[kepler-dev] kepler actor documentation

Nandita Mangal nmangal at sdsc.edu
Thu Apr 6 14:29:48 PDT 2006


Hi Matt,
Dan had faced a similar situation , when creating the parameter. 
Basically creating the parameter via :
 say  yyy = new StringParameter(this, "yyy param");
is not setting the display name to "yyy param", but a whole new 
parameter is being created. And hence in "Get Documentation",
"yyy param" has no information.
*I have included the conversation: " [kepler-dev] Documentation display 
in Kepler "

I will work on unescaping XML and showing it up with HTML formatting. 
Last time i tried this , i was having invalid XML in few Actors javadocs 
resulting in failing builds to generateDoc successfully...but i will 
work on it now.

thanks,
nandita.
______________________________________________________________________________________________________________
Edward,

   Thanks again for setting me straight on the parameter 'DisplayName' 
issue.

   I have one suggested change to your DocViewer code as a result of 
looking into this question. Currently the DocViewer window shows the 
parameter Name when documentation is displayed. To make the display 
agree with what is displayed in the Configuration dialog, it would seem 
useful to change the 'parameter.getName()' statement (line 302 of 
DocViewer.java) in the _getParameterEntries method to 
'parameter.getDisplayName()' (and this should change nothing if there is 
no DisplayName)

Dan Higgins

Edward A. Lee wrote:

>
> Dan:
>
> The code you give isn't right...  The statement:
>
>    numYPixels = new StringParameter(this, "Number of Y pixels in image");
>
> is not setting the display name to "Number of Y pixels in image",
> but rather is creating a new parameter named "Number of Y pixels in 
> image".
> The right way to set the display name is:
>
>    numYPixels = new StringParameter(this, "numYPixels");
>    numYPixels.setDisplayName("Number of Y pixels in image");
>
> Note that the former code violates our coding standards, and as
> a consequence, the actor will not clone correctly.  After the clone()
> operation, the numYPixels field of the new actor will actually be a
> reference to the numYPixels parameter of the original actor. That is,
> both actors will refer to the same parameter.  If you insist on having
> a field name different from the parameter name, then you should override
> the clone(Workspace) method to correctly set the field of the clone.
>
> That said, I'm not sure whether the documentation infrastructure is
> properly reflecting the possible distinction between the parameter
> name and the display name. There may be an actual bug there...
>
> Edward
>
>
> At 02:26 PM 3/14/2006, Dan Higgins wrote:
>
>> Hi All,
>>
>>     In adding documentation to the RExpression actor, I discoved a
>> problem with the documentation display in Kepler (and perhaps in
>> Ptolemy, also).
>>
>> Consider one parameter defined in the RExpression actor -
>>
>>    /**
>>     * The height of the output graphics bitmap in pixels
>>     */
>>     public StringParameter numYPixels;
>>
>> But later on, this parameter is created with the statement
>>
>>         numYPixels = new StringParameter(this, "Number of Y pixels in
>> image");
>>
>> Note that the display string "Number of Y pixels in image" is not the
>> same as the varible name 'numYPixels'
>>
>> The doclet generated xml file has the proper association between the
>> property and value; i.e.
>>
>> <property name="numYPixels">The height of the output graphics bitmap in
>> pixels</property>
>>
>> But when I display the doc xml in Kepler using 'Get Documentation' the
>> result is
>>
>> "Number of Y pixels in image - No description"
>>
>> The code to display the xml file replaces the 'numYPixels' name with its
>> display string 'Number of Y pixels in image' but it then loses the
>> value! This seems to happen whenever the parameter name is different
>> than the display string; when they agree, the descriptions are shown
>> properly.
>>
>> Dan
>>
>>
>>
 





Matt Jones wrote:

> I updated the EML200DataSource documentation this morning to try to 
> show what I thought would be a good level of granularity for the actor 
> documentation.  It now displays in Kepler if you run 'ant generateDoc' 
> before you run 'ant run-dev'.   However, for some reason the Parameter 
> documentation does not display -- it just says "No description".  Any 
> idea of how to fix this? The 'trigger' and 'output' ports inherited 
> from the superclass also show up as 'No port description', but I think 
> you already knew about that issue.
>
> Nandita's HTML version of the actor docs shows all of the 
> documentation, including the Parameter docs. However, it ignores the 
> HTML formatting that I embedded.  It seems it would be nice to be able 
> to deal with that because we need a way to provide some formatting in 
> longer documentation sections.  The html tags are probably being 
> escaped in the XML output and so probably are not being recognized by 
> the XSLT processor.  Nandita -- what do you think?
>
> Jing -- can you review the EML200DataSource documentation for content 
> to be sure I didn't make any mistakes -- I pretty much rewrote it all. 
> Thanks.
>
> Matt
>
> Dan Higgins wrote:
>
>> Matt,
>>    You may be interested in just seeing the list of actors currently 
>> in Kepler that were created at NCEAS under SEEK. [May be incomplete.]
>>
>> Dan
>>
>>
>> ------------------------------------------------------------------------
>>
>> Actors created by NCEAS
>>
>> -- Dan Higgins --
>> AddGrids
>> CVHull2Raster
>> GARPSummary
>> GISHullActorJ
>> GridRescaler
>> GridReset
>> MergeGrids
>> MessageDigestTest MetadataSource
>> ZipFiles
>> Rescaler
>> KeplerImageReader
>> ClimateChangeFileProcessor
>> ClimateFileProcessor
>> AscToRaw
>> URLToLocalFile
>> IJMacro
>> ImageJActor
>> ShowLocations
>> RExpression
>> ---
>> -- Chad --
>> GarpPrediction
>> GARPPresampleLayers
>> GARPAlgorithm
>> GDALFormatTranslator
>> GDALWarpandProjection FilesystemCacheGet
>> FilesystemCachePut
>> ---
>>
>> --Matt/Jing--
>> EML2Dataset
>> EcogridWriter
>> RBoxPlot
>> SummaryStatistics
>> ---
>
>



More information about the Kepler-dev mailing list