select DwcTaxonomy dwc where dwc.dwc.resource.id :=id
needs more joins than necessary if resource is in every table

help links for concepts or extensions should open in new window

curatorial and geospatial dwc extension are essential and used for most tapir output
models too. 
Also they are one2one mappings, so better integrate them into the occurrence core instead
of 
treating them as seperate extensions.

watch out for too large SQL statements to insert a single row

to link to evenlogs later on when the job is gone

use only strings for all dwc properties, but keep additional typed properties that are
tried to be 
filled during the upload. E.g. coordinates

toString, hasCode + equals need tests

 keep namespace with property not extension

just using the resource metadata

when adding an upload job additional jobs should be executed in the same running group:

a) create taxonomy
b) generate dwc star dump file

implement basic TAPIRlite darwin core services to be harvested by GBIF

The admin pages cannot be accessed even as admin.
A folder "admin" inside the exploded war creates chaos.

Removing the folder and all runs fine.
Remove this appfuse leftover!

add admin user with account id=1 that is the primary site administrator and used for
other 
metadata. Export as dbunit

What is the expected output? What do you see instead?

The drop down menus disappear sometimes - bug in the CSS / JS. 
Check with Appfuse to see if updated

Safari, and firefox

This needs fixed before launch proper

...

write webtests to add/edit/delete an occurrence resource

boolean flag for problems in core record

add upload jobs, show jobs, run upload adding logs

link a new page from the admin menu that gives an overview about the logs:

- latest
- by users
- by resources
- by source type/ids

add REST services for occurrences via REST plugin?

add gbif logo to header

write new csv file based import source. Needs to be uploaded in the webapp before mapping
can 
take place and a reference to this file should be stored in the resource metadata

[providertoolkit] INFO [Thread-16] WorkerPool.returnObject(75) | Returning worker, so
removing 
Job
Exception in thread "Thread-16" java.lang.StackOverflowError
    at org.gbif.provider.model.DarwinCore.hashCode(DarwinCore.java:659)
    at org.apache.commons.lang.builder.HashCodeBuilder.append(HashCodeBuilder.java:856)
    at org.gbif.provider.model.DarwinCoreLocation.hashCode(DarwinCoreLocation.java:339)
    at org.apache.commons.lang.builder.HashCodeBuilder.append(HashCodeBuilder.java:856)
    at org.gbif.provider.model.DarwinCore.hashCode(DarwinCore.java:661)
    at org.apache.commons.lang.builder.HashCodeBuilder.append(HashCodeBuilder.java:856)
    at org.gbif.provider.model.DarwinCoreLocation.hashCode(DarwinCoreLocation.java:339)
    at org.apache.commons.lang.builder.HashCodeBuilder.append(HashCodeBuilder.java:856)
    at org.gbif.provider.model.DarwinCore.hashCode(DarwinCore.java:661)

allow people to schedule jobs for certain times.
add time/date form to uploadAction

Also allow jobs to be repeated every x days, weeks or months
That information should be part of the job, so the job created a new job of
itself as the first thing before it launches.

Maybe make that a separate method of the Launchable interface?
so it runs in a different transaction?

Data Access Failure

Write operations are not allowed in read-only mode (FlushMode.NEVER/MANUAL): Turn your

Session into FlushMode.COMMIT/AUTO or remove 'readOnly' marker from transaction definition.

What steps will reproduce the problem?
1. viewing Pontaurus "occurrences by family" here
http://ipt.gbif.org/occResourceStatsByTaxon.html?resource_id=1&zoom=true&type=4
2. selecting view by "class" produces "Yikes"


What version of the provider software are you using? The version should be
displayed in the footer of any page.

Version 1.0beta


What browser are you using?

Firefox 3.0.3 on Windows XP

entering a db connection after creating a new occ resource produces a blank
screen

upload event gets created twice and with wrong data. fix uploadjob

EML is generated on the fly currently. Write all metadata changes to EML files which
are archived and 
point back to last version and the latest version

add gbif logo to header

show upload log events

if uploaded only once no graph is visible.
to change that always add a current datetime "fake upload" with the current record
num

validate collection codes against BCI webservices. Maybe also alow to register new collecitons
or 
institutions?

instead of authoring from scratch load existng document

allow mapping of values for controlled terms.
after mapping the concept try to do a select distinct so one can map original terms
to target terms

investigate e whether all jobs should be wrapped in a transaction. might not be needed
and causes 
troubles (DI) with the spring generated proxy objects

What steps will reproduce the problem?
1. view any record page, e.g.,
http://ipt.gbif.org/occDetail.html?resource_id=1&guid=2cd768ea-5b79-4599-a28a-ae97280ce69f

2. click on SourceID points to localhost


What version of the provider software are you using? The version should be
displayed in the footer of any page.

Version 1.0beta

What browser are you using?

Firefox 3.0.3 on Win XP

use drop down to select by institution or collection.
merge top10 taxa with other taxon class drop down
reduces number of charts

Future changes in the cache db after release 1.0 require an upgrade script to migrate
existing data 
folders

only 10 hits are shown so far

add geoserver so that
a) we have a map for all data per resource
b) a map per species
c) there is WFS per resource

currently all static files (data.zip. logo.gif, eml.xml, etc) are served through servlets,
mostly the 
DataAction one. It would be nice to not have to bother tomcat with this task.

for Data Resources (checklists & occurrences) the IPT should provide a button to update
the 
resource metadata with derived data. In order to prevend overwrites for manually added
good data, 
provide 3 checkboxes initially to overwrite taxonomic, geospatial or temporal data.

write portal interceptor that redirects to a "sorry, this resource is currently being
updated" page if 
the requested resource is running a job

investigate how to integrate BCI for all distinct collection codes.

http://www.silverbiology.com/services/biocol/basic.html

Currently it will not follow redirects etc. 
Pull out the simple version and use commons httpclient

use tablename.column_name for storing the concept mappings instead of the
colum index. So a changed SQL view statement doesnt crash an existing mapping.

Sites loc*
doesnt find anything. How do we handle this?

hardly anything can be seen correctly with IE6&7.
All blocks are wrongly placed. CSS and potentially some html has to be adapted to work
wiht IE too.
Major work

fixed values stored as columnnames not values. fix mapping action

resource browser in metadata/default sub-navigation