pywrapper-devel — PyWrapper Developers mailing list

[PyWrapper-devel] Python 2.5?

[Roger H. <ro...@td...>]

Hi All,

I am installing PyWrapper on my shiny new MacBook and (possibly  
stupidly) have installed the latest version of Python (2.5). This is  
the 'The current production version' according to Python.

I have hacked the adapt_installation.py and the start_server.py  
scripts so they don't complain about 2.5 and I now get as far as the  
web GUID.

Next problem is to get the MySQL driver for 2.5 but before I do this  
I was wondering whether I am really wasting my time and would be  
better off rolling back to 2.4 if I want to have a usable  (but not  
production) PyWrapper install to work with (I don't want to do major  
porting work). In other words are there known issues I am going to  
run into that are likely to be problematic.

If I should roll back to 2.4 any idea how I do it under OS X?

Thanks for your thoughts and help,

Roger

Re: [PyWrapper-devel] I was trying...

[Javier de la T. <ja...@gm...>]

Hi Milko,

Well it does not look like the ampersand problem. It looks like if  
the service is not really register on the service. you know that to  
uniquely identify a service you have the service name + the  
authority? Could it be that the authority you have is different?

Javier.

On 05/12/2006, at 15:14, Skofic A. Milko (Bioversity - HQ) wrote:

> ... to deregister a MOBY service and got the following error:
>
> Sorry, the service <b>TestServiceABCD</b> could not be  
> deregistered: The service specified by authority=eurisco.ecpgr.org  
> servicename=TestServiceABCD does not exist in the registry
>
> This is strange because it appears to be in the registry and it is  
> in the MOBY prefs of TAPIR. Do you think this could have something  
> to do with the problem of the ampersands? Should I manually  
> deregister the service (with dashboard)?
>
> Ciao!
>
> Milko

[PyWrapper-devel] TAPIR Developers Workshop

[Renato De G. <re...@cr...>]

TDWG and GBIF Call for participation

TAPIR Developers Workshop

The aim of the workshop is to advance the deployment of the TAPIR 
protocol and augment its community.

TAPIR, the TDWG Access Protocol for Information Retrieval, is being 
actively developed. There is a draft specification 
(http://www.tdwg.org/dav/subgroups/tapir/1.0/docs/) and a fully 
functional data provider available (http://www.pywrapper.org/). By 
the end of 2006 another data provider implementation and a TAPIR 
Network Builder's Guide should also be available.

The workshop will start with a training session and will be followed 
by hands-on development sessions in separate work groups. 

Outcomes of the workshop will include:

* Explanation of the main features and concepts of TAPIR to new 
developers.
* Amendments to the existing TAPIR documentation where needed.
* Development (at least partially) of new TAPIR client software 
including libraries and specific data harvesters to be used by 
biodiversity data networks. 
* Discussion, implementation and documenting LSID resolution service 
in existing TAPIR provider software.
* Creation of a generic query tool to interact with TAPIR providers.
* Creation of new output models and query templates to be used by 
TAPIR networks.
* Planning and implementation of interoperability unit tests for 
TAPIR providers.
* Production of additional development proposals related to TAPIR.

Dates: 12th to 15th of February 2007

Venue: GBIF, Copenhagen

How to participate:

Please contact Renato De Giovanni (renato [at] cria dot org dot br) 
before December 31st 2006 providing the following information:

* Your name, institution and location.
* What kind of development(s) you are planning or need to do in 
relation to TAPIR (what are the basic functionalities needed, what is 
the associated project, what is the expected timeline). 
* Whether your project/institution can help with the funding of your 
participation in this meeting.

Attendance is limited and participants will be prioritised according 
to the contribution they can make.

Support funding: All travel expenses for approved attendees will be 
covered by the TDWG Infrastructure Project (which in turn has been 
funded by the Gordon and Betty Moore Foundation). Travel will be in 
accordance with the GBIF travel guidelines 
(http://www.gbif.org/Support/travel).

For further information please contact Renato De Giovanni (renato 
[at] cria dot org dot br), Donald Hobern (dhobern [at] gbif dot org) 
or Lee Belbin (lee [at] tdwg dot org).

[PyWrapper-devel] Problem with MySQL library

[Javier de la T. <ja...@gm...>]

Hi all,

Just to not forget. I had to rollback to MySQLdb 1.2.0 library  
because 1.2.1b was not working in MySQL 5.
Just in case someone is suffering the same problem.

Javier.

[PyWrapper-devel] Fwd: [announce] Trac upgrade tonight

[Javier de la T. <ja...@gm...>]

Hi all,

We were going to release PyWrapper 3.0 today but considering that  
there is major update of our trac system today, look at the forwarded  
email, and considering that we have several extensions apart of the  
ones provided by our hosting service I think it is better to delay it  
until we know there is not going to be any trouble.

Meanwhile I thik Markus has started the releasing process and that  
the software is already available on the web site if anybody wants to  
try.

I will be checking tomorrow if there was any trouble with the update  
and try to bring things up as fast as possible.

Javier.

Begin forwarded message:

> From: "GeekISP Support" <su...@ge...>
> Date: 27 de noviembre de 2006 20:19:35 GMT+01:00
> To: ann...@ge...
> Subject: [announce] Trac upgrade tonight
>
> Dear GeekISP Customers,
>
> My apologies for the short notice, but I'm going to try and upgrade  
> the
> installed Trac instances to the 0.10.x series this evening,  
> probably after
> 7:30 or 8pm ET.
>
> During this time trac instances will be unavailable, but SVN and  
> all other
> services will continue to operate as normal.  I don't have a good  
> estimate
> on how long this will take, but if it takes 30 seconds per trac
> repository, then the upgrade will take roughly an hour.
>
> Moving to a 0.10.x Trac version will make it easier for me to  
> provide you
> with anti-spam plugins for trac.  I am looking into providing some
> anti-spam plugins in the default trac configuration.
>
> Regards,
> -- 
> Dave Steinberg
> http://www.geekisp.com/
> http://www.steinbergcomputing.com/

Re: [PyWrapper-devel] debian

[Patricia M. <p_m...@ya...>]

Hi Javier

Fred is using Debian routinely. I think he has about
104 servers running Debian at his new job ...

Please fill free to contact him at
fre...@gm... if you have specific
questions on Debian. 

Formally at BeBIF we had the previous versions of
PyWrapper running on Debian systems.


see you 

Pat 


--- Javier de la Torre <ja...@gm...> wrote:

> Hi Markus,
> 
> I dont have a Debian system to try this, but in the
> message on this  
> wiki page it looks more like if gcc is not
> available, although I know  
> this is very unlikely to be true.
> 
> I checked on the lxml and they say:
> 
> ----------
> If you have easy_install, you can run the following
> as super-user:
> 
> easy_install lxml
> 
> This has been reported to work on Linux, MacOS-X
> 10.4 and Windows, as  
> long as libxml2 and libxslt are properly installed.
> To compile and  
> install lxml without easy_install, please read how
> to build lxml from  
> source
> ------------
> 
> Could the problem with debian be that libxml2 and
> libxslt are not  
> properly installed? In mac os x they are availble by
> default, and  
> maybe in gento too, but debian is more elitist 
> choosing what to have  
> installed no?
> 
> Pyrex is only needed If you want to build lxml from
> SVN.
> 
> Is there any accesible debian machine out there I
> can have access to  
> take a look?
> 
> Javi.
> 
> On 23/11/2006, at 11:50, Döring, Markus wrote:
> 
> > Hey,
> > The debian problem is because lxml doesnt install
> out of the box on  
> > debian:
> >
> > http://trac.pywrapper.org/pywrapper/wiki/ 
> > InstallerTroubleshooting#Debian
> >
> > Not sure if we can fix that easily.
> > We might need to do a manual install into the
> system using apt-get...
> >
> > Markus
> 
> 
>
-------------------------------------------------------------------------
> 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
> _______________________________________________
> PyWrapper-devel mailing list
> PyW...@li...
>
https://lists.sourceforge.net/lists/listinfo/pywrapper-devel
> 



 
____________________________________________________________________________________
Do you Yahoo!?
Everyone is raving about the all-new Yahoo! Mail beta.
http://new.mail.yahoo.com

Re: [PyWrapper-devel] debian

[Patricia M. <p_m...@ya...>]

Hi Javier

Fred is using Debian routinely. I think he has about
104 servers running Debian at his new job ...

Please fill free to contact him at
fre...@gm... if you have specific
questions on Debian. 

Formally at BeBIF we had the previous versions of
PyWrapper running on Debian systems.


see you 

Pat 


--- Javier de la Torre <ja...@gm...> wrote:

> Hi Markus,
> 
> I dont have a Debian system to try this, but in the
> message on this  
> wiki page it looks more like if gcc is not
> available, although I know  
> this is very unlikely to be true.
> 
> I checked on the lxml and they say:
> 
> ----------
> If you have easy_install, you can run the following
> as super-user:
> 
> easy_install lxml
> 
> This has been reported to work on Linux, MacOS-X
> 10.4 and Windows, as  
> long as libxml2 and libxslt are properly installed.
> To compile and  
> install lxml without easy_install, please read how
> to build lxml from  
> source
> ------------
> 
> Could the problem with debian be that libxml2 and
> libxslt are not  
> properly installed? In mac os x they are availble by
> default, and  
> maybe in gento too, but debian is more elitist 
> choosing what to have  
> installed no?
> 
> Pyrex is only needed If you want to build lxml from
> SVN.
> 
> Is there any accesible debian machine out there I
> can have access to  
> take a look?
> 
> Javi.
> 
> On 23/11/2006, at 11:50, Döring, Markus wrote:
> 
> > Hey,
> > The debian problem is because lxml doesnt install
> out of the box on  
> > debian:
> >
> > http://trac.pywrapper.org/pywrapper/wiki/ 
> > InstallerTroubleshooting#Debian
> >
> > Not sure if we can fix that easily.
> > We might need to do a manual install into the
> system using apt-get...
> >
> > Markus
> 
> 
>
-------------------------------------------------------------------------
> 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
> _______________________________________________
> PyWrapper-devel mailing list
> PyW...@li...
>
https://lists.sourceforge.net/lists/listinfo/pywrapper-devel
> 



 
____________________________________________________________________________________
Yahoo! Music Unlimited
Access over 1 million songs.
http://music.yahoo.com/unlimited

Re: [PyWrapper-devel] debian

[Patricia M. <p_m...@ya...>]

Hi Javier

Fred is using Debian routinely. I think he has about
104 servers running Debian at his new job ...

Please fill free to contact him at
fre...@gm... if you have specific
questions on Debian. 

Formally at BeBIF we had the previous versions of
PyWrapper running on Debian systems.


see you 

Pat 


--- Javier de la Torre <ja...@gm...> wrote:

> Hi Markus,
> 
> I dont have a Debian system to try this, but in the
> message on this  
> wiki page it looks more like if gcc is not
> available, although I know  
> this is very unlikely to be true.
> 
> I checked on the lxml and they say:
> 
> ----------
> If you have easy_install, you can run the following
> as super-user:
> 
> easy_install lxml
> 
> This has been reported to work on Linux, MacOS-X
> 10.4 and Windows, as  
> long as libxml2 and libxslt are properly installed.
> To compile and  
> install lxml without easy_install, please read how
> to build lxml from  
> source
> ------------
> 
> Could the problem with debian be that libxml2 and
> libxslt are not  
> properly installed? In mac os x they are availble by
> default, and  
> maybe in gento too, but debian is more elitist 
> choosing what to have  
> installed no?
> 
> Pyrex is only needed If you want to build lxml from
> SVN.
> 
> Is there any accesible debian machine out there I
> can have access to  
> take a look?
> 
> Javi.
> 
> On 23/11/2006, at 11:50, Döring, Markus wrote:
> 
> > Hey,
> > The debian problem is because lxml doesnt install
> out of the box on  
> > debian:
> >
> > http://trac.pywrapper.org/pywrapper/wiki/ 
> > InstallerTroubleshooting#Debian
> >
> > Not sure if we can fix that easily.
> > We might need to do a manual install into the
> system using apt-get...
> >
> > Markus
> 
> 
>
-------------------------------------------------------------------------
> 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
> _______________________________________________
> PyWrapper-devel mailing list
> PyW...@li...
>
https://lists.sourceforge.net/lists/listinfo/pywrapper-devel
> 



 
____________________________________________________________________________________
Want to start your own business?
Learn how on Yahoo! Small Business.
http://smallbusiness.yahoo.com/r-index

Re: [PyWrapper-devel] debian

[Javier de la T. <ja...@gm...>]

Hi Markus,

I dont have a Debian system to try this, but in the message on this =20
wiki page it looks more like if gcc is not available, although I know =20=

this is very unlikely to be true.

I checked on the lxml and they say:

----------
If you have easy_install, you can run the following as super-user:

easy_install lxml

This has been reported to work on Linux, MacOS-X 10.4 and Windows, as =20=

long as libxml2 and libxslt are properly installed. To compile and =20
install lxml without easy_install, please read how to build lxml from =20=

source
------------

Could the problem with debian be that libxml2 and libxslt are not =20
properly installed? In mac os x they are availble by default, and =20
maybe in gento too, but debian is more elitist  choosing what to have =20=

installed no?

Pyrex is only needed If you want to build lxml from SVN.

Is there any accesible debian machine out there I can have access to =20
take a look?

Javi.

On 23/11/2006, at 11:50, D=F6ring, Markus wrote:

> Hey,
> The debian problem is because lxml doesnt install out of the box on =20=

> debian:
>
> http://trac.pywrapper.org/pywrapper/wiki/=20
> InstallerTroubleshooting#Debian
>
> Not sure if we can fix that easily.
> We might need to do a manual install into the system using apt-get...
>
> Markus

Re: [PyWrapper-devel] debian

[Javier de la T. <ja...@gm...>]

Hi Markus,

I suppose is the best to go straight to version 3... it is a pity, I =20
felt comfortable with 2.

In any case. I have added already versions to the trac system. I =20
havent include 3.1 and 3.2 to not confuse people. Lets wait to when =20
we have those releases to create the version in trac.

By the way that I talked the other day with the guy from the hosting =20
service and he told me that he will be soon updating all trac =20
systems. And this is great because the new version does not have =20
anymore the refresh problem with SVN and you can administer almost =20
everything from the web interface, no more command lines for these =20
kind of things.

Javi.

On 23/11/2006, at 13:19, D=F6ring, Markus wrote:

> Well,
> As noone contradicted to my re-versioning proposal I will update =20
> the trunk to become the number 3.0 release. So anytime you issue a =20
> ticket, try to use the version that is shown in http://=20
> search.biocase.org/tapir/about
>
> And we should start stating the exact build(=3Dsvn revision) that we =20=

> were using. This needs to be done in the ticket body as we cant add =20=

> new builds every time to trac.
>
> JAVI, can you please add new versions to the TRAC list of versions?
> I think we need 2.1 and 3.0 (should we can already include 3.1, 3.2?)
>
> Markus
>
>
>> -----Original Message-----
>> From: Skofic A. Milko (Bioversity - HQ) [mailto:m.s...@cg...]
>> Sent: Donnerstag, 23. November 2006 12:38
>> To: D=F6ring, Markus; Javier privat
>> Cc: Rajesh Sood
>> Subject: Re: debian
>>
>>
>> OK, thanks. I will see with the debian user to have him install it
>> manually, I guess there is enough information on http://
>> trac.pywrapper.org/pywrapper/wiki/InstallationGuide to do so.
>>
>> So, the only thing left now is accessing the IRRI registry.
>>
>> Regarding future enhancements, should I open tickets? I think
>> this is
>> the best method for you, if so, for which version should I post them?
>>
>> Thanks!
>>
>> Milko
>>
>> On Nov 23, 2006, at 11:50 , D=F6ring, Markus wrote:
>>
>>> Hey,
>>> The debian problem is because lxml doesnt install out of the box on
>>> debian:
>>>
>>> http://trac.pywrapper.org/pywrapper/wiki/
>>> InstallerTroubleshooting#Debian
>>>
>>> Not sure if we can fix that easily.
>>> We might need to do a manual install into the system using
>> apt-get...
>>>
>>> Markus
>>
>>

[PyWrapper-devel] Fwd: debian

[Javier de la T. <ja...@gm...>]

Begin forwarded message:

> From: "D=F6ring, Markus" <m.d...@BG...>
> Date: 23 de noviembre de 2006 11:50:42 GMT+01:00
> To: "Milko A. Skofic (IPGRI)" <m.s...@cg...>
> Cc: "Javier privat" <ja...@gm...>
> Subject: debian
>
> Hey,
> The debian problem is because lxml doesnt install out of the box on =20=

> debian:
>
> http://trac.pywrapper.org/pywrapper/wiki/=20
> InstallerTroubleshooting#Debian
>
> Not sure if we can fix that easily.
> We might need to do a manual install into the system using apt-get...
>
> Markus

Re: [PyWrapper-devel] Last tickets

[Javier de la T. <ja...@gm...>]

Hi Milko,

Ok, is weekend again and I will have time to take a look into  
PyWrapper... I am trying to catch up with my emails... here goes some  
answers.


On 22/11/2006, at 15:16, Skofic A. Milko (Bioversity - HQ) wrote:

> Hi, how are things going? We are close to the release, there are  
> only two things that need to be solved: installing on debian LINUX  
> and the ability to access the IRRI MOBY registry, once these are  
> closed we can release it. When do you thing this can be done?

Going to take a look when I get home. Once I find the way to trace  
the error shouldnt be difficult to solve if it is on the pywrapper side.

>
> I need to update a document concerning IP status of technology in  
> the GCP. What is the licence for TAPIR? Do you have a link to a  
> page that describes under which licence TAPIR can be distributed?

You can find the License in the LICENSE.TXT file that is distributed  
with the software. It says PyWrapper is released under the Mozilla  
Public License 1.1
Wouldnt be bad to have a wiki page with the same text... It just  
points to the licnse document http://www.mozilla.org/MPL/

It is still written there "The Original Code is BioCASE Provider  
Software.". I think we have to change this to PyWrapper. And we have  
to add the list of contributors!!

>
> I have a question: in the static models part of the data source  
> configuration file one lists all the concepts that must not be  
> repeated, but, since the majority of concepts is not repeated, why  
> didn't you do the opposite, listing those that can be repeated?

I think this is because this is comming from the BioCASE days with  
the dontRepeat command. The idea is that the software more or less  
handles fine when to repeat and do not repeat without having to set  
anything, it was just only sometimes that you had to actually set the  
dontRepeat here statement. So it was just like a tunning of the  
pywrapper algorithm when the software did not have enough information  
to handle by himself. Therefore there was only little places where  
you had to specify dont repeat.

Javier. 

[PyWrapper-devel] FW: debian

[<m.d...@BG...>]

-----Original Message-----
From: D=F6ring, Markus=20
Sent: Donnerstag, 23. November 2006 15:12
To: 'Rouard, Mathieu (INIBAP)'
Cc: Skofic, Milko (Bioversity)
Subject: RE: debian


Mathieu,
If you use the installer.sh script it compiles its own local python =
interpreter in tools/python. This interpreter only uses its own =
libraries.

If you want ot use the "system wide" libraries you might need to use the =
system wide interpreter in /usr/local or wherever it is. In that case =
you should NOT use the pywrapper installer.sh scipt but install the 4 =
required libraries manually. It should be on the wiki briefly: =
http://trac.pywrapper.org/pywrapper/wiki/InstallationGuide#BInstallIntoSy=
stem

If you need any more advice please let me know. Out network is being =
shut own in a few minutes so I might no be avilable til Monday!

Markus



> -----Original Message-----
> From: Rouard, Mathieu (INIBAP) [mailto:M.R...@CG...]
> Sent: Donnerstag, 23. November 2006 14:51
> To: D=F6ring, Markus
> Cc: Skofic, Milko (Bioversity)
> Subject: RE: debian
>=20
>=20
> Hi Marcus,
> =20
> Thanks for your answer. As I explained to Milko, apt-get
> install python2.4-lxml works fine on my debian server.=20
> However, when I try starting the tapir webserver, lxml is not=20
> detected and some errors occur. I think the script looks for=20
> these libraries locally whereas it is globally installed.
> =20
> So I copied the files as below
>=20
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1=20
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml=20
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml.egg-info=20
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml.egg-info/dependency_links.txt
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml.egg-info/native_libs.txt
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml.egg-info/not-zip-safe
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml.egg-info/PKG-INFO
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml.egg-info/SOURCES.txt
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml.egg-info/top_level.txt
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml/_elementpath.py
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml/_elementpath.pyc
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml/etree.so=20
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml/__init__.py
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml/__init__.pyc
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml/objectify.so
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml/sax.py=20
> /home/html/www_tapir/pywrapper/tools/python/lib/python2.4/site
> -packages/lxml1.1.1/lxml/sax.pyc
>=20
> I still have the same problem. Could you tell me which path
> the script expects?=20
>=20
> Warning! lxml is not yet installed (XSLT library). Some
> operations incl all BioMoby services won't work Traceback=20
> (most recent call last):
>   File "../webapp/start_server.py", line 60, in ?
>     import cherrypy, biocase.configuration, root, pywrapper
>   File "/home/html/www_tapir/pywrapper/webapp/root.py", line 17, in ?
>     import page, biocase.configuration
>   File "/home/html/www_tapir/pywrapper/webapp/page.py", line 4, in ?
>     import biocase.pywrapper
>   File=20
> "/home/html/www_tapir/pywrapper/lib/biocase/pywrapper/__init__
> .py", line 48, in ?
>     from biocase.pywrapper.pywrappercgi import PyWrapper,=20
> UnknownProtocolError
>   File=20
> "/home/html/www_tapir/pywrapper/lib/biocase/pywrapper/pywrappe
> rcgi.py", line 23, in ?
>     from biocase.pywrapper.protocols.BioMOBY.protocol import=20
> BioMOBYProtocol
>   File=20
> "/home/html/www_tapir/pywrapper/lib/biocase/pywrapper/protocol
> s/BioMOBY/protocol.py", line 17, in ?
>     from biocase.pywrapper.protocols.BioMOBY.request      =20
> import RequestClass
>   File=20
> "/home/html/www_tapir/pywrapper/lib/biocase/pywrapper/protocol
> s/BioMOBY/request.py", line 11, in ?
>     from biocase.tools.xmllibtools                     =20
> import transformXML
>   File=20
> "/home/html/www_tapir/pywrapper/lib/biocase/tools/xmllibtools.
> py", line 13, in ?
>     from lxml import etree
> ImportError: No module named lxml
> Error in atexit._run_exitfuncs:
> Traceback (most recent call last):
>   File=20
> "/home/html/www_tapir/pywrapper/tools/python/lib/python2.4/ate
> xit.py", line 24, in _run_exitfuncs
>     func(*targs, **kargs)
>   File=20
> "/home/html/www_tapir/pywrapper/tools/python/lib/python2.4/log
> ging/__init__.py", line 1333, in shutdown
>     h.close()
>   File=20
> "/home/html/www_tapir/pywrapper/tools/python/lib/python2.4/log
> ging/__init__.py", line 674, in close
>     del _handlers[self]
> KeyError: <logging.StreamHandler instance at 0xb78e64cc>
> Error in sys.exitfunc:
> Traceback (most recent call last):
>   File=20
> "/home/html/www_tapir/pywrapper/tools/python/lib/python2.4/ate
> xit.py", line 24, in _run_exitfuncs
>     func(*targs, **kargs)
>   File=20
> "/home/html/www_tapir/pywrapper/tools/python/lib/python2.4/log
> ging/__init__.py", line 1333, in shutdown
>     h.close()
>   File=20
> "/home/html/www_tapir/pywrapper/tools/python/lib/python2.4/log
> ging/__init__.py", line 674, in close
>     del _handlers[self]
> KeyError: <logging.StreamHandler instance at 0xb78e64cc>
> =20
> Thanks
> Mathieu
> =20
> Begin forwarded message:
>=20
>=20
> 	From: "D=F6ring, Markus" <m.d...@BG...>
> 	Date: November 23, 2006 11:50:42  GMT+01:00
> 	To: "Milko A. Skofic (IPGRI)" <m.s...@cg...>
> 	Cc: "Javier privat" <ja...@gm...>
> 	Subject: debian
>=20
> 	Hey,
> 	The debian problem is because lxml doesnt install out
> of the box on debian:
>=20
	=
http://trac.pywrapper.org/pywrapper/wiki/InstallerTroubleshooting#Debian

	Not sure if we can fix that easily.
	We might need to do a manual install into the system using apt-get...

	Markus

[PyWrapper-devel] Version 3.0

[<m.d...@BG...>]

Hi there,

I have updated (most of?) the wiki and the pywrapper source files to =
refer to version 3.0 as the new release target (hopefully to be released =
tomorrow or this weekend).=20

This way we hope to avoid any confusion with the previous biocase =
provider software (BPS) which stopped being developed in version 2.4 and =
also contained a pywrapper as its core(which had its own versioning; at =
that time the last pywrapper version was 1.7).

I would also like to ask everyone who is issuing tickets for pywrapper, =
see http://trac.pywrapper.org/pywrapper/wiki/Contribute, to state the =
version he was using and to give the full build number of pywrapper in =
the body text of the ticket. You can always find the build of your =
pywrapper installation in the about pane of the webapps. For the bgbm =
installation this is http://search.biocase.org/tapir/about


And many thanks to all of you that start getting into pywrapper.=20
It seems we are gaining momentum!

Markus
--
 Markus D=F6ring
 Botanic Garden and Botanical Museum Berlin Dahlem,
 Dept. of Biodiversity Informatics
 K=F6nigin-Luise-Str. 6-8, D-14191 Berlin
 Phone: +49 30 83850-284
 Email: m.d...@bg...
 URL: http://www.bgbm.org/BioDivInf/

Re: [PyWrapper-devel] Path to pywrapper.cgi is not correct

[Marc B. <mbr...@et...>]

Just found out that I forgot to add it in global configuration! Apologees
for this silly question.
 
Marc


  _____  

From: Marc Brugman [mailto:mbr...@et...] 
Sent: woensdag 22 november 2006 14:41
To: 'PyWrapper Developers mailing list'
Subject: Path to pywrapper.cgi is not correct


 Hi,
 
When I want to use the query tool of BPS 2.4.2 the path to pywrapper.cgi is
resolved to "http://localhost/biocase/pywrapper.cgi?". This is not correct.
It should be http://localhost:8080/biocase/pywrapper.cgi?
<http://localhost:8080/biocase/pywrapper.cgi> . (The Tapir version does this
correctly).  It is probably a config setting, but where to find it?
 
Thanks for your help!
 
Marc

____________________________________________________________

ETI BioInformatics

University of Amsterdam

Mauritskade 61

1092 AD Amsterdam

The Netherlands

Phone: +31 20 5257239

Fax: +31 20 5257238

Web:  <http://www.eti.uva.nl/> http://www.eti.uva.nl

____________________________________________________________

ETI facilitates the Netherlands GBIF Node:  <http://www.nlbif.nl/>
<http://www.nlbif.nleti/> http://www.nlbif.nl

ETI facilitates the Species2000 Annual Checklist ( <http://www.sp2000.org/>
http://www.sp2000.org/)

Bezoek eens  <http://www.soortenbank.nl/> http://www.soortenbank.nl

____________________________________________________________



 

[PyWrapper-devel] Path to pywrapper.cgi is not correct

[Marc B. <mbr...@et...>]

 Hi,
 
When I want to use the query tool of BPS 2.4.2 the path to pywrapper.cgi is
resolved to "http://localhost/biocase/pywrapper.cgi?". This is not correct.
It should be http://localhost:8080/biocase/pywrapper.cgi?
<http://localhost:8080/biocase/pywrapper.cgi> . (The Tapir version does this
correctly).  It is probably a config setting, but where to find it?
 
Thanks for your help!
 
Marc

____________________________________________________________

ETI BioInformatics

University of Amsterdam

Mauritskade 61

1092 AD Amsterdam

The Netherlands

Phone: +31 20 5257239

Fax: +31 20 5257238

Web:  <http://www.eti.uva.nl/> http://www.eti.uva.nl

____________________________________________________________

ETI facilitates the Netherlands GBIF Node:  <http://www.nlbif.nl/>
<http://www.nlbif.nleti/> http://www.nlbif.nl

ETI facilitates the Species2000 Annual Checklist ( <http://www.sp2000.org/>
http://www.sp2000.org/)

Bezoek eens  <http://www.soortenbank.nl/> http://www.soortenbank.nl

____________________________________________________________

 

Re: [PyWrapper-devel] [tdwg-tapir] Problem with Tapir and Outputmodel.

[<rp...@in...>]

Hi again,

I have updated my pywrapper to the last version in SVN.

I have tried the last instructions of Markus and the pywrapper works if
I use sequences instead of choices.

In the file Log-Test1.xml the response is OK. In the file Log-Test2.xml,
the pywrapper delete the nodes SpeciesRecordAuthors and
SpeciesRecordCollaborators.

I attach the file of two outputmodel (test1xml and test2.xml) and log
(Log-Test1.xml and Log-Test2.xml)

Thanks,

Ram=F3n.


D=F6ring, Markus escribi=F3:
> Ram=F3n,
> I am not sure what the reason is in your case.
> I would imagine it has to do with the choices. If more than one branch =
of a choce has mappings, Pywrapper tries to use the first one. In your ca=
se that's the "Text" one, for example SpeciesRecordCollaborators/Text. Co=
uld you try the model using sequences instead of chocies for a test? And =
can you send us the debug.log file too for your requests?
>
> By the way, in your models mapping section you have mapped the model st=
ructure to concepts defined by the same schema. We call this a canonical =
model and you can make your life easier by using the automapping=3D"true"=
 flag in the mapping section. Just like the canonical abcd model: http://=
rs.tdwg.org/tapir/cs/abcd2.06/model/abcd206.xml
>
> It maps every node of the schema to itself, using the namespace of the =
schema to create qualified concept ids.
>
> Markus
>
>
>  =20
>> -----Original Message-----
>> From: tdw...@li...=20
>> [mailto:tdw...@li...] On Behalf Of=20
>> Renato De Giovanni
>> Sent: Freitag, 17. November 2006 19:14
>> To: tdw...@li...
>> Subject: Re: [tdwg-tapir] Problem with Tapir and Outputmodel.
>>
>>
>> Hi Ram=F3n,
>>
>> Two quick comments about the output model:
>>
>> * At a first glance I thought there were still two mandatory=20
>> elements:
>>
>> /Species/Record/RecordInformation/TargetAudience/Audience
>> /Species/Record/RecordInformation/TargetAudience/Text
>>
>> But now I see that they are inside a choice of an optional element,=20
>> so in theory this should not be a problem.
>>
>> * It's curious that you mapped complex elements like Species, Record,=20
>> DataSource and others. Some are just grouping elements, others can be=20
>> seen as related to class instances. But as far as I know PyWrapper=20
>> doesn't handle this kind of mapping.
>>
>> Not sure if these things can help finding out the reasons for not=20
>> getting all expected results...=20
>>
>> Regards,
>> --
>> Renato
>>
>> On 17 Nov 2006 at 18:26, Ram=F3n P=E9rez wrote:
>>
>>    =20
>>> Hi Markus and Renato,
>>>
>>> Thanks very much for yours response. I have tried your=20
>>>      =20
>> suggestions but=20
>>    =20
>>> i
>>> don't resolve the problem.
>>>
>>> I have simplified the outputmodel and I have changed all=20
>>>      =20
>> attribute to
>>    =20
>>> optional.
>>>
>>> With the outputmodelpcore2.xml the response is:
>>>
>>> <RecordInformation>=20
>>> <GlobalUniqueIdentifier>GUIDS-1</GlobalUniqueIdentifier>
>>> <SpeciesRecordID>1--1.1</SpeciesRecordID>
>>> <Language>latin1</Language>
>>> <SpeciesRecordAuthors>
>>> <SpeciesRecordAutor>
>>> <Text>Alexander Rojas A.</Text>
>>> </SpeciesRecordAutor>
>>> <SpeciesRecordAutor>
>>> <Text>Jose Gonzalez</Text>
>>> </SpeciesRecordAutor>
>>> </SpeciesRecordAuthors>
>>> <SpeciesRecordCollaborators>
>>> <SpeciesRecordCollaborator>
>>> <FirstName>Francisco</FirstName>
>>> <LastName>Morales</LastName>
>>> </SpeciesRecordCollaborator>
>>> </SpeciesRecordCollaborators>=20
>>> <PublicationDate>2006-10-09</PublicationDate>
>>> <DateLastModified>2006-10-09</DateLastModified>
>>> <TargetAudience>
>>> <Text>1</Text>
>>> </TargetAudience>
>>> <Version>2.1</Version>
>>> </RecordInformation>
>>>
>>> With the outputmodelpcore3.xml, the response is:
>>>
>>>
>>> <RecordInformation>=20
>>> <GlobalUniqueIdentifier>GUIDS-1</GlobalUniqueIdentifier>
>>> <SpeciesRecordID>1--1.1</SpeciesRecordID>
>>> <Language>latin1</Language>=20
>>> <PublicationDate>2006-10-09</PublicationDate>
>>> <DateLastModified>2006-10-09</DateLastModified>
>>> <TargetAudience>
>>> <Text>1</Text>
>>> </TargetAudience>
>>> <Version>2.1</Version>
>>> </RecordInformation>
>>>
>>> The pywrapper deletes the SpeciesRecordAuthors and
>>> SpeciesRecordCollaborators nodes. This is the important=20
>>>      =20
>> information in=20
>>    =20
>>> the diagnostic:
>>>
>>> <diagnostic time=3D"2006-11-17T18:24:45.32" level=3D"warn">
>>> The element
>>>
>>>      =20
>> /Species/Record/RecordInformation/SpeciesRecordCollaborators/S
>> peciesRecordCollaborator/Record=20
>>    =20
>>> was limited 3 times. Some node instances have been lost.
>>> </diagnostic>
>>> -
>>> <diagnostic time=3D"2006-11-17T18:24:45.32" level=3D"warn">
>>> The element=20
>>>
>>>      =20
>> /Species/Record/RecordInformation/SpeciesRecordAuthors/Species
>> RecordAutor/Text=20
>>    =20
>>> was limited 5 times. Some node instances have been lost.
>>> </diagnostic>
>>>
>>> I attach a picture with the query result in the database. It doesn't
>>> exist null values.
>>>
>>> Thanks for all,
>>>
>>> Ram=F3n.
>>>      =20
>> _______________________________________________
>> tdwg-tapir mailing list
>> tdw...@li...=20
>> http://lists.tdwg.org/mailman/listinfo/tdwg-> tapir
>>
>>    =20
> _______________________________________________
> tdwg-tapir mailing list
> tdw...@li...
> http://lists.tdwg.org/mailman/listinfo/tdwg-tapir
>  =20

Re: [PyWrapper-devel] Pywrapper versions

[Marc B. <mbr...@et...>]

Hi Markus,

That would be a good idea. I am suffering from PyWrapper version =
dizzyness
myself.  Keep the BPS into the 2.x.x and continue the Tapir version with
3.x.x has my support. It is after all a major release.

Marc
____________________________________________________________
ETI BioInformatics
University of Amsterdam
Mauritskade 61
1092 AD Amsterdam
The Netherlands
Phone: +31 20 5257239
Fax: +31 20 5257238
Web: http://www.eti.uva.nl
____________________________________________________________
ETI facilitates the Netherlands GBIF Node: http://www.nlbif.nl
ETI facilitates the Species2000 Annual Checklist =
(http://www.sp2000.org/)
Bezoek eens http://www.soortenbank.nl
____________________________________________________________
=20

> -----Original Message-----
> From: pyw...@li...=20
> [mailto:pyw...@li...] On=20
> Behalf Of "D=F6ring, Markus"
> Sent: woensdag 22 november 2006 11:14
> To: Javier privat; Milko A. Skofic (IPGRI)
> Cc: pyw...@li...
> Subject: [PyWrapper-devel] Pywrapper versions
>=20
> Hey,
>=20
> I received some complaints that the pywrapper versions out=20
> there are confusing. So I looked at the situation again and=20
> they are right.=20
>=20
> The original biocase provider software has version 2.4.2=20
> right now. The problem here is that the included core=20
> wrapper, the pywrapper, was considered to be of version 1.7.x=20
> Thats why we started with the tapir pywrapper in version 2.0.=20
>=20
> But noone sees this wrapper version of biocase, you download=20
> BPS2.4.2. Thats why I released the last BPS2 version with=20
> only a single version for all components. Its only 2.4.2 now,=20
> including the pywrapper.=20
>=20
> Well, and the TAPIR pywrapper is in version 2.1 right now.=20
> Pretty confusing, isn't it?
> So what do we do about it? Raise the not-yet-released tapir=20
> pywrapper to version 3.0?
>=20
> Any feedback more than welcome.
>=20
> Markus
>=20
> --------------------------------------------------------------
> -----------
> Take Surveys. Earn Cash. Influence the Future of IT Join=20
> SourceForge.net's Techsay panel and you'll get the chance to=20
> share your opinions on IT & business topics through brief=20
> surveys - and earn cash=20
> http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge
&CID=3DDEVDEV
> _______________________________________________
> PyWrapper-devel mailing list
> PyW...@li...
> https://lists.sourceforge.net/lists/listinfo/pywrapper-devel
>=20

[PyWrapper-devel] Pywrapper versions

[<m.d...@BG...>]

Hey,

I received some complaints that the pywrapper versions out there are =
confusing. So I looked at the situation again and they are right.=20

The original biocase provider software has version 2.4.2 right now. The =
problem here is that the included core wrapper, the pywrapper, was =
considered to be of version 1.7.x Thats why we started with the tapir =
pywrapper in version 2.0.=20

But noone sees this wrapper version of biocase, you download BPS2.4.2. =
Thats why I released the last BPS2 version with only a single version =
for all components. Its only 2.4.2 now, including the pywrapper.=20

Well, and the TAPIR pywrapper is in version 2.1 right now. Pretty =
confusing, isn't it?
So what do we do about it? Raise the not-yet-released tapir pywrapper to =
version 3.0?

Any feedback more than welcome.

Markus

[PyWrapper-devel] Namespaces

[Skofic A. M. (B. - HQ) <m.s...@cg...>]

I was setting up a scj=EChema for use in TAPIR, I set some characters =20=

in the namespace to uppercase. TAPIR was unable to parse the alias =20
file, although I had triple-checked that the namespace names had the =20
same case. Now, is it an XML requirement to put namespaces in =20
lowercase or do you convert namespaces in lowercase? I had to change =20
all characters in the namespace to lowercase to make it work.

Thanks!

Milko=

[PyWrapper-devel] Altering the Alias of a table name in the datasource structure page crashes PyWrapper

[Marc B. <mbr...@et...>]

When I alter the Alias of a table name in the datasource structure page:
e.g. "//localhost/pywrapper/configtool/datasource/structure" and I try to
save  <http://trac.pywrapper.org/pywrapper/wiki/PyWrapper> PyWrapper (Tapir
beta version 2.1) crashes. The change is saved.

Marc Brugman

____________________________________________________________

ETI BioInformatics

University of Amsterdam

Mauritskade 61

1092 AD Amsterdam

The Netherlands

Phone: +31 20 5257239

Fax: +31 20 5257238

Web:  <http://www.eti.uva.nl/> http://www.eti.uva.nl

____________________________________________________________

ETI facilitates the Netherlands GBIF Node:  <http://www.nlbif.nl/>
<http://www.nlbif.nleti/> http://www.nlbif.nl

ETI facilitates the Species2000 Annual Checklist ( <http://www.sp2000.org/>
http://www.sp2000.org/)

Bezoek eens  <http://www.soortenbank.nl/> http://www.soortenbank.nl

____________________________________________________________

 

Re: [PyWrapper-devel] MOBY registries in PyWrapper

[Skofic A. M. (B. - HQ) <m.s...@cg...>]

I checked the data types, and they are there; I think it has to do  
with something else.

On Nov 21, 2006, at 8:27 , Javier de la Torre wrote:

> Uhmmm... I am going to have to trace this down with a tracer tool.  
> The python library I am using does not give you much information  
> when it crashes connecting to the server.
> But usually this means that there is no connection with this url or  
> it is limited. If there will be an error on biomoby then the  
> library catches the errors fine.
>
> You have to be careful that you have all the necessary data types  
> in the IRRI registry for the template that you are going to use,  
> but I doubt this is the problem because then you would get a  
> warning about it.
>
> I will let you know.
>
> Javier.
>
> On 20/11/2006, at 14:25, Skofic A. Milko (IPGRI) wrote:
>
>> I have tried to register a service in the IRRI registry, these are  
>> the parameters:
>> End-point: http://cropwiki.irri.org/cgi-bin/MOBY-Central.pl
>> Namespace: http://cropwiki.irri.org/MOBY/Central
>>
>> TAPIR returns the following error:
>> Sorry, the service was not registred. The BioMOBY end point and  
>> namespace you have configured is not accessible. Are you sure this  
>> is a MOBY registry?
>>
>> What should I ask the IRRI people to solve the problem?
>>
>> I took the end-point and namespace from Dashboard, so it should be  
>> the correct information.
>>
>> By the way, I registered in their registry the TapirProvider  
>> service type.
>>
>> On Nov 20, 2006, at 10:45 , Javier de la Torre wrote:
>>
>>> Aghh,
>>>
>>> Thats because your old biomoby-settings file is not now up to  
>>> date, the services now in the config file have information about  
>>> which registry was used to register them but in your case because  
>>> your files are not updated, see the empty parethesys close to the  
>>> service name? the configtool does not know from where to doregister.
>>>
>>> If you want send me your biomoby-settings file and I will update  
>>> it accordely to the new version, then it will work.
>>>
>>> Javi.
>>>
>>> On 20/11/2006, at 10:35, Skofic A. Milko (IPGRI) wrote:
>>>
>>>> I am trying now with the changes. I had to update the moby  
>>>> configuration files to get into the user interface.
>>>>
>>>> First of all, I think that having one registry per data source  
>>>> is fine, if you need to register a service in different  
>>>> registries why not create a new data source and point to the  
>>>> other registry from there? The need to be able to select a  
>>>> registry is crucial for us, but once it is selected I don't  
>>>> think it should change.
>>>>
>>>> Now for the tests, I wanted to deregister the current services  
>>>> before changing registry, but it doesn't allow me to do so, it  
>>>> says that:
>>>>
>>>> BioMOBY Settings
>>>>
>>>> The service <b>getTaxaOfBoundingBox</b> could not be  
>>>> deregistered. Is the central MOBY registry accesible?
>>>>
>>>> 1) Services you are already publishing and registry where you  
>>>> registered them:
>>>>
>>>> TestServiceABCD ()Deregister
>>>> getTaxaOfBoundingBox ()Deregister
>>>> Deregister all
>>>> 2) Considering the mappings that you already have done, you  
>>>> could be publishing:
>>>>
>>>> There are no other services available to register
>>>> General settings
>>>> BioMOBY registry end point
>>>>
>>>> BioMOBY registry Namespace
>>>>
>>>> bmttsUrl
>>>>
>>>> ServiceNs
>>>>
>>>> AuthURI
>>>>
>>>> ContactEmail
>>>>
>>>> databaseName
>>>>
>>>>
>>>> Save
>>>>
>>>>
>>>> It looks as if it cannot communicate with the registry end point  
>>>> (I left the Canada one).
>>>>
>>>> During the startup process of TAPIR I got a strange message:
>>>>
>>>> ERROR! from pywrapper: REQ_READ_MODEL_FAILED:::The requested  
>>>> model TCS 1.01 could not be read. Reason: Could not access  
>>>> external model component(s)
>>>> Traceback (most recent call last):
>>>>   File "/Library/WebServer/Services/pywrapper/lib/biocase/ 
>>>> pywrapper/model.py", line 252, in __readFile__
>>>>     self.parser.parse( file(filename) )
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/site-packages/_xmlplus/sax/expatreader.py", line 109,  
>>>> in parse
>>>>     xmlreader.IncrementalParser.parse(self, source)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/site-packages/_xmlplus/sax/xmlreader.py", line 123, in  
>>>> parse
>>>>     self.feed(buffer)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/site-packages/_xmlplus/sax/expatreader.py", line 216,  
>>>> in feed
>>>>     self._parser.Parse(data, isFinal)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/site-packages/_xmlplus/sax/expatreader.py", line 353,  
>>>> in start_element_ns
>>>>     AttributesNSImpl(newattrs, qnames))
>>>>   File "/Library/WebServer/Services/pywrapper/lib/biocase/tools/ 
>>>> saxhandler_extended.py", line 100, in startElementNS
>>>>     self.startSimpleElement(ns, localname.lower(), attributes)
>>>>   File "/Library/WebServer/Services/pywrapper/lib/biocase/ 
>>>> pywrapper/model_handler.py", line 69, in startSimpleElement
>>>>     parser.parse(attrs.get('location'))
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/site-packages/_xmlplus/sax/expatreader.py", line 103,  
>>>> in parse
>>>>     source = saxutils.prepare_input_source(source)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/site-packages/_xmlplus/sax/saxutils.py", line 523, in  
>>>> prepare_input_source
>>>>     f = urllib2.urlopen(source.getSystemId())
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/urllib2.py", line 130, in urlopen
>>>>     return _opener.open(url, data)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/urllib2.py", line 358, in open
>>>>     response = self._open(req, data)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/urllib2.py", line 376, in _open
>>>>     '_open', req)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/urllib2.py", line 337, in _call_chain
>>>>     result = func(*args)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/urllib2.py", line 1021, in http_open
>>>>     return self.do_open(httplib.HTTPConnection, req)
>>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>>> python2.4/urllib2.py", line 996, in do_open
>>>>     raise URLError(err)
>>>> URLError: <urlopen error (60, 'Operation timed out')>
>>>> WARNING! from pywrapper: The model http://rs.tdwg.org/tapir/cs/ 
>>>> tcs1.01/model/tcs101.xml could not be loaded. It might be corrupt.
>>>>
>>>> It looks like the TCS model is corrupt.
>>>>
>>>> Let me know if you need other information.
>>>>
>>>> Ciao!
>>>>
>>>> Milko
>>>>
>>>> On Nov 19, 2006, at 20:11 , Javier de la Torre wrote:
>>>>
>>>>> Hi Milko,
>>>>>
>>>>> Ok. I have implemented some sort of multiple biomoby registry  
>>>>> awareness in pywrapper.
>>>>>
>>>>> Now in the MOBY tab there are two new fields in the General  
>>>>> Settings:
>>>>>
>>>>> "BioMOBY registry end point" and "BioMOBY registry Namespace".
>>>>>
>>>>> This let you choose the registry you want to use. This  
>>>>> information is saved as a general setting for this datasource.  
>>>>> But preparing the work for the future now when you register a  
>>>>> service the information of the end point and the namespace is  
>>>>> stored in the service on the config files, thats allowing the  
>>>>> same service to be register in multiple biomoby registries.
>>>>>
>>>>> When you click on register a service the configtool uses the  
>>>>> registry specify at this moment on the general settings, but  
>>>>> when you click on Deregister it uses the registry that was used  
>>>>> for registration.
>>>>>
>>>>> By default I have left configured the CENTRAL MOBY REGISTRY in  
>>>>> Canada. Milko, if you think more people are going to use the  
>>>>> IRRI one I can change it, give me the end point and ns and I  
>>>>> change it on the default config file so that new installations  
>>>>> will appear with this.
>>>>>
>>>>> There are some limitations still that I hope will not prevent  
>>>>> us from delopying right now:
>>>>>
>>>>> -Althought the config tool stores the info on the registry that  
>>>>> was used for registering a service, the interface will still  
>>>>> not offer you the possibility to register the same service in  
>>>>> multiple registries. If the config tool detects that a certain  
>>>>> service has been already registered in a registry is not  
>>>>> offered as a potential service. In the config files is still  
>>>>> possible to configure the same service in multiple registries,  
>>>>> altought you would have to manually register them using  
>>>>> Dashboard or something like this...
>>>>>
>>>>> -The BMTTS file gives pywrapper a list of service templates,  
>>>>> but it does not tell you for what registry this templates are  
>>>>> designed. Different registries have different data types. So a  
>>>>> service template for the IRRI server does not necessarily have  
>>>>> to work with the Central one in Canada... Therefore is needed  
>>>>> that the BMTTS stores information about to which registry a  
>>>>> certain template was designed. Then the config tool should read  
>>>>> this info and react according to this...
>>>>>
>>>>> I have only tried it with the central moby registry in Canada,  
>>>>> but I suppose it should work with others. I havent tried  
>>>>> because I dont have the url of the IRRI registry and the  
>>>>> getCoordinatesOfTaxon service template will probably not work  
>>>>> with the data types there... Could you try Milko with yours?
>>>>>
>>>>>
>>>>> It is commited now.
>>>>>
>>>>> Javi.
>>>>
>>>
>>
>

Re: [PyWrapper-devel] MOBY registries in PyWrapper

[Javier de la T. <ja...@gm...>]

Uhmmm... I am going to have to trace this down with a tracer tool.  
The python library I am using does not give you much information when  
it crashes connecting to the server.
But usually this means that there is no connection with this url or  
it is limited. If there will be an error on biomoby then the library  
catches the errors fine.

You have to be careful that you have all the necessary data types in  
the IRRI registry for the template that you are going to use, but I  
doubt this is the problem because then you would get a warning about it.

I will let you know.

Javier.

On 20/11/2006, at 14:25, Skofic A. Milko (IPGRI) wrote:

> I have tried to register a service in the IRRI registry, these are  
> the parameters:
> End-point: http://cropwiki.irri.org/cgi-bin/MOBY-Central.pl
> Namespace: http://cropwiki.irri.org/MOBY/Central
>
> TAPIR returns the following error:
> Sorry, the service was not registred. The BioMOBY end point and  
> namespace you have configured is not accessible. Are you sure this  
> is a MOBY registry?
>
> What should I ask the IRRI people to solve the problem?
>
> I took the end-point and namespace from Dashboard, so it should be  
> the correct information.
>
> By the way, I registered in their registry the TapirProvider  
> service type.
>
> On Nov 20, 2006, at 10:45 , Javier de la Torre wrote:
>
>> Aghh,
>>
>> Thats because your old biomoby-settings file is not now up to  
>> date, the services now in the config file have information about  
>> which registry was used to register them but in your case because  
>> your files are not updated, see the empty parethesys close to the  
>> service name? the configtool does not know from where to doregister.
>>
>> If you want send me your biomoby-settings file and I will update  
>> it accordely to the new version, then it will work.
>>
>> Javi.
>>
>> On 20/11/2006, at 10:35, Skofic A. Milko (IPGRI) wrote:
>>
>>> I am trying now with the changes. I had to update the moby  
>>> configuration files to get into the user interface.
>>>
>>> First of all, I think that having one registry per data source is  
>>> fine, if you need to register a service in different registries  
>>> why not create a new data source and point to the other registry  
>>> from there? The need to be able to select a registry is crucial  
>>> for us, but once it is selected I don't think it should change.
>>>
>>> Now for the tests, I wanted to deregister the current services  
>>> before changing registry, but it doesn't allow me to do so, it  
>>> says that:
>>>
>>> BioMOBY Settings
>>>
>>> The service <b>getTaxaOfBoundingBox</b> could not be  
>>> deregistered. Is the central MOBY registry accesible?
>>>
>>> 1) Services you are already publishing and registry where you  
>>> registered them:
>>>
>>> TestServiceABCD ()Deregister
>>> getTaxaOfBoundingBox ()Deregister
>>> Deregister all
>>> 2) Considering the mappings that you already have done, you could  
>>> be publishing:
>>>
>>> There are no other services available to register
>>> General settings
>>> BioMOBY registry end point
>>>
>>> BioMOBY registry Namespace
>>>
>>> bmttsUrl
>>>
>>> ServiceNs
>>>
>>> AuthURI
>>>
>>> ContactEmail
>>>
>>> databaseName
>>>
>>>
>>> Save
>>>
>>>
>>> It looks as if it cannot communicate with the registry end point  
>>> (I left the Canada one).
>>>
>>> During the startup process of TAPIR I got a strange message:
>>>
>>> ERROR! from pywrapper: REQ_READ_MODEL_FAILED:::The requested  
>>> model TCS 1.01 could not be read. Reason: Could not access  
>>> external model component(s)
>>> Traceback (most recent call last):
>>>   File "/Library/WebServer/Services/pywrapper/lib/biocase/ 
>>> pywrapper/model.py", line 252, in __readFile__
>>>     self.parser.parse( file(filename) )
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/site-packages/_xmlplus/sax/expatreader.py", line 109,  
>>> in parse
>>>     xmlreader.IncrementalParser.parse(self, source)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/site-packages/_xmlplus/sax/xmlreader.py", line 123, in  
>>> parse
>>>     self.feed(buffer)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/site-packages/_xmlplus/sax/expatreader.py", line 216,  
>>> in feed
>>>     self._parser.Parse(data, isFinal)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/site-packages/_xmlplus/sax/expatreader.py", line 353,  
>>> in start_element_ns
>>>     AttributesNSImpl(newattrs, qnames))
>>>   File "/Library/WebServer/Services/pywrapper/lib/biocase/tools/ 
>>> saxhandler_extended.py", line 100, in startElementNS
>>>     self.startSimpleElement(ns, localname.lower(), attributes)
>>>   File "/Library/WebServer/Services/pywrapper/lib/biocase/ 
>>> pywrapper/model_handler.py", line 69, in startSimpleElement
>>>     parser.parse(attrs.get('location'))
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/site-packages/_xmlplus/sax/expatreader.py", line 103,  
>>> in parse
>>>     source = saxutils.prepare_input_source(source)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/site-packages/_xmlplus/sax/saxutils.py", line 523, in  
>>> prepare_input_source
>>>     f = urllib2.urlopen(source.getSystemId())
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/urllib2.py", line 130, in urlopen
>>>     return _opener.open(url, data)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/urllib2.py", line 358, in open
>>>     response = self._open(req, data)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/urllib2.py", line 376, in _open
>>>     '_open', req)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/urllib2.py", line 337, in _call_chain
>>>     result = func(*args)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/urllib2.py", line 1021, in http_open
>>>     return self.do_open(httplib.HTTPConnection, req)
>>>   File "/Library/WebServer/Services/pywrapper/tools/python/lib/ 
>>> python2.4/urllib2.py", line 996, in do_open
>>>     raise URLError(err)
>>> URLError: <urlopen error (60, 'Operation timed out')>
>>> WARNING! from pywrapper: The model http://rs.tdwg.org/tapir/cs/ 
>>> tcs1.01/model/tcs101.xml could not be loaded. It might be corrupt.
>>>
>>> It looks like the TCS model is corrupt.
>>>
>>> Let me know if you need other information.
>>>
>>> Ciao!
>>>
>>> Milko
>>>
>>> On Nov 19, 2006, at 20:11 , Javier de la Torre wrote:
>>>
>>>> Hi Milko,
>>>>
>>>> Ok. I have implemented some sort of multiple biomoby registry  
>>>> awareness in pywrapper.
>>>>
>>>> Now in the MOBY tab there are two new fields in the General  
>>>> Settings:
>>>>
>>>> "BioMOBY registry end point" and "BioMOBY registry Namespace".
>>>>
>>>> This let you choose the registry you want to use. This  
>>>> information is saved as a general setting for this datasource.  
>>>> But preparing the work for the future now when you register a  
>>>> service the information of the end point and the namespace is  
>>>> stored in the service on the config files, thats allowing the  
>>>> same service to be register in multiple biomoby registries.
>>>>
>>>> When you click on register a service the configtool uses the  
>>>> registry specify at this moment on the general settings, but  
>>>> when you click on Deregister it uses the registry that was used  
>>>> for registration.
>>>>
>>>> By default I have left configured the CENTRAL MOBY REGISTRY in  
>>>> Canada. Milko, if you think more people are going to use the  
>>>> IRRI one I can change it, give me the end point and ns and I  
>>>> change it on the default config file so that new installations  
>>>> will appear with this.
>>>>
>>>> There are some limitations still that I hope will not prevent us  
>>>> from delopying right now:
>>>>
>>>> -Althought the config tool stores the info on the registry that  
>>>> was used for registering a service, the interface will still not  
>>>> offer you the possibility to register the same service in  
>>>> multiple registries. If the config tool detects that a certain  
>>>> service has been already registered in a registry is not offered  
>>>> as a potential service. In the config files is still possible to  
>>>> configure the same service in multiple registries, altought you  
>>>> would have to manually register them using Dashboard or  
>>>> something like this...
>>>>
>>>> -The BMTTS file gives pywrapper a list of service templates, but  
>>>> it does not tell you for what registry this templates are  
>>>> designed. Different registries have different data types. So a  
>>>> service template for the IRRI server does not necessarily have  
>>>> to work with the Central one in Canada... Therefore is needed  
>>>> that the BMTTS stores information about to which registry a  
>>>> certain template was designed. Then the config tool should read  
>>>> this info and react according to this...
>>>>
>>>> I have only tried it with the central moby registry in Canada,  
>>>> but I suppose it should work with others. I havent tried because  
>>>> I dont have the url of the IRRI registry and the  
>>>> getCoordinatesOfTaxon service template will probably not work  
>>>> with the data types there... Could you try Milko with yours?
>>>>
>>>>
>>>> It is commited now.
>>>>
>>>> Javi.
>>>
>>
>

Re: [PyWrapper-devel] First impressions...

[Javier de la T. <ja...@gm...>]

We need this wiki page with instructions for different Os!
Could you copy this info into a wiki page?

Ciao

On 20/11/2006, at 17:06, D=F6ring, Markus wrote:

> Peter, Javi,
> Ive installed pywrapper with fastcgi/apache2 on debian. Was pretty =20
> easy when you knew where debian keeps the configuration files (its =20
> split in severals). Fastcgi comes with debian too, so you can use =20
> apt to install it.
>
> Here it is:
> http://search.biocase.org/tapir/index
>
> Markus
>
>
>> -----Original Message-----
>> From: pyw...@li...
>> [mailto:pyw...@li...] On
>> Behalf Of Javier privat
>> Sent: Sonntag, 19. November 2006 20:35
>> To: PyWrapper Developers mailing list
>> Subject: Re: [PyWrapper-devel] First impressions...
>>
>>
>> Hi Peter,
>>
>>
>>> I've installed from an anonymous checkout using both the standard
>>> install script and to the system using apt etc - both seem to work
>>> fine.  Unfortunately the instructions for using FastCGI do
>> not work on
>>> my system - I think this is because Kubuntu packages Apache in an
>>> awkward way.  I'll look at providing documentation at some point
>>> but for
>>> now I'll just use as a standalone server.  Out of interest
>> what is the
>>> distro of choice in the PyWrapper community?
>>>
>> I suppose dont really have one... but we should do a wiki
>> page with a
>> list of platforms where we have tested the software and
>> issues with it.
>>
>> I have tried with:
>> -Mac os X 10.4
>> -Red Hat 9
>> -Windows XP Pro
>> -Ubuntu
>>
>> But always using the cherrypy stand alone server, we need much more
>> testing and documentation on setting up the fastcgi way and the
>> redirection... And we need to start writing this on the wiki!
>>
>>> The next problem I have is that there is a considerable
>> delay between
>>> running start_server.py and the server actually coming online (I've
>>> timed it to 3mins 10secs).  Whilst in itself this isn't such a big
>>> deal
>>> (after all I'll only be (re)starting the server very
>> occasionally once
>>> I've got everything live!) but I'm worried this is indicative of a
>>> problem.  Can anyone give me their experiences on startup times?
>>
>> Markus, already answer this... In my case it takes around 20 seconds
>> when downloading and parsing things... I wonder why it takes so much
>> time for you...
>>
>>
>> Javi.
>>
>>
>> --------------------------------------------------------------
>> -----------
>> 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=3Djoin.php&p=3Dsourceforge
> &CID=3DDEVDEV
> _______________________________________________
> PyWrapper-devel mailing list PyW...@li...
> https://lists.sourceforge.net/lists/listinfo/pywrapper-devel
>
> ----------------------------------------------------------------------=20=

> ---
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to =20
> share your
> opinions on IT & business topics through brief surveys - and earn cash
> http://www.techsay.com/default.php?=20
> page=3Djoin.php&p=3Dsourceforge&CID=3DDEVDEV
> _______________________________________________
> PyWrapper-devel mailing list
> PyW...@li...
> https://lists.sourceforge.net/lists/listinfo/pywrapper-devel

Re: [PyWrapper-devel] First impressions...

[<m.d...@BG...>]

Peter, Javi,
Ive installed pywrapper with fastcgi/apache2 on debian. Was pretty easy =
when you knew where debian keeps the configuration files (its split in =
severals). Fastcgi comes with debian too, so you can use apt to install =
it.

Here it is:
http://search.biocase.org/tapir/index

Markus


> -----Original Message-----
> From: pyw...@li...=20
> [mailto:pyw...@li...] On=20
> Behalf Of Javier privat
> Sent: Sonntag, 19. November 2006 20:35
> To: PyWrapper Developers mailing list
> Subject: Re: [PyWrapper-devel] First impressions...
>=20
>=20
> Hi Peter,
>=20
>=20
> > I've installed from an anonymous checkout using both the standard=20
> > install script and to the system using apt etc - both seem to work=20
> > fine.  Unfortunately the instructions for using FastCGI do=20
> not work on=20
> > my system - I think this is because Kubuntu packages Apache in an=20
> > awkward way.  I'll look at providing documentation at some point
> > but for
> > now I'll just use as a standalone server.  Out of interest=20
> what is the
> > distro of choice in the PyWrapper community?
> >
> I suppose dont really have one... but we should do a wiki=20
> page with a =20
> list of platforms where we have tested the software and=20
> issues with it.
>=20
> I have tried with:
> -Mac os X 10.4
> -Red Hat 9
> -Windows XP Pro
> -Ubuntu
>=20
> But always using the cherrypy stand alone server, we need much more =20
> testing and documentation on setting up the fastcgi way and the =20
> redirection... And we need to start writing this on the wiki!
>=20
> > The next problem I have is that there is a considerable=20
> delay between=20
> > running start_server.py and the server actually coming online (I've=20
> > timed it to 3mins 10secs).  Whilst in itself this isn't such a big
> > deal
> > (after all I'll only be (re)starting the server very=20
> occasionally once
> > I've got everything live!) but I'm worried this is indicative of a
> > problem.  Can anyone give me their experiences on startup times?
>=20
> Markus, already answer this... In my case it takes around 20 seconds =20
> when downloading and parsing things... I wonder why it takes so much =20
> time for you...
>=20
>=20
> Javi.
>=20
>=20
> --------------------------------------------------------------
> -----------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the=20
> chance to share your opinions on IT & business topics through=20
> brief surveys - and earn cash=20
> http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge
&CID=3DDEVDEV
_______________________________________________
PyWrapper-devel mailing list PyW...@li...
https://lists.sourceforge.net/lists/listinfo/pywrapper-devel