XMILE test of `saveper`

[JamesPHoughton]

currently, the xmile version of the test for SAVEPER vs TIME STEP is the version autogenerated by @wasbridge . But, it doesn't actually test the concept that the save time could be different from the integration time.

Is there an XMILE equivalent of saveper? If not, I'll just delete that version of the test.

@alexprey, do you know about this in the XMILE standard?

[wasbridge]

Stella has a saveper I don't remember what actually made it into the spec. I can read that Monday unless you guys are in a rush.

…

On Fri, Jul 20, 2018, 5:37 PM James Houghton ***@***.***> wrote: currently, the xmile version of the test for SAVEPER vs TIME STEP is the version autogenerated by @wasbridge <https://github.com/wasbridge> . But, it doesn't actually test the concept that the save time could be different from the integration time. Is there an XMILE equivalent of saveper? If not, I'll just delete that version of the test. @alexprey <https://github.com/alexprey>, do you know about this in the XMILE standard? — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#48>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAzEakB8aj-8XvBIvZgoqRvBptNbGN2Sks5uIk2hgaJpZM4VZTYB> .

[alexprey]

For now I just copy value from dt, but I look again at specification and found that:
http://docs.oasis-open.org/xmile/xmile/v1.0/os/xmile-v1.0-os.html#_Toc439926042
So, I see that by xmile we can define not only period for export, but also select required variables that should be exported by simulation.
So I think that it should looks like this:

<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<xmile version="1.0" xmlns="http://docs.oasis-open.org/xmile/ns/XMILE/v1.0">
    <header>
        <vendor>James Houghton</vendor>
        <name>Teacup</name>
        <options>
            <uses_outputs/>
        </options>
        <product version="1.0">Hand Coded XMILE</product>
    </header>
    <sim_specs>
        <stop>30.0</stop>
        <start>0.0</start>
        <dt>0.125</dt>
    </sim_specs>
    <data>
       <export interval="1">
           <all/>
       </export>
    </data>
    <model>
        <variables>
            <flow name="Heat Loss to Room">
                <doc>Heat Loss to Room</doc>
                <eqn>("Teacup Temperature"-"Room Temperature")/"Characteristic Time"</eqn>
            </flow>
            <aux name="Room Temperature">
                <doc>Ambient Room Temperature</doc>
                <eqn>70</eqn>
            </aux>
            <stock name="Teacup Temperature">
                <doc>The average temperature of the tea and the cup</doc>
                <outflow>"Heat Loss to Room"</outflow>
                <eqn>180</eqn>
            </stock>
            <aux name="Characteristic Time">
                <eqn>10</eqn>
            </aux>
        </variables>
    </model>
</xmile>

See at data section, if interval is 0, that means we should export only final result.

[JamesPHoughton]

Thanks @alexprey - I made you a contributor on this repo - I'll let you commit the fix if you're happy to do so?

[alexprey]

Thank you @JamesPHoughton, I'm look at this later and try to write test for this case

[wasbridge]

Okay there is no saveper in the XMILE spec. Stella does it as isee:save_interval="1" As an attribute on <sim_specs/> On Fri, Jul 20, 2018 at 5:51 PM, Billy Schoenberg < billy.schoenberg@gmail.com> wrote:

…

Stella has a saveper I don't remember what actually made it into the spec. I can read that Monday unless you guys are in a rush. On Fri, Jul 20, 2018, 5:37 PM James Houghton ***@***.***> wrote: > currently, the xmile version of the test for SAVEPER vs TIME STEP is the > version autogenerated by @wasbridge <https://github.com/wasbridge> . > But, it doesn't actually test the concept that the save time could be > different from the integration time. > > Is there an XMILE equivalent of saveper? If not, I'll just delete that > version of the test. > > @alexprey <https://github.com/alexprey>, do you know about this in the > XMILE standard? > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub > <#48>, or mute the thread > <https://github.com/notifications/unsubscribe-auth/AAzEakB8aj-8XvBIvZgoqRvBptNbGN2Sks5uIk2hgaJpZM4VZTYB> > . >