[GeoNetwork-devel] CFV: Clustering and Improvements to workflow

Heikki_Doeleman · September 21, 2012, 2:25pm

dear PSC members,

please vote on these two proposals:

http://trac.osgeo.org/geonetwork/wiki/LoadBalanceable

This proposal makes it possible to cluster GeoNetwork, allowing for horizontal scaling.

http://trac.osgeo.org/geonetwork/wiki/ImprovedWorkflow

This proposal makes improvements to GeoNetwork’s metadata lifecycle, the workflow, it repairs the broken pessimistic locking in GeoNetwork and itroduces a graphical Metadata Difference Viewer.

We’re asking for a vote on both proposals together because the Clustering proposal needs the Improvements to Workflow to function correctly.

You can see the code in a GitHub branch here: https://github.com/heikkidoeleman/core-geonetwork/tree/clusteringworkflow. This code applies the changes of the two proposals to the version in the GeoNetwork master branch of a few days ago.

A very quick vote would be much appreciated.

Kind regards
Heikki Doeleman & Jose García

Ian_Allen · September 22, 2012, 12:20am

Just a few notes regarding the Improvements to workflow proposal.

- I did not see any information regarding the metadata data stored on the
filesystem such as thumbnails. I'm assuming that these would be stored in a
separate directory so that it does not conflict with the public data.

- If performing a metadata import from the admin section - 'metadata
insert", does this change in behaviour. would the "Overwrite metadata with
same UUID" overwrite the public version or would it create a new one in
draft mode? Should there be another option?

- Regarding the locking, it should probably also be possible for the user to
remove their own lock so that another user can continue the editing
(assuming they belong to the same group) so that the Reviewer does not need
to be involved in the process. This would be useful when there are multiple
content editors for the same record. But it may cause issues when trying to
identify the owner of the record if no one has a lock.

- It would also be nice to see a "locked" icon next to the records that are
locked by the current logged in user. And also a search option to locate all
current locked records. (but this can be a future request)

- I'm assuming there there would not be any svn revisions on the draft
content. And that there would only be a new svn version when the reviewer
accepts the changes and it is published since it is only at that time that
the public data is updated. Is my assumption correct?

For the record, if I was a voting member I would vote for it.

Good work

--
View this message in context: http://osgeo-org.1560.n6.nabble.com/CFV-Clustering-and-Improvements-to-workflow-tp5003579p5003701.html
Sent from the GeoNetwork developer mailing list archive at Nabble.com.

Heikki_Doeleman · September 22, 2012, 1:31am

thanks for your feedback, my answers inline:

On Sat, Sep 22, 2012 at 2:20 AM, ianwallen <ianwallen@anonymised.com> wrote:

Just a few notes regarding the Improvements to workflow proposal.

I did not see any information regarding the metadata data stored on the
filesystem such as thumbnails. I’m assuming that these would be stored in a
separate directory so that it does not conflict with the public data.

This is a good point that still needs to be addressed. We’re committed to making this work as expected very soon, but for the moment it doesn’t. Thanks for pointing this out.

If performing a metadata import from the admin section - 'metadata
insert", does this change in behaviour. would the “Overwrite metadata with
same UUID” overwrite the public version or would it create a new one in
draft mode? Should there be another option?

There has been no change to this, so it would (as now) overwrite the public version.

Regarding the locking, it should probably also be possible for the user to

remove their own lock so that another user can continue the editing
(assuming they belong to the same group) so that the Reviewer does not need
to be involved in the process. This would be useful when there are multiple
content editors for the same record. But it may cause issues when trying to
identify the owner of the record if no one has a lock.

In the current implementation, Editor users who own the lock can cancel the edit session but that also means their changes (the workspace copy) are deleted. If Symbolic Locking is enabled, Editor users who own the lock (or not) can transfer the lock to others. By default that’s not enabled so Editor lock owners can’t do that. It may be a good option indeed to allow Editor users who own the lock, to transfer it to others – as if they were Reviewers or as if Symbolic Locking was enabled.

It would also be nice to see a “locked” icon next to the records that are
locked by the current logged in user. And also a search option to locate all
current locked records. (but this can be a future request)

There is a lock icon in search results that shows which metadata are locked by any user of your groups. There is also an entry in the admin page to list all metadata that are either locked by you, or that you have the rights to transfer the lock from. Such ‘shortcut’ links can be modified or multiplied, as users think is useful.

I’m assuming there there would not be any svn revisions on the draft
content. And that there would only be a new svn version when the reviewer
accepts the changes and it is published since it is only at that time that
the public data is updated. Is my assumption correct?

Correct.

For the record, if I was a voting member I would vote for it.

Good work

Thanks !

Kind regards
Heikki Doeleman

–
View this message in context: http://osgeo-org.1560.n6.nabble.com/CFV-Clustering-and-Improvements-to-workflow-tp5003579p5003701.html
Sent from the GeoNetwork developer mailing list archive at Nabble.com.

How fast is your code?
3 out of 4 devs don\'t know how their code performs in production.
Find out how slow your code is with AppDynamics Lite.
http://ad.doubleclick.net/clk;262219672;13503038;z?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html

GeoNetwork-devel mailing list
GeoNetwork-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork

ticheler · September 22, 2012, 7:06am

+1 from me
Ciao, Jeroen

GeoCat Bridge for ArcGIS allows instant publishing of data and metadata on GeoServer and GeoNetwork. Visit http://geocat.net for details.
_________________________Jeroen Ticheler
GeoCat bv
Veenderweg 13
6721 WD Bennekom
Tel: +31 (0)6 81286572

Send from mobile phone.

Op 21 sep. 2012 om 16:25 heeft heikki <tropicano@anonymised.com> het volgende geschreven:

dear PSC members,

please vote on these two proposals:

http://trac.osgeo.org/geonetwork/wiki/LoadBalanceable

This proposal makes it possible to cluster GeoNetwork, allowing for horizontal scaling.

http://trac.osgeo.org/geonetwork/wiki/ImprovedWorkflow

This proposal makes improvements to GeoNetwork’s metadata lifecycle, the workflow, it repairs the broken pessimistic locking in GeoNetwork and itroduces a graphical Metadata Difference Viewer.

We’re asking for a vote on both proposals together because the Clustering proposal needs the Improvements to Workflow to function correctly.

You can see the code in a GitHub branch here: https://github.com/heikkidoeleman/core-geonetwork/tree/clusteringworkflow. This code applies the changes of the two proposals to the version in the GeoNetwork master branch of a few days ago.

A very quick vote would be much appreciated.

Kind regards
Heikki Doeleman & Jose García

Got visibility?
Most devs has no idea what their production app looks like.
Find out how fast your code is with AppDynamics Lite.
http://ad.doubleclick.net/clk;262219671;13503038;y?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html

GeoNetwork-devel mailing list
GeoNetwork-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork

sppigot · September 22, 2012, 7:07am

Heikki & Jose & Steven,

+1 from me.

Reviewed the proposals and have some brief notes/comments:

1. Load Balancing

- Really keen to see this proposal get up - helps us make much better use of machines with lots of processors etc
- Had some concerns about UUID handling and what was being proposed but I can see from using the test sites that you have not mixed the UUID for the db with the UUID/identifiers for the metadata - so looks ok!

2. Workflow

- Seems that any changes made to the metadata would not be recorded in the versioning/subversion repo until the metadata leaves the workspace table - this could be desirable as it avoids cluttering the versioning/subversion repo with lots of little changes.
- We could use the graphical diff comparison interface with a selector to compare different versions of a metadata record in versioning/subversion repo maybe in a future proposal
- I like the stricter controls on what can happen with state changes, the workspace concept to protect published metadata from changes etc and the better locking - nice

Cheers,
Simon

________________________________________
From: heikki [tropicano@anonymised.com]
Sent: Saturday, 22 September 2012 12:25 AM
To: Devel geonetwork-devel@lists.sourceforge.net
Subject: [GeoNetwork-devel] CFV: Clustering and Improvements to workflow

dear PSC members,

please vote on these two proposals:

http://trac.osgeo.org/geonetwork/wiki/LoadBalanceable

This proposal makes it possible to cluster GeoNetwork, allowing for horizontal scaling.

http://trac.osgeo.org/geonetwork/wiki/ImprovedWorkflow

This proposal makes improvements to GeoNetwork's metadata lifecycle, the workflow, it repairs the broken pessimistic locking in GeoNetwork and itroduces a graphical Metadata Difference Viewer.

We're asking for a vote on both proposals together because the Clustering proposal needs the Improvements to Workflow to function correctly.

You can see the code in a GitHub branch here: GitHub - heikkidoeleman/core-geonetwork at clusteringworkflow. This code applies the changes of the two proposals to the version in the GeoNetwork master branch of a few days ago.

A very quick vote would be much appreciated.

Kind regards
Heikki Doeleman & Jose García

sppigot · September 23, 2012, 2:25pm

Just one extra note: Could be an issue with Z3950 server ports - each servlet will try and start a Z server on the same port: resulting behaviour would depend on the config of servlets and machines (this is GN's Z server capability - this shouldn't affect GN's ability to harvest/search other Z3950 servers ie. when it is used as a Z client).

Who depends on GN as a Z server? I know that the Australian Spatial Data Directory as currently config'd does use GN as a Z server, but it could use a different protocol/interface to talk to GeoNetwork sites especially as JZKit in GN doesn't implement some of the hacks for getting all records from a Z server. Could be also that those who want GN as a Z server could switch to using the SRU/SRW http based Z capability (this shouldn't be affected by this proposal)?

Cheers,
Simon
________________________________________
From: Simon.Pigot@anonymised.com [Simon.Pigot@anonymised.com]
Sent: Saturday, 22 September 2012 5:07 PM
To: tropicano@anonymised.com; geonetwork-devel@lists.sourceforge.net
Subject: [ExternalEmail] Re: [GeoNetwork-devel] CFV: Clustering and Improvements to workflow

Heikki & Jose & Steven,

+1 from me.

Reviewed the proposals and have some brief notes/comments:

1. Load Balancing

- Really keen to see this proposal get up - helps us make much better use of machines with lots of processors etc
- Had some concerns about UUID handling and what was being proposed but I can see from using the test sites that you have not mixed the UUID for the db with the UUID/identifiers for the metadata - so looks ok!

2. Workflow

- Seems that any changes made to the metadata would not be recorded in the versioning/subversion repo until the metadata leaves the workspace table - this could be desirable as it avoids cluttering the versioning/subversion repo with lots of little changes.
- We could use the graphical diff comparison interface with a selector to compare different versions of a metadata record in versioning/subversion repo maybe in a future proposal
- I like the stricter controls on what can happen with state changes, the workspace concept to protect published metadata from changes etc and the better locking - nice

Cheers,
Simon

________________________________________
From: heikki [tropicano@anonymised.com]
Sent: Saturday, 22 September 2012 12:25 AM
To: Devel geonetwork-devel@lists.sourceforge.net
Subject: [GeoNetwork-devel] CFV: Clustering and Improvements to workflow

dear PSC members,

please vote on these two proposals:

http://trac.osgeo.org/geonetwork/wiki/LoadBalanceable

This proposal makes it possible to cluster GeoNetwork, allowing for horizontal scaling.

http://trac.osgeo.org/geonetwork/wiki/ImprovedWorkflow

This proposal makes improvements to GeoNetwork's metadata lifecycle, the workflow, it repairs the broken pessimistic locking in GeoNetwork and itroduces a graphical Metadata Difference Viewer.

We're asking for a vote on both proposals together because the Clustering proposal needs the Improvements to Workflow to function correctly.

You can see the code in a GitHub branch here: GitHub - heikkidoeleman/core-geonetwork at clusteringworkflow. This code applies the changes of the two proposals to the version in the GeoNetwork master branch of a few days ago.

A very quick vote would be much appreciated.

Kind regards
Heikki Doeleman & Jose García

------------------------------------------------------------------------------
How fast is your code?
3 out of 4 devs don\\\'t know how their code performs in production.
Find out how slow your code is with AppDynamics Lite.
http://ad.doubleclick.net/clk;262219672;13503038;z?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html
_______________________________________________
GeoNetwork-devel mailing list
GeoNetwork-devel@lists.sourceforge.net
geonetwork-devel List Signup and Options
GeoNetwork OpenSource is maintained at GeoNetwork - Geographic Metadata Catalog download | SourceForge.net

Francois_Prunayre · September 24, 2012, 6:02am

+1 for me. Nice improvements.

Francois

2012/9/23 Simon.Pigot@anonymised.com:

Just one extra note: Could be an issue with Z3950 server ports - each servlet will try and start a Z server on the same port: resulting behaviour would depend on the config of servlets and machines (this is GN’s Z server capability - this shouldn’t affect GN’s ability to harvest/search other Z3950 servers ie. when it is used as a Z client).

Who depends on GN as a Z server? I know that the Australian Spatial Data Directory as currently config’d does use GN as a Z server, but it could use a different protocol/interface to talk to GeoNetwork sites especially as JZKit in GN doesn’t implement some of the hacks for getting all records from a Z server. Could be also that those who want GN as a Z server could switch to using the SRU/SRW http based Z capability (this shouldn’t be affected by this proposal)?

Cheers,
Simon

From: Simon.Pigot@anonymised.com [Simon.Pigot@anonymised.com]
Sent: Saturday, 22 September 2012 5:07 PM
To: tropicano@anonymised.com; geonetwork-devel@anonymised.coms.sourceforge.net
Subject: [ExternalEmail] Re: [GeoNetwork-devel] CFV: Clustering and Improvements to workflow

Heikki & Jose & Steven,

+1 from me.

Reviewed the proposals and have some brief notes/comments:

Load Balancing

Really keen to see this proposal get up - helps us make much better use of machines with lots of processors etc

Had some concerns about UUID handling and what was being proposed but I can see from using the test sites that you have not mixed the UUID for the db with the UUID/identifiers for the metadata - so looks ok!

Workflow

Seems that any changes made to the metadata would not be recorded in the versioning/subversion repo until the metadata leaves the workspace table - this could be desirable as it avoids cluttering the versioning/subversion repo with lots of little changes.

We could use the graphical diff comparison interface with a selector to compare different versions of a metadata record in versioning/subversion repo maybe in a future proposal

I like the stricter controls on what can happen with state changes, the workspace concept to protect published metadata from changes etc and the better locking - nice

Cheers,
Simon

From: heikki [tropicano@anonymised.com]
Sent: Saturday, 22 September 2012 12:25 AM
To: Devel geonetwork-devel@lists.sourceforge.net
Subject: [GeoNetwork-devel] CFV: Clustering and Improvements to workflow

dear PSC members,

please vote on these two proposals:

http://trac.osgeo.org/geonetwork/wiki/LoadBalanceable

This proposal makes it possible to cluster GeoNetwork, allowing for horizontal scaling.

http://trac.osgeo.org/geonetwork/wiki/ImprovedWorkflow

This proposal makes improvements to GeoNetwork’s metadata lifecycle, the workflow, it repairs the broken pessimistic locking in GeoNetwork and itroduces a graphical Metadata Difference Viewer.

We’re asking for a vote on both proposals together because the Clustering proposal needs the Improvements to Workflow to function correctly.

You can see the code in a GitHub branch here: https://github.com/heikkidoeleman/core-geonetwork/tree/clusteringworkflow. This code applies the changes of the two proposals to the version in the GeoNetwork master branch of a few days ago.

A very quick vote would be much appreciated.

Kind regards
Heikki Doeleman & Jose García

How fast is your code?
3 out of 4 devs don\'t know how their code performs in production.
Find out how slow your code is with AppDynamics Lite.
http://ad.doubleclick.net/clk;262219672;13503038;z?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html

GeoNetwork-devel mailing list
GeoNetwork-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork

Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://ad.doubleclick.net/clk;258768047;13503038;j?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html

GeoNetwork-devel mailing list
GeoNetwork-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork

Heikki_Doeleman · September 24, 2012, 8:23am

hi Simon,

as we are very unfamiliar with Z39.50 in GeoNetwork, we haven’t considered this in the clustering implementation. It would be great if someone (hint) who knows something about that could take a look and see if any additional changes are needed for this, in a clustered environment.

If you want we can discuss this on IRC or Skype, let us know.

thanks and kind regards
Heikki Doeleman

On Sun, Sep 23, 2012 at 4:25 PM, <Simon.Pigot@anonymised.com> wrote:

Just one extra note: Could be an issue with Z3950 server ports - each servlet will try and start a Z server on the same port: resulting behaviour would depend on the config of servlets and machines (this is GN’s Z server capability - this shouldn’t affect GN’s ability to harvest/search other Z3950 servers ie. when it is used as a Z client).

Who depends on GN as a Z server? I know that the Australian Spatial Data Directory as currently config’d does use GN as a Z server, but it could use a different protocol/interface to talk to GeoNetwork sites especially as JZKit in GN doesn’t implement some of the hacks for getting all records from a Z server. Could be also that those who want GN as a Z server could switch to using the SRU/SRW http based Z capability (this shouldn’t be affected by this proposal)?

Cheers,
Simon

From: Simon.Pigot@anonymised.com [Simon.Pigot@anonymised.com]
Sent: Saturday, 22 September 2012 5:07 PM
To: tropicano@anonymised.com; geonetwork-devel@anonymised.comourceforge.net
Subject: [ExternalEmail] Re: [GeoNetwork-devel] CFV: Clustering and Improvements to workflow

Heikki & Jose & Steven,

+1 from me.

Reviewed the proposals and have some brief notes/comments:

Load Balancing

Really keen to see this proposal get up - helps us make much better use of machines with lots of processors etc

Had some concerns about UUID handling and what was being proposed but I can see from using the test sites that you have not mixed the UUID for the db with the UUID/identifiers for the metadata - so looks ok!

Workflow

Seems that any changes made to the metadata would not be recorded in the versioning/subversion repo until the metadata leaves the workspace table - this could be desirable as it avoids cluttering the versioning/subversion repo with lots of little changes.

We could use the graphical diff comparison interface with a selector to compare different versions of a metadata record in versioning/subversion repo maybe in a future proposal

I like the stricter controls on what can happen with state changes, the workspace concept to protect published metadata from changes etc and the better locking - nice

Cheers,
Simon

From: heikki [tropicano@anonymised.com]
Sent: Saturday, 22 September 2012 12:25 AM
To: Devel geonetwork-devel@lists.sourceforge.net
Subject: [GeoNetwork-devel] CFV: Clustering and Improvements to workflow

dear PSC members,

please vote on these two proposals:

http://trac.osgeo.org/geonetwork/wiki/LoadBalanceable

This proposal makes it possible to cluster GeoNetwork, allowing for horizontal scaling.

http://trac.osgeo.org/geonetwork/wiki/ImprovedWorkflow

This proposal makes improvements to GeoNetwork’s metadata lifecycle, the workflow, it repairs the broken pessimistic locking in GeoNetwork and itroduces a graphical Metadata Difference Viewer.

We’re asking for a vote on both proposals together because the Clustering proposal needs the Improvements to Workflow to function correctly.

You can see the code in a GitHub branch here: https://github.com/heikkidoeleman/core-geonetwork/tree/clusteringworkflow. This code applies the changes of the two proposals to the version in the GeoNetwork master branch of a few days ago.

A very quick vote would be much appreciated.

Kind regards
Heikki Doeleman & Jose García

How fast is your code?
3 out of 4 devs don\'t know how their code performs in production.

Find out how slow your code is with AppDynamics Lite.
http://ad.doubleclick.net/clk;262219672;13503038;z?

http://info.appdynamics.com/FreeJavaPerformanceDownload.html

GeoNetwork-devel mailing list
GeoNetwork-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at http://sourceforge.net/projects/geonetwork

Francois_Prunayre · September 27, 2012, 7:21am

Hi,

I've been testing the workflow proposal on one of the clustering node
and have some more comments on the workflow:
* if you create a record, go to privilege and publish to internet,
the record is not yet published - because it's not approved - so maybe
privileges should not be accessible for a workspace copy ? or a
warning should be displayed in the privilege.
* when changing status to approved, it looks like the metadata is
published to all automatically - the metadata should keep its
privileges, even if it's approved it could be a non public (for
internet) record. Any options to configure that ? "approved" should
not mean "publish to internet" I think.
* it could be nice to provide an option to disable/bypass the feature
where the save action behave as currently (so maybe copying the
workspace copy to the metadata when saving) and all the new options in
menu are hidden - this keep the user interface not too complex with
less options which may be not of interest for all. With the workflow,
the number of clicks to make a change is doubled and we probably have
some user who would like to make changes with less click ... with
inline text editing feature for example.
* about menu, at some point we should maybe group actions by category
because the list of choice is sometime very long and user may be lost.
So grouping editing/viewing actions together at least.

Cheers.

Francois

2012/9/24 Francois Prunayre <fx.prunayre@anonymised.com>:

+1 for me. Nice improvements.

Francois

2012/9/23 <Simon.Pigot@anonymised.com>:

Just one extra note: Could be an issue with Z3950 server ports - each
servlet will try and start a Z server on the same port: resulting behaviour
would depend on the config of servlets and machines (this is GN's Z server
capability - this shouldn't affect GN's ability to harvest/search other
Z3950 servers ie. when it is used as a Z client).

Who depends on GN as a Z server? I know that the Australian Spatial Data
Directory as currently config'd does use GN as a Z server, but it could use
a different protocol/interface to talk to GeoNetwork sites especially as
JZKit in GN doesn't implement some of the hacks for getting all records from
a Z server. Could be also that those who want GN as a Z server could switch
to using the SRU/SRW http based Z capability (this shouldn't be affected by
this proposal)?

Cheers,
Simon
________________________________________
From: Simon.Pigot@anonymised.com [Simon.Pigot@anonymised.com]
Sent: Saturday, 22 September 2012 5:07 PM
To: tropicano@anonymised.com; geonetwork-devel@lists.sourceforge.net
Subject: [ExternalEmail] Re: [GeoNetwork-devel] CFV: Clustering and
Improvements to workflow

Heikki & Jose & Steven,

+1 from me.

Reviewed the proposals and have some brief notes/comments:

1. Load Balancing

- Really keen to see this proposal get up - helps us make much better use
of machines with lots of processors etc
- Had some concerns about UUID handling and what was being proposed but I
can see from using the test sites that you have not mixed the UUID for the
db with the UUID/identifiers for the metadata - so looks ok!

2. Workflow

- Seems that any changes made to the metadata would not be recorded in the
versioning/subversion repo until the metadata leaves the workspace table -
this could be desirable as it avoids cluttering the versioning/subversion
repo with lots of little changes.
- We could use the graphical diff comparison interface with a selector to
compare different versions of a metadata record in versioning/subversion
repo maybe in a future proposal
- I like the stricter controls on what can happen with state changes, the
workspace concept to protect published metadata from changes etc and the
better locking - nice

Cheers,
Simon

________________________________________
From: heikki [tropicano@anonymised.com]
Sent: Saturday, 22 September 2012 12:25 AM
To: Devel geonetwork-devel@lists.sourceforge.net
Subject: [GeoNetwork-devel] CFV: Clustering and Improvements to workflow

dear PSC members,

please vote on these two proposals:

http://trac.osgeo.org/geonetwork/wiki/LoadBalanceable

This proposal makes it possible to cluster GeoNetwork, allowing for
horizontal scaling.

http://trac.osgeo.org/geonetwork/wiki/ImprovedWorkflow

This proposal makes improvements to GeoNetwork's metadata lifecycle, the
workflow, it repairs the broken pessimistic locking in GeoNetwork and
itroduces a graphical Metadata Difference Viewer.

We're asking for a vote on both proposals together because the Clustering
proposal needs the Improvements to Workflow to function correctly.

You can see the code in a GitHub branch here:
https://github.com/heikkidoeleman/core-geonetwork/tree/clusteringworkflow.
This code applies the changes of the two proposals to the version in the
GeoNetwork master branch of a few days ago.

A very quick vote would be much appreciated.

Kind regards
Heikki Doeleman & Jose García

------------------------------------------------------------------------------
How fast is your code?
3 out of 4 devs don\\\'t know how their code performs in production.
Find out how slow your code is with AppDynamics Lite.
http://ad.doubleclick.net/clk;262219672;13503038;z?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html
_______________________________________________
GeoNetwork-devel mailing list
GeoNetwork-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at
http://sourceforge.net/projects/geonetwork

------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://ad.doubleclick.net/clk;258768047;13503038;j?
http://info.appdynamics.com/FreeJavaPerformanceDownload.html
_______________________________________________
GeoNetwork-devel mailing list
GeoNetwork-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geonetwork-devel
GeoNetwork OpenSource is maintained at
http://sourceforge.net/projects/geonetwork