From: zze-M2S F. V e. RD-BIZZ-S. <vfr...@or...> - 2007-02-20 14:08:56
|
Hi, I post another thread because my features are well inserted now, ie. their primary key (a column called "id") is valued with the database sequence (not with IDs generated by geoserver). Altough I'm not sure why it's working now (moving all my feature tables to the "public" schema was very profitable for me).=20 But... if i look into the Set returned by the featureStore.addFeatures(), I can see a HashSet where the keys are the IDs (initialy?) generated by geoserver, eg: newesig:esig_points.9223372036854775791 <=3D "new" + my namespace for = this feature + "." + an ID based on a unknown sequence, but not on my DB sequence!=20 So what could I do to retrieve the "real" FIDs, ie. the ones which are really used (at the end) for the insertion and which are based on my database sequence?=20 Another relative question: why a Set and not a List is returned? Ordering seems very important since a whole features collection could have been inserted, so the client should retrieve a collection of relative FIDs which should be ordered in the same way than the feature collection was.. If the ordering is not preserved, it's not very easy from the client side to know which ID is relative to which feature! Thanks a lot, Vincent.=20 PS: yeap this is my 3rd mail account on this mailing list since SourceForge ML servers are still mad with me... :( |