[Geotools-devel] Fun with filter negation in sql stores

Hi, I've recently stumbled into a fun problem with our sql encoding. Say you have the following filter (which, while odd, is legal): not(attribute > 3) our sql translator translates it to: not(attribute > 3) Ahh.. all good no? No! The behavior of our in memory filtering, and database filtering,

Re: [Geotools-devel] Fun with filter negation in sql stores

On Mon, Feb 3, 2014 at 10:50 AM, Andrea Aime wrote: > Anyone has better ideas? :-) > We could also have a look at the attribute metadata and see if the column > is nullable, or not. > Speaking of which: http://docs.oracle.com/javase/7/docs/api/java/sql/DatabaseMetaData.html#getColumns(java.lang.S

[Geotools-devel] [jira] (GEOT-4688) SimplifyingFilterVisitor should simplify and-ing/or-ing of negated filters

Title: Message Title Andrea Aime created an issue

[Geotools-devel] [jira] (GEOT-4689) auto-wrap option breaks before/after parentheses

Title: Message Title Michaël Michaud created an issue

[Geotools-devel] Build failed in Jenkins: geotools-master #138

See Changes: [andrea.aime] [GEOT-4688] SimplifyingFilterVisitor should simplify and-ing/or-ing of negated filters -- [...truncated 16913 lines...] Tests run: 0, Failures: 0, Errors: 0, Skip

Re: [Geotools-devel] Build failed in Jenkins: geotools-master #138

I'm on it Cheers Andrea On Mon, Feb 3, 2014 at 11:30 AM, wrote: > See > > Changes: > > [andrea.aime] [GEOT-4688] SimplifyingFilterVisitor should simplify > and-ing/or-ing of negated filters > > --

[Geotools-devel] [jira] (GEOT-4690) WFS store won't ask for the attributes needed to evaluate post filters

Title: Message Title Andrea Aime created an issue

[Geotools-devel] Jenkins build is back to normal : geotools-master #139

See -- Managing the Performance of Cloud-Based Applications Take advantage of what the Cloud has to offer - Avoid Common Pitfalls. Read the Whitepaper. htt

Re: [Geotools-devel] Fun with filter negation in sql stores

Am I missing something? Could we represent not( attribute > 3 ) as attribute <= 3 ? It would involve some logic to identify the appropriate inverse of our filters, but it may be worth it. Jody Garnett On Mon, Feb 3, 2014 at 8:50 PM, Andrea Aime wrote: > Hi, > I've recently stumbled into a fun

Re: [Geotools-devel] Fun with filter negation in sql stores

On Mon, Feb 3, 2014 at 1:27 PM, Jody Garnett wrote: > Am I missing something? Could we represent not( attribute > 3 ) as > attribute <= 3 ? > > It would involve some logic to identify the appropriate inverse of our > filters, but it may be worth it. > Yes, that is also a way, but more work would

Re: [Geotools-devel] PerceptualDiff on Boundless Jenkins

No problem. Installed. On Sun, Feb 2, 2014 at 8:37 PM, Ben Caradoc-Davies < ben.caradoc-dav...@csiro.au> wrote: > Justin, > > can we enable PerceptualDiff on Boundless Jenkins. It would catch failures > like this: > https://jira.codehaus.org/browse/GEOT-4687 > > Kind regards, > > -- > Ben Carado

[Geotools-devel] Two bugs for parsing ArcGIS Server 10.1 WFS GetFeature response

Hi all, (Totally new to Geotools community so sorry if I am posting at wrong place and please help to move it to where it belongs to.) While using Geotools in our project, I found two bugs when parsing WFS GetFeature response from ArcGIS Server 10.1. 1) attribute "gml:id" is missing. Unlike a st

Re: [Geotools-devel] Fun with filter negation in sql stores

I admit de morgan is a pain to implement, but I would find the result more intuitive then null checks suddenly appearing. Jody Garnett On Mon, Feb 3, 2014 at 11:48 PM, Andrea Aime wrote: > On Mon, Feb 3, 2014 at 1:27 PM, Jody Garnett wrote: > >> Am I missing something? Could we represent not( a

Re: [Geotools-devel] Fun with filter negation in sql stores

On Mon, Feb 3, 2014 at 9:41 PM, Jody Garnett wrote: > I admit de morgan is a pain to implement, but I would find the result more > intuitive then null checks suddenly appearing. > Easier to read, yes, easier to index, maybe, but ... more intuitive? How do you explain a end user that he wrote "no

Re: [Geotools-devel] Two bugs for parsing ArcGIS Server 10.1 WFS GetFeature response

Patches are easiest to submit via github pull request. You can also attach the patch to a JIRA issue. Be sure to include a test case (it makes the difference between your patch quickly getting reviewed vs a developer having to set up a test case to prove that the patch does something before it can

Re: [Geotools-devel] Fun with filter negation in sql stores

FWIW the simplifier in my CSS converter just adds (or removes) the IS NOT NULL logic. I've tried it a few ways and this seems the only way to be consistent across two-state and three-state logic systems. Might be nice if GeoTools were to emulate three-state logic for in-memory systems, but I'm no

Re: [Geotools-devel] Fun with filter negation in sql stores

Hi David, Is this something new? I'm asking bacause I got into this exactly out of CSS style filters, that do contain negations, without any null handling, and as a result upend up not painting the features with null values. I dont have the style handy now, but I can share it tomorrow if you want

Re: [Geotools-devel] Fun with filter negation in sql stores

It's actually something quite old, one of the first issues I encountered with the project was the inconsistency between shapefile and postgres backends. I forgot that I haven't added this behavior back to the converter since I rewrote the simplifier a while back. On Mon, Feb 3, 2014 at 4:51 PM,

[Geotools-devel] Reminder: GeoTools / GeoServer Meeting at 13:00 UTC

GeoTools / GeoServer Meeting at 13:00 UTC on Skype as usual. Make yourself known if you want to attend but do not usually do so: http://www.timeanddate.com/worldclock/fixedtime.html?msg=GeoTools+%2F+GeoServer+Meeting+2014-02-04&iso=20140204T13&ah=1 -- Ben Caradoc-Davies Software Engineer CSIRO

[Geotools-devel] Vous n'avez pas encore répondu à l'invitation de asma ouqalli

Invitation Dogfinance Invitation Dogfinance