[kepler-dev] [kepler-users] SDF Firing Order

[Dan Higgins]

Edward,
    Thank you very much for your offer. I fully agree that it is 
benefical to have fewer actors, especially if the differences are minor 
and could be incorporated into a single actor.

    I am not sure exactly why the FileWrite actor was created (since it 
was built by the Geon group). Hopefully, other groups participating in 
Kepler will take note of your comments.

Dan Higgins
NCEAS

---

Edward A. Lee wrote:
> Dan (an everybody else):
>
> I think we all benefit if there are fewer actors with more
> functionality.  I'm very willing to (try to) keep up with request
> to include functionality into Ptolemy II (such as the output port
> of FileWrite, which I presume is there precisely to address firing
> order issues that Ken has raised... if so, this is totally worthwhile).
>
> I confess that I don't keep up with all the updates to the Kepler
> tree, but please, don't hesitate to send me requests for enhancements
> in the Ptolemy tree.  I will try to keep up with those, at least...
> When we can reduce the total number of actors, we all benefit...
>
> So for example, rather than creating a minor variant of an actor we
> have, send me your suggested modification. I do want to take a look
> for a sanity check, or have someone else take a look (we've
> definitely gotten some off-the-wall enhancement
> requests), but supporting Kepler is a top priority, so we'll
> do what we can...
>
> Edward
>
>
> At 12:29 PM 11/26/2007, Dan Higgins wrote:
>   
>> Hi Ken,
>>    The error you report was occurring in the Kepler nightly build over 
>> the weekend and then disappeared. I assume it was a bug fixed in one of 
>> the recent Ptolemy code updates. I suggest you download a new version of 
>> the Ptolemy.
>>
>>   
>>
>> ptolemy.actor.lib.io is a package in Ptolemy, not a class. LineWriter is a class in that package and FileWrite extends that class. Any extension runs some of the code in the class it is extended from. So Kepler is running org.geon.FileWrite whihch inherits most of its methods from LineWriter. You can always run the LineWriter actor directly if you do not need the output port added in FileWrite.
>>
>> Dan
>>
>>
>> Kenneth Evans wrote:
>>     
>>> Hi,
>>>
>>> As I said before, in Kepler when I use a FileWriter from the list at the
>>> left, the Description says org.geon.FileWrite, but the code that is running
>>> is really in ptolemy.actor.lib.io.  I know this from debugging.  So I am
>>> confused.
>>>
>>> I am trying to avoid writing actors that add little details like "an port
>>> that outputs the file path" as I think it leads to more confusion.  However,
>>> that approach doesn't solve the problem of having things work as you want
>>> (or need) them to.  Fortunately, Edward fixed the FileWriter, at least the
>>> one in ptolemy.actor.lib.io.LineWriter.  It's nicest when it happens that
>>> way.
>>>
>>> I also note that is a nice fix, but I am still having problems with firing
>>> order with SDF and now with DDF.
>>>
>>> Also my ptII build in Eclipse now has errors (after doing a CVS update to
>>> get Edward's changes):
>>>
>>> Severity and Description      Path    Resource        Location
>>> Creation Time Id
>>> The method _isCloseToIfNilArrayElement(Token, Token, double) is undefined
>>> for the type PublisherTest    ptII/ptolemy/actor/lib  PublisherTest.java
>>> line 219      1196107937711   1944703
>>>
>>> Severity and Description      Path    Resource        Location
>>> Creation Time Id
>>> The method _isCloseToIfNilRecordElement(Token, Token, double) is undefined
>>> for the type PublisherTest    ptII/ptolemy/actor/lib  PublisherTest.java
>>> line 224      1196107937711   1944704
>>>
>>> Severity and Description      Path    Resource        Location
>>> Creation Time Id
>>> The method getFileName(FileParameter) in the type FileReader is not
>>> applicable for the arguments (FilePortParameter)
>>> ptII/ptolemy/codegen/c/actor/lib/io   LineWriter.java line 93
>>> 1196107933539 1944499
>>>
>>> Do I need to do something?  I do seem to get the fixed FileWriter and have
>>> successfully changed the workflow example I previously sent.
>>>
>>> Thanks,
>>>  
>>>         -Ken
>>>
>>> -----Original Message-----
>>> From: Dan Higgins [mailto:higgins at nceas.ucsb.edu] 
>>> Sent: Monday, November 26, 2007 12:01 PM
>>> To: Edward A. Lee
>>> Cc: Kenneth Evans; kepler-dev at ecoinformatics.org
>>> Subject: Re: [kepler-dev] [kepler-users] SDF Firing Order
>>>
>>> Just for anyone's interest, the 'org.geon.fileWrite' class in Kepler 
>>> extends ptolemy.actor.lib.io.LineWriter to add an port that outputs the 
>>> file path.
>>>
>>> Dan Higgins
>>>
>>>
>>> _______________________________________________
>>> Kepler-dev mailing list
>>> Kepler-dev at ecoinformatics.org
>>> http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-dev
>>>   
>>>       
>> -- 
>> *******************************************************************
>> Dan Higgins                                  higgins at nceas.ucsb.edu
>> http://www.nceas.ucsb.edu/   Ph: 805-893-5127
>> National Center for Ecological Analysis and Synthesis (NCEAS) Marine Science Building - Room 3405
>> Santa Barbara, CA 93195
>> *******************************************************************
>>
>> _______________________________________________
>> Kepler-dev mailing list
>> Kepler-dev at ecoinformatics.org
>> http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-dev
>>     
>
> ------------ 
> Edward A. Lee
> Chair of EECS and Robert S. Pepper Distinguished Professor
> 231 Cory Hall, UC Berkeley, Berkeley, CA 94720-1770
> phone: 510-642-0253, fax: 510-642-2845
> eal at eecs.Berkeley.EDU, http://www.eecs.berkeley.edu/Faculty/Homepages/lee.html  
>
>
>