From noreply at rubyforge.org  Tue Aug  4 08:53:27 2009
From: noreply at rubyforge.org (noreply at rubyforge.org)
Date: Tue,  4 Aug 2009 08:53:27 -0400 (EDT)
Subject: [BioCatalogue-developers] [Task #2688] Look to changing the session
	store to memcache
Message-ID: <20090804125328.027161858127@rubyforge.org>

Task #2688 has been updated. 

Project: BioCatalogue
Subproject: Pilot
Summary: Look to changing the session store to memcache
Complete: 0%
Status: Open

Description: Currently sessions are stored in the db. This is quite slow (but more secure than cookies).

The best solution is to use memcache as the session store.

Follow-Ups:

-------------------------------------------------------
Date: 2009-08-04 12:53
By: jits

Comment:
>From Thomas:

"we can do it yes, but we should try it on the test site first (maybe have an idea of what should be the memcached size if we handle sessions as well, or is not important ?)."

-------------------------------------------------------
For more info, visit:

http://rubyforge.org/pm/task.php?func=detailtask&project_task_id=2688&group_id=6901&group_project_id=12540

From jits at cs.man.ac.uk  Wed Aug  5 04:35:00 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Wed, 05 Aug 2009 09:35:00 +0100
Subject: [BioCatalogue-developers] PHP in Ruby
Message-ID: <4A794434.9010406@cs.man.ac.uk>

Hi Eric,

I came across this yesterday:

Phuby - A PHP Runtime from within Ruby
http://github.com/tenderlove/phuby/tree/master

Might be useful in integrating Dan's PHP WSDL Parser!

Cheers,
Jits

From jits at cs.man.ac.uk  Mon Aug 10 05:57:27 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Mon, 10 Aug 2009 10:57:27 +0100
Subject: [BioCatalogue-developers] [Fwd: [Fwd: eScience09 workshop and
	main track submission.]]
In-Reply-To: <4A7FE4F6.6000607@ebi.ac.uk>
References: <4A7FE4F6.6000607@ebi.ac.uk>
Message-ID: <4A7FEF07.9020700@cs.man.ac.uk>

Not to worry, the paper isn't for a workshop.

Cheers,
Jits


Eric Nzuobontane wrote:
> Hi Jits,
> I have received this mail about the submission made to the eScience 
> workshop.  I suppose you received it as well. Could you please follow 
> up on the request?
>
> Eric
>
>
> ------------------------------------------------------------------------
>
> Subject:
> [Fwd: eScience09 workshop and main track submission.]
> From:
> Thomas Laurent <tlaurent at ebi.ac.uk>
> Date:
> Mon, 10 Aug 2009 09:31:02 +0100
> To:
> Eric Nzuobontane <ericnzuo at ebi.ac.uk>
>
> To:
> Eric Nzuobontane <ericnzuo at ebi.ac.uk>
>
>
> Hi Eric,
>
> I received this email today, did you get it as well ? I think it was 
> probably intended to you.
>
> Cheers,
> Thomas
>
> -------- Original Message --------
> Subject: eScience09 workshop and main track submission.
> Date: Sat, 8 Aug 2009 00:28:13 -0700
> From: <cmt at microsoft.com>
> Reply-To: david.wallom at oerc.ox.ac.uk
> To: tlaurent at ebi.ac.uk
>
> Dear Thomas Laurent,   Paper#: 104  Title: The BioCatalogue: Curating 
> Web Services in the Life Sciences
>
> Please can you ensure that if the paper you have submitted is intended 
> for one of the workshops that you clearly add to the title the name of 
> the workshop before the real title.
> e.g. SECURITY. The BioCatalogue: Curating Web Services in the Life 
> Sciences if it is for the security workshop.
>
> We have left the system open to allow for this editing until Monday.
>
> regards
>
> David & Paul


From jits at cs.man.ac.uk  Mon Aug 10 11:09:27 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Mon, 10 Aug 2009 16:09:27 +0100
Subject: [BioCatalogue-developers] [Fwd: Re: [Taverna-users] EMBOSS services
	on biocatalogue - runresponse]
Message-ID: <4A803827.5000801@cs.man.ac.uk>

FYI...

Sounds like some changes to the Soaplab 2 services..... :-( as if things 
weren't tricky already....

Jits


-------- Original Message --------
Subject: 	Re: [Taverna-users] EMBOSS services on biocatalogue - runresponse
Date: 	Mon, 10 Aug 2009 16:03:01 +0100 (BST)
From: 	uludag at ebi.ac.uk
Reply-To: 	taverna-users at lists.sourceforge.net
To: 	taverna-users at lists.sourceforge.net
References: 
<351EEAB3365AFF4BAE56D8DB1926C7AF8AD050A006 at UOS-CL-EX7-L1.soton.ac.uk>



> I am trying to use emboss services such as emma in taverna to make
> workflows.
> I am using the operation 'run'. Does this then need to be connected to
> another operation in order to get the multiple alignment results? The
> inputs required for many others seems to be 'jobID', but instead of this
> the output of run is 'runresponse'.


This is my mistake. We were using metro for Soaplab-2.2 but couple of days
before the release we have discovered some problems and reverted back to
using latest jax-ws directly available as maven dependency. Version of
jax-ws available through metro maven dependency was the latest version of
jax-ws and was allowing Soaplab know which operation called but the one
included in Soaplab-2.2 doesn't let Soaplab know which operation called.
For this reason we deleted the workaround for Taverna problem of not
recognizing xsd:extended structures (we then were able to know the
operation name based on the class name of the input request). I
unfortunately forgot this change when we had the Soaplab/EMBOSS updates
recently. Please use taverna soaplab plugin that uses the Soaplab generic
interface for now, As Alan and Paul suggested. We will remove the
xsd:extend in the input structures as it was before the latest update,
hopefully within next few days.

Regards,
Mahmut




------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with 
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
taverna-users mailing list
taverna-users at lists.sourceforge.net
taverna-users at lists.sourceforge.net
Web site: http://www.taverna.org.uk
Mailing lists: http://www.taverna.org.uk/taverna-mailing-lists/


From jits at cs.man.ac.uk  Tue Aug 11 12:13:10 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Tue, 11 Aug 2009 17:13:10 +0100
Subject: [BioCatalogue-developers] sandbox.biocatalogue.org
Message-ID: <4A819896.7090407@cs.man.ac.uk>

Hi Thomas,

I'm thinking of setting up sandbox.biocatalogue.org as a permanent 
sandbox environment for API development and testing. This is different 
from test.biocatalogue.org in that features would be stable but new API 
endpoints etc can be tested by external users. Also, this will allow 
people developing applications on the API to test their apps on a 
sandbox so they don't hit the live site. We have exactly this set up for 
myExp too.

Would you be able to set up a CNAME entry for the biocatalogue.org 
domain as below:

sandbox.biocatalogue.org => rpc232.cs.man.ac.uk

Cheers,
Jits

From noreply at rubyforge.org  Fri Aug 14 09:44:43 2009
From: noreply at rubyforge.org (noreply at rubyforge.org)
Date: Fri, 14 Aug 2009 09:44:43 -0400 (EDT)
Subject: [BioCatalogue-developers] [Task #2777] Log all filters acces on
	/services
Message-ID: <20090814134443.DACCF185812F@rubyforge.org>

Task #2777 has been updated. 

Project: BioCatalogue
Subproject: Pilot
Summary: Log all filters acces on /services
Complete: 0%
Status: Open

Description: ... incl any search queries used on /services. This should also record a "search" event log.

-------------------------------------------------------
For more info, visit:

http://rubyforge.org/pm/task.php?func=detailtask&project_task_id=2777&group_id=6901&group_project_id=12540

From jits at cs.man.ac.uk  Sun Aug 16 13:58:07 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Sun, 16 Aug 2009 18:58:07 +0100
Subject: [BioCatalogue-developers] SOAPlab server issues!
Message-ID: <4A8848AF.7020300@cs.man.ac.uk>

Hi Eric,

Please treat this as highest priority.

I've been doing the Soaplab Server UI work (feel free to do an svn 
update) and have come across a few major issues! We need to address 
these urgently (except maybe 2.2) so that we get the UI fix in place 
asap and improve the user experience.

1. In the app/models/soaplab_server.rb mode file I've added a couple of 
FIXME's that you need to look at and fix.

2. Seems like we have a couple of major issues with the integrity of 
some of the Soaplab services data on the live site:

2.1. I've been working with a local copy of the db and noticed there are 
duplicate services for some of them. Example: 
http://www.biocatalogue.org/search?q=showfeat. I've noticed that the 
services are essentially the same but with different WSDLs, which is 
valid according to the duplication rule we have set up. My concern is 
that they are exactly the same service and so the annotations get spread 
out between the two. This issue is not crucial to fix for the UI thing I 
have committed, BUT it would be good to sort it out now if we can. Any 
ideas?

2.2. THIS IS CRUCIAL - A large set of the SoapServices that are soaplab 
services do not have a corresponding Relationship object to link the 
SoapService to the SoaplabServer object. I am using this Relationship in 
the UI fix to display more relevant info, so it is crucial that we sort 
this data issue out so that the UI fix is as effective as possible on 
all soaplab services.

Jits

From ericnzuo at ebi.ac.uk  Mon Aug 17 04:35:45 2009
From: ericnzuo at ebi.ac.uk (Eric Nzuobontane)
Date: Mon, 17 Aug 2009 09:35:45 +0100
Subject: [BioCatalogue-developers] SOAPlab server issues!
In-Reply-To: <4A8848AF.7020300@cs.man.ac.uk>
References: <4A8848AF.7020300@cs.man.ac.uk>
Message-ID: <4A891661.4080003@ebi.ac.uk>

Hi Jits,
I will look at this immediately and get back to you later.

Eric


Jiten Bhagat wrote:
> Hi Eric,
>
> Please treat this as highest priority.
>
> I've been doing the Soaplab Server UI work (feel free to do an svn 
> update) and have come across a few major issues! We need to address 
> these urgently (except maybe 2.2) so that we get the UI fix in place 
> asap and improve the user experience.
>
> 1. In the app/models/soaplab_server.rb mode file I've added a couple 
> of FIXME's that you need to look at and fix.
>
> 2. Seems like we have a couple of major issues with the integrity of 
> some of the Soaplab services data on the live site:
>
> 2.1. I've been working with a local copy of the db and noticed there 
> are duplicate services for some of them. Example: 
> http://www.biocatalogue.org/search?q=showfeat. I've noticed that the 
> services are essentially the same but with different WSDLs, which is 
> valid according to the duplication rule we have set up. My concern is 
> that they are exactly the same service and so the annotations get 
> spread out between the two. This issue is not crucial to fix for the 
> UI thing I have committed, BUT it would be good to sort it out now if 
> we can. Any ideas?
>
> 2.2. THIS IS CRUCIAL - A large set of the SoapServices that are 
> soaplab services do not have a corresponding Relationship object to 
> link the SoapService to the SoaplabServer object. I am using this 
> Relationship in the UI fix to display more relevant info, so it is 
> crucial that we sort this data issue out so that the UI fix is as 
> effective as possible on all soaplab services.
>
> Jits


From jits at cs.man.ac.uk  Mon Aug 17 04:45:32 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Mon, 17 Aug 2009 09:45:32 +0100
Subject: [BioCatalogue-developers] SOAPlab server issues!
In-Reply-To: <4A891661.4080003@ebi.ac.uk>
References: <4A8848AF.7020300@cs.man.ac.uk> <4A891661.4080003@ebi.ac.uk>
Message-ID: <4A8918AC.3010406@cs.man.ac.uk>

Thanks Eric.

Eric Nzuobontane wrote:
> Hi Jits,
> I will look at this immediately and get back to you later.
>
> Eric
>
>
> Jiten Bhagat wrote:
>> Hi Eric,
>>
>> Please treat this as highest priority.
>>
>> I've been doing the Soaplab Server UI work (feel free to do an svn 
>> update) and have come across a few major issues! We need to address 
>> these urgently (except maybe 2.2) so that we get the UI fix in place 
>> asap and improve the user experience.
>>
>> 1. In the app/models/soaplab_server.rb mode file I've added a couple 
>> of FIXME's that you need to look at and fix.
>>
>> 2. Seems like we have a couple of major issues with the integrity of 
>> some of the Soaplab services data on the live site:
>>
>> 2.1. I've been working with a local copy of the db and noticed there 
>> are duplicate services for some of them. Example: 
>> http://www.biocatalogue.org/search?q=showfeat. I've noticed that the 
>> services are essentially the same but with different WSDLs, which is 
>> valid according to the duplication rule we have set up. My concern is 
>> that they are exactly the same service and so the annotations get 
>> spread out between the two. This issue is not crucial to fix for the 
>> UI thing I have committed, BUT it would be good to sort it out now if 
>> we can. Any ideas?
>>
>> 2.2. THIS IS CRUCIAL - A large set of the SoapServices that are 
>> soaplab services do not have a corresponding Relationship object to 
>> link the SoapService to the SoaplabServer object. I am using this 
>> Relationship in the UI fix to display more relevant info, so it is 
>> crucial that we sort this data issue out so that the UI fix is as 
>> effective as possible on all soaplab services.
>>
>> Jits
>


From jits at cs.man.ac.uk  Mon Aug 17 07:09:55 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Mon, 17 Aug 2009 12:09:55 +0100
Subject: [BioCatalogue-developers] SOAPlab server issues!
In-Reply-To: <4A8848AF.7020300@cs.man.ac.uk>
References: <4A8848AF.7020300@cs.man.ac.uk>
Message-ID: <4A893A83.6070909@cs.man.ac.uk>

Hi Eric,

After running this past a couple of users, some new issues have been 
brought up:

3. For the SoaplabServer, we are currently storing the WSDL location 
(which is fine) but not the endpoint. We really need to store and show 
the endpoint for the Soaplab Server as this is what is most useful in 
tools like Taverna.

4. We need to record what Soaplab version the services are (ie: the v1 
untyped or v2 typed ones).

Cheers,
Jits


Jiten Bhagat wrote:
> Hi Eric,
>
> Please treat this as highest priority.
>
> I've been doing the Soaplab Server UI work (feel free to do an svn 
> update) and have come across a few major issues! We need to address 
> these urgently (except maybe 2.2) so that we get the UI fix in place 
> asap and improve the user experience.
>
> 1. In the app/models/soaplab_server.rb mode file I've added a couple 
> of FIXME's that you need to look at and fix.
>
> 2. Seems like we have a couple of major issues with the integrity of 
> some of the Soaplab services data on the live site:
>
> 2.1. I've been working with a local copy of the db and noticed there 
> are duplicate services for some of them. Example: 
> http://www.biocatalogue.org/search?q=showfeat. I've noticed that the 
> services are essentially the same but with different WSDLs, which is 
> valid according to the duplication rule we have set up. My concern is 
> that they are exactly the same service and so the annotations get 
> spread out between the two. This issue is not crucial to fix for the 
> UI thing I have committed, BUT it would be good to sort it out now if 
> we can. Any ideas?
>
> 2.2. THIS IS CRUCIAL - A large set of the SoapServices that are 
> soaplab services do not have a corresponding Relationship object to 
> link the SoapService to the SoaplabServer object. I am using this 
> Relationship in the UI fix to display more relevant info, so it is 
> crucial that we sort this data issue out so that the UI fix is as 
> effective as possible on all soaplab services.
>
> Jits
>


From ericnzuo at ebi.ac.uk  Mon Aug 17 08:10:03 2009
From: ericnzuo at ebi.ac.uk (Eric Nzuobontane)
Date: Mon, 17 Aug 2009 13:10:03 +0100
Subject: [BioCatalogue-developers] SOAPlab server issues!
In-Reply-To: <4A893A83.6070909@cs.man.ac.uk>
References: <4A8848AF.7020300@cs.man.ac.uk> <4A893A83.6070909@cs.man.ac.uk>
Message-ID: <4A89489B.5080807@ebi.ac.uk>

Hi Jits,
I have checked in some fixes on trunk for the initial issues you raised 
with soaplab. Could check these please!

>
> After running this past a couple of users, some new issues have been 
> brought up:
>
> 3. For the SoaplabServer, we are currently storing the WSDL location 
> (which is fine) but not the endpoint. We really need to store and show 
> the endpoint for the Soaplab Server as this is what is most useful in 
> tools like Taverna.
I need to make some clarification here.
I believe there are many ways that the services registration for soaplab 
in Biocatalogue could have been done. We decided to re-use the soap 
infrastructure which we had in place already. The way I chose to achieve 
this is to ask the user to supply the meta service in his/her soaplab 
server, and it only the wsdl of this service which is recorded in the 
soaplab model for the moment. What I am calling the meta service here is 
a service in soaplab server which can be queried to get information 
about the other services, and this is what I do. The question is,  how 
do you know what this service is if all you have is the endpoint to the 
soaplab server? I am interested to know how Taverna  uses just the 
endpoint to get to all the analyses!
>
> 4. We need to record what Soaplab version the services are (ie: the v1 
> untyped or v2 typed ones).
OK, but is it sufficient to assume that the typed/untyped state can be 
determined from a URL only?

Eric

>
>
> Jiten Bhagat wrote:
>> Hi Eric,
>>
>> Please treat this as highest priority.
>>
>> I've been doing the Soaplab Server UI work (feel free to do an svn 
>> update) and have come across a few major issues! We need to address 
>> these urgently (except maybe 2.2) so that we get the UI fix in place 
>> asap and improve the user experience.
>>
>> 1. In the app/models/soaplab_server.rb mode file I've added a couple 
>> of FIXME's that you need to look at and fix.
>>
>> 2. Seems like we have a couple of major issues with the integrity of 
>> some of the Soaplab services data on the live site:
>>
>> 2.1. I've been working with a local copy of the db and noticed there 
>> are duplicate services for some of them. Example: 
>> http://www.biocatalogue.org/search?q=showfeat. I've noticed that the 
>> services are essentially the same but with different WSDLs, which is 
>> valid according to the duplication rule we have set up. My concern is 
>> that they are exactly the same service and so the annotations get 
>> spread out between the two. This issue is not crucial to fix for the 
>> UI thing I have committed, BUT it would be good to sort it out now if 
>> we can. Any ideas?
>>
>> 2.2. THIS IS CRUCIAL - A large set of the SoapServices that are 
>> soaplab services do not have a corresponding Relationship object to 
>> link the SoapService to the SoaplabServer object. I am using this 
>> Relationship in the UI fix to display more relevant info, so it is 
>> crucial that we sort this data issue out so that the UI fix is as 
>> effective as possible on all soaplab services.
>>
>> Jits
>>


From jits at cs.man.ac.uk  Mon Aug 17 08:25:27 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Mon, 17 Aug 2009 13:25:27 +0100
Subject: [BioCatalogue-developers] SOAPlab server issues!
In-Reply-To: <4A89489B.5080807@ebi.ac.uk>
References: <4A8848AF.7020300@cs.man.ac.uk> <4A893A83.6070909@cs.man.ac.uk>
	<4A89489B.5080807@ebi.ac.uk>
Message-ID: <4A894C37.8080500@cs.man.ac.uk>

Hi Eric,

Eric Nzuobontane wrote:
> Hi Jits,
> I have checked in some fixes on trunk for the initial issues you 
> raised with soaplab. Could check these please!

Looks good to me, thanks.

>
>>
>> After running this past a couple of users, some new issues have been 
>> brought up:
>>
>> 3. For the SoaplabServer, we are currently storing the WSDL location 
>> (which is fine) but not the endpoint. We really need to store and 
>> show the endpoint for the Soaplab Server as this is what is most 
>> useful in tools like Taverna.
> I need to make some clarification here.
> I believe there are many ways that the services registration for 
> soaplab in Biocatalogue could have been done. We decided to re-use the 
> soap infrastructure which we had in place already. The way I chose to 
> achieve this is to ask the user to supply the meta service in his/her 
> soaplab server, and it only the wsdl of this service which is recorded 
> in the soaplab model for the moment. What I am calling the meta 
> service here is a service in soaplab server which can be queried to 
> get information about the other services, and this is what I do. The 
> question is,  how do you know what this service is if all you have is 
> the endpoint to the soaplab server? I am interested to know how 
> Taverna  uses just the endpoint to get to all the analyses!

I understand the decision we took and how it works - we discussed this 
in a LOT of detail :-) and I'm convinced it's the right decision given 
all the factors.

Isn't the meta service just what we call the SoaplabServer (or can we 
least be abstract it like that for our purposes)? In which case we could 
just store the endpoint which we get from the meta service WSDL in the 
SoaplabServer model, right?

Taverna does exactly what you are doing in your code - it calls the 
operations from the meta service, to figure out the internals.

>>
>> 4. We need to record what Soaplab version the services are (ie: the 
>> v1 untyped or v2 typed ones).
> OK, but is it sufficient to assume that the typed/untyped state can be 
> determined from a URL only?

I have no idea to be honest. My guess would be, no. Is there a way of 
determining it from the meta service's WSDL? Could you ask Mahmut about 
this?

Jits

>
> Eric
>
>>
>>
>> Jiten Bhagat wrote:
>>> Hi Eric,
>>>
>>> Please treat this as highest priority.
>>>
>>> I've been doing the Soaplab Server UI work (feel free to do an svn 
>>> update) and have come across a few major issues! We need to address 
>>> these urgently (except maybe 2.2) so that we get the UI fix in place 
>>> asap and improve the user experience.
>>>
>>> 1. In the app/models/soaplab_server.rb mode file I've added a couple 
>>> of FIXME's that you need to look at and fix.
>>>
>>> 2. Seems like we have a couple of major issues with the integrity of 
>>> some of the Soaplab services data on the live site:
>>>
>>> 2.1. I've been working with a local copy of the db and noticed there 
>>> are duplicate services for some of them. Example: 
>>> http://www.biocatalogue.org/search?q=showfeat. I've noticed that the 
>>> services are essentially the same but with different WSDLs, which is 
>>> valid according to the duplication rule we have set up. My concern 
>>> is that they are exactly the same service and so the annotations get 
>>> spread out between the two. This issue is not crucial to fix for the 
>>> UI thing I have committed, BUT it would be good to sort it out now 
>>> if we can. Any ideas?
>>>
>>> 2.2. THIS IS CRUCIAL - A large set of the SoapServices that are 
>>> soaplab services do not have a corresponding Relationship object to 
>>> link the SoapService to the SoaplabServer object. I am using this 
>>> Relationship in the UI fix to display more relevant info, so it is 
>>> crucial that we sort this data issue out so that the UI fix is as 
>>> effective as possible on all soaplab services.
>>>
>>> Jits
>>>
>


From jits at cs.man.ac.uk  Mon Aug 17 08:40:17 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Mon, 17 Aug 2009 13:40:17 +0100
Subject: [BioCatalogue-developers] SOAPlab server issues!
In-Reply-To: <4A894C37.8080500@cs.man.ac.uk>
References: <4A8848AF.7020300@cs.man.ac.uk>
	<4A893A83.6070909@cs.man.ac.uk>	<4A89489B.5080807@ebi.ac.uk>
	<4A894C37.8080500@cs.man.ac.uk>
Message-ID: <4A894FB1.5030009@cs.man.ac.uk>

I should clarify this...


> Taverna does exactly what you are doing in your code - it calls the 
> operations from the meta service, to figure out the internals. 

Taverna doesn't do *exactly* what you are doing - it doesn't use the 
WSDL because it already knows which operations to append to the endpoint 
to call. Eg:

url = endpoint + "/getAvailableAnalyses"

... etc, which follows the SoapLab operations.

Actually, Mahmut wrote the SoapLab activity plugin for Taverna :-)

Jits


Jiten Bhagat wrote:
> Hi Eric,
>
> Eric Nzuobontane wrote:
>> Hi Jits,
>> I have checked in some fixes on trunk for the initial issues you 
>> raised with soaplab. Could check these please!
>
> Looks good to me, thanks.
>
>>
>>>
>>> After running this past a couple of users, some new issues have been 
>>> brought up:
>>>
>>> 3. For the SoaplabServer, we are currently storing the WSDL location 
>>> (which is fine) but not the endpoint. We really need to store and 
>>> show the endpoint for the Soaplab Server as this is what is most 
>>> useful in tools like Taverna.
>> I need to make some clarification here.
>> I believe there are many ways that the services registration for 
>> soaplab in Biocatalogue could have been done. We decided to re-use 
>> the soap infrastructure which we had in place already. The way I 
>> chose to achieve this is to ask the user to supply the meta service 
>> in his/her soaplab server, and it only the wsdl of this service which 
>> is recorded in the soaplab model for the moment. What I am calling 
>> the meta service here is a service in soaplab server which can be 
>> queried to get information about the other services, and this is what 
>> I do. The question is,  how do you know what this service is if all 
>> you have is the endpoint to the soaplab server? I am interested to 
>> know how Taverna  uses just the endpoint to get to all the analyses!
>
> I understand the decision we took and how it works - we discussed this 
> in a LOT of detail :-) and I'm convinced it's the right decision given 
> all the factors.
>
> Isn't the meta service just what we call the SoaplabServer (or can we 
> least be abstract it like that for our purposes)? In which case we 
> could just store the endpoint which we get from the meta service WSDL 
> in the SoaplabServer model, right?
>
> Taverna does exactly what you are doing in your code - it calls the 
> operations from the meta service, to figure out the internals.
>
>>>
>>> 4. We need to record what Soaplab version the services are (ie: the 
>>> v1 untyped or v2 typed ones).
>> OK, but is it sufficient to assume that the typed/untyped state can 
>> be determined from a URL only?
>
> I have no idea to be honest. My guess would be, no. Is there a way of 
> determining it from the meta service's WSDL? Could you ask Mahmut 
> about this?
>
> Jits
>
>>
>> Eric
>>
>>>
>>>
>>> Jiten Bhagat wrote:
>>>> Hi Eric,
>>>>
>>>> Please treat this as highest priority.
>>>>
>>>> I've been doing the Soaplab Server UI work (feel free to do an svn 
>>>> update) and have come across a few major issues! We need to address 
>>>> these urgently (except maybe 2.2) so that we get the UI fix in 
>>>> place asap and improve the user experience.
>>>>
>>>> 1. In the app/models/soaplab_server.rb mode file I've added a 
>>>> couple of FIXME's that you need to look at and fix.
>>>>
>>>> 2. Seems like we have a couple of major issues with the integrity 
>>>> of some of the Soaplab services data on the live site:
>>>>
>>>> 2.1. I've been working with a local copy of the db and noticed 
>>>> there are duplicate services for some of them. Example: 
>>>> http://www.biocatalogue.org/search?q=showfeat. I've noticed that 
>>>> the services are essentially the same but with different WSDLs, 
>>>> which is valid according to the duplication rule we have set up. My 
>>>> concern is that they are exactly the same service and so the 
>>>> annotations get spread out between the two. This issue is not 
>>>> crucial to fix for the UI thing I have committed, BUT it would be 
>>>> good to sort it out now if we can. Any ideas?
>>>>
>>>> 2.2. THIS IS CRUCIAL - A large set of the SoapServices that are 
>>>> soaplab services do not have a corresponding Relationship object to 
>>>> link the SoapService to the SoaplabServer object. I am using this 
>>>> Relationship in the UI fix to display more relevant info, so it is 
>>>> crucial that we sort this data issue out so that the UI fix is as 
>>>> effective as possible on all soaplab services.
>>>>
>>>> Jits
>>>>
>>
>
> _______________________________________________
> BioCatalogue-developers mailing list
> BioCatalogue-developers at rubyforge.org
> http://rubyforge.org/mailman/listinfo/biocatalogue-developers


From jits at cs.man.ac.uk  Wed Aug 19 06:27:57 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Wed, 19 Aug 2009 11:27:57 +0100
Subject: [BioCatalogue-developers] Sandbox is ALIVE
Message-ID: <4A8BD3AD.20904@cs.man.ac.uk>

http://sandbox.biocatalogue.org is the new girl in town.

She might not be reliable, clever, athletic or even pretty, but she's 
serves her purpose well - her machine readable frontend can be used by 
anyone to test out their deepest, darkest ideas over the wire*.

Treat her nice, and she may reward you.

Thanks to Thomas for helping her on her feet.

Jits

* In normal English terms that means the sandbox is for testing out any 
new API features and for anyone developing apps over the API to use for 
testing, instead of hitting the live site.

From noreply at rubyforge.org  Wed Aug 19 10:13:20 2009
From: noreply at rubyforge.org (noreply at rubyforge.org)
Date: Wed, 19 Aug 2009 10:13:20 -0400 (EDT)
Subject: [BioCatalogue-developers] [Task #2778] possibility to add
	annotation inline to the page ...
Message-ID: <20090819141320.6E92018581FC@rubyforge.org>

Task #2778 has been updated. 

Project: BioCatalogue
Subproject: Pilot
Summary: possibility to add annotation inline to the page ...
Complete: 0%
Status: Open

Description: I've really had a bad and frustrating experiment trying to annotate the operation of this service (http://www.biocatalogue.org/services/2048-wsblastpgpservice_414364#operations).
Anytime i submit a description the page just goes back to the top. i have to scroll down to find where i was and what i've just added.
 Can the page just return to the field i have added the description to when i click submit? Please double check...

Franck

-------------------------------------------------------
For more info, visit:

http://rubyforge.org/pm/task.php?func=detailtask&project_task_id=2778&group_id=6901&group_project_id=12540

From jits at cs.man.ac.uk  Fri Aug 21 07:16:55 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Fri, 21 Aug 2009 12:16:55 +0100
Subject: [BioCatalogue-developers] [Fwd: Re: external access from
	pweb-3a]
In-Reply-To: <4A8E65BD.5050700@ebi.ac.uk>
References: <4A8E65BD.5050700@ebi.ac.uk>
Message-ID: <4A8E8227.2000402@cs.man.ac.uk>

Thanks for the update Eric.

Jits


Eric Nzuobontane wrote:
> In relation to the new WSDL parsing work, there are issues with access 
> external urls from our web servers because going out is via proxy. I 
> have asked Dan for a fix as my googling did not lead me to any quick 
> fix and I am no php expert.  Dan has left the EMBRACE project already 
> as I understood, but he promised to make a fix available. So, there 
> might be a bit of a delay for this work.
>
> Eric
>
>
> ------------------------------------------------------------------------
>
> Subject:
> Re: external access from pweb-3a
> From:
> Jenny Martin <jenny at ebi.ac.uk>
> Date:
> Thu, 20 Aug 2009 17:35:50 +0100
> To:
> Eric Nzuobontane <ericnzuo at ebi.ac.uk>
>
> To:
> Eric Nzuobontane <ericnzuo at ebi.ac.uk>
>
>
> I cannot find a way of getting the $_GET method to use a proxy.  
> Perhaps you can ask a PHP expert ;-)
>
> You could use the PHP cURL libraries.  I see it has options for 
> CURLOPT_PROXY and CURLOPT_NOPROXY.  Maybe we can set these variables 
> in the Apache environment so they are not bound into the code.
>
>
>   Using curl to Query Remote Servers
>
>
>     Curl and proxies
>
> As with all other full featured browsers curl has support for proxies. 
> Proxy servers are buffers between the requesting client and the web 
> server. Proxy servers are used for a variety of reasons including 
> companies restricting web access to people wanting to appear anonymous.
>
> There are a few curl options to set when using a proxy. First to 
> enable use of a proxy in curl use the option CURLOPT_HTTPPROXYTUNNEL. 
> Second set the proxy with the option CURLOPT_PROXY. If you need to set 
> authentication information use the option CURLOPT_PROXYUSERPWD. 
> CURLOPT_PROXYUSERPWD expects a string in the format of user:password. 
> HTTP proxies are default, to use a SOCKS proxy use the 
> CURLOPT_PROXYTYPE option.
>
> Example:
> | <? 
> $ch = curl_init();
> curl_setopt($ch, CURLOPT_URL, 'http://www.example.com');
> curl_setopt($ch, CURLOPT_HEADER, 1);
> curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
> curl_setopt($ch, CURLOPT_HTTPPROXYTUNNEL, 1);
> curl_setopt($ch, CURLOPT_PROXY, 'fakeproxy.com:1080');
> curl_setopt($ch, CURLOPT_PROXYUSERPWD, 'user:password');
> $data = curl_exec();
> curl_close($ch);
>  ?>
> |
>
> Eric Nzuobontane wrote:
>> Hi Jenny,
>> about what I mentioned over lunch, that is, having access to external 
>> sites over http from pweb-3a  from a php script.
>>
>> here is an example of a url
>>
>> http://pweb-3a:15080/WSDLUtils/WSDLUtils.php?method=parse&wsdl_uri=http://bioinformatics.istge.it:8080/axis/services/cabri_getz_services.getCabriBacteriaEntriesByIds?wsdl 
>>
>>
>>
>> Eric
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> BioCatalogue-developers mailing list
> BioCatalogue-developers at rubyforge.org
> http://rubyforge.org/mailman/listinfo/biocatalogue-developers
>   


From noreply at rubyforge.org  Wed Aug 26 04:38:58 2009
From: noreply at rubyforge.org (noreply at rubyforge.org)
Date: Wed, 26 Aug 2009 04:38:58 -0400 (EDT)
Subject: [BioCatalogue-developers] [Task #2781] Implements custom feed
	builder
Message-ID: <20090826083859.1E9951858201@rubyforge.org>

Task #2781 has been updated. 

Project: BioCatalogue
Subproject: Pilot
Summary: Implements custom feed builder
Complete: 0%
Status: Open

Description: Some people may want to subscribe to feeds for specific filters.

Build a very easy to use UI that allows people to specify the exact filters they require and then give them the URL to it. This can potentially all be done in JavaScript at client side.

-------------------------------------------------------
For more info, visit:

http://rubyforge.org/pm/task.php?func=detailtask&project_task_id=2781&group_id=6901&group_project_id=12540

From noreply at rubyforge.org  Wed Aug 26 04:39:11 2009
From: noreply at rubyforge.org (noreply at rubyforge.org)
Date: Wed, 26 Aug 2009 04:39:11 -0400 (EDT)
Subject: [BioCatalogue-developers] [Task #2781] Implement custom feed builder
Message-ID: <20090826083911.C5C7F185824E@rubyforge.org>

Task #2781 has been updated. 

Project: BioCatalogue
Subproject: Pilot
Summary: Implement custom feed builder
Complete: 0%
Status: Open

Description: Some people may want to subscribe to feeds for specific filters.

Build a very easy to use UI that allows people to specify the exact filters they require and then give them the URL to it. This can potentially all be done in JavaScript at client side.

-------------------------------------------------------
For more info, visit:

http://rubyforge.org/pm/task.php?func=detailtask&project_task_id=2781&group_id=6901&group_project_id=12540

From jits at cs.man.ac.uk  Wed Aug 26 12:29:06 2009
From: jits at cs.man.ac.uk (Jiten Bhagat)
Date: Wed, 26 Aug 2009 17:29:06 +0100
Subject: [BioCatalogue-developers] roadmap update
In-Reply-To: <d0ee827d0908260925x62c4d4fu40ec0a502727ee8@mail.gmail.com>
References: <d0ee827d0908260925x62c4d4fu40ec0a502727ee8@mail.gmail.com>
Message-ID: <4A9562D2.7090807@cs.man.ac.uk>

Looks fine to me, thanks Franck.

We need to to also start fleshing out the quarters for 2010. I guess the 
process will be started in the next F2F.

Cheers,
Jits


Franck Tanoh wrote:
> please check it out before i make it public:
> http://www.biocatalogue.org/wiki/doku.php?id=public_staging:roadmap
>
>  **Franck
> -- 
> "Science knows no country, because knowledge belongs to humanity, and 
> is the torch which illuminates the world." - Louis Pasteur
> ------------------------------------------------------------------------
>
> _______________________________________________
> BioCatalogue-developers mailing list
> BioCatalogue-developers at rubyforge.org
> http://rubyforge.org/mailman/listinfo/biocatalogue-developers
>   


From tlaurent at ebi.ac.uk  Wed Aug 26 12:32:19 2009
From: tlaurent at ebi.ac.uk (Thomas Laurent)
Date: Wed, 26 Aug 2009 17:32:19 +0100
Subject: [BioCatalogue-developers] roadmap update
In-Reply-To: <d0ee827d0908260925x62c4d4fu40ec0a502727ee8@mail.gmail.com>
References: <d0ee827d0908260925x62c4d4fu40ec0a502727ee8@mail.gmail.com>
Message-ID: <4A956393.8060309@ebi.ac.uk>

That's fine by me.

Thomas

Franck Tanoh wrote:
> please check it out before i make it public:
> http://www.biocatalogue.org/wiki/doku.php?id=public_staging:roadmap
> 
>  **Franck
> -- 
> "Science knows no country, because knowledge belongs to humanity, and is 
> the torch which illuminates the world." - Louis Pasteur
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> BioCatalogue-developers mailing list
> BioCatalogue-developers at rubyforge.org
> http://rubyforge.org/mailman/listinfo/biocatalogue-developers