soaplab-users — Forum for Soaplab users and developers

[Soaplab-users] EBI SOAPLab server for EMBOSS-4.0 applications

[Tim T. <tim...@ed...>]

Hi,

Is the EBI SOAPLab server for EMBOSS-4.0 applications available to  
download so I can install it locally?

http://www.ebi.ac.uk/soaplab/emboss4/index.html

The link to documentation on the specialized version of soaplab is  
broken on the above page.

Cheers, Tim

[Soaplab-users] emboss-services-2004-06-16 and emboss 4?

[Tim T. <tim...@ed...>]

Hi,

Does emboss-services-2004-06-16.tar.gz work with emboss 4?
Or do I still have to use emboss 2.8.0?

Thanks,

Tim

Re: [Soaplab-users] question regarding images

[Mahmut U. <ul...@eb...>]

Hi Sirisha,

If you select the workflow output for the graph produced by your dot
service, in your Taverna Advanced Model Explorer and add the 'image/png'
MIME type using the 'Metadata for ...' -> 'MIME Types' tabs then Taverna
should be able to display the graph.

Regards,
Mahmut

> I have wrapped the dot graph layout tool successfully using Soaplab, 
> using the ACD file provided. When I supply a dot file to the service, 
> the output is displayed as something like this:
> 
> application/octet-stream
> urn:lsid:net.sf.taverna:dataItem:1910c00b... etc
> 
> and the image can only be viewed by saving the result somewhere locally, 
> and then opening it with a suitable program.
> 
> However it would be better if the resulting image could be viewed 
> directly in the results pane, much like the Dilbert comic book example 
> workflow which is provided with Taverna.

[Soaplab-users] question regarding images

[Sirisha G. <si...@my...>]

I have wrapped the dot graph layout tool successfully using Soaplab, 
using the ACD file provided. When I supply a dot file to the service, 
the output is displayed as something like this:

application/octet-stream
urn:lsid:net.sf.taverna:dataItem:1910c00b... etc

and the image can only be viewed by saving the result somewhere locally, 
and then opening it with a suitable program.

However it would be better if the resulting image could be viewed 
directly in the results pane, much like the Dilbert comic book example 
workflow which is provided with Taverna.

Any tips on achieving this would be great :)

Many thanks,

Sirisha

This message has been checked for viruses but the contents of an attachment
may still contain software viruses, which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

Re: [Soaplab-users] first run works, second throws an exception

[Mahmut U. <ul...@eb...>]

Hi Martin,

> When have you last checked-out Soaplab code, was it recently or
> sometimes ago? Are you using Soaplab from CVS, or from binary
> distribution?

I thought the same possibility of CVS version of Soaplab being broken
and just executed the EMBOSS tutorial workflow 4 times from the Taverna
1.4 workbench and all returned successfully (using the EBI Soaplab
server).

Regards,
Mahmut

Re: [Soaplab-users] first run works, second throws an exception

[Martin S. <mar...@gm...>]

Hi Marcus,

When have you last checked-out Soaplab code, was it recently or sometimes
ago? Are you using Soaplab from CVS, or from binary distribution?

Martin

-- 
Martin Senger
   email: mar...@gm...
   skype: martinsenger

[Soaplab-users] first run works, second throws an exception

[Marcus E. <er...@in...>]

Hey everyone,

I have been developing webservices with soaplab 1.1 for a while now, but 
suddenly encountered a problem. After I started the Applab server and 
Tomcat I can execute a webservice that is supposed to return an 
"outfile" without any problems. However right afterwards when I try to 
execute the same webservice I get an error. If I then restart Tomcat I 
can execute the webservice once more without that error so there must be 
something left over from the execution that prevents a successful 2nd 
execution. The result to return is created in both cases, so the problem 
shows up only when its time to return the file to Taverna.

Here is the error message that Taverna (1.4) throws:

Details: org.embl.ebi.SoaplabShare.SoaplabException: 
org.embl.ebi.SoaplabShare.S
oaplabException: Internal communication failed. (in fetchResults) 
org.omg.CORBA.
COMM_FAILURE:   vmcid: SUN  minor code: 208 completed: Maybe
Details: org.embl.ebi.SoaplabShare.SoaplabException: Internal 
communication fail
ed. (in fetchResults) org.omg.CORBA.COMM_FAILURE:   vmcid: SUN  minor 
code: 208
completed: Maybe
        at 
org.embl.ebi.SoaplabServer.applab.JobAppLab.fetchResults(JobAppLab.ja
va:697)
        at 
org.embl.ebi.SoaplabServer.applab.JobAppLab.getResults(JobAppLab.java
:667)
        at 
org.embl.ebi.SoaplabServer.applab.AnalysisWSAppLabImpl.getSomeResults
(AnalysisWSAppLabImpl.java:769)
        at 
org.embl.ebi.SoaplabServer.applab.AnalysisWSAppLabImpl.getSomeResults
(AnalysisWSAppLabImpl.java:768)
        at sun.reflect.GeneratedMethodAccessor50.invoke(Unknown Source)
        at 
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAcces
sorImpl.java:25)
        at java.lang.reflect.Method.invoke(Method.java:585)

        at 
org.embl.ebi.SoaplabServer.applab.AnalysisWSAppLabImpl.logAndThrow(An
alysisWSAppLabImpl.java:163)
        at 
org.embl.ebi.SoaplabServer.applab.AnalysisWSAppLabImpl.getSomeResults
(AnalysisWSAppLabImpl.java:771)
        at 
org.embl.ebi.SoaplabServer.applab.AnalysisWSAppLabImpl.getSomeResults
(AnalysisWSAppLabImpl.java:768)
        at sun.reflect.GeneratedMethodAccessor50.invoke(Unknown Source)
        at 
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAcces
sorImpl.java:25)
        at java.lang.reflect.Method.invoke(Method.java:585)

        at 
org.apache.axis.message.SOAPFaultBuilder.createFault(SOAPFaultBuilder
.java:222)
        at 
org.apache.axis.message.SOAPFaultBuilder.endElement(SOAPFaultBuilder.
java:129)
        at 
org.apache.axis.encoding.DeserializationContext.endElement(Deserializ
ationContext.java:1087)
        at 
org.apache.xerces.parsers.AbstractSAXParser.endElement(Unknown Source
)
        at 
org.apache.xerces.impl.XMLNSDocumentScannerImpl.scanEndElement(Unknow
n Source)
        at 
org.apache.xerces.impl.XMLDocumentFragmentScannerImpl$FragmentContent
Dispatcher.dispatch(Unknown Source)
        at 
org.apache.xerces.impl.XMLDocumentFragmentScannerImpl.scanDocument(Un
known Source)
        at org.apache.xerces.parsers.XML11Configuration.parse(Unknown 
Source)
        at org.apache.xerces.parsers.XML11Configuration.parse(Unknown 
Source)
        at org.apache.xerces.parsers.XMLParser.parse(Unknown Source)
        at org.apache.xerces.parsers.AbstractSAXParser.parse(Unknown Source)
        at javax.xml.parsers.SAXParser.parse(Unknown Source)
        at 
org.apache.axis.encoding.DeserializationContext.parse(Deserialization
Context.java:227)
        at org.apache.axis.SOAPPart.getAsSOAPEnvelope(SOAPPart.java:696)
        at org.apache.axis.Message.getSOAPEnvelope(Message.java:435)
        at 
org.apache.axis.handlers.soap.MustUnderstandChecker.invoke(MustUnders
tandChecker.java:62)
        at org.apache.axis.client.AxisClient.invoke(AxisClient.java:206)
        at org.apache.axis.client.Call.invokeEngine(Call.java:2784)
        at org.apache.axis.client.Call.invoke(Call.java:2767)
        at org.apache.axis.client.Call.invoke(Call.java:2443)
        at org.apache.axis.client.Call.invoke(Call.java:2366)
        at org.apache.axis.client.Call.invoke(Call.java:1812)
        at 
org.embl.ebi.escience.scuflworkers.soaplab.SoaplabTask.execute(Soapla
bTask.java:159)
        ... 7 more

thanks in advance,
Marcus

Re: [Soaplab-users] new and improved applab

[Mahmut U. <ul...@eb...>]

Hi,

One issue we later noticed that there are number of scripts in the
SOAPLAB_HOME/ws directory that also use Applab jar files. One should
either update the CLASSPATH definitions in these scripts or should keep
copies of the previous versions of the Applab jar files in the
SOAPLAB_HOME/lib directory for the use of these scripts. As these
scripts are not for starting the server, memory leaks are not important
when using these scripts.

Regards,
Mahmut

> > - with the improved Applab server - is it sufficient to simply 
> > kill my current server, replace the 4 jar files you mentioned in the 
> > previous post with the new "applab.jar" and then restart the 
> > server/redeploy my services?
> 
> No need to redeploy the services, restarting should be enough.
> 
> If you use the run-AppLab-server script please make sure you delete the
> following 4 lines,
> 
> CLASSPATH=SOAPLAB_HOME/lib/adaptors.jar:{CLASSPATH}
> CLASSPATH=SOAPLAB_HOME/lib/AppLabClient.jar:${CLASSPATH}
> CLASSPATH=SOAPLAB_HOME/lib/AppLabServer.jar:${CLASSPATH}
> CLASSPATH=SOAPLAB_HOME/lib/AppLabShare.jar:${CLASSPATH}
> 
> and add the following line instead.
> 
> CLASSPATH=SOAPLAB_HOME/lib/applab.jar:${CLASSPATH}
> 
> Developer version of the run-AppLab-server does have a generic mechanism
> for building the CLASSPATH so it doesn't need to have the above change.
> I think you are probably not using the developer version of the script.

Re: [Soaplab-users] new and improved applab

[Mahmut U. <ul...@eb...>]

Hi Sirisha,

> Wonderful! I have just run a data-intensive workflow that always 'broke' 
> the applab server and it has completed successfully all the way through 
> for the first time! I am going to do a few more runs this afternoon to 
> make sure, but preliminary results look good :)

I carbon copy this message to the group in order to encourage people to
use the new version of applab jar file.

> One thing I would like to do is check the heap usage after running a 
> certain number of jobs

I use jconsole for this purpose. The last line of the run-Applab-server
script should normally be the following line.

java -Dcfg=Server.cfg -mx856m Server Applications.xml "$@"

If you add the "-Dcom.sun.management.jmxremote" option to this line
Applab should next time start with JMX support.

java -Dcfg=Server.cfg -Dcom.sun.management.jmxremote -mx856m Server
Applications.xml "$@"


After restarting Applab server with JMX support if you start jconsole it
will automatically detect the Applab process with JMX support and then
you can easily monitor heap usage and other characteristics of the
Applab process.

Regards,
Mahmut


> > Hi Sirisha,
> > 
> > Changes in the run-Applab-server script updates the CLASSPATH for the
> > Applab server but the deployment mechanism has its own definition of
> > CLASSPATH.For the other mechanisms use, such as the deployment
> > mechanism, you can keep the previous 4 jar files together with the
> > applab.jar file in the Soaplab lib directory.
> > 
> > Since you update the CLASSPATH in the run-Applab-server script it will
> > not use the previous 4 jar files but will use the new applab.jar file
> > but the other mechanism will continue to use the previous jar files.
> > 
> > Actually it is best to change CLASSPATH definition in all places. I will
> > look at this and will email to the group about this.
> > 
> > Please let me know if anything is not clear or not working.
> > 
> > Mahmut
> > 
> > 
> > 
> > On Fri, 2006-11-03 at 13:20 +0000, Sirisha Gollapudi wrote:
> >> I've just realised, the class not found error seems to be something to 
> >> do with AppLabShare which I had deleted both the jar file and reference 
> >> to the jar file in run-applab-server script...
> >>
> >> Mahmut Uludag wrote:
> >>> Hi Sirisha,
> >>>
> >>>> - with the improved Applab server - is it sufficient to simply 
> >>>> kill my current server, replace the 4 jar files you mentioned in the 
> >>>> previous post with the new "applab.jar" and then restart the 
> >>>> server/redeploy my services?
> >>> No need to redeploy the services, restarting should be enough.
> >>>
> >>> If you use the run-AppLab-server script please make sure you delete the
> >>> following 4 lines,
> >>>
> >>> CLASSPATH=SOAPLAB_HOME/lib/adaptors.jar:{CLASSPATH}
> >>> CLASSPATH=SOAPLAB_HOME/lib/AppLabClient.jar:${CLASSPATH}
> >>> CLASSPATH=SOAPLAB_HOME/lib/AppLabServer.jar:${CLASSPATH}
> >>> CLASSPATH=SOAPLAB_HOME/lib/AppLabShare.jar:${CLASSPATH}
> >>>
> >>> and add the following line instead.
> >>>
> >>> CLASSPATH=SOAPLAB_HOME/lib/applab.jar:${CLASSPATH}
> >>>
> >>> Developer version of the run-AppLab-server does have a generic mechanism
> >>> for building the CLASSPATH so it doesn't need to have the above change.
> >>> I think you are probably not using the developer version of the script.
> >>>
> >>> Regards,
> >>> Mahmut

Re: [Soaplab-users] new and improved applab

[Mahmut U. <ul...@eb...>]

Hi Sirisha,

> - with the improved Applab server - is it sufficient to simply 
> kill my current server, replace the 4 jar files you mentioned in the 
> previous post with the new "applab.jar" and then restart the 
> server/redeploy my services?

No need to redeploy the services, restarting should be enough.

If you use the run-AppLab-server script please make sure you delete the
following 4 lines,

CLASSPATH=SOAPLAB_HOME/lib/adaptors.jar:{CLASSPATH}
CLASSPATH=SOAPLAB_HOME/lib/AppLabClient.jar:${CLASSPATH}
CLASSPATH=SOAPLAB_HOME/lib/AppLabServer.jar:${CLASSPATH}
CLASSPATH=SOAPLAB_HOME/lib/AppLabShare.jar:${CLASSPATH}

and add the following line instead.

CLASSPATH=SOAPLAB_HOME/lib/applab.jar:${CLASSPATH}

Developer version of the run-AppLab-server does have a generic mechanism
for building the CLASSPATH so it doesn't need to have the above change.
I think you are probably not using the developer version of the script.

Regards,
Mahmut

[Soaplab-users] new and improved applab

[Sirisha G. <si...@my...>]

Hi Mahmut - with the improved Applab server - is it sufficient to simply 
kill my current server, replace the 4 jar files you mentioned in the 
previous post with the new "applab.jar" and then restart the 
server/redeploy my services? I am thrilled this improvement has been 
made, I was beginning to despair over my services taking up so much 
memory and eventually running out all together!

Many thanks :)


This message has been checked for viruses but the contents of an attachment
may still contain software viruses, which could damage your computer system:
you are advised to perform your own checks. Email communications with the
University of Nottingham may be monitored as permitted by UK legislation.

Re: [Soaplab-users] Error in creating XML parser Parser class

[Martin S. <mar...@gm...>]

Hi,

Probably your tomcat does not have XML parser. You need to put  files
xercesImpl.jar and xercesSamples.jar into
<your-tomcat-home>/common/endorsed, and restart tomcat. (The files are
widely available, e.g. in soaplab/lib - well, perhaps xercesSampels.jar is
not needed.)

Let me know if this helps.
Cheers,
Martin

-- 
Martin Senger
   email: mar...@gm...
   skype: martinsenger

[Soaplab-users] Error in creating XML parser Parser class

[Per L. <per...@sb...>]

Dear list!

I recently did a fresh install of everything required to run soaplab (ie 
apache, tomcat and axis). Installation was fine and deployment of 
default services also worked. However, when I tried to access the 
classic.helloworld service using ./ws/run-analysis-client -name 
classic.helloworld -w -r it returned this message: (same thing happens 
when i try to invoke the service using a client such as taverna). I have 
run into this before, but can't remember what the remedy was :-) (Axis 
seem to be working fine, for example)

Cheers
/Per

Fault string: org.embl.ebi.SoaplabShare.SoaplabException: Internal 
communication failed. (in cacheMetadata) 
embl.ebi.AppLabShare.AppLabException: Error in creating XML parser 
Parser class: 
org.apache.xerces.parsers.SAXParser(org.apache.xerces.parsers.SAXParser)
Details: embl.ebi.AppLabShare.AppLabException: Error in creating XML 
parser Parser class: 
org.apache.xerces.parsers.SAXParser(org.apache.xerces.parsers.SAXParser)
        at 
embl.ebi.AppLabShare.AppLabMetadataAccessorImpl._load(AppLabMetadataAccessorImpl.java:156)
        at 
embl.ebi.AppLabShare.AppLabMetadataAccessorImpl.load(AppLabMetadataAccessorImpl.java:138)
        at 
embl.ebi.AppLabShare.AppLabMetadataAccessorImpl.<init>(AppLabMetadataAccessorImpl.java:49)
        at 
org.embl.ebi.SoaplabServer.applab.AnalysisWSAppLabImpl.cacheMetadata(AnalysisWSAppLabImpl.java:385)
        at 
org.embl.ebi.SoaplabServer.applab.AnalysisWSAppLabImpl.getInputSpec(AnalysisWSAppLabImpl.java:350)
        at 
org.embl.ebi.SoaplabServer.applab.AnalysisWSAppLabImpl.getInputSpec(AnalysisWSAppLabImpl.java:349)
        at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
        at sun.reflect.NativeMethodAccessorImpl.invoke(Unknown Source)
        at sun.reflect.DelegatingMethodAccessorImpl.invoke(Unknown Source)
        at java.lang.reflect.Method.invoke(Unknown Source)

Fault code:   
{http://schemas.xmlsoap.org/soap/envelope/}Server.userException
Fault actor:  null
When calling:
        http://localhost:8080/axis/services/classic.helloworld
        getInputSpec

Re: [Soaplab-users] acd2xml

[Peter E. <P....@dk...>]

On Tue, 26 Sep 2006, Martin Senger wrote:

> > Can't use an undefined value as an ARRAY reference at ./generator/
> > acd2xml line 218.
> 
> 
> The problem is with the newer version of Perl (you are using). I have a
> fix for this bug but have not committed it yet (actually I thought I did
> but apparently not, sorry for that).

I copied the fix a few months ago to:

http://soaplab.sourceforge.net/jar-storage/acd2xml.tar.gz

However, in order to update your local cache, you must run the Ant target 
"gather". (or use "build.sh" instead of "build-dev.sh").

Kind regards,

	Peter

Re: [Soaplab-users] acd2xml

[Martin S. <mar...@gm...>]

> Can't use an undefined value as an ARRAY reference at ./generator/
> acd2xml line 218.


The problem is with the newer version of Perl (you are using). I have a
fix for this bug but have not committed it yet (actually I thought I did but
apparently not, sorry for that). Here it is: please edit
your generator/acd2xml perl script by adding the following line:

   $app{'input'} = [] unless defined $app{'input'};

Put the line just before the line:

   my ($input);  # a reference to a hash

This should fix it.

Martin

-- 
Martin Senger
   email: mar...@gm...
   skype: martinsenger

[Soaplab-users] acd2xml

[allan <al...@im...>]

Hi
I'm trying to use soaplab in a miRNA research project but have run  
into a problem with acd2xml

allan 09:43 AM> ./generator/acd2xml -d -l test.xml miranda
Processing miranda...
         using /home/allan/projects/soap2/analysis-interfaces/ 
metadata/miranda.acd
Can't use an undefined value as an ARRAY reference at ./generator/ 
acd2xml line 218.

I have  tried some of the other acd files in 'metadata' as well, but  
they all give the same error. I hope somebody has a suggestion to  
what can go wrong.

Best wishes

Allan Lind-Thomsen

[Soaplab-users] Applab is now more stable

[Mahmut U. <ul...@eb...>]

Hi,

I just wanted to say that "Too many open files" problem and most of the
important memory leaks in AppLab has been fixed.

Fixed version of AppLab is being used by the EBI Soaplab server since
last Friday evening.

Using an input sequence of about 400Kb size results now a heap usage of
about 20Mb after 600 jobs while this generates about 520Mb heap usage
with using the previous version of AppLab.

New version of AppLab can be downloaded from Soaplab web site at the
following address.

http://soaplab.sourceforge.net/applab.jar

This jar file replaces the following 4 jar files.

- ApplabServer.jar
- AppLabShare.jar
- AppLabClient.jar
- adaptors.jar

Regards,
Mahmut

Re: [Soaplab-users] Multiple input files with filelist

[Mahmut U. <ul...@eb...>]

Hi Andrew,

Glad that the modification works. I'm now sending this message to Tom as
well. I hope Tom would update the related file.

Meantime, I have been looking at memory leaks in Applab/Soaplab servers
and finally had very encouraging results yesterday on Applab servers. I
continue working on it to further reduce the leaks.

Regards,
Mahmut


On Thu, 2006-08-31 at 12:04 +0200, Andrew Emerson wrote:
> Hi Mahmut
> 
> Looks like the modification to taverna works!
> Many thanks. What are the chances of convincing the Taverna developers to 
> include the changes in the next release of Taverna?
> 
> thanks again
> Andrew
> 
> 
> 
> At 03:40 PM 8/29/2006, you wrote:
> >Hi Andrew,
> >
> > > Yes, in fact my problem was the missing :: on the command line and now it
> > > works.
> > > Final question, do you know if it is possible in Taverna to execute a
> > > soaplab web service with the file list ?
> >
> >No, but it seems that having the following additional lines in
> >SoaplabTask.java at line 82 fixes the problem.
> >
> >   if (parameterValue instanceof ArrayList){
> >     ArrayList v = (ArrayList)parameterValue;
> >     byte[][] realValue = new byte [v.size()][];
> >     parameterValue = v.toArray(realValue);
> >   }
> >
> >Taverna doesn't have the notion of a special file that includes a list
> >of input file names, similar to what we have in Soaplab AnalysisClient.
> >In taverna, however we need to specify each input file individually. You
> >can do this in the 'Run Workflow' window by first selecting 'New List'
> >rather than 'New Input' then you need to load each file by selecting
> >'New Input' and then 'Load'.
> >
> >Regards,
> >Mahmut
> 
> ------------------------
> Dr Andrew Emerson
> CINECA (High Performance Systems)
> via Magnanelli, 6/3
> 40033 Casalecchio di Reno (BO)-ITALY
> tel: +39-051-6171653, fax: +39-051-6132198
> e-mail: a.e...@ci... 
> 

Re: [Soaplab-users] Multiple input files with filelist

[Mahmut U. <ul...@eb...>]

Hi Andrew,

> Yes, in fact my problem was the missing :: on the command line and now it 
> works.
> Final question, do you know if it is possible in Taverna to execute a 
> soaplab web service with the file list ?

No, but it seems that having the following additional lines in
SoaplabTask.java at line 82 fixes the problem.

  if (parameterValue instanceof ArrayList){
    ArrayList v = (ArrayList)parameterValue;
    byte[][] realValue = new byte [v.size()][];
    parameterValue = v.toArray(realValue);
  }

Taverna doesn't have the notion of a special file that includes a list
of input file names, similar to what we have in Soaplab AnalysisClient.
In taverna, however we need to specify each input file individually. You
can do this in the 'Run Workflow' window by first selecting 'New List'
rather than 'New Input' then you need to load each file by selecting
'New Input' and then 'Load'.

Regards,
Mahmut

Re: [Soaplab-users] Multiple input files with filelist

[Andrew E. <a.e...@ci...>]

Hi Mahmut

Thanks for the reply.
Yes, in fact my problem was the missing :: on the command line and now it 
works.
(there was a memory problem when you tried on our server: I dont know why, 
but we are changing servers soon so I wont investigate).

Final question, do you know if it is possible in Taverna to execute a 
soaplab web service with the file list ?

many thanks
Andrew


At 11:28 AM 8/8/2006, you wrote:
>Hi Andrew,
>
>An example command line to supply list of files should look like as
>follows.
>
>run/run-analysis -e http://vega.cineca.it:8082/axis/services -name
>classic.listfile list ::files.txt -w -r
>
>In the above example the files.txt file contains the names of the input
>files, each file name being on a separate line. Please note the '::'
>characters before the file name, the client program use this prefix to
>distinguish the filelist options from ordinary options.
>
>The filelist option seems working fine in my local soaplab server.
>
>When I submit the above example command to your server I receive an
>"Internal Server Error". Can you please check the error message in your
>tomcat log files?
>
>Thanks,
>Mahmut
>
>
>On Fri, 2006-07-28 at 12:22 +0200, Andrew Emerson wrote:
> > Dear All
> >
> > Sorry for the repost but I am still having problems passing an
> > unknown/unspecified number of input files to soaplab using the filelist
> > data type as described in the Soaplab FAQ.
> >
> > Here is the error I get when using the run/run-analysis soap client:
> >
> > run/run-analysis -e http://vega.cineca.it:8082/axis/services -name
> > classic.listfile \
> > list file1 file2 file3 \
> > -w -r
> >
> > embl.ebi.utils.GException: ===ERROR===
> > Fault details:
> > [stackTrace: null]
> > [hostname: null]
> > Fault string: org.embl.ebi.SoaplabShare.SoaplabException:
> > CONFLICTING_INPUT. list * (Parameter :  (CORBA related problem) Unknown
> > TypeCode kind)
> >
> > Fault 
> code:   {http://schemas.xmlsoap.org/soap/envelope/}Server.userException
> > Fault actor:  null
> > When calling:
> >          http://vega.cineca.it:8082/axis/services/classic.listfile
> >          createAndRun
> > ===========
> >
> >
> > My ACD file looks like this:
> >
> > appl: ListTest [
> >    documentation: "File input test"
> >    groups: "Classic"
> >    nonemboss: "Y"
> >    executable: "listfile.pl"
> > ]
> >
> > filelist: list [
> >     additional:"Y"
> >     qualifier: "list"
> > ]
> >
> > outfile: output  [
> >    additional: "Y"
> >    default: "stdout"
> > ]
> >
> > (My application merely prints out the contents of the file passed to it).
> >
> > I must be doing something stupid but I can't see it.
> > Any help appreciated
> >
> > best wishes
> > Andy Emerson
> >
> >
> > ------------------------
> > Dr Andrew Emerson
> > CINECA (High Performance Systems)
> > via Magnanelli, 6/3
> > 40033 Casalecchio di Reno (BO)-ITALY
> > tel: +39-051-6171653, fax: +39-051-6132198
> > e-mail: a.e...@ci...
> >
> >
> > -------------------------------------------------------------------------
> > Take Surveys. Earn Cash. Influence the Future of IT
> > Join SourceForge.net's Techsay panel and you'll get the chance to share 
> your
> > opinions on IT & business topics through brief surveys -- and earn cash
> > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> > _______________________________________________
> > Soaplab-users mailing list
> > Soa...@li...
> > https://lists.sourceforge.net/lists/listinfo/soaplab-users

------------------------
Dr Andrew Emerson
CINECA (High Performance Systems)
via Magnanelli, 6/3
40033 Casalecchio di Reno (BO)-ITALY
tel: +39-051-6171653, fax: +39-051-6132198
e-mail: a.e...@ci... 

Re: [Soaplab-users] Multiple input files with filelist

[Mahmut U. <ul...@eb...>]

Hi Andrew,

An example command line to supply list of files should look like as
follows.

run/run-analysis -e http://vega.cineca.it:8082/axis/services -name
classic.listfile list ::files.txt -w -r

In the above example the files.txt file contains the names of the input
files, each file name being on a separate line. Please note the '::'
characters before the file name, the client program use this prefix to
distinguish the filelist options from ordinary options.

The filelist option seems working fine in my local soaplab server.

When I submit the above example command to your server I receive an 
"Internal Server Error". Can you please check the error message in your
tomcat log files?

Thanks,
Mahmut


On Fri, 2006-07-28 at 12:22 +0200, Andrew Emerson wrote: 
> Dear All
> 
> Sorry for the repost but I am still having problems passing an 
> unknown/unspecified number of input files to soaplab using the filelist 
> data type as described in the Soaplab FAQ.
> 
> Here is the error I get when using the run/run-analysis soap client:
> 
> run/run-analysis -e http://vega.cineca.it:8082/axis/services -name 
> classic.listfile \
> list file1 file2 file3 \
> -w -r
> 
> embl.ebi.utils.GException: ===ERROR===
> Fault details:
> [stackTrace: null]
> [hostname: null]
> Fault string: org.embl.ebi.SoaplabShare.SoaplabException: 
> CONFLICTING_INPUT. list * (Parameter :  (CORBA related problem) Unknown 
> TypeCode kind)
> 
> Fault code:   {http://schemas.xmlsoap.org/soap/envelope/}Server.userException
> Fault actor:  null
> When calling:
>          http://vega.cineca.it:8082/axis/services/classic.listfile
>          createAndRun
> ===========
> 
> 
> My ACD file looks like this:
> 
> appl: ListTest [
>    documentation: "File input test"
>    groups: "Classic"
>    nonemboss: "Y"
>    executable: "listfile.pl"
> ]
> 
> filelist: list [
>     additional:"Y"
>     qualifier: "list"
> ]
> 
> outfile: output  [
>    additional: "Y"
>    default: "stdout"
> ]
> 
> (My application merely prints out the contents of the file passed to it).
> 
> I must be doing something stupid but I can't see it.
> Any help appreciated
> 
> best wishes
> Andy Emerson
> 
> 
> ------------------------
> Dr Andrew Emerson
> CINECA (High Performance Systems)
> via Magnanelli, 6/3
> 40033 Casalecchio di Reno (BO)-ITALY
> tel: +39-051-6171653, fax: +39-051-6132198
> e-mail: a.e...@ci...  
> 
> 
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys -- and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Soaplab-users mailing list
> Soa...@li...
> https://lists.sourceforge.net/lists/listinfo/soaplab-users

[Soaplab-users] Multiple input files with filelist

[Andrew E. <a.e...@ci...>]

Dear All

Sorry for the repost but I am still having problems passing an 
unknown/unspecified number of input files to soaplab using the filelist 
data type as described in the Soaplab FAQ.

Here is the error I get when using the run/run-analysis soap client:

run/run-analysis -e http://vega.cineca.it:8082/axis/services -name 
classic.listfile \
list file1 file2 file3 \
-w -r

embl.ebi.utils.GException: ===ERROR===
Fault details:
[stackTrace: null]
[hostname: null]
Fault string: org.embl.ebi.SoaplabShare.SoaplabException: 
CONFLICTING_INPUT. list * (Parameter :  (CORBA related problem) Unknown 
TypeCode kind)

Fault code:   {http://schemas.xmlsoap.org/soap/envelope/}Server.userException
Fault actor:  null
When calling:
         http://vega.cineca.it:8082/axis/services/classic.listfile
         createAndRun
===========


My ACD file looks like this:

appl: ListTest [
   documentation: "File input test"
   groups: "Classic"
   nonemboss: "Y"
   executable: "listfile.pl"
]

filelist: list [
    additional:"Y"
    qualifier: "list"
]

outfile: output  [
   additional: "Y"
   default: "stdout"
]

(My application merely prints out the contents of the file passed to it).

I must be doing something stupid but I can't see it.
Any help appreciated

best wishes
Andy Emerson


------------------------
Dr Andrew Emerson
CINECA (High Performance Systems)
via Magnanelli, 6/3
40033 Casalecchio di Reno (BO)-ITALY
tel: +39-051-6171653, fax: +39-051-6132198
e-mail: a.e...@ci...  

Re: [Soaplab-users] ACD question

[Marc L. <Mar...@DE...>]

Thanks a lot for this intensive investigation !

Regards,
Marc
 =20

> -----Original Message-----
> From: Mahmut Uludag [mailto:ul...@eb...]=20
> Sent: Friday, June 30, 2006 5:03 PM
> To: Marc Logghe
> Cc: soa...@li...
> Subject: Re: [Soaplab-users] ACD question
>=20
> Hi Marc,
>=20
> On Tue, 2006-06-27 at 16:08 +0200, Marc Logghe wrote:
> > I thought I finally got it. But it seems I am lost again. Or is it=20
> > only due to some inconsistency in the system ?
> > Tried to use that trick again, as described below. But this=20
> time with=20
> > a default value set to 'true'. Guess what. It does not work at all.=20
> > Just the opposite behaviour.
> > Consider the acd:
> >=20
> > boolean: show_header  [
> >   additional: "Y"
> >   default: "Y"
> >   comment: "defaults"
> >   comment: "method --noheader"
> > ]
> >=20
> > The behaviour I expected was:
> > If the user sets show_header to 'true', the --noheader is=20
> not put on=20
> > the command line. In all other cases (e.g. show_header =3D>=20
> false) the=20
> > passed value is different from the default and thus,=20
> --noheader is put=20
> > on the command line.
> > It happened to be completely the way around.
> > On the other hand, if this was put in the acd (like in the old post
> > below) it behaves as expected.
> >=20
> > boolean: no_header  [
> >   additional: "Y"
> >   default: "N"
> >   comment: "defaults"
> >   comment: "method --noheader"
> > ]
> >=20
> > Meaning: if no_header is set to 'true', --noheader is added to the=20
> > command line, if no_header is ommitted or set to 'false',=20
> the option=20
> > is not added to the command line because 'false' is the=20
> default value.
> >=20
> > Why did it not work in the other way (the first example) ?
> > What is the system behind it ????
>=20
> In AppLab, having representation of Boolean parameter values=20
> in the command line are basically decided by the following=20
> piece of code in the Parameter class.=20
>=20
> <<<<<
> if (strVal =3D=3D null)
>   value =3D (paramDef.is (NODISPLAY) ? UUtils.is=20
> (paramDef.dflt) : false); else
>   value =3D UUtils.is (strVal);
> if (! value) return new String[0];
> >>>>>
>=20
> This can be represented by the following pseudo code.
>=20
> <<<<<
> If the parameter value is null
> Then the value is set
>   to the default value if the parameter is of type nondisplay
>   otherwise the value is set to false
>=20
> If the parameter value is false
> Then do not include any representation on the command line
> >>>>>
>=20
> The default parameter values are only used if the client=20
> didn't provide any value and the parameter is of type=20
> 'nondisplay'. Even this doesn't guarantee that the parameter=20
> will have an explicit representation on the command line=20
> because if the final parameter value is 'false' then it is=20
> automatically excluded from the command line. There is no=20
> comparison of the default value and the actual value as you=20
> expected above.
>=20
> In order to find out the above piece of code I needed to=20
> debug both Applab and Soaplab servers in Eclipse. It seems=20
> that Eclipse debugger works fine with debugging multiple=20
> programs at the same time even when one of the programs=20
> basically is the Tomcat server. The only tricky bit regarding=20
> the debugging of the Soaplab servers, you manually need to=20
> remove the soaplab.jar file from your axis directory and add=20
> your source code based soaplab project to the classpath of=20
> the tomcat server using the launch configuration manager=20
> provided by Eclipse.
>=20
> Regards,
> Mahmut
>=20
>=20
>=20

Re: [Soaplab-users] ACD question

[Mahmut U. <ul...@eb...>]

Hi Marc,

On Tue, 2006-06-27 at 16:08 +0200, Marc Logghe wrote:
> I thought I finally got it. But it seems I am lost again. Or is it only
> due to some inconsistency in the system ?
> Tried to use that trick again, as described below. But this time with a
> default value set to 'true'. Guess what. It does not work at all. Just
> the opposite behaviour.
> Consider the acd:
> 
> boolean: show_header  [
>   additional: "Y"
>   default: "Y"
>   comment: "defaults"
>   comment: "method --noheader"
> ]
> 
> The behaviour I expected was:
> If the user sets show_header to 'true', the --noheader is not put on the
> command line. In all other cases (e.g. show_header => false) the passed
> value is different from the default and thus, --noheader is put on the
> command line.
> It happened to be completely the way around.
> On the other hand, if this was put in the acd (like in the old post
> below) it behaves as expected.
> 
> boolean: no_header  [
>   additional: "Y"
>   default: "N"
>   comment: "defaults"
>   comment: "method --noheader"
> ]
> 
> Meaning: if no_header is set to 'true', --noheader is added to the
> command line, if no_header is ommitted or set to 'false', the option is
> not added to the command line because 'false' is the default value.
> 
> Why did it not work in the other way (the first example) ?
> What is the system behind it ????

In AppLab, having representation of Boolean parameter values in the
command line are basically decided by the following piece of code in the
Parameter class. 

<<<<<
if (strVal == null)
  value = (paramDef.is (NODISPLAY) ? UUtils.is (paramDef.dflt) : false);
else
  value = UUtils.is (strVal);
if (! value) return new String[0];
>>>>>

This can be represented by the following pseudo code.

<<<<<
If the parameter value is null
Then the value is set
  to the default value if the parameter is of type nondisplay
  otherwise the value is set to false

If the parameter value is false
Then do not include any representation on the command line
>>>>>

The default parameter values are only used if the client didn't provide
any value and the parameter is of type 'nondisplay'. Even this doesn't
guarantee that the parameter will have an explicit representation on the
command line because if the final parameter value is 'false' then it is
automatically excluded from the command line. There is no comparison of
the default value and the actual value as you expected above.

In order to find out the above piece of code I needed to debug both
Applab and Soaplab servers in Eclipse. It seems that Eclipse debugger
works fine with debugging multiple programs at the same time even when
one of the programs basically is the Tomcat server. The only tricky bit
regarding the debugging of the Soaplab servers, you manually need to
remove the soaplab.jar file from your axis directory and add your source
code based soaplab project to the classpath of the tomcat server using
the launch configuration manager provided by Eclipse.

Regards,
Mahmut

Re: [Soaplab-users] How to quote arguments ?

[Mahmut U. <ul...@eb...>]

Hi Marc,

> It seemed that trying to verify the quoting played some tricks on me: no
> quotes in the soaplab report + dump of @ARGV (the soaplab wrapped
> command is a perl script) does not show quotes neither. But it seems to
> work, as the output to stdout is catched.
> I mean, if the command would end with a ';', then it would be impossible
> to redirect the stdout to a file, right ?
> It is the difference between
> 'command --arg1 blah --arg2 blah ; > outfile' => outfile is empty.
> and
> 'command --arg1 blah --arg2 blah ";" > outfile' => outfile has content

The ';' characters on the command line don't cause any problem because
the actual command lines generated by Applab don't include the output
direction as in the above example and are executed using the Perl's exec
function. Outputs of the command  line jobs are collected by the
launchers by listening the stdout and the stderr ports of the child
processes that actually execute the command lines.

Regards,
Mahmut