[Auscope-geosciml] GeoSciML Version 2.x problems and subversion cleanup [SEC=UNCLASSIFIED]

Oliver.Raymond at ga.gov.au Oliver.Raymond at ga.gov.au
Wed Oct 21 02:39:50 EDT 2009


Hi Steve,

I think we are on the same page.

“The GeoSciML/trunk/model directory and GeoScimL/trunk/schema directories are both deprecated at this point, an as I understand things, could be removed.”  -  I agree

------------------------------------------------------------------

I have figured out how to get the GSML v2 uml (XMI) files to work properly in EA.  They seem to only work properly (ie; all the links to GML3.1, ISO and OGC) when coupled with the old version of HollowWorld.  I got a copy of the old HollowWorld package from the unversioned EAP at www.geosciml.org<http://www.geosciml.org/> and I am using it for making the new GeoSciML v2.1 model.

A question about classMaps (does FullMoon create these?).  The v2.0 UML model currently has these classMap tags:
          - cgiUtilities package = https://www.seegrid.csiro.au/subversion/GeoSciML/trunk/model/ClassMap_CGI_Utilities.xml
          - geologicTime package = https://www.seegrid.csiro.au/subversion/FullMoon/trunk/resources/type-mapping-util/ClassMap_GeoTime_V2.xml
          - GeoSciML package = empty

Should the v2.1 tags point to a more stable location, like:
          - cgiUtilities: http://www.geosciml.org/cgiutilities/1.0/ClassMap_CGI_Utilities.xml  or  https://www.seegrid.csiro.au/subversion/GeoSciML/tags/2.1/model/ClassMap_CGI_Utilities.xml
          - geologicTime: http://www.geosciml.org/geotime/2.1/ClassMap_GeoTime_V2.xml  or  https://www.seegrid.csiro.au/subversion/GeoSciML/tags/2.1/model/ClassMap_GeoTime_V2.xml
          - GeoSciML: http://www.geosciml.org/geosciml/2.1/ClassMap_GeoSciML_V2.xml  or  https://www.seegrid.csiro.au/subversion/GeoSciML/tags/2.1/model/ClassMap_GeoSciML_V2.xml

Ollie

------------------------------------------------------------------------------------------------
Ollie Raymond
National Advice,  Maps and Standards Project
Geoscience Australia

Address: GPO Box 378, Canberra, ACT, 2601, Australia | ABN: 80 091 799 039
Ph: (02) 62499575 | Fax: (02) 62499992 | Email: Oliver.Raymond at ga.gov.au
Web:  http://www.ga.gov.au/minerals/research/national/nat_maps/nat_geol_maps.jsp<http://www.ga.gov.au/geoscience/national>

Google Map<http://maps.google.com/maps?f=q&hl=en&geocode=&q=canberra+australia&ie=UTF8&ll=-35.344028,149.158362&spn=0.007684,0.016404&t=h&z=17&iwloc=addr&om=1>

-- This message was created with 100% recycled electrons --


-----Original Message-----
From: auscope-geosciml-bounces at lists.arcs.org.au [mailto:auscope-geosciml-bounces at lists.arcs.org.au] On Behalf Of stephen richard
Sent: Wednesday, 21 October 2009 4:34 PM
Cc: auscope-geosciml at lists.arcs.org.au
Subject: Re: [Auscope-geosciml] GeoSciML Version 2.x problems and subversion cleanup [SEC=UNCLASSIFIED]

My understanding is that the trunk is the current, cutting edge development directory. Branches are forks from that main development for experimentation or update/maintenance of previous development states. Thus any work we do on fixing v2 is now a branch activity. The v3 work is trunk. The GeoSciML/trunk/model directory and GeoScimL/trunk/schema directories are both deprecated at this point, an as I understand thinks, could be removed. The model directory contents should be the xmi model files for EA that predate the August/Sept 2009 refactoring of the model. The schema directory contains xsd's generated from the UML in Nov 2008 (using the dates on the files), which appear to be post rc3 tag, and pre _final tag (4/15/09). It looks like the current tag is in the tags/2.0.2/2.0.x directory dated 5/19/2009. My guess is the GeoSciML/trunk/schema schemas were generated at some in between stage, never put in a tag, and then superseded but not removed or updated.

The actual models that we're working on are in the borehole, core, fossil, geochemistry, specimen, timescale, and utilities directories, with the working v3 xmi in the trunk/uml subdirectory of each of those. The GeoSciML/model directory is a subversion-managed directory that mirrors all of these xmi files in one place ('external') to make EA version configuration easier.

the other directories in Geosciml/Trunk -- Documents, instances, ModelTemplate, TEstbed3, tools all may contain useful files, but should probably be sorted through and either moved to other places or removed? Not sure what best practice here would be, except to cul stuff that's no longer any use. Most of the stuff in documents I put there, so I'll clean that up. The ModelTemplates are probably still useful for EA users. Tools seems to be getting updated for Geoserver work? Testbed3, I'm not sure--any one claim this stuff?

steve

Bruce.Simons at dpi.vic.gov.au<mailto:Bruce.Simons at dpi.vic.gov.au> wrote:

> “Shouldn't we delete GeoSciML/model?” - this directory contains the externalled copies of the refactored v3 xmi’s.  Not absolutely required, but it makes using Enterprise Architect a lot easier with the new multiple packages.

Yes, but shouldn't they be in GeoSciML/branches/3.0_rc1?

Bruce

GeoScience Victoria
EARTH RESOURCES DIVISION
Department of Primary Industries
Melbourne, Victoria
AUSTRALIA
Ph: +61-3-9658 4502
Fax: +61-3-9658 4555
Mobile: +61 429 177155

<Oliver.Raymond at ga.gov.au><mailto:Oliver.Raymond at ga.gov.au>

21/10/2009 01:55 PM

To

<Bruce.Simons at dpi.vic.gov.au><mailto:Bruce.Simons at dpi.vic.gov.au>

cc

<simon.cox at jrc.ec.europa.eu><mailto:simon.cox at jrc.ec.europa.eu>, <steve.richard at azgs.az.gov><mailto:steve.richard at azgs.az.gov>

Subject

RE: GeoSciML Version 2.x problems and subversion cleanup [SEC=UNCLASSIFIED]










“Shouldn't we delete GeoSciML/model?” - this directory contains the externalled copies of the refactored v3 xmi’s.  Not absolutely required, but it makes using Enterprise Architect a lot easier with the new multiple packages.


-----Original Message-----
From: Bruce.Simons at dpi.vic.gov.au<mailto:Bruce.Simons at dpi.vic.gov.au> [mailto:Bruce.Simons at dpi.vic.gov.au]
Sent: Wednesday, 21 October 2009 1:48 PM
To: Raymond Oliver
Cc: simon.cox at jrc.ec.europa.eu<mailto:simon.cox at jrc.ec.europa.eu>; steve.richard at azgs.az.gov<mailto:steve.richard at azgs.az.gov>
Subject: Re: GeoSciML Version 2.x problems and subversion cleanup [SEC=UNCLASSIFIED]


Just to emphasise Ollie's points, we have models in the following subversion (https://www.seegrid.csiro.au/subversion/GeoSciML) locations:
1. GeoSciML/branches/ (versions 2.0, 2.x, etc)
2. GeoSciML/model (version 3.x?)
3. GeoSciML/tags/ (versions 0.9, 1.0, 2.0, 2.0_rc3 etc)
4. GeoSciML/trunk/model (version ?)

For clarity
Shouldn't we be doing 3.0_rc1 and 2.1 in GeoSciML/branches?
Shouldn't we delete GeoSciML/model?
Shouldn't GeoSciML/trunk be the current release version (2.0)?

Bruce

GeoScience Victoria
EARTH RESOURCES DIVISION
Department of Primary Industries
Melbourne, Victoria
AUSTRALIA
Ph: +61-3-9658 4502
Fax: +61-3-9658 4555
Mobile: +61 429 177155
<Oliver.Raymond at ga.gov.au><mailto:Oliver.Raymond at ga.gov.au>

21/10/2009 01:17 PM


To

<simon.cox at jrc.ec.europa.eu><mailto:simon.cox at jrc.ec.europa.eu>, <steve.richard at azgs.az.gov><mailto:steve.richard at azgs.az.gov>

cc

<Bruce.Simons at dpi.vic.gov.au><mailto:Bruce.Simons at dpi.vic.gov.au>

Subject

GeoSciML Version 2.x problems and subversion cleanup [SEC=UNCLASSIFIED]














Hi Steve and Simon,

Bruce and I are puzzled by some subversion issues.  I hope you can help us out...

A preliminary question: What is the difference between a “tag” and a “branch”.  My understanding is that tag a date-stamped version which should not be edited?  Whereas a branch can be edited in an environment removed from the trunk. Correct?

---------------------------------

Version 2.1 editing

The refactoring of HollowWorld.xml over the last year has affected the GeoSciML v2.0 UML model in the subversion.  I cannot find a copy of the v2.0 UML as released on the subversion.  The UML model in www.seegrid.csiro.au\subversion\GeoSciML\tags\2.0_final<http://www.seegrid.csiro.au/subversion/GeoSciML/tags/2.0_final> is not the final v2.0 release UML.  It contains links to ISO O&M and GML 3.2 that are not in the v2.0 release schemas.  Hence the ../branches/2.x/model (which I presume was derived from v2.0_final) also has these ISO/GML problems.

1.  I suggest we replace the subversion v2.0_final UML model with the one from www.geosciml.org/geosciml/2.0<http://www.geosciml.org/geosciml/2.0>/eap.  In the absence of a clean subversion copy of the final v2.0 UML to make v2.1, I will be using the v2.0 UML that I have got from www.geosciml.org<http://www.geosciml.org/> to create GeoScimL v2.1 and I will put it on the subversion at ../branches/2.1.

2.  The www.geosciml.org<http://www.geosciml.org/> v2.0 external packages register points to class maps at https://www.seegrid.csiro.au/subversion/GeoSciML/trunk/model/.  But ../GeoSciML/trunk/model is not a stable v2.0 release model path.  Should the paths to the v2.0 release class maps point to somewhere stable on www.geosciml.org<http://www.geosciml.org/> instead?

----------------------------------

Version 3.0 editing

As instructed on the Twiki, my v3 GeoSciML package points at www.seegrid.csiro.au\subversion\GeoSciML\model\geosciml.xml<http://www.seegrid.csiro.au/subversion/GeoSciML/model/geosciml.xml>.  And the app schemas that are being updated are:

www.seegrid.csiro.au\subversion\GeoSciML\borehole\trunk\uml\Borehole.xml<http://www.seegrid.csiro.au/subversion/GeoSciML/borehole/trunk/uml/Borehole.xml>
www.seegrid.csiro.au\subversion\GeoSciML\core\trunk\uml\GeoSciML-Core.xml<http://www.seegrid.csiro.au/subversion/GeoSciML/core/trunk/uml/GeoSciML-Core.xml>
www.seegrid.csiro.au\subversion\GeoSciML\timescale\trunk\uml\geotimescale.xml<http://www.seegrid.csiro.au/subversion/GeoSciML/timescale/trunk/uml/geotimescale.xml>
www.seegrid.csiro.au\subversion\GeoSciML\vocabulary\trunk\uml\vocabulary.xml<http://www.seegrid.csiro.au/subversion/GeoSciML/vocabulary/trunk/uml/vocabulary.xml>
etc
etc

1. Do we need www.seegrid.csiro.au\subversion\GeoSciML\trunk<http://www.seegrid.csiro.au/subversion/GeoSciML/trunk> any more?  Following the refactoring, it appears to not be used for anything anymore.  Can we delete it so that users are not confused with where the latest model is?

2. Do we need www.seegrid.csiro.au\subversion\GeoSciML\modules<http://www.seegrid.csiro.au/subversion/GeoSciML/modules> any more?  It appears to be an earlier attempt at refactoring which contains old (~August 2009, pre-Quebec) versions of the refactored model.  Can we delete it?

----------------------------------

Cheers,
Ollie

------------------------------------------------------------------------------------------------
Ollie Raymond
National Advice,  Maps and Standards Project
Geoscience Australia

Address: GPO Box 378, Canberra, ACT, 2601, Australia | ABN: 80 091 799 039
Ph: (02) 62499575 | Fax: (02) 62499992 | Email: Oliver.Raymond at ga.gov.au<mailto:Oliver.Raymond at ga.gov.au>
Web:  http://www.ga.gov.au/minerals/research/national/nat_maps/nat_geol_maps.jsp

Google Map<http://maps.google.com/maps?f=q&hl=en&geocode=&q=canberra+australia&ie=UTF8&ll=-35.344028,149.158362&spn=0.007684,0.016404&t=h&z=17&iwloc=addr&om=1>

-- This message was created with 100% recycled electrons --



________________________________
Ÿԟiǀ&,ޘڭǿ㓔㓲ܥx[1](dNPЂ8}4ӀQSLSNȳ{aN57ڱૉS+‑jwZ&!魲fr�❺+b {ajجꬢvr߉ק&^eʚ⾩^皝ߨʩʸ߅8ԅ8ԟiǀ&6Zڟۡ靭Ꞧǝi֩稭ʦ颱^w񎵱N57ڱૉpبۡz⢼+۞ʧ魡ayʩʸڰ꿔㓼SO󏔣S,ޘSM��⪒-ˡz쉸ܨ~؞碦'ڜ社ڝڞ޾*?㓼SO󏔣S}‑[1]ȳ{b­ʋj��筆+‑w镦zz-랝׫o*޶ꧺfץzע蛪.̬隝מڨɩ򲊢zƨޞZب+‑vڮ稚kƭz뺜؞w讦ܢ{Zw{aǦj)㓼󅸔ߩǀ&rhMW稞ȝzay鯊'魭稭꫊{b
aǦj)㓼󅸔ߩǀ&,ޛ񎵹۲Ͼv짳ϝ)

W���(�ȝz�ay�'���m���k�{b�
a�Ǧj)�S���8��i�[1]���ȳx �S��)���n�<�۲�
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengeospatial.org/pipermail/geosciml/attachments/20091021/23f677f2/attachment.htm>


More information about the GeoSciML mailing list