Dont worry,
As I said I hacked it on the client side. And yes I did not manage to actually run models but I am a little bit lost there so I was trying to make sure that the rest is working...
When do you get back from your little country? How is extrema going? I miss it!
Cheers.
On 12/07/2007, at 17:39, Tim Sutton wrote:
Hi Renato & Javi
A quick note from an icafe in extreme... Renato I discovered there is a problem with the xml coming back from teh svm alg - open it with ffox to see the error. Javi for now you can resolve teh issue by simply removing /usr/local/lib/openmodeller/libomsvm.so . Also it seems the cron job for the om service is not running - in the parallels image the crontab for www-data is set up but I think you may need to check the cron configs to ensure that www-data is allowed to run cronjobs.
Hope that helps!
Regards
Tim
2007/7/9, Javier de la Torre jatorre@gmail.com:
I am using the OMWS that Tim left me in a parallels image. I believe it is the last version as it is taken from repository and compiled... In any case I am not sure at all.
I just made a hack on my side and now is working so I will wait for Tim to come back and figure out this as I dont know how openmodeller has been installed on this debian that I have.
Cheers.
Javier.
On 09/07/2007, at 19:18, Renato De Giovanni wrote:
Hello Javi,
Can you tell me which omws service address are you using?
The library does escape the special characters &, < and >, but I've just noticed that double quotes were not being escaped (fixed now). The library returns content in latin1. However, the soap server
will
always return content in utf-8.
I tested with the "official" omws service (http://modeller.cria.org.br/cgi-bin/om_soap_server.cgi) and it
seems
to be working with the most recent list of algorithms. I've dumped the response and copied to a static document if you want to check:
http://www.cria.org.br/~renato/omws_alg_metadata_dump.xml
Maybe you're using another service address with an older version of the library? Or maybe there's something doing some conversions between the real raw XML you receive and your parser?
Regards,
Renato
On 9 Jul 2007 at 13:19, Javier de la Torre wrote:
Hey all,
Finally I had to hack two things to handle incorrect messages from OMWS. I could not find where the metadata about the algorithms is stored so I hack in the biogeosdi code... that is what I made:
//HACK BECUASE OMWS IS RETURNING INVALID XML $rawXml = str_replace("&oldid","&oldid",$rawXml); $rawXml = str_replace("& Shawe","& Shawe",$rawXml); $rawXml = utf8_encode($rawXml); //HACK FINISHEDThe first two are to handle the not scaped parethesis. The
third one
is to encode the result as UTF8 because if not is not valid XML without a encoding definition, which the response does not
include.
If the encoding is not defined I think the default is UTF8 and
this
seems to be latin1, but in any case I would really recommend being explicit about the encoding bein used.
Cheers.
biogeosdi mailing list biogeosdi@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/biogeosdi
biogeosdi mailing list biogeosdi@lists.tdwg.org http://lists.tdwg.org/mailman/listinfo/biogeosdi
-- Tim Sutton QGIS Project Steering Committee Member - Release Manager Visit http://qgis.org for a great open source GIS openModeller Desktop Developer Visit http://openModeller.sf.net for a great open source ecological niche modelling tool Home Page: http://tim.linfiniti.com Skype: timlinux Irc: timlinux on #qgis at freenode.net