incubator-clerezza-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Reto Bachmann-Gmuer <r...@apache.org>
Subject Re: EasyGraph: suggestion on what to include in the release and how
Date Tue, 28 Jun 2011 13:33:58 GMT
On Mon, Jun 27, 2011 at 2:57 PM, Henry Story <henry.story@bblfish.net>wrote:

>
> On 27 Jun 2011, at 11:08, Reto Bachmann-Gmuer wrote:
>
> Hi Henry and all,
>
> I suggest to do the following for the release:
>
> - We add the construct allowing to add properties to a node with node --
> property --> object and its inverse -<- property -- subject to RichGraphNode
> - We add a trait EzGraph (or EasyGraph) that provides by default the
> implicit conversions as when importing a Preamble(new SimpleMGraph) object,
> a graph other than a new SimpleMGraph can be specified by overriding a field
> - We add the ability for creating string-referenceable bnodes to the
> EzGraph Trait
>
>
> I can look into it.
>
> But don't you think this would be better left for after your release?
>
Okay, a  bit a pity we couldn't get more of the good stuff in but its good
to have a release.

I removed the dependency in the platform shellcustomizer to EasyGraph. Could
you remove the other dependencies (foaffssl.test and accountcontrolpanel) as
well as other code you added an that's not yet ready for release? (hope the
add-contact feature can stay at leasts in its bare-bone version of adding
contacts by their WebIds).

Cheers,
Reto


>  No code currently uses the projected RichGraphNode, so we would not know
> if it is a good thing to enhance it. RichGraphNode is currenlty used mostly
> in SSP pages where read only behavior is required. So I am not sure if
> adding write behavior to a read only framework is such a good idea.
>
> Making the seperation clear, and making it possible to combine them would
> be good, but that requires working on traits, and doing more work there.
>
> And postpone the other suggested features like support for these unicode
> characters avaliable on mac
>
>
> Are they not available on other OSes?
>
> If you look at the scalaz code you will see they are full of unicode
> characters.
>
> and the new language features to after the release where they should be
> tackled in issue-branches. I think tgether with what we already have with
> ZZ-577 this would allow us to have the most important improvements
> introduced by the EasyGraph issue.
>
>
> I suppose the best thing would be to create a release branch, and start
> tightening things up there. As mentioned previously
> you could remove all my code from it and get your release now.
>
> Henry
>
>
>
>
> What do you think?
>
> Cheers,
> Reto
>
> On Wed, Jun 22, 2011 at 5:46 PM, Reto Bachmann-Gmuer <reto@apache.org>wrote:
>
>>
>>
>> On Wed, Jun 22, 2011 at 5:38 PM, Henry Story <henry.story@bblfish.net>wrote:
>>
>>>
>>> On 22 Jun 2011, at 17:23, Reto Bachmann-Gmuer wrote:
>>>
>>> To make debugging easier it woul be good to have tests of smaller sizes
>>> pass. So I recommend that before writing a test for an 18 triple-graph write
>>> a couple of test with only between  1 and 3 triples.
>>>
>>>
>>> yes, that's just what I was doing now, and am having trouble there.
>>>
>>> But at least we can see that the notation looks good.
>>>
>>
>> I though I try with what I thought to be the easiest possible case:
>>
>>     private val miniGraph: Graph = {
>>         val henry = new UriRef("http://bblfish.net/#hjs")
>>         val reto = new UriRef("https://farewellutopia.com/user/me#me")
>>         val gr = new SimpleMGraph
>>         gr.add(new TripleImpl(reto,FOAF.knows, henry))
>>         gr.getGraph
>>     }
>>
>>     @Test
>>     def miniUsingAsciiArrows {
>>         import org.apache.clerezza.rdf.scala.utils.EasyGraph._
>>         val ez = new EasyGraph()
>>         // example using arrows
>>         (
>>             "https://farewellutopia.com/user/me#me".uri -- FOAF.knows -->
>> "http://bblfish.net/#hjs".uri
>>         )
>>         Assert.assertEquals("Both graphs should have the same
>> size",largerGraph.size, ez.size)
>>         Assert.assertEquals("Both graphs should contain exactly the same
>> triples",miniGraph, ez.getGraph)
>>     }
>>
>> But this doesn't even compile, how is one supposed to write this
>> single-triple graph with ez-graph?
>>
>> Cheers,
>> Reto
>>
>>
>>
>>>
>>> Henry
>>>
>>>
>>>    Social Web Architect
>>> http://bblfish.net/
>>>
>>>
>>
>
> Social Web Architect
> http://bblfish.net/
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message