Re: [las_users] configuring LAS "properties" in xml-files

[Roland Schweitzer <Roland.Schweitzer@xxxxxxxx>]

Hi Martin,

These issues with inheritance of properties are bugs.  The logic was to 
add the properties from the data set and variable only if the property 
did not already exist, when in fact these properties should be replace 
an existing property.  This is fixed and will be in the next release.  
If you're feeling brave you can apply this patch:

Index: JavaSource/gov/noaa/pmel/tmap/las/jdom/LASConfig.java
===================================================================
--- JavaSource/gov/noaa/pmel/tmap/las/jdom/LASConfig.java    (revision 
12024)
+++ JavaSource/gov/noaa/pmel/tmap/las/jdom/LASConfig.java    (working copy)

@@ -2246,10 +2249,8 @@
                     String propName = prop.getChildTextNormalize("name");
                     String propValue = 
prop.getChildTextNormalize("value");
                     String currentValue = group.get(propName);
-                    if (currentValue == null) {
+                    if (propValue != null && propName != null ) {
                         group.put(propName, propValue);
-                    } else if (currentValue.equals("default")) {
-                        group.put(propName, propValue);
                     }
                 }

@@ -2272,11 +2273,9 @@
                     String propName = prop.getChildTextNormalize("name");
                     String propValue = 
prop.getChildTextNormalize("value");
                     String currentValue = group.get(propName);
-                    if (currentValue == null) {
+                    if (propValue != null && propName != null ) {
                         group.put(propName, propValue);
-                    } else if (currentValue.equals("default")) {
-                        group.put(propName, propValue);
-                    }
+                    }
                 }


The reset button problem is a separate issue, also a known bug.

Roland

On 04/13/2011 05:18 AM, Martin Schmidt wrote:
> Sorry, wrong topic - so I resend ...
>
> Good morning,
> I am working on the design of a LAS using version 7.2. I would like to 
> ask some questions concerning
> the behaviour of the "plot options". I am trying to set the defaults 
> of plot levels, palettes and so on in the
> xml-files.
>
> Following the manual there should be a hierarchie of "properties", but 
> I cannot get this working.  The LAS is
>
> phy-51.io-warnemuende.de
>
> Here a piece of the xml-file:
>
> <datasets>
> <Iowtopo1 name="IOW Belt Sea Topography 1 n.m."
>             url="/u/las/las-daten/IOW-topgraphie/iowtopo1_rev02.nc"
>         doc="http://www.io-warnemuende.de/research/de_iowtopo.html";>
> <properties>
> <ferret>
> <palette>no_red</palette>
> <land_type>none</land_type>
> <do_contour>0</do_contour>
> <fill_type>shade</fill_type>
> </ferret>
> </properties>
> <variables>
> <LANDMASK-Iowtopo1 name="prescribed landmask" units="no units" 
> url="#LANDMASK">
> <link match="/lasdata/grids/grid-XT_I-YT_J-Iowtopo1" />
> <properties>
> <ferret>
> <land_type>none</land_type>
> <palette>inverse_greyscale</palette>
> <fill_levels>(-1,1,1)</fill_levels>
> </ferret>
> </properties>
> </LANDMASK-Iowtopo1>
> ...
> <Z_NEAR-Iowtopo1 name="nearest data point height/depth" units="m" 
> url="#Z_NEAR">
> <link match="/lasdata/grids/grid-XT_I-YT_J-Iowtopo1" />
> <properties>
> <ferret>
> <land_type>none</land_type>
> <palette>land_sea</palette>
> <fill_levels>(-100)(-50,50,5)(100)</fill_levels>
> </ferret>
> </properties>
> </Z_NEAR-Iowtopo1>
>
> The problems are for example:
> - there is allways a land contour in the plots until explicitely 
> switched off via the "set plot options" menue. land_type is ignored.
> - the palette of   LANDMASK should be inverse_greyscale but no_red is 
> used. Changing this in "plot options", also
>   the palettes of other variables are changed, for example sea 
> "nearest data point height". Also setting "default" in
>   "set plot options" does not have any effect. Also RESET does nothing
>
> There are other examples of the same kind.
>
> Is this a know issue and are there any methhods to fix it?
>
> Greetings,
> Martin