Good idea!!!
I see there's even a JDBC driver for it:
http://www.zentus.com/sqlitejdbc/
so I may take a look at it, but even if I can add SQLite support to the SISDB
plugin,
it will be read-only like any other OJ DataStore.
But the great thing would be the adoption of a light DB, such as
From the JDBC point of view any data type _should_ be equal,
so even a geometric data type _should_ be equal to all others,
but I don't know in which manner SpatiaLite is implemented on top of SQLite,
so it may as well not work...
Bye
Paolo Rizzi
Da: Martin
Hi Jukka,
thank you very much for pointing out this!!!
In fact I used / as separator before the db name instead of :!!!
I don't remember why...
Probably I didn't bother to test it very much since it was working with the
complete TNS...
I'll fix it as soon as possibile (when I'll have an
I was missing it... :-)
First of all you should check your TNS with some other Oracle client,
to ensure the connection can actually be made with those parameters.
There are lots of reason why a connection to Oracle could not be extabilished...
Also check the OpenJump connection dialog, keeping
You can't be sure it's actually a Java bug...
Maybe newer versions of Java has changed something that made an existing old OJ
bug to surface...
I tested an old OJ version (1.0) and sure enough the problem is there with Java
6.10.
I don't know why it happens (and actually don't care in this
I didn't tested it very extensively, but it seems to give no problem.
It's just a matter of trying it for a while and see what happens.
Bye
Paolo Rizzi
Da: Sunburned Surveyor [mailto:sunburned.surve...@gmail.com]
Inviato: gio 18/12/2008 18.59
A: OpenJump
I'm happy to be helpful!!! :-)
But remember that the check on coordinates.size() avoids the exception,
but it doesn't solve the problem.
Someway it has to be ensured that each coordinate is added to the array
before it is used.
I don't know how to ensure this...???
Even if both methods were
I didn't debugged the code, but the exception says:
java.lang.IndexOutOfBoundsException: Index: 0, Size: 0
so it seems that the coordinates ArrayList is empty when
trying to access it's first element in:
coordinates.get(0);
Defensive programming would suggest something like:
Refresh layers is indeed used with DataStore based layers,
the ones added with Layers/Add DataStore Layers.
It would be great if it worked with other kind of layers too,
expecially layers created with Layers/Run DataStore query,
but also layers created from shape files may be refreshed, why not???
The distance of the point from the origin of the line sounds like
a job for JTS linear referencing capabilities. See class:
LengthIndexedLine
method:
double indexOf(Coordinate pt)
Computes the minimum index for a point on the line.
or
double project(Coordinate
I read your blog and, as always, it has a distinctive courageous taste,
so I really hope you can attain your proposed goals!!!
Here follows my modest opinion (I'm not subscribed to blogspot).
I'm very sorry because I don't contribute to OJ at all, while you are
spending lots of energy in it!!!
Well, kind of funny anyway... :-)
I can't understand why you wrote:
select SDO_TUNE.EXTENT_OF('SCHEMA.MYTABLE','MYUSERNAME') from dual t;
it's the 'MYUSERNAME' part that makes me wonder...
The syntax is supposed to be:
SDO_TUNE.EXTENT_OF(
table_name IN VARCHAR2,
column_name IN VARCHAR2
I tryies using the '-' so I got access to the default schema for the
logged-in user.
Maybe it's this the culprit, something in how your server/database/schena/user,
etc. are configured...???
Try with this SQL from any Oracle client:
select sys_context( 'userenv', 'current_schema' ) from dual
I'm glad to see the '-' dash solved the schema issue (I think a space should
have been good too,
but I'm too sleepy now to check)...
The error seems to indicate that the function MDSYS.SDO_TUNE or something alike
is not recognized,
so it may not be available (not installed) or it may be
The Db field should contain only the database name, SER1 in your example.
The JDBC URL is then built by the plugin itself.
I'm not an Oracle expert, but with the two servers I worked with, the only way
to get
a JDBC connection (from the plugin or from any other Java program), was to use
a
I never looked at this code, but from you wrote here
I, personally, would leave the interface in place.
Even if now only the TaskFrame class implements the
Task.TaskNameListener interface who can tell if in
some near or distant future some other class would need to???
An interface specify a
I use the ?: a lot!!!
I agree that an if/else is more readable, but in some situations the ?: is
better.
For example I use it to normalize method parameters:
void mmm(int i,String[] sss)
{
i = i 0 ? 0 : i;
sss = sss == null ? new String[0] : sss;
...
}
The above is
Yes, but I don't believe that would add anything to readability...
I think the key is not about everybody using the same style,
but about each one using a consistent style.
If one always use the same style, even a non-programmer it's perfectly able
of understanding it after a couple of
If I can tell my opinion the whole configuraiton should be in the xml file.
I don't like the way some plugins are hardly configured in Java code.
There should be nothing really basic and hardly configured, because it
will be impossible to remove that and, besides, what is really basic
depends
From a _very_ brief search it seems to be a yet-to-be-resolved bug
on Eclipse side:
https://bugs.eclipse.org/bugs/show_bug.cgi?id=19602
A patch exists, but you have to apply it for yourself, or wait for version
3.4M3.
I have no idea if any external solution exists to print Java source
You wrote: ...the old DataStore class
Are you sure it is indeed _old_???
It is a while I didn't actually look at OJ source,
but I seem to remember it was an alternative to the DataSource.
That is: DataSouce for files and DataStore for databases.
But I may have misunderstood or things may have
So, thanks for the thanks!!!
(I'm the developer of the SIS database plugin).
I'll give a try to Reeders' one when I'll find time to.
Bye
Paolo Rizzi
-Messaggio originale-
Da: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] conto di
Rahkonen Jukka
Inviato: venerdì 30 maggio 2008 13.11
Or better yet /openjump/lib/ext/profiles...
so that each user can choose among different profiles if needed.
Bye
Paolo Rizzi
-Messaggio originale-
Da: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] conto di
Sunburned Surveyor
Inviato: giovedì 29 maggio 2008 2.05
A: OpenJump develop
I never tried my SISDB PostGIS/Oracle plugin with 10 milions records,
but it always worked flawlessy up to a few hundred thousands,
so you may give it a try.
Obviously, beeing it a DataStore, it will be read-only.
Regarding the max features option, I don't think it would be useful,
because the
I may be wrong, and I'm too lazy to actually test it,
but I think the last return null is indeed reached.
The catch kind of kills the exception, it eats it up,
so, with the exception gone, execution resumes right after
the catch block, and the return null _is_ reached.
That's why the compiler
A UUID is one of those 128 bits strings used to uniquely and universally
identify something
that can be generated without the need for a central authority.
If you look inside a Windows Registry you'll find lots of them, because they're
used to
identify interfaces and objects by Microsoft OLE.
If you want an automagically universal unique ID, you can use a UUID
stored inside a suitable char field. You'll surely find some
Java util to create it. It's not as small and performant as an integer,
but it's guaranteed to be unique. A simple integer qith it's 2^31
upper limit, may not be enough
Even for a single user, some sort of multiple profile support could be useful.
Maybe you're performing different jobs with OJ, so it may be handy to have
different configurations, preferences, database connections, etc.
And for logging too, a single user may like or need to impersonate
I'm not talking about compiling but about running.
If you have a Java class file compiled with 1.5
you can run it with 1.4 (AFAIK, at least).
Bye
Paolo Rizzi
-Messaggio originale-
Da: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] conto di
Andreas Schmitz
Inviato: martedì 25 settembre
Very sad from Sun...!!!
Just for fun I did a little searching and found a couple of tools
capable to weaving 1.5 bytecode into 1.4:
http://retroweaver.sourceforge.net/
http://retrotranslator.sourceforge.net/
Never tried them so I have no idea about if and how they work.
But if
In Windows simply using:
start anyfile.pdf
will open the associated viewer for PDF files.
I have no idea for Linux and MacOS.
Maybe Java itself have something to abstract OS differencies...???
Bye
Paolo Rizzi
-Messaggio originale-
Da: [EMAIL PROTECTED]
[mailto:[EMAIL
May I suggest to introduce a profile concept,
so that each user can define more than one configuration set
and decide which one to use when launching OJ???
Bye
Paolo Rizzi
-Messaggio originale-
Da: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED]
conto di Paul
Austin
Inviato: lunedì
But if you store the configuration stuff inside the user's home directory,
each user wil have only one set of configuration available, or not???
OK, the user may always user some other more explicit mean to specify
configurations, but...
What about subdirectories of the user's home dir. Each
Java classloading is a _very_ delicate thing.
I wouldn't change anything if not necessary.
Since it now works (I think it works) I see no reason to change it.
It may as well be that it has no parent for some specific reason.
Also the Java spec for ClassLoader.getParent() says:
...Returns the
You can try the SIS DataBase Plug-In.
It can be downloaded from:
http://sourceforge.net/project/showfiles.php?group_id=118054package_id=217237
Instructions are inside file SISDB.txt.
It works for add a Datastore query too and it also works with Oracle
(albeit it's not fully debugged
People don't like user lists...
Only developers (but not always) like them...
This makes me wonder if developers are indeed people or not...
Jokes apart, people don't like surveys too...
I think a forum would be much more appreciated.
Also a wish-list or feature requests form may prove useful.
Remember that spatial enabled RDBMS (like PostGIS or Oracle)
are already capable of doing projection, so one solution
is to pass a SRID code as a parameter to a Datastore connection
and let the Datastore driver do the conversion on the fly.
This makes me think that I can do it already with no
I would add the possibility to save Datastore query definitions
inside JMP files along with normal layers.
This would be _very_ useful, at least for me...
Also the Datastore API should be enhanced to support saving.
Even a simple Save To Datastore... would be enough
(a real _update_ would be
I think you should separate developer concerns from user concerns.
For developers, I think 1.5 or even 1.6 could be OK for _compiling_.
Users may still have 1.4 installed and I wouldn't force them to upgrade
just because of _running_ OJ.
Since generics and the like are onyl syntactic sugar,
mhm... actually
it is not just our decission if we use external libs. I use one which is
now only available for 1.5
stefan
Yes, of course one can not renounce to a needed functionality just to stay
compatible with 1.4.
Just be sure to do it only if there's no other (or no easy) way to do it.
Feature - An in memory representation of user's data, which can have
geometry and attribution. Question: can we use the Feature concept for
data without geometry or should Feature be a subtype of a DataObject
interface that has the same interface but without the
Geometry methods.
If I
I like that post and I agree with you!!!
A GIS is nothing more than a database in the end,
something that OGC seems to forget sometime...
Bye
Paolo Rizzi
Da: [EMAIL PROTECTED] per conto di Sunburned Surveyor
Inviato: gio 07/06/2007 21.07
A: OpenJUMP Dev
I agree with Martin!!!
Also I can't understand Surveyor really well here...
Are you thinking about using shapefiles to persist the FeatureCache???
If yes, it seems a little odd, given all the limits shapfiles have
(field names length, data tyeps, etc.).
If not, pardon me!!!
Anyway, to do the
Please no...
Writing software it's already a hard job, expecially working software...
There's no way to adopt good coding standards without using tools
that it enable it.
If you work in a big organization they may have a corporate IDE
using corporate standards and, in that case, following those
I can't completely understand what you're saying...
Using the Datastore API you can open several layers from
a single connection...
You can try my PostGIS/Oracle plugin from:
http://sourceforge.net/project/showfiles.php?group_id=118054package_id=217237
But you pointed out something I never
Before casting you should always check that the instance is indeed what you
think, like this:
if( needToCast instanceof IRendererFactoryTool )
IRendererFactoryTool casted = (IRendererFactoryTool) needToCast
and this saves you from nulls too, because the instanceof will be false if
You can design your label layer as it was only meant to be used manually,
that is you can add labels by hand however you like.
Then you can create a sort of wizard that's capable of automatically do
what you're doing by hand, that is it's capable to draw labels using data
from somewhere, for
I updated the SIS DataBase Plug-In with a couple of fixes.
It can be downloaded from:
http://jump-pilot.cvs.sourceforge.net/jump-pilot/SISJUMP/
Instructions are inside:
http://jump-pilot.cvs.sourceforge.net/jump-pilot/SISJUMP/release/sisjump/SISDB.txt
Actually to _run_ it you
Hi Michael,
thank you very much for pointing this out!!!
I never tested it with an unsized VARCHAR, always used some VARCHAR(XXX)...
I'll fix it as soon as possible!!!
Bye
Paolo Rizzi
-Messaggio originale-
Da: [EMAIL PROTECTED]
[mailto:[EMAIL PROTECTED] conto di
Michaël Michaud
I would vote for having converters in and out of the GT
Feature model as
required. It's only really needed for I/O, right? For other
functionality, use the raw functionality provided by GT (such as
coordinate transformations) and develop a JUMP-specific API
on top of it.
One of
If I remember well most of the new things that Java 5 added to Java 4 are just
syntactic sugar.
Generics, for example, are compiled into standard bytecode.
This means that, very often, you can compile a Java 5 source code with a Java 5
compiler
and then execute it with a Java 4 runtime.
So,
Hi Paolo,
Nice to hear you, you surely have a good experience about
database-access.
I like the idea of an embedded database.
I have made some tests in the past with jump+hsqldb, but I didn't get
any memory improvement (I have not much dbms knowledge and I may have
done mistakes
Stefan: Yes, I have tried the SISDB plugIn but you can not
save a table yet :-(
I did not have the time yet to add the saving part...
But I also don't know how to add it...
The SISDB plugin is a Datastore, so you can add a layer
through the Layers/Add Dataset menu.
But *JUMP doesn't define any
Hi Paolo,
if you try the PostGIS-PlugIn V.1.2.1 from
http://sourceforge.net/project/showfiles.php?group_id=118054
and you make Save dataset as...
you will find behind Format: a ComboBox
where you can save a PostGIS Table.
I dont't know how it is works but maybe
you can have a look at
This post is _very_ interesting.
It clarifies the distinction between plugins and extensions.
In my mind they were the same, and I was always wondering
why there existed the two of them.
Now you say plugins are present in the basic package while
extensions are third-party...
In my mind I
Hi Landon, Plan A seems good, and I like how you wrote it down so clean!!!
It only seems a little too simple and I always fear solutions that seems too
simple...
But often they prove to be winners in the end!!!
I have one little change and two requirements to suggest for Plan A.
.) The
I read the post on your blog, but it seems to be a crucial detail missing...
A plugin doesn't usually need another plugin itself, it probably needs what
the other plugin do.
Maybe a plugin needs to attach something to another plugin menu, or use another
plugin services.
So, for the second
Hi Landon,
I read your PDF and found it good, but here follows my idea.
I think there should exist _no_ different code bases at all, there has to be
only _one_ core!!!
I think all organizations developing JUMP versions would be better served off
by sharing efforts
into a common project,
I may be completely wrong here, but is seems that your geometry is encoded
as WKB (Well Known Binary), but the driver is trying to access it as WKT (Well
Known Text).
Inside PostgisValueConverterFactory there's this code:
...
public ValueConverter getConverter(ResultSetMetaData rsm, int
Looking back inside jump-workbench-1.0.1.jar from OpenJUMP 1.0.1
I can see a language directory with a few .properties files, and one of them
is jump_en.properties.
If this file is missing from the JAR you are creating, that's probably why
you're getting this exception.
Bye
Paolo Rizzi
You can delete the CVS files after downloading.
If you want to keep the working copies working
copy them to another directory first.
With TortoiseCVS (I don't know how to do that with
other CVS clients but it is surely possible),
right-click on the directory you want ot remove CVS files from
(do
61 matches
Mail list logo
