Hi Andrea, Jody,
My memory of it is very hazy. I just looked through my old emails and
found the one below which mentions the motivation for dispose in Layer
which, I suppose, carried through to MapContent.
On 7 June 2010 at 23:36, Jody Garnett jody.garn...@gmail.com wrote:
Had a good session
Hi Steve,
Good Morning Michael,
Good morning from Sydney
Thanks for the quick response and solution! I'll try the patch with the
10.4 release.
Great. Please let me know how you get on with it.
Michael
--
Hi Steve,
Ah yes, a couple of items from my long shoulda but didn't list...
Looking at the GEOT-3952 (Andrea's patch) I don't see or recall any
reason why that wasn't applied other than the ambition to replace the
whole thing at some stage. I can apply the patch to the current
development branch
Just my 2c, but I thought it might not have been the best thing to
close that pull request. I can easily imagine submitting a PR,
getting feedback about the need for further work from the core devs,
and then not getting round to it for a few months (or years !) because
of other things. I don't
No, I don't think there's any irritation Brett, or any implication
that you were not being helpful. It's just highlighted something that
probably needs to be discussed and formalized a bit.
Michael
On 21 July 2013 19:25, Brett Walker brett.wal...@geometryit.com wrote:
Hi,
I'm sorry if this
/forum/#!forum/jai-tools
Michael
On 18 July 2013 15:54, Andrea Aime andrea.a...@geo-solutions.it wrote:
On Thu, Jul 18, 2013 at 5:55 AM, Michael Bedward michael.bedw...@gmail.com
wrote:
It was GeoSolutions that urgently wanted the bug fixes to JAITools
that are contained in v1.3.1
1660272
mob: +39 333 8128928
http://www.geo-solutions.it
http://twitter.com/geosolutions_it
---
On Wed, Jul 17, 2013 at 3:28 AM, Brett Walker
brett.wal...@geometryit.com wrote:
-Original Message-
From: Michael Bedward
On 18 July 2013 13:39, Ben Caradoc-Davies ben.caradoc-dav...@csiro.au wrote:
Note that it breaks GeoServer wps-core on master and 2.3.x:
Build failure in wps-core CoverageUtilities caused by jaitools upgrade
https://jira.codehaus.org/browse/GEOS-5892
On 18/07/13 08:41, Michael Bedward wrote
On 18 July 2013 14:43, Brett Walker brett.wal...@geometryit.com wrote:
It lost the ability to supply a default value. I wasn't impressed with it.
Concerning GS - I know nothing :)
Concerning JAITools and your not being impressed - I suggest you
discuss any concerns with Simone since he
Hi Brett,
Thanks for this work.
Yes, there were substantial changes to RangeLookupTable and related
classes to fix thread safety, so it's no surprise that you had to hack
things at the GeoTools end. However, I don't know the
CoverageUtilities code at all - it would be best to ask Simone, Andrea
Hi Akos,
Perhaps next time use a smaller demo image :)
I'd call this option tick marks since that is a widely used term in
numerical plotting and tick or tickmark seems easier to picture
than subscale.
I think this would be a very useful addition for the grid module
(GeoTools 10.x). If you
If the Google lawyer(s) could provide a list of some OS licences that
are currently in favour for collaborative work, then there would be
some context for discussion between the GeoTools PMC and the OSGeo
Board.
Michael
On 30 January 2013 12:26, Jody Garnett jody.garn...@gmail.com wrote:
Thanks
On 23/01/13 10:06, Andrea Aime wrote:
Humm... I believe that for this bug there is no way to stop the thread in
case the
bug shows up again.
If so, I would not go for the trouble of setting up a separate thread I
think, just
use the main one, and if the bug shows up again, the build will
Could you make it a daemon thread ?
Michael
On 23 January 2013 20:57, Ákos Maróy a...@maroy.hu wrote:
On 23/01/13 10:34, Andrea Aime wrote:
The thing is, you cannot kill a thread in java, it has to die on its own
(the method that supposedly does it is deprecated and afaik does nothing).
On 23 January 2013 21:12, Andrea Aime andrea.a...@geo-solutions.it wrote:
Daemon threads are known for not blocking the JVM shutdown, however, not
sure if
that holds true also for active ones
Mmm... thought I was doing that in Jiffle with a daemon polling
thread. Probably a false memory.
:
Michael Bedward
Components:
main
Created:
22/Jan/13 2:44 AM
Hi all,
Sorry - the build failure is my fault. Using a JTS 1.13 method on the
8.x branch which is still working with JTS 1.12.
What do you think would be best: upgrade 8.x to JTS 1.13 (which has
been fine for 9.x for the last month or so) or just downgrade my unit
tests ?
Michael
On 22 January
Since it's getting late here I've just edited the test class to remove
the offending JTS 1.13 method call and fix the build.
Michael
On 22 January 2013 21:38, Michael Bedward michael.bedw...@gmail.com wrote:
Hi all,
Sorry - the build failure is my fault. Using a JTS 1.13 method on the
8.x
I think that's a good point Andrea. Let's leave 8.x with JTS 1.12
making GeoTools 9.x just that little bit sweeter.
Michael
On 23 January 2013 05:52, Andrea Aime andrea.a...@geo-solutions.it wrote:
Sorry for the late answer.
One one side, I believe upgrading JTS is low risk, so it's most
Hi Brett, Mikel,
Could other devs please look at
org.geotools.coverage.grid.io.AbstractGridCoverage2DReader.decimationOnReadingControl
and confirm or deny that there is no possibility of a rounding error,
underflow or double to int conversion that results in a negative number!
Mikel, it
move
gt-grid over to extension land I am happy to kick the geotools release out
as soon as you are ready.
--
Jody Garnett
On Thursday, 17 January 2013 at 10:44 PM, Michael Bedward wrote:
On 17 January 2013 22:47, Andrea Aime andrea.a...@geo-solutions.it wrote:
+1
Thanks Andrea
Mind
On 17 January 2013 18:08, Andrea Aime andrea.a...@geo-solutions.it wrote:
It seems to me it's better put in another module. gt-main has become a rat
nest with a ton of
things past and present and nobody feels like being the real maintainer of
it.
Yep - let's not bury the vector grid stuff in
On 17 January 2013 22:47, Andrea Aime andrea.a...@geo-solutions.it wrote:
+1
Thanks Andrea
Mind, the proposal should also say who's the maintainer of the module
Oops - just added note about me being maintainer (although happy if
someone else wants it).
Michael
On 16 January 2013 20:46, Jody Garnett jody.garn...@gmail.com wrote:
If you have any pending work to sneak in before the deadline, please wrap it
up this week.
I hate to mention this for fear of sounding like a broken record, but
any chance of promoting gt-grid (vector grids) ? It gets used
Hi Jody,
So lets try this, write up a change request (faster then deciding if it is
needed).
There is this old one which I could edit if that's ok.
Question: This vector grid functionality is likely to be reused by other
modules right? If so it would be better in gt-main … extensions tend
I changed a few words...
http://docs.codehaus.org/display/GEOTOOLS/Vector+grids
Anything more to say ?
On 17 January 2013 14:43, Jody Garnett jody.garn...@gmail.com wrote:
There is this old one which I could edit if that's ok.
Yeah found it also
Aside: would it be possible for the addresses for posting to the lists
to be in the geotools domain, e.g. geotools-u...@geotools.org ?
Michael
On 9 January 2013 11:22, Jody Garnett jody.garn...@gmail.com wrote:
We raise a ticket, and talk to the system admin committee.
I had a look at allure
Hi Andrea, all,
Martin has just released JTS version 1.13 and maven artifacts
(binaries, sources and javadocs) have just landed at Central.
I've managed to build GeoTools master against the new version with
only one tiny change required in the render module. Details are in
this issue:
Thanks Andrea.
I'm a bit wary of merging on the stable series, maybe we should wait a bit
and see how things go on master before a backport?
Yep - that sounds like the best way to go.
Michael
--
LogMeIn Rescue:
Cool - thanks Jody.
Michael
On 15 December 2012 17:58, Jody Garnett jody.garn...@gmail.com wrote:
uDig is stuck on the last milestone release until QA is finished (so perhaps
after I return from Holidays?). Please upgrade to JTS 1.13 without us … and
thanks for asking.
--
Jody Garnett
:
Task
Assignee:
Michael Bedward
Components
Hi Andrea, Jody,
Please see my comment on this one:
https://jira.codehaus.org/browse/GEOT-4345?focusedCommentId=315573#comment-315573
Michael
On 13 December 2012 01:15, Pierre Beule (JIRA) j...@codehaus.org wrote:
Pierre
On 12 December 2012 09:06, Jody Garnett jody.garn...@gmail.com wrote:
If we stole some of the language from here ( or here ) would it be
appropriate? Or even just GeoTools is an open source Java library that
provides tools for geospatial data. from our website.
The reason the scope of our
On 12 December 2012 12:36, Ben Caradoc-Davies
ben.caradoc-dav...@csiro.au wrote:
Framework is a bit too generic in my opinion. I think we need to include the
word software to make it clear that we are not talking about a physical
thing, which I think, from my reading of Frank's email, is
Hi Jody,
Sounds like it might be a lot of work ? I'd imagine there are too
many fiddly cases in the code to be able to push it all through a
shell script. But it would be nice for the sources to be consistent
with recommended usage which I assume is to use create() rather than
constructor.
Priority:
Major
Reporter:
Michael Bedward
On 26 November 2012 20:17, Andrea Aime andrea.a...@geo-solutions.it wrote:
I am under the impression none of the documents above convey two important
bits:
- the need to discuss on the mailing list any non trivial change, and in
particular any API changes
- the presence of tests for changes
Hi Kevin,
Thanks for this. Just had a quick look at your patch. Hope to have a
proper look later.
I'm particularly interested in the rasterizing elements. Do you see
these as eventually replacing the existing VectorToRasterProcess ?
That class has a list of long-standing issues and needs either
David,
This list is for discussions directly relating to the development of
the GeoTools library only.
Michael
On 20 November 2012 03:25, David Shi davidg...@yahoo.co.uk wrote:
Hi,
Is there a class to do the following?
to replace an arraylist of Point2D with a set of Points between min
On 9 November 2012 11:38, Jody Garnett jody.garn...@gmail.com wrote:
For geotools I would like to get SLD files in place for this dataset,
and start to use it in our tutorials.
Yes - excellent idea Jody
Michael
--
Thanks Jody. A huge amount of work there !
A while ago I updated the feature tutorial to avoid using
FeatureCollection.add, but instead of using
DefaultFeatureCollection.add I explicitly created a List to accumulate
features and then wrapped that in a ListFeatureCollection when writing
to the
On 1 November 2012 09:03, Jody jody.garn...@gmail.com wrote:
I chose the same solution as you for the branch!
Using a List and then ListFeatureCollection.
Cool. Great minds :)
So could DefaultFeatureCollection.add be deprecated ?
Michael
On 26 October 2012 06:24, Gabriel Roldan grol...@opengeo.org wrote:
re xerces, we'd also need to exclude it from the JTS dependency:
[INFO] | +- com.vividsolutions:jts:jar:1.12:compile
[INFO] | | \- xerces:xercesImpl:jar:2.7.1:compile (version managed from
2.4.0)
The JTS jars from Maven
Thanks for that explanation Justin.
I know Martin still tries to cater for Java 1.5 but xerces is not
used in any of the core classes that make up the JTS maven artifacts.
Michael
On 26 October 2012 10:40, Justin Deoliveira jdeol...@opengeo.org wrote:
Btw, i did this on the branch i pushed
On 23 October 2012 21:35, Andrea Aime andrea.a...@geo-solutions.it wrote:
Yep
Done
Michael
--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free
:
Michael Bedward
This message is automatically generated by JIRA
Hello,
I'm seeing the test error below when building the master branch (mac
osx 10.6.8, java 1.6.0_29). Hudson seems to be happy so perhaps it's
just me - but since the error mentions Envelope I wondered if this was
connected to the work that Jody has been doing recently ?
Michael
Running
Sorry, I think that was a false alarm. I thought I was on master but
was actually on a branch where with the the jsr-275 dependency
replaced by jscience (GEOT-4290). I just did a build on master and it
was fine.
Michael
On 17 October 2012 12:15, Michael Bedward michael.bedw...@gmail.com wrote
Your question is not about the development of GeoTools David. General
questions about using the library should be directed to the user list
instead:
https://lists.sourceforge.net/lists/listinfo/geotools-gt2-users
Michael
On 14 October 2012 01:56, David Shi davidg...@yahoo.co.uk wrote:
How do I
On 8 October 2012 22:02, Ákos Maróy a...@maroy.hu wrote:
what am I doing wrong?
You're using Java 7 which isn't yet supported by GeoTools.
Sorry about that.
Michael
--
Don't let slow site performance ruin your
On 9 October 2012 10:00, Brett Walker brett.wal...@geometryit.com wrote:
I wonder if comments made on a jira case can be posted to the dev list? Is
this possible? The original case can be. Why not comments?
Hello Brett,
I think it would overload the list. There is presently the option of
On Monday, 8 October 2012 at 5:03 AM, Andrea Aime wrote:
Actually, what I'm proposing is that we create a separate, parallel, mailing
list
for builds that are not the main one, the one that people get their head
off if they break.
Seems like a very good idea.
Aside, the jdbc builds seem to
Bummer - looks like this one is my fault
Michael
On 4 October 2012 14:43, Hudson moni...@openplans.org wrote:
See http://hudson.opengeo.org/hudson/job/geotools-8.x-nightly/90/
--
[...truncated 17369 lines...]
274432/?
278528/?
282624/?
286720/?
Michael Bedward
Ah - the static create methods were removed from all (I think) image
operations in jaitools 1.3.
Andrea: The fix is trivial so I'll do it directly and reopen GEOT-4276
to record the changes so you can double check for sanity.
Sorry about that.
Michael
On 4 October 2012 16:01, Michael Bedward
I just had a quick look at the Zhao-Saalfeld paper as well as the
review paper of Shi and Cheung:
http://citeseer.ist.psu.edu/viewdoc/summary?doi=10.1.1.95.5882
Yes - plenty to work through there :)
One thing I don't understand, and I'm hoping you can explain, is how
one guards against
Yes I guess it will be fine for many situations, so we could add the
simplification option to the jaitools operation, using DP for the
moment, with a note in the javadocs about it being the user's
responsibility to check for a valid result - thus washing our hands :)
Here is a jaitools issue.
Yep - good idea Martin.
I guess we could add an extra parameter to the image operation to
control the degree of simplification, with a default value.
Did you use a DP simplifier ?
Michael
On 4 October 2012 10:02, Martin Davis mda...@opengeo.org wrote:
Is there any way the simplification can
:
Bug
Assignee:
Michael Bedward
Created
Not my fault :)
Michael
On 3 October 2012 14:55, Hudson moni...@openplans.org wrote:
See http://hudson.opengeo.org/hudson/job/geotools-master/4560/changes
Changes:
[mbedward] GEOT-4276 update jaitools dependency to 1.3.0
Tests in error:
On 2 October 2012 08:22, Jody Garnett jody.garn...@gmail.com wrote:
Micheal Bedward pointed out that the use of add is common in examples and
when staging content - DefaultFeatureCollection works in this case as a drop
in replacement.
Someone mention my name ?
You know, I actually think it's
Priority:
Minor
Reporter:
Michael Bedward
:
Improvement
Assignee:
Michael Bedward
Components
not; it is a workaround for projects using maven copy-dependencies
to correct the jars after they have been downloaded into a folder.
Jody
--
Jody Garnett
On Tuesday, 11 September 2012 at 11:56 AM, Michael Bedward wrote:
Just trying to follow this...
The change means that jars will be deployed as xxx
Hi Jody
Yeah something is a pain here - and what is most annoying is that if you
ever build geotools locally you get back to versions saying SNAPSHOT :(
Yep, I think that's the only choice with maven 3: deployed jars get a
timestamp while locally built jars don't. I queried this on the
On 8 September 2012 04:15, Frank Gasdorf fg...@users.sourceforge.net wrote:
Just to clarify what's going on:
- it seems to me that prior 9-SNAPSHOT all deployments were done with
maven versions 3.x
Hi,
The issue was time-stamped artifacts with Maven 3 was discussed on the
dev list a while
On 31 August 2012 22:35, Andrea Aime andrea.a...@geo-solutions.it wrote:
For the record, I would be +1 on publishing jars there.
Wondering, is there any size limit? GeoTools is a large library, if we are
going
to push a release out every month the size requirements will go up fast
Good
On 31 August 2012 16:47, Joachim Van der Auwera joac...@geosparc.com wrote:
They are not (at least that was the situation a couple of months ago). We
have staged many Geomajas artefacts through oss.sonatype.org which have
repository tags and references to dependencies (like GeoTools) which are
Hi Jody,
Sorry, don't know the answer to your question but just wondering if
there are still plans to publish to maven central at some stage ?
Michael
On 30 August 2012 15:17, Jody Garnett jody.garn...@gmail.com wrote:
I am doing a preflight check prior to releasing uDig, and tried switching
On 30 August 2012 16:42, Jody Garnett jody.garn...@gmail.com wrote:
I was under the impression the easiest thing would be to deploy stable
releases to codehaus (which we are already set up with) and watch it
replicate out from there?
Sounds good if codehaus does that.
I've used the Sonatype
On 31 August 2012 14:55, Joachim Van der Auwera joac...@geosparc.com wrote:
There is indeed a rule that there should not be any external
repositories in the pom, but that rule is not enforced.
Similarly, it is advisable that all dependencies are themselves on
central, but that rule is also not
On 23 August 2012 20:30, Andrea Aime andrea.a...@geo-solutions.it wrote:
it would be nice to have an SPI for that too, maybe have the factory lookup in
META-INF/services and look up the class names, but not going through the
official SPI where you actually get a single instance of the class,
:
Major
Reporter:
Michael Bedward
Hello Martin and all,
Since we have a vector grids module (gt-grid) my first preference
would be the term raster. However, since that is only an unsupported
module I'd also be fine with changing grid to lattice in its code
and docs to allow the term grid to be used purely for rasters within
Hi Justin,
As the author of the vector to raster bits I'd be happy to see that
factory folded into another. It is not separate for any good reason.
Michael
On 14 August 2012 09:23, Justin Deoliveira jdeol...@opengeo.org wrote:
Hi all,
Recently on the geoserver list a thread was started
I'm out of my depth here, but wouldn't a new marker
UnsimplifiableFunction interface, of which the VolatileFunction
interface could be a descendant, lead to a much simpler and more
understandable system now and in the long term ?
Michael
On 9 July 2012 14:02, Jody Garnett jody.garn...@gmail.com
On 5 July 2012 14:20, Justin Deoliveira jdeol...@opengeo.org wrote:
(2) I got all the new branches from the geotools repo with git fetch.
What is the right way to push branches into my forked personal remote on
github? Check them out and then push them? Is their an easier way to
mass-update a
Hi Justin,
is there a recommended work-flow for keeping our developer forks in
sync with the canonical branches ?
Michael
On 3 July 2012 11:09, Justin Deoliveira jdeol...@opengeo.org wrote:
Hi all,
The 8.x branch has been cut.
https://github.com/geotools/geotools/tree/8.x
And along
that helps.
-Justin
On Mon, Jul 2, 2012 at 7:40 PM, Michael Bedward michael.bedw...@gmail.com
wrote:
Hi Justin,
is there a recommended work-flow for keeping our developer forks in
sync with the canonical branches ?
Michael
On 3 July 2012 11:09, Justin Deoliveira jdeol...@opengeo.org wrote
On 1 July 2012 17:20, Andrea Aime andrea.a...@geo-solutions.it wrote:
That is, how is one's fork kept in synch with the canonical one, I guess
it's not
something automatic?
Very pleased that you asked that question Andrea.
And then to sync up my master to upstream I do:
git pull upstream
On 30 June 2012 16:47, Andrea Aime andrea.a...@geo-solutions.it wrote:
Master wise, if we want to release a milestone, we just tag the master?
However this will require a out of the pool
period, even if a small one.
My reading of Justin's proposed scheme was that milestones would never
be
Hello Victor,
I just grabbed the project zip and am testing a build against an empty
local repository.
Will report back when maven finishes downloading the internet.
Michael
On 29 June 2012 17:55, victor@csiro.au wrote:
I am leaving the release over the weekend as we don’t do releases
On 29 June 2012 19:22, Andrea Aime andrea.a...@geo-solutions.it wrote:
Will report back when maven finishes downloading the internet.
Ah ha, got you! That's why internet is _so slow_ today! :-p
Yes - if there's anything you want I've got a copy of it on my machine now :)
Meanwhile, happy to
Many thanks for these docs Justin - they are very helpful.
Should we be using the https or ssh protocols to clone the repo or
doesn't it matter ? Both are shown in the clone examples on the
'Working with Git' page.
Michael
On 29 June 2012 06:06, Justin Deoliveira jdeol...@opengeo.org wrote:
Dear all,
I recently chatted to Jody about the vector grids module (gt-grids)
which has been sitting in unsupported for some time. The module is
described here:
http://docs.geotools.org/latest/userguide/unsupported/grid.html
There is an old proposal for it to be promoted here:
On 19 June 2012 17:26, Jody Garnett jody.garn...@gmail.com wrote:
I have used the geotools rendering engine to draw under SWT before; the best
technique is to render into a buffered image with the ARGB byte order; and
then hijack that as the contents of your SWT Image (which you can then draw
On 20 June 2012 07:40, ChenFred dragonslayer2...@hotmail.com wrote:
Thank you Michael for your suggestion. That is correct. an Overview docking
inside the MainView is definitely what we want. I think a pair of linked
MapViewports might be the way to go. I will dig into it.
Fred
Fred, Jody,
Project:
GeoTools
Priority:
Minor
Reporter:
Michael Bedward
David,
The JTS library includes the standard Douglas-Peucker algorithm:
http://tsusiatsoftware.net/jts/javadoc/com/vividsolutions/jts/simplify/DouglasPeuckerSimplifier.html
as well as a topology preserving algorithm:
:
Bug
Affects Versions:
8.0-RC1
Assignee:
Michael Bedward
I use maven 2 by preference due to its much smaller memory use on my
modest machine.
More importantly, I think Andrea mentioned he is still using maven 2.
Michael
--
Live Security Virtual Conference
Exclusive live event
Priority:
Major
Reporter:
Michael Bedward
On 25 May 2012 13:56, Ben Caradoc-Davies ben.caradoc-dav...@csiro.au wrote:
Are there any other subversion properties whose absence will be missed
when we migrate to git?
We lose the svn source and URL keywords. The source keyword is used by
the GeoTools javadoc plugin to add the module
On 23 May 2012 19:04, Andrea Aime andrea.a...@geo-solutions.it wrote:
I would not be surprised, the build takes a long time, I normally try to get
some work
done while the build is running too :-p
Think of the gt-swing tests as your opportunity to relax :)
Michael
Excellent !!! Thanks Justin.
I wonder if Jody was fiddling, I mean attempting to multi-task, too ?
Michael
On 22 May 2012 13:46, Justin Deoliveira jdeol...@opengeo.org wrote:
Interesting. Nevermind... i take it back, the module builds now :) I think I
see what is happening. The FEST stuff
Could it be a Maven 3 problem ? Both FileUtils and Artifact are maven
classes so should find their way to the repo with the rest of the
plexus stuff.
Michael
On 23 May 2012 14:25, Jody Garnett jody.garn...@gmail.com wrote:
And again with the clean build from bundled source code; against an
I just tried building jar collector against an empty repo with maven 2
and it succeeded - so the finger of suspicion points to maven 3 !
Michael
On 23 May 2012 15:47, Jody Garnett jody.garn...@gmail.com wrote:
Could it be a Maven 3 problem ? Both FileUtils and Artifact are maven
classes so
I'm unable to reproduce this. Using latest trunk code, all tests pass
on my system (osx 10.6.8, Java 1.6.0_29) including the JTextReporter
tests that you cite as failing.
Are the tests failing for anyone else ?
Michael
(Assert.java:74)
at org.junit.Assert.assertTrue(Assert.java:37)
at org.junit.Assert.assertTrue(Assert.java:46)
at org.geotools.swing.tool.PanToolTest.dragPanMap(PanToolTest.java:84)
On Mon, May 21, 2012 at 1:29 AM, Michael Bedward michael.bedw...@gmail.com
wrote:
I'm unable
What OS is that on Jody ?
Michael
On 21 May 2012 01:11, Jody Garnett jody.garn...@gmail.com wrote:
With H2DataStore working again (thanks Ben!) I was able to make it a bit
further
Next up with have gt-swing module:
1)
1 - 100 of 1256 matches
Mail list logo
