On Sat, Nov 3, 2012 at 1:09 AM, Jody Garnett wrote:
> Justin has advised me to ask for a code freeze ( rather than delay
> patches ) while the feature collection cleanup patch is applied.
>
> Instead can I ask that we exercise caution on any patches using
> feature collection. There are a few diff
Hey Jody,
took some time to go over the pull request at
https://github.com/geotools/geotools/pull/44/files (it's the right
one, yes?)
- Like that we're dropping the FeatureCollections "factory" methods
- The example here is scary:
https://github.com/geotools/geotools/pull/44/files#L2R66
Basi
> > Instead can I ask that we exercise caution on any patches using
> > feature collection. There are a few difficult sections ( especially
> > GML parsing ) that I would ask everyone to steer clear of.
> >
>
> -1, this is too vague in two ways:
> * no clear list of modules that we'd need to stay
Jody Garn
Thanks for the review/feedback Andrea!
I am having a hard time clicking on your links and finding which line you are
referring to - the page is too big (14 MB) and my browser gives up waiting.
Saving the file locally may work ...
> took some time to go over the pull request at
> https://github
Updated the pull request in response to feedback from Andrea and Micheal.
- DataUtilities.close( Iterator )
- DataUtilities.first( SimpleFeatureCollection ): SimpleFeature
- DataUtilities.first( FeatureCollection ): F
- DataUtilities.list( FeatureCollection ): List // Already existed
- DataUtil
On Sat, Nov 3, 2012 at 3:36 PM, Jody Garnett wrote:
> Updated the pull request in response to feedback from Andrea and Micheal.
>
> - DataUtilities.close( Iterator )
> - DataUtilities.first( SimpleFeatureCollection ): SimpleFeature
> - DataUtilities.first( FeatureCollection ): F
> - DataUtilities.
On Sat, Nov 3, 2012 at 4:19 PM, Andrea Aime
wrote:
> It would be nice to see how the whole looks now, but it seems you are making
> changes in another branch.
> If you had committed them in the same branch where you made the original
> pull request github would have updated the pull request with i
Ákos Maró
Agree with Andrea.
Imo this proposal is still potentially far from being "complete", at least
if we consider having GeoServer work with it a pre-requisite. I am working
through them now but there are a couple of significant issues in GeoServer.
SecuredFeatureCollection and PropertyValueCollection.
On Sat, Nov 3, 2012 at 5:35 PM, Justin Deoliveira wrote:
> Agree with Andrea.
>
> Imo this proposal is still potentially far from being "complete", at least
> if we consider having GeoServer work with it a pre-requisite. I am working
> through them now but there are a couple of significant issues
On Sat, Nov 3, 2012 at 10:55 AM, Andrea Aime
wrote:
> On Sat, Nov 3, 2012 at 5:35 PM, Justin Deoliveira
> wrote:
> > Agree with Andrea.
> >
> > Imo this proposal is still potentially far from being "complete", at
> least
> > if we consider having GeoServer work with it a pre-requisite. I am
> wor
On Sat, Nov 3, 2012 at 7:27 PM, Justin Deoliveira wrote:
>> SecuredFeatureCollection wise I believe we can just drop all the extra
>> methods
>> Jody removed from FeatureCollection,
>> don't think that GeoServer uses any of them but wanted to make sure
>> security
>> could not be broken by acciden
Ok, I had to make a few changes on top of yours to make GeoServer happy
with these changes. I pushed up them to a barnch in my repo. If they look
ok i will merge them into the main cleanup branch.
https://github.com/geotools/geotools/pull/46
On Fri, Nov 2, 2012 at 5:31 PM, Jody Garnett wrote:
Here it is. I pushed it up to the main repository in case you want to make
changes Andrea.
https://github.com/geoserver/geoserver/tree/fc_cleanup
Indeed the part i was unsure about was the secure feature collection /
secure feature source. I basically dropped all the unnecessary methods from
Se
Thanks Justin, I just wen rover them now and they look great, and have merged
them into the main clean up branch.
Sorry I took the ValueCollectionType / PropertyValueCollection change in the
wrong direction - it drove me nuts figuring out what it was doing.
--
Jody Garnett
On Sunday, 4 Nov
> Hum.. a few things are clear by looking at the commit diff anyways:
> - commented out code:
> https://github.com/geotools/geotools/commit/c0fd151d0265aa53b696010d0d995344c6533104#L6R1149
> - given how generics work, isn't this method redundand with respect
> the one right below it?
> https://gith
Bah that email got away from me - lets try again.
> - commented out code:
> https://github.com/geotools/geotools/commit/c0fd151d0265aa53b696010d0d995344c6533104#L6R1149
>
>
Yes this was a utility method when we removed the FeatureResults class; time
for it to die (now removed).
> - given ho
> > - RasterAsPointFeatureCollection now extends BaseFeatureCollection in order
> > to avoid the use of WrappingIterator
>
> Hum... ok, Wondering what level of overlap there is now between
> BaseFeatureCollection
> and that processing oriented feature collection base I've made some time ago
Chec
An idea for being extra cautious on the secure feature collection / secure
feature collection implementations.
Implement the drop methods, as *private final* to catch out any code that may
be trying to subclass and duck the security check, and implement them with a
throws UnsupportedOperationE
20 matches
Mail list logo
