

Thank you guys! I did not expect PROCESS=SYSTEM in SCOOP applications, as the first what SCOOP noobs have to learn is that regions have their own processors :-) Best, Karsten

Hi Larry Seems reasonable, except that 'twin' can be superficial and I suspect that separate objects are "deeply separate", so at least a deep_twin should be used. Following your example, and dusting off some old memories, there used to be a kind of global once (vs per process) that, with a lot

Yes! A wholesale transfer of the object from one Region-Processor to another is in this case what I am after. Which I think means the compiler has to statically prove that all references to the object in the sender or source Region-Processor are nonexistent or safely disconnected, yes? On Tue,

Hi Try with once ("PROCESS") Vänligen Anders Persson BSharp AB Linked in profile +46 763 17 23 25 On Tue, 25 Feb 2020 at 15:54, wrote: > Dear Alexander, > > I successfully implemented your suggestion #3: > horse1, horse2, … horseN ===> state of all horses <=== printer >

Hi Or maybe the possibility to transfer an object from one region to another. Vänligen Anders Persson BSharp AB Linked in profile +46 763 17 23 25 On Tue, 25 Feb 2020 at 16:57, Larry Rix wrote: > This also makes me wonder why one could not just > > 1. Receive

Another observation I can do is that your comment could be welcome into the sed_container class to understand what it does and whats its purpose. Thx anyway for the feeds! On Tue, Feb 25, 2020 at 12:48 PM Larry Rix wrote: > This spurs an observation. > > The

Actually—I did this notation wrong. 'l_local_thing := a_arg.twin_from_separate' where my_feature (a_arg: separate ANY) -- blah-comment ... local l_local_thing: like a_arg do l_local_thing := a_arg.twin_from_separate ... end

This also makes me wonder why one could not just 1. Receive a reference from another SCOOP Region 2. Create a twin of it (as long as it can be twinned) 3. Give the new Recipient SCOOP Region-Process control over the newly minted (twinned) reference. To put it more succinctly—within a call of

Hi all, installed ES 19.12, registered myself as a new user, never received a confirmation email about my registration and unable to login in any way, is this right? Thank you. ---------------------------------------------------------------- -- Dott. Ing. Luca Paganotti -- Via dei Giardini 9 --

This spurs an observation. The mechanism seems simple enough: 1. Serialize an object to a string representation. 2. Use the string representation to create a new string object in the Target Region. 3. Constitute (deserialize) a new object in the target Region from the newly created string

THERE IT IS!

[image: image.png]

Good morning! The PE_IMAGE_FILE is of no real consequence. It is just an Object (Class) and could be ANY-thing at all.

To add more information to my previous answer From the library $ISE_EIFFEL/unstable/library/scoop_patterns you can use CP_SED_CONTAINER [G] The principle is to "import" from another scoop processor a serialized representation of the object. Quite similar to the `make_from_separate` from STRING.

Dear Alexander, I successfully implemented your suggestion #3: horse1, horse2, … horseN ===> state of all horses <=== printer (see http://t1p.de/b64o folder scoop_horses.zip\2020-02-25). Storage for exchange of the 'state of all horses' between all horses and the reporter is currently

Hi, You may also look at the library $ISE_EIFFEL/unstable/library/scoop_patterns especially the examples/sed_container that demonstrates the use of CP_SED_CONTAINER [G] On Tue, Feb 25, 2020 at 3:04 PM Gachoud Philippe wrote: > As a complement, the READABLE_STRING_32.make_

Does my answer not help you? do Current.url := a_image.url -- Current can be removed but for the example Current.extension := a_image.extension ...... end as I don't know your PE_IMAGE_FILE class On Tuesday, February 25, 2020 at 11:12:52 AM UTC-3, Larry Rix wrote: > > [image:

Hi Larry, probably some more capable of SCOOP could understand and answer you better than me, but I'm trying anyway. As far as I understood your question, your question is to create a twin from a separate object to a non separate one, am I correct? Maybe it helps l_horse: HORSE l_horse_sep:

Is there a specific routine on a specific class I need to look at? I have both SED_CONTAINER_EXAMPLE and CP_SED_CONTAINER open and looking at them. That stated—I do see the require 'execution_completed: a_server.completed' on 'wait_for_server_completion', which appears to be a "wait condition"