I have a few patches I just did for 2.1.x - the bugs are still in 2.2.x
and trunk. I'd like to move the patches forward, but I'm unable to
build 2.2.x.
I'm also stalled moving geoserver from 2.1.x to 2.2.x and have been for
a while. I'm actually using geoserver quite a bit; this is prime
te
Jody Garnett wrote:
Jesse Eichar wrote:
Jody said he had some people working on the problem so I haven't been
looking into it. I'll see if he is still on this.
Yes oliver is working on this, I hope to hear back tomorrow myself.
Fraid we were stuck in a meeting all day today...
On a related
Okay, I've found a pretty nasty bug in the PostGIS driver.
My dataset is a table with about 250 rows in it. Each row has the VMAP0
dataset polygon of a country in. Each row is large.
If you look closely at the JDBC datastore, you'll notice that it will
fetch 200 rows at a time - which means i
Thanks a lot, oliver.
Changes should be commited to the 2.2.x AND trunk branches.
Putting it in 2.2.x means that everyone using the stable branch gets
your changes and putting it in trunk means that your changes will make
it to the next stable version (and for the folks currently using trunk).
Most of the connection pooling is actually done by the Postgresql JDBC
driver, but there's a proxy object on top of it thats hard to look inside.
I talked to the Postgresql JDBC mailing list and they said "oops" - it
should reset the stream so its in a consistent state. Dont know when
this wi
Folks told me to go to maven 2 for 2.2.x. I did, but I'm getting
different errors.
I have JAI installed. I verified that
"com/sun/media/imageio/stream/FileChannelImageInputStream" is in
"C:\j2sdk1.4.2_07\jre\lib\ext\jai_imageio.jar".
Any idea whats wrong?
dave
-
(I picked up yesterday's changes, but 'no go')
With Maven 2, I'm getting the same missing-JAI library-like error as
before in plugin/image. It builds/tests fine with Maven 1, so I'm
betting this is some type of Maven 2 configuration problem.
==
java.lang.NoClassDefFoundError:
com/sun/media/
I spent a lot of time looking at whats was going on, but that code is
*extreamly* difficult to read. As far as I can tell, it looks like it
has one thread reading the shapefile and another writing to it. That
should *always* throw an exception as the write operation seems to
replace the origi
+
| Executing clean,jar:install geotools-oraclespatial
| Memory: 34M/59M
+
Attempting to download ojdbc-14.jar.
WARNING: Failed to download ojdbc-14.jar.
BUILD FAILED
I looked at the project.xml and I see:
Cory,
You added to TransformerBase.java (commit message "GEOT-806: XML
Indentation fixed", line 71):
if (indentation > -1) {
tFactory.setAttribute("indent-number", new
Integer(indentation));
}
This is causing:
Caused by: java.lang.IllegalArgumentException: Not supp
I've been able to get 2.2.x working with maven 1. Still no go with maven 2.
For those of you who are interested, there's a geoserver branch
("branches/1_3_0a_geotools_2_2_x") that is the current trunk of
geoserver (to be 1.3.1 soonish) moved to 2.2.x.
dave
-
I've noticed a few problems with an older version of the Postgresql JDBC
driver.
This had to do with detecting encoding of data in the database. I've
been using:
http://jdbc.postgresql.org/download/postgresql-8.1-405.jdbc3.jar
and have found that it works better.
I'd like to suggest that w
Martin Desruisseaux wrote:
I already posted an email suggesting the same (for different reason)
two weeks ago. Since I got no objection, I already upgrated the Maven
2 build on trunk to the following version:
8.1-404.jdbc3
This is 404 instead of 405 only because the 404 version was the o
Martin Desruisseaux wrote:
So the compilation goes well but the execution failed. Could it be
that you have 2 Java installation on yours machine? Our Maven build is
configured in order to fork a JVM before to run the tests, and it may
happen that the JVM forked by Maven is not the same than
Paul Ramsey wrote:
On 13-Mar-06, at 11:08 AM, [EMAIL PROTECTED] wrote:
ps. Jody -- you put udig on its own confluence so I'd like to hear your
experience.
It was "as advertised". Take the backup XML file and upload, content
appears.
Excellent. Did you try to move over JIRA issues?
I noticed this change.
I think this is the incorrect interpretation since there is a
test in Filter.
dave
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media.
David Blasby wrote:
I noticed this change.
I think this is the incorrect interpretation since there is a
test in Filter.
oops, I noticed that is interpreted as
"[ interstate = null ]"
which causes exceptions because the right hand side of the "=" is 'null&#x
I'm unable to build shapefile using maven 1 or maven 2. Same problem as
before.
dave
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
Paul Ramsey wrote:
If so, how do I represent "" in XML?
Thats the problem -- in 2.2.1 meant "", now (2.2) it
means "null".
You can test for null with PropertyIsNull, so the 2.2 implementation is
almost certainly wrong since there's no way to test for "" anymore.
dave
--
http://jira.codehaus.org/browse/GEOT-832
Jody Garnett wrote:
David Blasby wrote:
I'm unable to build shapefile using maven 1 or maven 2. Same problem
as before.
We applied a patch to close the readers ... so it must be a different
problem then we fixed? Can you give Jesse a Jira n
Jody, I modified my program so it looks like this (using a
DefaultTransaction ):
private void process() throws Exception
{
DefaultTransaction trans = new DefaultTransaction();
FeatureWriter aWriter = ds.getFeatureWriter(table_name,trans);
while (aWriter.ha
Jody,
I looked more into the featurewriting stuff that we talked about last
night -- here's what I found.
I think there's a problem in how "Feature" implements "equals". Here's
where it came up.
I wrote a tiny program that updates a table in my database:
/**
* similiar to:
* update s
This problem has been going for almost 2 months. I've only been able to
build 2.2.x ONCE.
I say we just delete the test case - this is been going on WAY too long
and cost me way too much time.
dave
---
This SF.Net email is sponsored by x
>That is horrible, and now I need to think :-(
>Ah - please reverse aWriter.close() and trans.commit().
Thats not going to do anything since I can see that its not in a
transaction when its in the while() loop.
IE. while its processing I can open another connection to the database
and see the
Okay, this still doesnt work.
I stepped through the code and the connection to postgresql is still in
auto-commit mode which would explain why its not working.
private void process() throws Exception
{
DefaultTransaction trans = new DefaultTransaction();
FeatureWriter a
Okay, maven 1 builds are not working at the moment for oracle spatial
and maven 2 doesnt have the eclipseAll option.
dave
---
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and m
David Blasby wrote:
Okay, maven 1 builds are not working at the moment for oracle spatial
and maven 2 doesnt have the eclipseAll option.
Hay - I just deleted the plugin/oraclespatial directory and now things
work. This is a *great* fixing technique!
dave
Speeding up PostGIS WMS
I found that drawing data from postgis was way too slow. I needed to be
sped up to be twice as fast. I've accomplished this. Here's the changes:
1. get spatial data from postgis in base64 instead of octal
This was only 2 lines difference and was definately the best
The "PostgisFeatureCollectionOnlineTest" seems to assume that there is a
"bc_voting_areas" dataset in the test database.
The PostGIS Online Tests are supposed to auto-populate the database with
the required data. This test is ONLY available via the RR postgis
installation.
I also noticed t
The PostgisPermissionOnlineTest doesnt work on a local database and it
just hangs when it runs remotely.
It appears to be accessing a "restricted" table in the database, so this
looks like another "missing data" problem.
There's no real comments in the test case, so I have no idea what its
s
Jessie,
I've been optimizing the PostGIS driver, and I speeded up the reading of
a set of coordinates. I use the ObjectInputStream.bytesToDoubles()
NATIVE function which takes a byte[] and writes to a double[]. You can
probably do the same thing with the shapefile driver. On my test of
100
I've upgraded the StreamingRenderer so that it smarter about sending
in the datastore. Jessie did the original work, I just
generalized it a bit.
You can set a renderer hint "maxFiltersToSendToDatastore" (Integer) to
limit the number of Filters that get sent to the datastore. The default
i
(sending this out quickly for the IRC meeting)
Okay, I've been fixing up the StreamingRenderer the last little while,
and here's what I think needs to be done. Most of these are just little
clean-up to get rid of the evidence of a long period of people slapping
stuff into it.
A) The path
Justin,
I tried to move a patch forward to CompareFilterImpl, but its changed
too much on trunk. You were the one who made the changes so I'm sending
the patch to you because I dont know what the new implementation does.
It appears to still have the same bug if one or both of the objs are
n
Changes
-
NOTE: test cases are mostly turned off for "maven 2" and non-functional
in "maven 1". I ran the tests in eclipse and they pass.
NOTE: I had awful trouble moving my changes from 2.2.x to trunk.
As I mentioned last week, I made a bunch of changes to improve
performance and fi
Andrea Aime wrote:
Hum I'm wondering what changes did you made. It's easy to make
decimation more efficient
by making it more aggressive, but a properly working decimation
generates results that cannot
be distinguished from not having applied decimation at all.
When I coded it, I made sure
Jody,
My plan to not generate FIDs for Rendering only had a 5% speed
increase. I'm not sure why the profiler was off by so much - perhaps IO
is a limiting factor.
Well, the next biggest time sink seems to be calculating hashcodes for
the Symbolizers (and their sub-objects).
I was thinking
Saul,
I noticed this too. I added the SQL92 like-encoder a little while ago
and tested in in postgis. I dont have oracle, mysql, or SDE here to
test, so I didnt add it to them. I did, however, send a message out
asking folks to test and add it to the filter capabilities if it worked.
I th
Jody,
There's no setTransaction() method in the FeatureSource.
The only way I can see to set a property for a Query operation is to
convert the FeatureSource to a FeatureStore and then call
setTransaction() on it. I was a bit worried about this not working
because I have my server set to not
>>Either way I feel like using Transaction hints to do rendering stuff
is stretching the design a slight bit too far. Perhaps put hints in
Query itself?
Thats what I first suggested, but jody said the way to go was to put it
in Transaction.
But, then it affect all future Queries on that Tra
Okay, I finished investigating making Queries w/o FIDs.
According to the profiler, I was expecting an approximately 15-25% speed
increase.
The actual results were not as expected (for reading and rendering
100,000 lines - average 83 points/line in the original data, very high
decimation duri
Cory Horner wrote:
Hi folks,
We've been getting strange errors in uDig when rendering from PostGIS
(and possibly others). Everything appeared to draw correctly, but
some exceptions occurred including this:
Caused by: java.lang.ArrayIndexOutOfBoundsException: 1
at
org.geotools.renderer.
ull?
Return the zero-width box - it still has data in it. It could be doing
some type of clipping...
dave
Cory Horner wrote:
David Blasby wrote:
This has happened for a long time, but now the decimation is more
aggressive so it occurs more often.
I havent been able to get the Geoserver WMS
Just a reminder - jesse, martin, alessio, simone, and anyone else who's
interested.
http://timeanddate.com/worldclock/fixedtime.html?day=1&month=5&year=2006&hour=22&min=0&sec=0&p1=37
That 1:00pm PST.
We're meeting on "freenode" #geotools.
dave
--
a patch for this to 8.0, 8.1, and HEAD cvs branches and new
>official releases should hopefully be out soon.
dave
David Blasby wrote:
Okay, I've found a pretty nasty bug in the PostGIS driver.
My dataset is a table with about 250 rows in it. Each row has the
VMAP0 dataset polygon of a
IF YOU ARE INTERESTED IN GRID COVERAGES OR THINK GRID COVERAGE WORK WILL
AFFECT YOU - PLEASE SPEAK UP NOW.
We didnt have enough time to discuss all the details, so we're having
another meeting next week. Also, we've been discussing merging in the
Grid Coverage renderer to trunk, which could a
Martin (and others),
Thanks a lot for this update, its given me a lot of info.
dave
---
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Do
During the Grid Coverage/Renderer improvements IRC today, we talked
about merging the grid coverage branch onto trunk.
See:
http://docs.codehaus.org/display/GEOTOOLS/2006/05/01/Renderer+IRC
and the " Renderer improvements meeting part 2 (Grid Coverge Merge to
Trunk)" message thread.
I know almos
(resending because my email was having problems)
My rough idea of a project plan for the FM branch would be:
- write test cases for everything up to "SimpleFeature"
- port shapefile and postgis over to use SimpleFeature
- slam the change onto trunk, and explore the rest of the API as
aspects of
(resending because my email was having problems)
Martin Desruisseaux wrote:
- The GO-1 module has basic classes for a new renderer with more
rigourous axis support.
Martin,
I've heard about the GO-1 renderer, but I dont know anything about it.
Could you give a few sentances on what it is?
Jody Garnett wrote:
We spent the day fighting with Maven (heck it is just like normal
geotools development). The goal is to get the FM branch ready for
group development, rather then just one person hacking in eclipse.
Thanks to those who showed up, the positive feedback on the
TypeBuilder w
Here's a quick question for you, jody. When is udig planning on
transitioning to the FM? I'm not sure if you have a enough info to
answer that, but what does your gut say?
dave
---
Using Tomcat but need to do more? Need to support web serv
Cory Horner wrote:
As mentioned in the meeting:
http://docs.codehaus.org/display/GEOTOOLS/Tasks+for+2.2.0+release
...are the tasks needed to be done! Please review and edit to your
liking.
I think you should add "PostGIS datastore" to Milestone 3 because
Geotools isnt really useful withou
Martin Desruisseaux wrote:
- The GO-1 module has basic classes for a new renderer with more
rigourous axis support.
Martin,
I've heard about the GO-1 renderer, but I dont know anything about it.
Could you give a few sentances on what it is?
dave
---
My rough idea of a project plan for the FM branch would be:
- write test cases for everything up to "SimpleFeature"
- port shapefile and postgis over to use SimpleFeature
- slam the change onto trunk, and explore the rest of the API as
aspects of the orphaned complex-feature branch trickle in o
When I modified the PostGIS driver to (a) use Base64 and (b) optimize
WKB parsing, I added two file:
http://svn.geotools.org/geotools/trunk/gt/plugin/postgis/src/org/geotools/data/postgis/attributeio/Base64.java
http://svn.geotools.org/geotools/trunk/gt/plugin/postgis/src/org/geotools/data/postg
I also noticed that the intersection() function at the very bottom of
LabelCacheDefault.java is very similiar (but not exact) to Sun's
Rectangle.intersection() function.
I've replaced this function with a call to JTS's Envelope.intersection()
function. I believe this happened because JTS 1.6
Jessie,
As we talked about in the IRC.
The intersection() function at the very bottom of the
LabelCacheDefault.java should be:
//djb: replaced because old one was from sun's Rectangle class
private Envelope intersection(Envelope e1,Envelope e2)
{
Envelope r = e1.intersection(e
I thought about this problem before -- the main issue is that the
anti-aliasing causes a lot of "extra" colours to be created.
The original colour reduction algorithm was to find the first 256
colours in the image and use those colours in the colour map. This
completely failed to work after antia
I'll look into the problem this afternoon.
The timezone handling in sql lite is very limited and confusing.
Cheers,
David
___
GeoTools-Devel mailing list
GeoTools-Devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/geotools-devel
nt to timestamp without time zone"
>
> Cheers
> Andrea
>
>
> On Tue, Mar 10, 2020 at 7:01 PM David Blasby
> wrote:
>
>> I'll look into the problem this afternoon.
>>
>> The timezone handling in sql lite is very limited and confusing.
>
Hi,
I was trying to use the Natural Earth GeoPackage (countries) - link below.
However, I found the rendering was very "wrong";
[image: image.png]
This triangle is actually 3 small islands - it's represented in the
GeoPackage as a Multipolygon (one polygon for each island).
This works fine in QG
Oh - I reported the issue to the Natural Earth people -
https://github.com/nvkelso/natural-earth-vector/issues/343
On Tue, May 12, 2020 at 5:11 PM David Blasby
wrote:
> Hi,
>
> I was trying to use the Natural Earth GeoPackage (countries) - link below.
>
> However, I found the ren
es even when it is not needed.
> --
> Jody Garnett
>
>
> On Wed, 13 May 2020 at 00:38, Andrea Aime
> wrote:
>
>> Goin back to the original mail
>>
>> On Wed, May 13, 2020 at 2:12 AM David Blasby
>> wrote:
>>
>>>
>>>
>>>
Hi,
I've been working in the GT codebase and I'd like to have a bit more
permission. I cannot re-run failed builds and I cannot ask for PR reviews.
Thanks a lot,
Dave
___
GeoTools-Devel mailing list
GeoTools-Devel@lists.sourceforge.net
https://lists.so
Issue Type: Bug
Components: core filter
Affects Versions: 2.2.0
Environment: all
Reporter: david blasby
Attachments: partial_support_for_codelists.patch
Currently, there is no support for CodeList (i.e. GeoAPI enumerated values) in
Filter.
This patch
: all
Reporter: david blasby
Attachments: handle_empty_strings_better.patch
Currently if you have a or a ...one or more
spaces... in your Filter, the Literal gets replaced with a null.
This causes all sorts of problems.
This patch changes the parser so that it will create
Title: Message Title
David Blasby created
Title: Message Title
David Blasby created
69 matches
Mail list logo
